Courses

@mastersthesis{uuid:91f200f7-4966-4504-bc83-5a87e5550a91,

title = {Real-time Vision-based Autonomous Navigation of MAV in Dynamic Environments},

author = {Jiahao Lin},

url = {http://resolver.tudelft.nl/uuid:91f200f7-4966-4504-bc83-5a87e5550a91},

year  = {2019},

date = {2019-01-01},

school = {TU Delft Mechanical, Maritime and Materials Engineering; TU Delft Cognitive Robotics},

abstract = {Safe navigation in unknown environments is a challenging task for autonomous Micro Aerial Vehicle (MAV) systems. Previous works generally avoid obstacles by assuming that the environment is static. The purpose of this thesis work is to develop a MAV system that can navigate autonomously and safely in dynamic environments. We present an onboard vision-based approach for the avoidance of moving obstacles in dynamic environments. This approach uses a state-of-art visual odometry algorithm to estimate the pose of MAV and an efficient obstacle sensing method based on stereo image pairs to estimate the center position, velocity, and size of the obstacles. Considering the uncertainties of the estimations, a chance-constrained Model Predictive Controller (MPC) is applied to achieve robust collision avoidance. The method takes into account the MAV’s dynamics, state estimation and the obstacle sensing results ensuring that the collision probability between the MAV and each obstacle is below a specified threshold. The proposed approach is implemented on a designed experimental platform that consists of a quadrotor, a depth camera, and a single-board computer, and is successfully tested in a variety of environments, showing effective online collision avoidance of moving obstacles.},

note = {de Croon, Guido (mentor); Alonso Mora, Javier (mentor); Ferrari, Riccardo M.G. (graduation committee); Zhu, Hai (mentor); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:99f41ef5-f2c9-4a0a-9b89-0245e106f6de,

title = {State Estimation and Optimal Control for Racing Drones: In search of control algorithms for competing against human pilots},

author = {Nilay Sheth},

url = {http://resolver.tudelft.nl/uuid:99f41ef5-f2c9-4a0a-9b89-0245e106f6de},

year  = {2019},

date = {2019-01-01},

school = {TU Delft Electrical Engineering, Mathematics and Computer Science},

abstract = {The e-sport of drone-racing involves human pilots to race against time. Recently, drone races have also gone fully-autonomous. As a result, these agile robotic platforms not only pose challenges of flying fast to the participating pilots but also create challenges for the flight control computers. As a result, the concept of autonomous drone racing has gained significant attention from research groups around the world. These races aim to push the boundaries of perception and control algorithms, while simultaneously mitigating the real-world uncertainty of execution on autonomous systems. While perception algorithms face challenges due to limited feature detection, high motion blur and computational requirements, control algorithms face challenges of convergence to the desired trajectories that are planned out in the race arena. <br/>This thesis addresses the challenge of control for racing, which is responsible for guiding the drone to design and track desired trajectories for fast flights. The control sub-modules of racing drones are responsible for generating trajectories for fastest possible flights and also for obeying these generated commands. Additionally, the requirement of limited algorithm complexity is added to match the philosophy of computationally efficient algorithms at the Micro Air Vehicle Laboratory. However, to address the requirements of these control sub-modules, the prerequisite of accurate state estimation always persists. Assigning control actions to a robot without information on the current state of the robot is rather unwise. As a result, this thesis first aims to perform accurate state estimation before designing controllers for time-optimal trajectory tracking. Again, another constraint of using only a single sensor (i.e. the Inertial Measurement Unit) is added to make the drone race in GPS denied environments. As a result, the goal of the thesis is two-fold i.e. making accurate state estimators while using limited sensors and designing optimal controllers for taking the quickest trajectory through the arena. To achieve the goal of accurate state estimation, existing techniques are studied. Several features from each of these methods are selected to design a new estimator. To achieve the goal of time-optimal trajectory generation, firstly, the flaws of traditional control methods are pointed out. A new optimal-control technique is proposed, which makes use of fundamental principles dating back several decades. This principle is then fused along with present-day optimization solvers. Finally, the proposed state estimation and control algorithm are compared against prior (benchmarked) techniques in the area. Compared to existing optimal control techniques, the proposed algorithm leads to faster trajectories and consumes less computational power onboard.},

note = {de Croon, Guido (mentor); de Wagter, Christophe (mentor); Langendoen, Koen (mentor); Zuñiga Zamalloa, Marco (mentor); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:cc607dbb-7116-4c9f-991f-988d832833a9,

title = {Wing deformation measurements of the DelFly II in different flight conditions},

author = {Dorian Heitzig},

url = {http://resolver.tudelft.nl/uuid:cc607dbb-7116-4c9f-991f-988d832833a9},

year  = {2019},

date = {2019-01-01},

school = {TU Delft Aerospace Engineering},

abstract = {This study investigates the wing deformation of the flapping-wing micro air vehicle (MAV) DelFly II in various flight configurations. Experiments were carried out with the MAV tethered in a windtunnel test section. To determine the best suited measurement approach, a trade-off study was carried out which showed that a point tracking approach with background illumination is most suitable. The therefore used high-speed camera pair and illumination were mounted on the same rotating frame with the DelFly, which allowed adequate viewing axes of the wings at for all pitch angles. Processing was done a purpose-build algorithm, allowing 136 points per wing to be measured simultaneously with an average lost point ratio of 3.4 % and an estimated accuracy of 0.25 mm. Results of hovering flight show some previously unnoticed behaviors. First, it was noted that the upper and lower wing on each side do not deform purely symmetric but show some considerable asymmetric behavior like heave and camber production. Furthermore, the upper wing shows a torsional wave and recoil behavior at faster flapping frequencies, which was shown to be beneficial in insect flight. Lastly, it was found that an air-buffer remains present between the wing surfaces at all times of the clap-and-peel motion (apart from the root trailing edge). This air-buffer increases once freestream velocity is added, which is investigated during the climbing flight study. Here, the reduced angle of attack of the wing is assumed to reduce the wing loading at faster climb, resulting in lower deformations outside the clap-and-peel motion. The isolated effect of a body pitch angle is also studied. Here, the asymmetrical freestream direction results in larger asymmetries such as wing alignment with the freestream direction and reduced camber and even camber reversal during the upstroke. In forward flight the pitch angle is changed simultaneously with the flapping frequency and freestream velocity. Due to the non-linear properties the wing deforms not directly as a superposition of the individual effects. Deviations are mostly present in increased asymmetry in incidence angle, while the camber behaves more linear and the clap-and-peel motion also remains relatively unchanged. The torsional wave and recoil are here however reduced. Descending flight was also tested. Velocities below 1m/s result in relatively minor deformation changes, while faster descent leads to large flapping frequency fluctuations, making interpretation of the results impossible.},

note = {van Oudheusden, B.W. (mentor); De Breuker, R. (graduation committee); de Wagter, C. (graduation committee); Olejnik, D.A. (mentor); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:ce520f94-bd3c-41a5-9ddd-edfdf6ead35e,

title = {Monocular Optical Flow based Attitude Estimation in Micro Aerial Vehicles: A Bio-Inspired Approach},

author = {Abhishek Chatterjee},

url = {http://resolver.tudelft.nl/uuid:ce520f94-bd3c-41a5-9ddd-edfdf6ead35e},

year  = {2019},

date = {2019-01-01},

school = {TU Delft Aerospace Engineering},

abstract = {The exceptional flight capabilities of insects have long amazed and inspired researchers and roboticists striving to make Micro Aerial Vehicles (MAVs) smaller and more agile. It is well known that optical flow plays a prominent role in insect flight control and navigation, and hence it is being increasingly investigated for applications in flying robots as well. However, optical flow based strategies for estimation and stabilization of orientation remain obscure in literature. In this report, we introduce a novel state estimation algorithm based on optical flow measurements and the knowledge of efference copies. The proposed technique estimates the following states of a flying robot (constrained to move with three degrees of freedom): roll angle, rate of change of roll angle, horizontal and vertical components of velocity and height. The estimator only utilizes the knowledge of control inputs and optical flow measurements obtained from a downward looking monocular camera. Through non-linear observability analysis, we theoretically prove the feasibility of estimating the attitude of a MAV using ventral flow and divergence measurements. Based on the findings of the observability analysis, an extended Kalman filter state estimator is designed and its performance is verified in simulations and through flight data recorded on a real flying robot. To the best of our knowledge, the introduced strategy is the first attitude estimation technique that utilizes monocular optical flow as the only sensory information.<br/><br/>Besides the investigation on optical flow based attitude estimation technique, this thesis presents a comprehensive literature survey on the main topics relevant to the work.},

note = {de Croon, G.C.H.E. (mentor); Olejnik, D.A. (graduation committee); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:38547b1d-0535-4b30-a348-67ac40c7ddcc,

title = {Online Reinforcement Learning for Flight Control: An Adaptive Critic Design without prior model knowledge},

author = {Dave Kroezen},

url = {http://resolver.tudelft.nl/uuid:38547b1d-0535-4b30-a348-67ac40c7ddcc},

year  = {2019},

date = {2019-01-01},

school = {TU Delft Aerospace Engineering; TU Delft Control & Simulation},

abstract = {Online Reinforcement Learning is a possible solution for adaptive nonlinear flight control. In this research an Adaptive Critic Design (ACD) based on Dual Heuristic Dynamic Programming (DHP) is developed and implemented on a simulated Cessna Citation 550 aircraft. Using an online identified system model approximation, the method is independent of prior model knowledge. The agent consists of two Artificial Neural Networks (ANNs) which form the Adaptive Critic Design and is supplemented with a Recursive Least Squares (RLS) online model estimation. The implemented agent is demonstrated to learn a near optimal control policy for different operating points, which is capable of tracking pitch and roll rate while actively minimizing the sideslip angle in a faster than real-time simulation. Providing limited model knowledge is shown to increase the learning, performance and robustness of the controller.},

note = {van Kampen, E. (mentor); de Croon, G.C.H.E. (graduation committee); Mitici, M.A. (graduation committee); Pan, W. (graduation committee); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:a37b96b7-90de-4f22-99ce-87cc97d414d9,

title = {Longitudinal grey-box model identification of a tailless flapping-wing MAV based on free-flight data},

