Publications

@mastersthesis{uuid:bd671c3b-afc3-4215-a724-dd69512f4715,

title = {Adaptive dynamic incremental nonlinear control allocation: An actuator fault-tolerant control solution for high-performance aircraft},

author = {Noah Stam},