author = {Jorgen Nijboer},

url = {http://resolver.tudelft.nl/uuid:a37b96b7-90de-4f22-99ce-87cc97d414d9},

year  = {2019},

date = {2019-01-01},

school = {TU Delft Aerospace Engineering},

abstract = {Tailless flapping wing micro aerial vehicles (FMWAV) are known for their light weight and agility. However, given the fact that these FWMAVs have been recently developed, their flight dynamics have not yet been fully explained. In this paper we will develop local time-averaged longitudinal grey-box models based on closed-loop system identification techniques, where free-flight experimental data, obtained from the DelFly Nimble, is used to estimate and validate the local grey-box models. With these models we can take the first steps towards fully understanding the flight dynamics of tailless FWMAVs. The consequence of the tailless configuration is inherent instability and therefore tailless FWMAVs are generally more complex, compared to its tailed counterpart, and require a active feedback control system. The active feedback control system introduces additional challenges to the system identification process since it follows that feedback control works against the objectives of system identification. Dynamic effects that play a major role when studying the dynamic behaviour of FWMAVs are the sub-flap and the flap cycle-averaged effects. However, in this paper, we are only interested in modelling the flap cycle-averaged (time-averaged) effects of the DelFly Nimble. Based on this approach, grey-box models were estimated and validated for airspeeds near hover condition 0 m/s, up to 1.0 m/s forward flight. Despite the complexity of the system, we were able to obtain low-order local models that are both efficient and accurate (R2 values up to 0.92) to predict the flight dynamic behaviour of the DelFly Nimble and can therefore be used for stability analysis, simulation and control design.},

note = {de Visser, Coen (mentor); Karasek, Matej (mentor); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:57cfce9e-75aa-42f1-bed4-64b64b927fb4,

title = {Onboard Visual Control of a Quadcopter MAV performing a Landing Task: on a Platform of Unknown Size and Location},

author = {Jari Blom},

url = {http://resolver.tudelft.nl/uuid:57cfce9e-75aa-42f1-bed4-64b64b927fb4},

year  = {2019},

date = {2019-01-01},

school = {TU Delft Aerospace Engineering; TU Delft Control & Simulation},

abstract = {Vision based control allows Micro Air Vehicles (MAV) to move autonomously in GPS-denied environments, for example in indoor applications. An open issue in this field is landing on an unknown platform. The difficulty in visual control w.r.t. such an unknown platform, is a lack of scale. Without knowledge of the scale of offsets and object sizes (without height knowledge from GPS) it is difficult to determine an appropriate response from the controller. A control algorithm is designed to fit these requirements using an adaptation of an optical flow divergence based landing scheme, combined with an Image Based Visual Servoing approach applied to features in the Virtual Camera. The approach leads to satisfactory behavior in Gazebo simulations. It results in a robust controller for a range of starting heights and divergence settings.},

note = {de Croon, G.C.H.E. (graduation committee); Scheper, K.Y.W. (mentor); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:0cd4f9e3-faf3-4e65-995c-7dd401c8da4c,

title = {Decentralized Stochastic Optimal Control for a Swarm of Micro Aerial Vehicles},

author = {Bianca Bendriş},

url = {http://resolver.tudelft.nl/uuid:0cd4f9e3-faf3-4e65-995c-7dd401c8da4c},

year  = {2019},

date = {2019-01-01},

school = {TU Delft Aerospace Engineering},

abstract = {In this work, we model a multi-robot formation planning and control task as an optimization problem, which we solve on-line and in a decentralized manner using the Stochastic Optimal Control (SOC) framework. Typically, the solution of a SOC problem requires solving the Hamilton-Jacobi-Bellman (HJB) equation for all system states and controls. However, this operation becomes intractable when high-dimensional systems are used. In recent years, advances on a certain type of SOC problem, which can be efficiently solved by sampling from a diffusion process have been presented and are better known as path integral (PI) control. We build upon this theory and implement a decentralized formulation of the PI algorithm to compute the optimal controls of real Micro Aerial Vehicles (MAVs) flying in formation using solely on-board computational resources. One challenging aspect of the PI control method is the efficient sampling of useful trajectories. It is not clear how to guide the samples towards the optimal states. To this end, we propose a probe enhanced importance sampling (PEIS) method which performs a coarse exploration of the state space with the objective of identifying an optimal guiding trajectory around which the samples are taken. The feasibility of the proposed method is shown by means of simulation and real-hardware experiments with up to four MAVs in an indoor environment.},

note = {de Croon, Guido (mentor); McGuire, Kimberly (graduation committee); Kappen, Bert (graduation committee); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:ec42642a-5872-4177-b514-c3679f6b4055,

title = {Persistent Surveillance of a Greenhouse: Evolved neural network controllers for a swarm of UAVs},

author = {Thomas Fijen},

url = {http://resolver.tudelft.nl/uuid:ec42642a-5872-4177-b514-c3679f6b4055},

year  = {2019},

date = {2019-01-01},

school = {TU Delft Mechanical, Maritime and Materials Engineering; TU Delft Delft Center for Systems and Control},

abstract = {With the growing population the agricultural industry needs to find and implement new methods for enhancing food production. Using a Micro Aerial Vehicle (MAV) in Precision Agriculture (PA) offers a large number of benefits such as enabling the farmer to create targeted strategies to increase crop yield, reduced waste and halt the spread of diseases. Despite these advantages, the use of MAVs, particularly in greenhouses, is still very limited. To this end, this thesis seeks to combine, improve and implement existing strategies to solve the persistent surveillance task for a swarm of MAVs operating in a greenhouse environment.<br/>Broadly speaking, the persistent surveillance task seeks to find the optimal paths for a swarm of MAVs such that every point within the Mission Space (MS) is visited and they must minimise the time between successive visits. This will ensure that the MAVs are able fly through the entire greenhouse to collect up-to-date data about all the crops and the local environment. Naturally, on a physical system one has to deal with the limited flight times of the MAVs. This factor becomes very important to the effectiveness of the solution and is critical to the continuous operation of the MAVs.<br/>In literature, many methods have be proposed to solve this task, but the majority are still only tested in simulation. As a result, many works do not consider some physical constraints that will be applied to the system during implementation in a real-world setting. For example, in most cases the authors do not consider the limited fuel available to the agents or they do not consider a practical alternative indoor positioning system to GPS. In this work the problem has been divided into two main sub-tasks, namely; the persistent surveillance task and the refuelling task.<br/>For the persistent surveillance task it was decided to implement a reactive controller, in the form of an evolved Neural Network (NN), which was run on-board the MAVs. The NN used positional information from the other members of the swarm along with limited environmental information to supply its MAV with a command velocity. These NN controllers could achieve coverage levels of over 95% while simultaneously avoiding collisions between 8 MAVs in a 25m x 25m MS. Later, this method was shown to be robust to failures and scalable in terms of both MS and swarm size.<br/>When dealing with the fuel constraints, a Behaviour Tree (BT) was used to determine when the MAV should return to the depot. Surprisingly, when combined with the NN controllers the system experienced an increase in performance across all the defined metrics. No MAV failed due to low fuel levels, coverage increased to 97.41%, average cell age to 52.39s and the number of tests were no collisions were recorded more than doubled. This increase in performance was attributed to the fact that the refuelling periodically drew the MAV towards the centre of the MS. This is counter to the evolved behaviours of the NN where the MAVs would mainly focus their attention around the edges of the MS.<br},

note = {Keviczky, T. (mentor); de Croon, G.C.H.E. (mentor); Coppola, M. (graduation committee); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:82cb0f68-061e-4346-b536-a35a61621e51,

title = {Detecting Empty Wireframe Objects on Micro-Air Vehicles: Applied for Gate Detection in Autonomous Drone Racing},

author = {Philipp Dürnay},

url = {http://resolver.tudelft.nl/uuid:82cb0f68-061e-4346-b536-a35a61621e51},

year  = {2018},

date = {2018-01-01},

school = {TU Delft Electrical Engineering, Mathematics and Computer Science},

abstract = {Autonomous MAV are an emerging technology that supports a wide range of applications such as medical delivery or finding survivors in disaster scenarios. As flying in such missions is difficult the robust estimation of an MAV's state within its environment is crucial to ensure safe operation. In indoor scenarios, cameras are one of the predominant choices for state estimation sensors. This requires Computer Vision algorithms to interpret the obtained high dimensional signal. An application that allows the competitive evaluation of control and state estimation algorithms is MAV Racing such as the IROS 2018 Autonomous Drone Race. Thereby a race court consisting of several race gates has to be followed. For a fast flight during such a race court the detection of the racing gates with a camera can be used in a high level control loop. As these objects consist only of small structures that are spread across large parts of the image, this gives rise to a challenging Object Detection problem. In recent years CNN showed promising results on various vision tasks. However, due to their computational complexity the deployment on mobile devices remains a challenge. Furthermore, CNN typically require a vast amount of training data. Finally, the objects typically studied in Object Detection consist of solid and complex features which is not the case for racing gates. Therefore, this work defines the class of EWFO and studies their detection on MAV with YoloV3. Thereby, the training data is created with a graphical engine. We are interested in how to detect EWFO with a CNN on a MAV, using synthetic data. We conduct several simple experiments about EWFO in simulation and compare their detection to more filled objects. Subsequently experiments in a more challenging environment such as an MAV race are conducted. The experiments show how EWFO are harder to detect than filled objects as the detector can be confused to patterns present in the empty part. Particularly for larger objects the detection performance decreases. We give several recommendations on how to generate data for the detection of EWFO on MAV. These include how to add variations in background as well as the camera placement. Finally, we study the incorporation of image augmentation techniques to transfer the detector to the real world. We can report that especially modelling lens distortion improves the performance on the real data. Nevertheless, a reality gap remains that can not fully be explained. Furthermore, different architectures are studied for the detection of EWFO. It can be seen how a relatively shallow network of 9 layers can be used for the detection of EWFO on MAV. A further reduction in weights leads to a gradual decrease in performance. Based on the gained insights the deployment of a detector on the example system JeVois is studied. A detection performance/speed trade-off is evaluated. The final detector achieves 32% average precision at a frame rate of 12 Hz on a real world test set created during this work. The gained insights can be used to deploy the detector in a control loop for MAV. This ensures the safe flight through a racing court of an autonmous drone race. The gained insights about the detection of EWFO can be transferred to objects with similar properties},

note = {Tax, D.M.J. (mentor); de Croon, G.C.H.E. (mentor); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:bc31203c-1956-4665-a92a-8203881f22ce,

title = {Attitude modeling of the DelftaCopter: a system identification approach},

author = {Joost Meulenbeld},

url = {http://resolver.tudelft.nl/uuid:bc31203c-1956-4665-a92a-8203881f22ce},

year  = {2018},

date = {2018-01-01},

school = {TU Delft Aerospace Engineering; TU Delft Control & Simulation},

abstract = {Previous years have seen a rise in the use of Unmanned Aerial Vehicles (UAVs). Reaching a large endurance and range while being able to perform Vertical Take-Off and Landing (VTOL) landings allows a broad range of applications. For this purpose the DelftaCopter (DC) was developed, a tilt-body tailsitter UAV. It hovers using a single helicopter rotor for lift and transitions to forward flight by pitching its body down by 90°. In this forward flight state, wings generate the lift, while the helicopter rotor now provides thrust. The single rotor is more efficient than using multiple smaller rotors and helicopter swashplate is used for attitude and speed control. The heavy single helicopter rotor introduces significant gyroscopic moments, as is the case for all helicopters. In contrast with normal helicopters, the DC has a heavy fuselage putting the attitude dynamics between a helicopter and aircraft. In previous research, a controller based on a model incorporating the rotor as a rotating cylinder was implemented. This controller was unable to counteract the gyroscopic pitch-roll coupling, leading to the question of this thesis: how should the DC be modeled to allow control design. <br/>In this thesis, the previous model is called the Cylinder Dynamics (CD) model, and is compared with another model from literature. The latter model, in this thesis called the Tip-Path Plane (TPP) model, includes the flapping dynamics through the tip-path plane dynamics and is also a linear state-space model. In flight tests, chirps were used to cover a broad frequency range. Fitting both the CD and TPP models on this flight test data, it is shown that the CD model lacks accuracy in the high-frequency area, while the TPP is able to accurately model these dynamics. This shows that the flapping dynamics are important to the attitude dynamics of the DC. An Linear Quadratic Regulator (LQR) controller was implemented based on the fitted TPP model, and shows adequate tracking performance, further validating the applicability of the model to the DC. For forward flight, extensions to the hover models are proposed. The extension including the elevator and aerodynamic damping is shown to simulate key dynamics of the DC in forward flight with reasonable accuracy. The parameters and eigenfrequencies of this model are not significantly different from the hover model. Therefore it can be concluded that the gyroscopic effect plays an important role in forward flight attitude dynamics. Another extension which estimates of angle of attack and sideslip using high-pass filtered rotational rates, yields better accuracy, but significantly changes the model parameters also present in the hover model. More research with angle of attack and sideslip vanes could validate this modeling approach. It was also found that for a new version of the DC with a smaller, more quickly rotating rotor, the modeling done before resulted in much worse fits. It was shown that the CD and TPP model response is much more comparable for this version. Control performance also suffers due to this lower accuracy model fit. Further research is required to understand why this is the case.},

note = {de Wagter, C. (mentor); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:28fad2a0-4b4c-47f4-8930-01708f4b52d1,

title = {Hear-and-avoid for UAVs using convolutional neural networks},

author = {Dirk Wijnker},

url = {http://resolver.tudelft.nl/uuid:28fad2a0-4b4c-47f4-8930-01708f4b52d1},

year  = {2018},

date = {2018-01-01},

school = {TU Delft Aerospace Engineering},

abstract = {We investigate how an Unmanned Air Vehicle (UAV) can detect manned aircraft with a single microphone. In particular, we create an audio data set in which UAV ego-sound and recorded aircraft sound can be mixed together, and apply convolutional neural networks to the task of air traffic detection. Due to restrictions on flying UAVs close to aircraft, the data set has to be artificially produced, so the UAV sound is captured separately from the aircraft sound. The aircraft data set is collected at Lelystad airport by capturing flyovers with a microphone array. It is mixed with UAV recordings, during which labels are given indicating whether the mixed recording contains aircraft audio or not. The mixed recordings are the input for a model that determines whether an aircraft is present or not. The model is a CNN which uses the features MFCC, spectrogram or Mel spectrogram as input. For each feature the effect of UAV/aircraft amplitude ratio, the type of labeling, the window length and the addition of third party aircraft sound database recordings is explored. The results show that the best performance is achieved using the Mel spectrogram feature. The performance increases when the UAV/aircraft amplitude ratio is decreased, when the time window is increased or when the data set is extended with aircraft audio recordings from a third party sound database. It is not desirable to train the model on distant approaches and test them on nearby approaches as the performance then drops. The results also prove that the performance increases the closer the aircraft is. Although the currently presented approach has a number of false positives and false negatives, that is still too high for real-world application, this study indicates multiple paths forward that can lead to an interesting performance. In addition, the data set is provided as open access, allowing the community to contribute to the improvement of the detection task.},

note = {van Dijk, Tom (mentor); de Croon, G.C.H.E. (mentor); de Wagter, C. (mentor); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:b0394f21-302c-484d-a6dc-031f5860c521,

title = {Flight control and collision avoidance for quadcopter and flapping wing MAVs using only optical flow: Theory, Simulation and Experiment},

author = {Daan Vrede},

url = {http://resolver.tudelft.nl/uuid:b0394f21-302c-484d-a6dc-031f5860c521},

year  = {2018},

date = {2018-01-01},

school = {TU Delft Mechanical, Maritime and Materials Engineering},

abstract = {Both quadcopter Micro Aerial Vehicles (MAVs) and Flapping Wing MAVs (FWMAVs) are constrained in Size, Weight and Processing power (SWaP) in order to achieve flight tasks that include attitude and velocity stabilisation, as well as obstacle avoidance. <br/>Conventional sensory and control approaches, such as those relying on inertial, visual and Global Positioning System (GPS) sensors, can fulfil these tasks using sensor fusion. However such approaches do not score well in terms of SWaP criteria. <br/>Very simple proportional feedback control laws using single optical flow vectors from very basic high frame-rate low-resolution cameras provide a promising path to achieve aforementioned tasks. <br/>This thesis shows that in theory these control laws are well suited for stabilising a FWMAV, and could be used for a high-drag adapted quadcopter MAV within bounds. Simulations confirm these findings and illustrate robustness to noise and additional emergent behaviour such as sideways wall avoidance and trajectory following, however simulations also show that disparity between walls can lead to unintended rotational behaviour during vertical translation. <br/>The system is tested in experiment on a quadcopter-like setup with onboard processing, using only ADNS 9800 computer mouse optical flow sensors for flight control. Results show that the system behaves similarly to simulation, however the sensory configuration used is highly dependent on texture in environment and light conditions. <br/>For future work it is recommended to investigate optical flow sensors in more detail to obtain reliable output on a vibrating platform (such as a FWMAV) in a broader range of texture and light conditions. Preliminary results from theory, simulation and experiment indicate that the addition of derivative feedback could strongly enhance performance on a quadcopter MAV and remove the requirement for high drag.},

note = {Goosen, J.F.L. (mentor); de Croon, G.C.H.E. (mentor); Breedveld, P. (mentor); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:c75ab7d9-e711-4e0c-93ea-ff092e2e9131,

title = {Estimation of ego-motion velocities from single static images},

author = {Yeshwanth Napolean},

url = {http://resolver.tudelft.nl/uuid:c75ab7d9-e711-4e0c-93ea-ff092e2e9131},

year  = {2018},

date = {2018-01-01},

school = {TU Delft Aerospace Engineering},

abstract = {Velocity estimation based on visual information is a well- researched topic. Traditional approaches usually rely on how a given feature or features change between successive images in a sequence. However, a single static image might contain motion information that could potentially be lever- aged to estimate the optical flow. It can be hypothesized that motion blur and context of the scene are two sources of mo- tion information in static images. This research work has two main goals, one is to investigate the prospect of using a learning-based framework to model a mapping directly to camera ego-motion velocity. The second goal is to ana- lyze the contributing features in learning such a mapping. Experiments show that the model is able to learn velocity based on context of the scene but performs better when in- put images contain motion blur.<br},

note = {de Croon, G.C.H.E. (mentor); van Gemert, J.C. (graduation committee); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:82cfe98f-46ad-42b4-8021-082fbae1f740,

title = {Passive Localization of Robots with Ambient Light},

author = {Danielle Werff},

url = {http://resolver.tudelft.nl/uuid:82cfe98f-46ad-42b4-8021-082fbae1f740},

year  = {2018},

date = {2018-01-01},

school = {TU Delft Electrical Engineering, Mathematics and Computer Science},

abstract = {A lot of research is been being done on Visible Light Communication (VLC), which has shown to be of interest for many applications, such as localization. Since localization based on VLC requires active modulation of light sources, this limits the amount of light sources that can be used for localization. Furthermore, in some situations there might not even be a controllable light source present (for example outdoors). To extend the use of light-based localization schemes, this thesis looks into a way to achieve the same result as current VLC localization methods in a passive manner, i.e. without control of the light sources. <br/><br/>Previous work has been done on passive ambient light-based localization by Wang et al.: objects are equipped with unique barcodes, that reflect ambient light in a distinct manner. The reflected light is received by photosensors, from which their ID is obtained. However, this work has focused on identifying large-sized objects in one dimension. Using the same principle for localization of small-sized objects, and in two dimensions, are open challenges that this thesis addresses . <br/><br/>The work presented here forms a proof-of-concept of a passive light-based localization system for two-dimensional, real-time tracking of small-sized objects. In order to achieve this, a special enclosure has been designed, giving simple photosensors the ability to distinguish small-sized objects without compromising their FOV. With this enclosure, a single photosensor can detect barcodes down to 7 cm in size in the test set-up, while distinguishing up to three different IDs. A particle filter has been implemented to combine detections from different photosensors into a single estimate of an object’s location. <br/><br/>The localization system is designed around the robots designed by a MSc student at the Embedded Systems group at TU Delft. By moving these robots at a speed of 15.4 cm/s in a straight line through the test set-up, a localization error of 4.8 cm is obtained. The distance between the robots and the sensor equals 20 cm.},

note = {Zuñiga Zamalloa, Marco (mentor); Pawelczak, Przemek (mentor); de Croon, Guido (graduation committee); Langendoen, Koen (graduation committee); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:d3bd3ba9-61aa-4d48-b13d-aead118b8015,

title = {Vector Field Based Path Following for UAVs using Incremental Nonlinear Dynamic Inversion},

author = {Suresh Sharma},

url = {http://resolver.tudelft.nl/uuid:d3bd3ba9-61aa-4d48-b13d-aead118b8015},

year  = {2018},

date = {2018-01-01},

school = {TU Delft Aerospace Engineering; TU Delft Control & Operations},

abstract = {This work presents a vector field based path following method to be used by Multirotor Unmanned Aerial Vehicles (UAVs). The desired path to be followed is a smooth planar path defined in its implicit form. The vector field around the desired path is then constructed using the implicit function, such that the integral curves of the vector field converge to the path. The algorithm takes into account the future change in the trajectory as well as the current state of the UAV in order to calculate the desired linear acceleration, which is then tracked using the Incremental Nonlinear Dynamic Inversion (INDI) controller in the autopilot. The implementation also allows for the velocity of the UAV to be controlled independently. The efficiency of the algorithm is demonstrated using real world flight tests, and the performance is shown to be better than the<br/>traditional carrot-chasing controller.},

note = {Smeur, E.J.J. (mentor); Chu, Q. P. (mentor); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:3794f912-f141-4fa7-aff4-464598958e94,

title = {Vision-based Autonomous Drone racing in GPS-denied Environments},

author = {Michaël Ozo},

url = {http://resolver.tudelft.nl/uuid:3794f912-f141-4fa7-aff4-464598958e94},

year  = {2018},

date = {2018-01-01},

school = {TU Delft Aerospace Engineering},

abstract = {High-speed autonomous flight of Micro Air Vehicles has gained much attention in recent years. However, flight in complex GPS-denied environments still poses a serious challenge. One scenario which contains these elements is drone racing, where pilots have to fly complex tracks at high speed, often in an indoor environment. In this work we therefore present an MAV capable of autonomously flying such a drone race track. The system has to operate in a GPS-denied environment, hence a visual navigation method is employed. Position is recovered from gate detections based on a novel least-squares method, while heading is estimated using an optimization based method. Experiments show that both methods have a higher accuracy than the standard P3P pose estimation method. Furthermore, a state estimation filter is designed to fuse the visual measurements with IMU measurements, by using an EKF with drag based prediction model. For high-level control different motion primitives are linked, which allow the MAV to fly the track without having a detailed on-board map. The overall approach does not rely on SLAM or Visual odometry, which results in low computational complexity. Also, it does not rely on downward optical flow velocity measurements, which enables it to work even in low texture environments.},

note = {de Croon, G.C.H.E. (mentor); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:717e7e15-94c3-47ac-a348-12f5c2275aa2,

title = {A minimal longitudinal dynamic model of a tailless flapping wing robot},

author = {Karl Kajak},

url = {http://resolver.tudelft.nl/uuid:717e7e15-94c3-47ac-a348-12f5c2275aa2},

year  = {2018},

date = {2018-01-01},

school = {TU Delft Aerospace Engineering; TU Delft Control & Operations},

abstract = {Tailless flapping wing micro air vehicles (FWMAVs) have<br/>the potential of providing efficient flight at small scale,<br/>with considerable agility. However, this agility also brings<br/>significant control challenges, which are exacerbated by<br/>the fact that the aerodynamics and dynamics of flapping<br/>wing robots are still only partly understood.<br/>In this article, we propose a novel, minimal dynamic<br/>model that is not only validated with experimental data,<br/>but also able to predict the consequences of various important<br/>design changes. Specifically, the model captures<br/>the flapping cycle averaged longitudinal dynamics of a<br/>tailless flapping wing robot, taking into account the main<br/>aerodynamic effects. The model is validated for airspeeds<br/>up to 3.5 m/s (when the forward velocity starts to approximate<br/>the wing velocities). It successfully predicts the effects<br/>of changes to the center of mass and flight at different<br/>pitch angles. Hence, the presented model forms an<br/>important step in accelerating the control design of flapping<br/>wing robots - which can now be done to a greater<br/>extent in simulation. In order to illustrate this, we have<br/>used the model to improve our control design, resulting in<br/>a change of the maximal stable speed of the tailless DelFly<br/>Transformer from 4 m/s to 7 m/s.},

note = {Karasek, M. (mentor); Chu, Q. P. (mentor); de Croon, G.C.H.E. (mentor); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:28774acc-89e7-44a8-be43-b06d2db0056c,

title = {Online Trajectory Planning and Control of a MAV Payload System in Dynamic Environments: A Non-Linear Model Predictive Control Approach},

author = {Nikhil Potdar},

url = {http://resolver.tudelft.nl/uuid:28774acc-89e7-44a8-be43-b06d2db0056c},

year  = {2018},

date = {2018-01-01},

school = {TU Delft Aerospace Engineering},

abstract = {Micro Aerial Vehicles (MAVs) are increasingly being used for aerial transportation in remote and urban spaces where portability can be exploited to reach previously inaccessible and inhospitable spaces. Current approaches to MAV swung payload system path planning have primarily focused on pre-generating (agile) collision-free, or conservative minimal-swing trajectories in static environments. However, these approaches have failed to address the prospect of online re-planning in uncertain and dynamic environments which is a prerequisite for real-world deployability. This article describes a novel Non-Linear Model Predictive Controller (NMPC) for online, agile and closed-loop local trajectory planning and control addressing the limitations mentioned of contemporary approaches. We integrate the controller in a full system framework and demonstrate the algorithm’s effectiveness in simulation and experimental studies. Results show the scalability and adaptability of our method to various dynamic setups with repeatable performance over several complex tasks which include flying through a narrow opening and avoiding moving humans.},

note = {de Croon, G.C.H.E. (mentor); Alonso Mora, J. (mentor); de Visser, C.C. (graduation committee); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:6a3c6f9c-1634-4575-844c-510092f73dc6,

title = {On-board Range-based Relative Localization: For Leader-Follower Flight of Micro Aerial Vehicles},

author = {Steven Helm},

url = {http://resolver.tudelft.nl/uuid:6a3c6f9c-1634-4575-844c-510092f73dc6},

year  = {2018},

date = {2018-01-01},

school = {TU Delft Aerospace Engineering; TU Delft Control & Simulation},

abstract = {In this paper a range-based relative localization solution is proposed and demonstrated in practice. The approach is based on wireless range measurements between robots, along with the communication of their velocities, accelerations, yaw rates, and height. It can be implemented on many robotic platforms without the need for dedicated sensors. With respect to previous work, we remove the dependency on a common heading reference between robots. The main advantage of this is that it makes the relative localization approach independent of magnetometer readings, which are notoriously unreliable in an indoor environment. A theoretical observability analysis shows that it may also have two disadvantages: the motion of the robots must meet more stringent conditions and the relative localization method becomes more susceptible to noise on the range measurements. However, simulation results have shown that in the presence of significant magnetic disturbances that are common to indoor environments, removing the heading dependency is beneficial. We conclude the paper by implementing the heading-independent method on real Micro Aerial Vehicles (MAVs) and performing leader-follower flight in an indoor environment. Despite the observability analysis showing leader-follower flight to be an especially difficult task, we still manage to successfully fly for over 3 minutes with two fully autonomous followers using only on-board sensing.},

note = {de Croon, G.C.H.E. (mentor); McGuire, K.N. (graduation committee); Coppola, M. (graduation committee); Chu, Q. P. (graduation committee); Verhoeven, C.J.M. (graduation committee); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:07bc64ba-42e3-4aa7-ba9b-ac0ac4e0e7a1,

title = {Deep Reinforcement Learning for Goal-directed Visual Navigation},

author = {Máté Kisantal},

url = {http://resolver.tudelft.nl/uuid:07bc64ba-42e3-4aa7-ba9b-ac0ac4e0e7a1},

year  = {2018},

date = {2018-01-01},

school = {TU Delft Aerospace Engineering; TU Delft Control & Operations},

abstract = {Safe navigation in a cluttered environment is a key capability for the autonomous operation of Micro Aerial Vehicles (MAVs). This work explores a (deep) Reinforcement Learning (RL) based approach for monocular vision based obstacle avoidance and goal directed navigation for MAVs in cluttered environments. We investigated this problem in the context of forest flight under the tree canopy.<br/><br/>Our focus was on training an effective and practical neural control module, that is easy to integrate into conventional control hierarchies and can extend the capabilities of existing autopilot software stacks. This module has the potential to greatly improve the autonomous capabilities of MAVs, and their applicability for many interesting real world use-cases. We demonstrated training this module in a visually highly realistic virtual forest environment, created with a state-of-the-art computer game engine.},

note = {de Croon, G.C.H.E. (mentor); van Hecke, K.G. (mentor); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:ed1025e8-66ad-4e71-ba8e-771f511e9022,

title = {Ready for detection: Stair-detecting in depth images using spatial features and Adaboosting},

author = {Frerik Andriessen},

url = {http://resolver.tudelft.nl/uuid:ed1025e8-66ad-4e71-ba8e-771f511e9022},

year  = {2018},

date = {2018-01-01},

school = {TU Delft Aerospace Engineering},

abstract = {Space exploration could be significantly aided by combining the disciplines of machine learning and computer vision, but these disciplines need to be developed further for specific space-related applications to have merit. One of the applications for space exploration is the detection of certain structures designating areas of interest. This thesis demonstrates a method of structure-detecting that is applied to staircases. In addition to incorporating certain physical features, like other algorithms have done, the proposed algorithm (Step-1) also takes into account the spatial relation between these features, in order to increase its robustness. Looking at a staircase from the front, the distances between each step become warped, as they are further away from the observer. This exponential spatial distortion is known as a ’chirp’. Step-1 tries to match a chirp-waveform to every edge along a straight line randomly drawn through an image, and based on that match classify the image as containing a staircase or not. The random lines are then weighted based on their effectiveness using Adaboost, which are finally combined to obtain a final classification. The results show potential but there are still some issues to be addressed. However, once the algorithm has been upgraded it could aid space exploration by being applied to satellite images and autonomous rovers.},

note = {Sundaramoorthy, P.P. (mentor); Cervone, A. (graduation committee); de Croon, G.C.H.E. (graduation committee); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:4a03785d-8822-4548-a24a-7b23b1a4232c,

title = {Uncertainty Estimation in Vision-Aided Robot Teleoperation System},

author = {Shubham Vyas},

url = {http://resolver.tudelft.nl/uuid:4a03785d-8822-4548-a24a-7b23b1a4232c},

year  = {2018},

date = {2018-01-01},

school = {TU Delft Aerospace Engineering; TU Delft Space Engineering},

abstract = {Teleoperation allows the use of human intelligence and decision making in remote tasks which are too dangerous for humans to perform. Technologies such as force feedback and haptic guidance have shown to increase task efficiency during teleoperation. In an unmodeled environment, sensors provide input for haptic guidance or present extra information about the environment to the user in order to make decisions and to perform the tasks. These sensors come with inherent errors and uncertainties which propagate through the teleoperation system. The absence of knowledge of these errors has been shown to cause deterioration in the task performance. These errors can further cause the application of forces on the environment by the robot without the knowledge of the user while using haptic guidance. Thereby, the strategies being used to increase task performance can have some adverse hidden effects. Thus, it crucial to have an understanding of the behaviour of the errors and uncertainties in the system. It is considered critical for making decisions about how the robot system can be controlled and used to manipulate objects remotely.<br/><br/>In this thesis, a novel framework for estimating the uncertainties in a vision-aided teleoperation system in real-time is introduced. The uncertainty estimate can then be used by the control system or communicated to the user. Methods to use the uncertainty estimate for haptic guidance and for user display are proposed. Furthermore, the thesis analyzes the behaviour of the uncertainties in the system and the sensitivity of the system to individual component errors. It evaluates the uncertainties in individual components of the system and implements an uncertainty model for each of them. It then provides a method to propagate these uncertainty models through the system. This results in a final uncertainty estimate in the frame of reference of interest for the task. Experiments were performed to validate the component uncertainty models, the propagation method, and the system as a whole. Additionally, an inverse of the propagation method is also conceived so as to obtain the component accuracy specification from system uncertainty requirements. This can be used in the design of future teleoperation systems.},

note = {Verhoeven, C.J.M. (mentor); Krueger, Thomas (mentor); Schiele, A. (mentor); Gill, E.K.A. (graduation committee); de Croon, G.C.H.E. (graduation committee); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:34ce9384-9352-41bc-99ce-2a54bd1f3361,

title = {Informative Path Planning for Search and Rescue using a UAV},

author = {Ajith Anil Meera},

url = {http://resolver.tudelft.nl/uuid:34ce9384-9352-41bc-99ce-2a54bd1f3361},

year  = {2018},

date = {2018-01-01},

school = {TU Delft Mechanical, Maritime and Materials Engineering},

abstract = {Target search in an obstacle filled environment is a practically relevant challenge in robotics that has a huge impact in the society. The wide range of applications include searching for victims in a search and rescue operation, detecting weeds in precision agriculture, patrolling borders for military and navy, automated census of endangered species in a forest etc. An efficient target search algorithm provides a data acquisition platform with least human intervention, thus improving the quality of life of humans. This thesis aims at introducing a general path planning algorithm for UAVs flying at different heights in an obstacle filled environment, searching for targets in the ground field. An adaptive informative path planning (IPP) algorithm is introduced that simultaneously trade off between area coverage, field of view, height dependent sensor performance and obstacle avoidance. It plans under uncertainties in the sensor measurements at varying heights, and is robust against wrong target detections. It generates an optimal fixed horizon plan in the form of a 3D minimum-snap trajectory that maximizes the information gain in minimum flight time by providing maximum area coverage, without any collision with the obstacles. The resulting planner is modular in terms of the mapping strategy, environment complexity, different target, changes in the sensor model and optimizer used. The planner is tested against varying environmental complexities, demonstrating its capability in handling a wide range of possible environments. The planner outperforms other planners like non-adaptive IPP planner, coverage planner and random sampling planner, by demonstrating the fastest decrease in map error while flying for a fixed time budget. A proof of concept for the algorithm is provided through real experiments by running the algorithm on a UAV flying inside a lab environment, searching for targets lying on the ground. All the targets were successfully found and mapped by the algorithm, demonstrating its applicability in a real-life target search problem.},

note = {Siegwart, Roland (mentor); Wisse, M. (mentor); Popović, Marija (mentor); Millane, Alexander (mentor); Pan, W. (graduation committee); Alonso Mora, J. (graduation committee); Mohajerin Esfahani, P. (graduation committee); Delft University of Technology (degree granting institution); ETH Zürich (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:aa13959b-79b9-4dfc-b5e0-7c501d9d3e2f,

title = {Neuromorphic Computing of Event-Based Data for High-Speed Vision-Based Navigation},

author = {Fede Paredes Valles},

url = {http://resolver.tudelft.nl/uuid:aa13959b-79b9-4dfc-b5e0-7c501d9d3e2f},

year  = {2018},

date = {2018-01-01},

school = {TU Delft Aerospace Engineering; TU Delft Control & Simulation},

abstract = {The combination of Spiking Neural Networks and event-based vision sensors holds the potential of highly efficient and high-bandwidth optical flow estimation. This thesis presents, to the best of the author’s knowledge, the first hierarchical spiking architecture in which motion (direction and speed) selectivity emerges in a biologically plausible unsupervised fashion from the stimuli generated with an event-based camera. A novel adaptive neuron model and Spike-Timing-Dependent Plasticity formulation are at the core of this neural network governing its spike-based processing and learning, respectively. After convergence, the neural architecture exhibits the main properties of biological visual motion systems: feature extraction and local and global motion perception. To assess the outcome of the learning, a shallow conventional Artificial Neural Network is trained to map the activation traces of the penultimate layer to the optical flow visual observables of ventral flows. The proposed solution is validated for simulated event sequences with ground truth measurements. Experimental results show that accurate estimates of these parameters can be obtained over a wide range of speeds.},

note = {de Croon, G.C.H.E. (mentor); Scheper, K.Y.W. (mentor); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:ff167b06-bbaf-4897-b76c-9f246e50eadb,

title = {Directed Increment Policy Search for Behavior Tree Task Performance Optimization: Crossing the Reality Gap},

author = {Steven Leest},

url = {http://resolver.tudelft.nl/uuid:ff167b06-bbaf-4897-b76c-9f246e50eadb},

year  = {2017},

date = {2017-01-01},

school = {TU Delft Aerospace Engineering},

abstract = {Robotic behavior policies learned in simulation suffer from a performance degradation once transferred to a real-world robotic platform. This performance degradation originates from discrepancies between the real-world and simulation environment, referred to as the reality gap. To cross the reality gap, this papers presents a simple reinforcement learning algorithm named Directed Increment Policy Search (DIPS). DIPS is a form of episodic model-free policy search which leverages the interpretable structure and the coupling of the Behavior Tree (BT) parameters to reduce the number of required real-world evaluations. Additionally, DIPS does not require a form of reward function crafting and is robust to hyper-parameter settings. DIPS is evaluated on a simulated model of the DelFly Explorer which is tasked to perform a window fly-through maneuver. It is demonstrated that DIPS efficiently and effectively improves the BT behavior policy performance for three simulated environments with increasingly large reality gaps. We believe DIPS can generalize to other behavior representation methods and tasks due to the inherent coupling between behavior and environment experienced by embodied robots.},

note = {van Kampen, E. (mentor); de Croon, G.C.H.E. (mentor); Scheper, K.Y.W. (mentor); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:9494dbb1-9da0-44de-b401-02e297a3689c,

title = {Structure-from-Motion Hazard Detection for Autonomous Planetary Landings},

author = {Alessandro Vanzella},

url = {http://resolver.tudelft.nl/uuid:9494dbb1-9da0-44de-b401-02e297a3689c},

year  = {2017},

date = {2017-01-01},

school = {TU Delft Aerospace Engineering},

abstract = {Future space exploration missions on solar system bodies will require landing safely and precisely, with an accuracy of ~100 m at touchdown. This accomplishment is made challenging by vehicle design limitations, the dearth of onboard situational awareness, and the limited knowledge of the variability of the landing terrain. To date, only the Chinese Chang’e-3 has implemented hazard detection and avoidance capabilities, within its Guidance, Navigation, and Control (GN&C) subsystem, therefore being able to actively adjust its trajectory. On the contrary, the majority of the space landers only had the ability to execute autonomously a small series of simple and programmed commands. Therefore, past missions have essentially landed "blind" in regions deemed relatively safe, forcing landing site selection to be capability-limited rather than scientifically driven. In this thesis, hazard detection was investigated as a mean to increase autonomy for planetary landings and to further decrease the risk of a landing failure, employing equipment readily available on space missions. The analysis has been limited to the framework of Structure-from-Motion (SfM) where the input images are acquired from a single moving camera and thus the scene is reconstructed from the resulting video sequence. A software package was developed and tested to compute depth maps from adjacent descent images, captured at half altitude from one to the other. The basic pinhole camera model was selected to address the measurement taken from synthetic surface images, rendered in the Planet and Asteroid Natural scene Generation Utility (PANGU). To assess the hazardousness of the terrain, hazard maps are computed combining slope, roughness, and shadow information. In contrast to the results of the Jet Propulsion Laboratory (JPL) NASA, it has been shown that rocks and boulders are not well resolved from shape recovery with both low- and high-elevation image pairs. Thus, their presence on the surface has been accounted through an adapted version of theHarris Corner detector directly on the input images. Two different mission scenarios were simulated: 1) a perfect vertical motion forward along the camera pointing direction and 2) a 45° angle dropping trajectory for a more realistic approaching descent phase, with a 40° imaging sensor line-of-site offset. Furthermore, the limitations of the developed algorithm were tested under ordinary operative conditions. For the former scenario, the results show that the overall quality of the recovered depth maps does not appear adequate enough for landing site selection. As a matter of fact, the locations around the image centre can not be correctly assessed. This represents a significant problem since these locations are the most convenient in terms of distance and guidance costs. On the contrary, the latter descent sequence indicates that below 300maltitude the software is a suitable candidate for hazard detection, with total correct detection on average >94% and the percentage of undetected hazards below the allowable maximum 1%. To assess the algorithm robustness to errors in camera position, a Monte Carlo simulation was performed. Thereupon, random uncertainties within the interval [-0.5 0.5] meters were taken into account for the altitude of both camera poses. The errors for the computed Digital Elevation Model (DEM) are bounded to the maximum allowable only when both altitudes are affected by small deviation of similar magnitude and same sign (approximately 10 cm), peaking to 250%-300% increase for the other values of the considered interval. Moreover, concerning the robustness to errors in camera orientation, deviations of the camera pointing direction were considered only along the plane containing both the normal to the surface and the camera axis. Already differences greater than +0.05°, in the imaging sensor line-of-site, are responsible for exorbitant errors in the DEM for all altitudes. These results clearly indicate that the developed SfMalgorithmis not suitable as a stand-alone method for hazard detection and landing site selection.},

note = {Woicke, S. (mentor); Mooij, E. (mentor); de Croon, G.C.H.E. (graduation committee); Visser, P.N.A.M. (graduation committee); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:faf5d4fb-5785-4d27-9d52-0b09214f3a6a,

title = {Fusion of stereo and monocular depth estimates in a self-supervised learning context},

author = {Diogo Tomás Cardoso Rézio Martins},

url = {http://resolver.tudelft.nl/uuid:faf5d4fb-5785-4d27-9d52-0b09214f3a6a},

year  = {2017},

date = {2017-01-01},

school = {TU Delft Aerospace Engineering},

abstract = {We study how autonomous robots can better evaluate distances by fusing depth estimates from both stereo vision and a convolutional neural network (CNN) that processes a single still image. The main contribution is a novel fusion method that preserves high confidence stereo estimates, while leveraging the CNN estimates in the low-confidence regions. The main concern with such a fusion scheme is that the CNN may work on the training set, but will degrade significantly in the operational environment. Therefore, we also show that the performance of the monocular estimator in the operational environment improves if stereo vision provides supervised targets in a self-supervised learning (SSL) fashion. The merging framework is implemented on-board of a Parrot SLAMDunk<br/>and tested in real world scenarios, providing more reliable depth maps for use in autonomous navigation.},

note = {de Croon, G.C.H.E. (mentor); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:82c91d74-6c01-4718-a574-221df210f01a,

title = {Low-memory Visual Route Following for Micro Aerial Vehicles in Indoor Environments},

author = {Tom Dijk},

url = {http://resolver.tudelft.nl/uuid:82c91d74-6c01-4718-a574-221df210f01a},

year  = {2017},

date = {2017-01-01},

school = {TU Delft Mechanical, Maritime and Materials Engineering},

abstract = {This thesis presents a visual route following method that minimizes memory consumption to the point that even Micro Aerial Vehicles (MAV) equipped with only a simple microcontroller can traverse distances of a few hundred meters. Existing Simultaneous Localization and Mapping (SLAM) algorithms are too complex for use on a microcontroller. Instead, the route is modeled by a sequence of snapshots that can be followed back using a combination of visual homing and odometry. Three visual homing methods are evaluated to find and compare their memory efficiency. Of these methods, Fourier-based homing performed best: it still succeeds when snapshots are compressed to less than twenty bytes. Visual homing only works from a small region surrounding the snapshot, therefore odometry is used to travel longer distances between snapshots. The proposed route following technique is tested in simulation and on a Parrot AR.Drone 2.0. The drone can successfully follow long routes with a map that consumes only 17.5 bytes per meter.<br},

note = {McGuire, K.N. (mentor); de Croon, G.C.H.E. (mentor); Campoy Cervera, P. (mentor); Jonker, P.P. (mentor); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:6e3ce742-a974-491d-97c2-1cafc090b3d9,

title = {Vision-based stabilization of micro quadrotors},

author = {T. I. Braber},

url = {http://resolver.tudelft.nl/uuid:6e3ce742-a974-491d-97c2-1cafc090b3d9},

year  = {2017},

date = {2017-01-01},

school = {TU Delft Mechanical, Maritime and Materials Engineering; TU Delft Delft Center for Systems and Control},

abstract = {On-board stabilization of quadrotors is often done using an Inertial Measurement Unit (IMU), aided by additional sensors to combat the IMU drift. For example, GPS readings can aid when flying outdoors, or when flying in GPS denied environments, such as indoors, visual information from one or more camera modules can be used. <br/>A single downwards facing camera however cannot determine the absolute height of the quadrotor, leaving the results from the Optical Flow (OF) up to scale. To estimate the velocity of the quadrotor an additional range sensor, such as an Ultrasonic Sensor (US), is used to solve this scaling problem.<br/>These solutions are difficult to scale down to micro quadrotors as the platform becomes too small to fit and lift additional sensors. Therefore stabilizing a quadrotor with a single camera and IMU only would pave the way for the development of even smaller quadrotors. This master thesis presents an adaptive control strategy to stabilize a micro quadrotor in all<br/>three axes using only an IMU and a monocular camera. This is achieved by extending the stability based approach for a single, vertical, axis by De Croon in Distance estimation with efference copies and optical flow maneuvers: a stability-based strategy[1]. This stability based method ncreases the control gain in the visual feedback loop until the quadrotor detects it is oscillating by detecting that the covariance of the given thrust inputs and the measured divergence passes a threshold. Next the height can be estimated using the predetermined relationship between gain and height at which these self-induced oscillations occur and proper gains can be set for the estimated height. <br/>An analysis is done in simulation to present proof of concept of the stabilization method in three axis and to determine the effects of scaling and the effects of varying effective Frames per Second (FPS) caused by computations. It was shown that the adaptive gain strategy can stabilize the simulated quadrotor and prevent it from drifting. Furthermore, the control gains were scaled such that the effects of scaling a quadrotor could be mostly negated, though at about a tenth of the scale the simulated noise had such an influence that the scaled gains could not negate it anymore. Furthermore, the minimum effective FPS required to stabilize an ARDrone 2 was determined to be 15 FPS, and it was shown that an increase in effective FPS aids stabilizing the smaller scale quadrotors that became unstable due to the scaling effects.<br/>Furthermore, flights on an Parrot ARDrone 2 and Parrot Bebop are performed to show the usability of this control strategy in real life. It was shown that both quadrotors could achieve stable hover without drifting at multiple heights, using various strategies.<br},

note = {Babuska, Robert (mentor); de Croon, Guido (mentor); de Wagter, Christophe (mentor); de Bruin, Tim (graduation committee); Bregman, Sander (graduation committee); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:d00ade0b-7350-4d28-baf7-55d56a185032,

title = {Human Control Performance in Solving Multi-UAV Dynamic Vehicle Routing Problems Using an Ecological Interface},

author = {Niek Klein Koerkamp},

url = {http://resolver.tudelft.nl/uuid:d00ade0b-7350-4d28-baf7-55d56a185032},

year  = {2017},

date = {2017-01-01},

school = {TU Delft Aerospace Engineering; TU Delft Control & Simulation},

abstract = {Real-time optimization of Vehicle Routing Problems during mission operations raises concerns regarding reliability of obtaining a solution and solution time. Improvements in control performance by having a human-in-the-loop might be possible by leveraging human visual pattern recognition qualities. By developing an ecological interface, supporting the operator in controlling multiple Unmanned Aerial Vehicles in a simulated payload delivery mission, and by conducting a human-in-the-loop experiment, interface effectiveness and human control performance in Dynamic Vehicle Routing Problems was investigated. Results show the ecological interface offers good support and scales well with problem size. Results also show participants can in some cases achieve solutions faster and more reliably compared to an optimization algorithm, although generally yielding less efficient solutions. Having a human-in-the-loop can thus offer improved control performance over relying on pure automation, especially in time critical situations.},

note = {Borst, C. (mentor); de Croon, G.C.H.E. (graduation committee); van Paassen, M.M. (graduation committee); Mulder, Max (graduation committee); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:015f322f-9d86-4717-b2e2-74cf25bfa70c,

title = {Stability-based Scale Estimation of Monocular SLAM for Autonomous Quadrotor Navigation},

author = {Seong Hun Lee},

url = {http://resolver.tudelft.nl/uuid:015f322f-9d86-4717-b2e2-74cf25bfa70c},

year  = {2017},

date = {2017-01-01},

school = {TU Delft Aerospace Engineering},

abstract = {We propose a novel method to deal with the scale ambiguity in monocular SLAM based on control stability. We analytically show that (1) using unscaled state feedback from monocular SLAM for control can lead to system instability, and (2) there is a unique linear relationship between the absolute scale of the SLAM system and the control gain at which instability arises. Using this property, our method estimates the scale by adapting the gain and detecting self-induced oscillations. Unlike conventional monocular approaches, no additional metric sensors are used for scale estimation. We demonstrate the ability of our system to estimate the scale for performing autonomous indoor navigation with a low-cost quadrotor MAV.},

note = {de Croon, G.C.H.E. (mentor); Hoekstra, J.M. (graduation committee); Kooij, J.F.P. (graduation committee); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:0333e720-8105-4429-abcd-430c0d04f031,

title = {Thermal Modelling and Thermal Control Optimisation of the mN-μHEMPT},

author = {Nacho Granero Moneva},

url = {http://resolver.tudelft.nl/uuid:0333e720-8105-4429-abcd-430c0d04f031},

year  = {2017},

date = {2017-01-01},

school = {TU Delft Aerospace Engineering; TU Delft Space Engineering},

abstract = {Airbus Friedrichshafen is working on the development of a milliNewton HEMPT (High Efficiency<br/>Multistage Plasma Thruster): an electrostatic thruster concept suitable for small satellite<br/>propulsion. An engineering model, the mN-μHEMPT, has been built and tested in vacuum,<br/>generating thrust levels in the range of 1 to 5 mN. Although the working principle is understood,<br/>there is still uncertainty in the loss process, in particular the heat transfer in the plasma-wall<br/>interaction. An efficient heat management is crucial for the operation of the thruster, as the<br/>performance of the magnets is severely hindered after reaching 250ºC. With this in mind, the<br/>present thesis aims to produce the first thermal model of the mN-μHEMPT, with which a detailed<br/>thermal analysis can be carried out. The model validation strategy, based on correlation<br/>to testing results, makes it possible to overcome the uncertainty regarding the thermal losses.<br/>By simulating the operation of the thruster in extreme load cases in a Low Earth Orbit, its<br/>thermal performance is assessed, resulting in a detailed understanding of the temperature<br/>evolution and heat propagation through the different components. This information is then<br/>used to improve the performance by implementing design modifications. The result of the<br/>thesis is a thermal model validated to within 1.65ºC as mean deviation, predicting a maximum<br/>temperature of 180ºC at the magnet stack during operation. The application of a boron nitride<br/>coating to the radiator and the decoupling of the heat losses at the magnet stack and at the<br/>anode thanks to a second radiator, results in a maximum temperature of the magnet stack<br/>of 85ºC. In conclusion, the thermal performance of the mN-μHEMPT is analysed for the first<br/>time, and the design modifications proposed become a successful improvement.},

note = {Cervone, A. (mentor); Hey, Franz Georg (mentor); Zandbergen, B.T.C. (graduation committee); de Croon, G.C.H.E. (graduation committee); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:de84d7b7-52f7-4c0f-a00f-d7a6c244a678,

title = {Enhanced Sparse Depth Reconstruction Using Edge and Temporal Information: An Application to Micro Air Vehicles},

author = {Tobias Heil},

url = {http://resolver.tudelft.nl/uuid:de84d7b7-52f7-4c0f-a00f-d7a6c244a678},

year  = {2017},

date = {2017-01-01},

school = {TU Delft Aerospace Engineering; TU Delft Control & Operations},

abstract = {The reconstruction of dense depth maps is of great value to resource-constrained Mirco Air Vehicles (MAVs), in the pursuit of achieving autonomous flight with a high situational awareness. Most MAVs implement sensing methods which provide a sparse depth map, limiting their capabilities significantly. This article introduces two novel methods to enhance existing depth reconstruction algorithms in terms of geometric reconstruction, depth approximation and computational time. The first contribution is the introduction of a novel method that includes edge information from the image-domain into the depth-regularization problem. This to enhance the retrieval of the complete scene geometry. The second contribution is a novel scheme which includes temporal information in the reconstruction approach, allowing extremely sparse depth scenes to be reconstructed. By estimating the geometric transformation with optical flow, previous depth reconstructions can be used as initial solutions for the current depth-regularization problem. Empirical results show a consistent reduction reconstruction error, while at the same time reducing the computational time. Qualitative estimation shows significant improvement in the retrieval of scene geometry.},

note = {de Croon, G.C.H.E. (mentor); Gao, Zhi (graduation committee); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:ec5df737-cad0-4b24-89a0-75d45ebac51b,

title = {Velocity Templates for Dense Swarms of Flying Robots},

author = {Wilco Vlenterie},

url = {http://resolver.tudelft.nl/uuid:ec5df737-cad0-4b24-89a0-75d45ebac51b},

year  = {2017},

date = {2017-01-01},

school = {TU Delft Aerospace Engineering; TU Delft Control & Operations},

abstract = {In the near future many tasks could be performed by swarms of flying robots. To successfully implement multiple of these swarms in the same airspace they will have to be decentralised, autonomously cope with high densities and even resolve conflicting objectives of other swarms, while remaining controllable by operators through high-level objectives. This article introduces a novel swarming approach dubbed "Velocity Templates" based on artificial potential fields. These global fields represent the objectives of the swarm, which are balanced with local interaction. Different fields are considered leading to still or sustained motion swarms where conflicting objectives between sub-groups or multiple swarms are gracefully resolved. The approach is implemented on groups of 2 and 4 Parrot Bebop UAVs, using an efficient on-board vision algorithm to locate neighbours and a motion tracking system for guidance. The experiments show promising results for further outdoor tests assessing the scalability of the proposed approach.},

note = {Chu, Q. P. (mentor); de Croon, G.C.H.E. (mentor); Remes, B.D.W. (mentor); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:edbb8630-d1ad-4230-b4cd-f593e81622b2,

title = {Differential Dynamic Programming for Aerial Robots using an Aerodynamics Model},

author = {Nicolás Omar Abuter Grebe},

url = {http://resolver.tudelft.nl/uuid:edbb8630-d1ad-4230-b4cd-f593e81622b2},

year  = {2017},

date = {2017-01-01},

school = {TU Delft Aerospace Engineering; TU Delft Control & Simulation},

abstract = {State of the art trajectory generation schemes for quadrotors assume a simple dynamic model. They neglect aerodynamic effects such as induced drag and blade flapping and assume that no wind is present. In order to overcome this limitation, this thesis investigates a trajectory optimization scheme based upon Differential Dynamic Programming (DDP). There are various software-implementations of the DDP scheme. For future deployment on robotic hardware the software is required to be computationally efficient, written in C++ and to be open-source. A library named GCOP, which was developed at the John Hopkins University, fulfills these requirements and is used. Before implementing the solver, a full model of the Crazyflie Nano Quadcopter is identified experimentally. The solver is validated, normalized and the performance is benchmarked. This method yields reliable minimum control-effort trajectories. A control scheme is proposed and studied in Monte-Carlo simulations. Itis robust and able to handle large modelling errors in mass and moment of inertia while ensuring minimal error on the final state.},

note = {de Wagter, C. (mentor); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:18dee61c-9828-430a-9d71-5a12586da89c,

title = {Studying the effect of the tail on the dynamics of a flapping-wing MAV},

author = {F. G. J. Rijks},

url = {http://resolver.tudelft.nl/uuid:18dee61c-9828-430a-9d71-5a12586da89c},

year  = {2017},

date = {2017-01-01},

school = {Delft University of Technology},

abstract = {The effects of horizontal tail geometry and position on longitudinal flapping-wing micro aerial vehicle dynamics were studied using wind tunnel and free-flight experiments. Linearised models were used to analyse the effect on the dynamic properties of the ornithopter. Results show higher steady-state velocity and increased pitch damping for increased tail surface area and aspect ratio. The maximum span width of the tail surface is also found to play an important role in determining dynamic behaviour, in particular when the distance between the tail surface and the flapping wings is large. Steady-state conditions can be predicted accurately using linear functions of tail geometry for this ornithopter. Predicting dynamic behaviour is more complicated and requires further study. However, the observed trends in some of the model parameters suggest that future models explicitly including the tail geometry may be used to design flapping-wing robots with desirable dynamic properties.},

note = {De Visser, C.C. (mentor); Karásek, M. (mentor); Armanini, S.F. (mentor)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:ffa1ec41-3930-4dfe-b454-e11c3517a7f4,

title = {Vertical Landing for Micro Air Vehicles using Event-Based Optical Flow},

author = {B. J. Pijnacker Hordijk},

url = {http://resolver.tudelft.nl/uuid:ffa1ec41-3930-4dfe-b454-e11c3517a7f4},

year  = {2016},

date = {2016-01-01},

school = {Delft University of Technology},

abstract = {Small flying robots can perform landing maneuvers using bio-inspired optical flow by maintaining a constant divergence. However, optical flow is typically estimated from frame sequences recorded by standard miniature cameras. This requires processing full images on-board, limiting the update rate of divergence measurements, thus the speed of the control loop and the robot. Event-based cameras overcome these limitations by only measuring pixel-level brightness changes at microsecond temporal accuracy, hence providing an efficient mechanism for optical flow estimation. This thesis presents, to the best of our knowledge, the first research integrating event-based optical flow estimation into the control loop of a flying robot. We extend an existing 'local plane fitting' algorithm to obtain an improved and more computationally efficient optical flow estimation method, valid for a wide range of optical flow velocities. This method is validated for real event sequences. In addition, a method for estimating the divergence from event-based optical flow is introduced, which accounts for the aperture problem. The developed algorithms are implemented in a constant divergence landing controller on-board of a quadrotor. Flight tests demonstrate that, using event-based optical flow, accurate divergence estimates can be obtained over a wide range of speeds. This enables the quadrotor to perform very fast landing maneuvers.},

note = {de Croon, G.C.H.E. (mentor)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:8efab9c5-e78b-40ff-ab37-a563366d22f9,

title = {The Exploring DelFly: How to increase the indoor explored area of the DelFly Explorer by means of computationally efficient routing decisions?},

author = {C. R. Fonville},

url = {http://resolver.tudelft.nl/uuid:8efab9c5-e78b-40ff-ab37-a563366d22f9},

year  = {2016},

date = {2016-01-01},

school = {Delft University of Technology},

abstract = {Small robots, such as Micro Aerial Vehicles, form an increasingly popular field of interests in research, industry and the consumer market. The autonomous capabilities of these systems keep evolving and one of the main research goals is to reach full autonomy. However, this is often achieved at the cost of growing hardware demands. In this study a computationally light and efficient way to enhance autonomous on-board exploration capabilities for the DelFly Explorer, a 20-gram flapping wing Micro Aerial Vehicle (FWMAV), is presented. Both theory and new insights were combined to design an exploration algorithm for the on-board stereo-vision system. The algorithm primarily consists of a disparity map based decision tree, different exploration phases and computationally light odometry. Computer simulations proved the effectiveness of the algorithm to enable autonomous exploration capabilities for the FWMAV system. Initial flight tests also show that the proposed algorithm increases its exploration capabilities and form a foundation for future research.},

note = {de Croon, G.C.H.E. (mentor)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{Pijnacker2016,

title = {Vertical Landing for Micro Air Vehicles using Event-Based Optical Flow},

author = {B J Pijnacker Hordijk},

url = {http://resolver.tudelft.nl/uuid:ffa1ec41-3930-4dfe-b454-e11c3517a7f4},

year  = {2016},

date = {2016-01-01},

address = {Delft, NL},

school = {Delft University of Technology},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:b76bd35d-9d56-472e-8ff8-35fd453b6a49,

title = {Incremental Nonlinear Dynamic Inversion and Multihole Pressure Probes for Disturbance Rejection Control of Fixed-Wing Micro Air Vehicles},

author = {E. S. Van der Sman},

url = {http://resolver.tudelft.nl/uuid:b76bd35d-9d56-472e-8ff8-35fd453b6a49},

year  = {2016},

date = {2016-01-01},

school = {Delft University of Technology},

abstract = {Maintaining stable flight during high turbulence intensities is challenging for fixed-wing micro air vehicles. Two methods have been identified to improve the disturbance rejection performance of the MAV: incremental nonlinear dynamic inversion and phase-advanced pitch probes. Incremental nonlinear dynamic inversion uses the angular acceleration measurements to counteract disturbances. Multihole pressure probes measure the incoming flow angle and velocity ahead of the wing in order to react to gusts before an inertial response has occurred. The performance of incremental nonlinear dynamic inversion is compared to a traditional proportional integral derivative controller with and without the multihole pressure probes. The attitude controllers are tested by performing autonomous wind tunnel flights and stability augmented outdoor flights. This thesis shows that nonlinear dynamic inversion improves the disturbance rejection performance of fixed-wing MAVs compared to traditional proportional integral derivative controllers.},

note = {Chu, Q.P. (mentor); Remes, B. (mentor); Smeur, E.J.J. (mentor)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:55f9ab7a-2651-4a90-93a0-a8c9ddc7c6a9,

title = {Self-Supervised Monocular Distance Learning on a Lightweight Micro Air Vehicle},

author = {K. Lamers},

url = {http://resolver.tudelft.nl/uuid:55f9ab7a-2651-4a90-93a0-a8c9ddc7c6a9},

year  = {2016},

date = {2016-01-01},

school = {Delft University of Technology},

abstract = {This thesis presents all the work performed in developing a novel method for estimating distances on a flapping wing micro air vehicle using a monocular camera. These distance estimates are useful for providing a way to avoid collisions while flying indoors. The proposed method is based on a self-supervised learning algorithm that uses a short range impact detector to learn camera based long range distance estimates. The first part of this thesis contains an extended version of the paper on this topic as was submitted to the 2016 International Conference on Intelligent Robots and Systems (IROS). The second part contains the preliminary thesis that was preparatory to the final work and gives an in-depth overview of the state-of-the-art of different aspects of the problem as found in literature.},

note = {Hoekstra, J.M. (mentor); De Croon, G.C.H.E. (mentor); Tijmons, S. (mentor); Guo, J. (mentor)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{Fonville2016,

title = {The Exploring DelFly: How to increase the indoor explored area of the DelFly Explorer by means of computationally efficient routing decisions?},

author = {C R Fonville},

url = {http://resolver.tudelft.nl/uuid:8efab9c5-e78b-40ff-ab37-a563366d22f9},

year  = {2016},

date = {2016-01-01},

address = {Delft, NL},

school = {Delft University of Technology},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:424ead9b-50be-4e80-94a9-d041a1418dd3,

title = {The influence of a bottom camera in indoor ground-segmentation based obstacle avoiding performance for MAVs},

author = {M. W. M. Kuijpers},

url = {http://resolver.tudelft.nl/uuid:424ead9b-50be-4e80-94a9-d041a1418dd3},

year  = {2016},

date = {2016-01-01},

school = {Delft University of Technology},

note = {de Wagter, C. (mentor); de Croon, G.C.H.E. (mentor)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{Lamers2016b,

title = {Self-Supervised Monocular Distance Learning on a Lightweight Micro Air Vehicle},

author = {K Lamers},

url = {http://resolver.tudelft.nl/uuid:55f9ab7a-2651-4a90-93a0-a8c9ddc7c6a9},

year  = {2016},

date = {2016-01-01},

address = {Delft, NL},

school = {Delft University of Technology},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:3704b044-b9bf-454a-8678-0d140bd1d308,

title = {Actuator Saturation Handling using Weighted Optimal Control Allocation Applied to an INDI Controlled Quadcopter},

author = {D. C. Höppener},

url = {http://resolver.tudelft.nl/uuid:3704b044-b9bf-454a-8678-0d140bd1d308},

year  = {2016},

date = {2016-01-01},

school = {Delft University of Technology},

abstract = {Incremental Nonlinear Dynamic Inversion provides a high performance attitude controller for multi-rotor Micro Aerial Vehicles by providing very good disturbance rejection capabilities. Flights conducted with a quadcopter revealed undesired pitch and rolling motions which occurred simultaneously with actuator saturation for instantaneous yaw angle reference tracking commands. Constrained control allocation methods can increase the system's performance by providing an effective strategy to prioritize control objectives, and redistribute control effort accordingly. Weighted Least Squares control allocation makes the constrained control allocation problem a quadratic optimization problem. An iterative solver based on the computationally efficient active-set algorithm finds the optimal control distribution for a weighted control objective. In this paper the Weighted Least Squares control allocator is used to overcome two challenges 1) increase performance by applying prioritization between control objectives and redistribute control effort accordingly, accounting for the actuator limits 2) enable flight when flying with severely compromised actuator(s). Real-world flight experiments are performed and show a significant increase in performance for high load yaw maneuvers, and enabled a quadcopter to perform controlled flight with a severely compromised actuator},

note = {de Wagter, C. (mentor)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:382dec56-7789-40df-af28-f2e61de99fad,

title = {Attitude control- and stabilisation moment generation of the DelFly using Wing Tension Modulation},

author = {R. M. J. Janssen},

url = {http://resolver.tudelft.nl/uuid:382dec56-7789-40df-af28-f2e61de99fad},

year  = {2016},

date = {2016-01-01},

school = {Delft University of Technology},

note = {Karasek, M. (mentor)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:58a4c285-e3b6-4bf0-b885-2908077e9b02,

title = {Tracking and Following a Moving Person Onboard a Small Pocket Drone},

author = {T. Duro},

url = {http://resolver.tudelft.nl/uuid:58a4c285-e3b6-4bf0-b885-2908077e9b02},

year  = {2016},

date = {2016-01-01},

school = {Delft University of Technology},

abstract = {This paper presents a vision based strategy, designed to work fully onboard a small pocket drone, for autonomously tracking and following a person. Flying a drone is not an easy task, usually requiring a trained pilot, with the presented system it is possible to use a drone for filming or taking pictures from previously inaccessible places without the need for a person controlling the aircraft. Such framework is comprised by two main components, a tracker and a control system. The tracker has the function of estimating the position of the person that is being followed, while the control system gets the drone near that person. Limited by payload weight, power consumption and processing power the system results in a delicate balance between these constraints. The main contributions of this paper are the comparison between two state-of-the-art visual trackers running on paparazzi, Struck and KCF, as well as the control system that uses the tracker’s output location to perform the person following task. Then a new tracker is developed to be as computationally light as possible so that it can run onboard a small pocket drone, based on HOG feature extraction, it uses logistic regression to train a detector on the appearance of a person.},

note = {De Croon, G. (mentor); De Wagter, C (mentor); Meertens, R. (mentor)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:4b4c4e4b-5e45-4166-bd2c-f35a1e495c6a,

title = {Learning Depth from Single Monocular Images Using Stereo Supervisory Input},

author = {J. Paquim},

url = {http://resolver.tudelft.nl/uuid:4b4c4e4b-5e45-4166-bd2c-f35a1e495c6a},

year  = {2016},

date = {2016-01-01},

school = {Delft University of Technology},

abstract = {Stereo vision systems are often employed in robotics as a means for obstacle avoidance and navigation. These systems have inherent depth-sensing limitations, with significant problems in occluded and untextured regions, leading to sparse depth maps. We propose using a monocular depth estimation algorithm to tackle these problems, in a Self-Supervised Learning (SSL) framework. The algorithm learns online from the sparse depth map generated by a stereo vision system, producing a dense depth map. The algorithm is designed to be computationally efficient, for implementation onboard resource-constrained mobile robots and unmanned aerial vehicles. Within that context, it can be used to provide both reliability against a stereo camera failure, as well as more accurate depth perception, by filling in missing depth information, in occluded and low texture regions. This in turn allows the use of more efficient sparse stereo vision algorithms. We test the algorithm offline on a new, high resolution, stereo dataset, of scenes shot in indoor environments, and processed using both sparse and dense stereo matching algorithms. It is shown that the algorithm’s performance doesn’t deteriorate, and in fact sometimes improves, when learning only from sparse, high confidence regions rather than from the computationally expensive, dense, occlusion-filled and highly post-processed dense depth maps. This makes the approach very promising for self- supervised learning on autonomous robots.},

note = {de Croon, G.C.H.E. (mentor)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:e5604e9a-c241-4236-83dd-5fc823e7e284,

title = {Mission Planning for Sensor Network Deployment using a Fleet of Drones},

author = {P. Goyal},

url = {http://resolver.tudelft.nl/uuid:e5604e9a-c241-4236-83dd-5fc823e7e284},

year  = {2016},

date = {2016-01-01},

school = {Delft University of Technology},

abstract = {Various methods for route planning of on-road vehicles to serve transportation requests have been developed in the literature in order to reduce transportation and operational costs. The applicability and thus development of these methods is primarily motivated by the field of application. This article deals with the mission planning for a fleet of drones to deploy sensors in a network. In particular, they are conceived to complete the task of delivering geophones in the seismic surveys. Unlike conventional on-road vehicles used for delivery purposes, every drone in the fleet is constrained to make a frequent return trip back to the depot to pick-up a new payload and restore its battery. A centralized planner is proposed in this article due to this constraint. The problem of planning is decomposed into two phases: route formation and route scheduling. The first phase is handled using the extensive formulation of Multi-Trip Vehicle Routing Problem (MTVRP) aiming at minimizing the overall journey time. A heuristic method is also proposed for this phase which provides near-optimal solutions in a computationally efficient manner. The second phase of the planning algorithm deals with the unaddressed problem of depot congestion arising due to the frequent visits of each drone to the depot. This problem is expressed in the form of a Mixed-Integer Linear Program (MILP) that can be solved using available software. This phase is computationally intensive and comparatively slow which restricts the usage of this mission planner in the re-planning phase to the cases involving longer journeys with limited number of routes. The results from a flight-test are also presented in order to demonstrate the mission planner.},

note = {Hoekstra, J.M. (mentor); Blacquiere, G. (mentor); de Croon, G.C.H.E. (mentor); Smeur, E.J.J. (mentor)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}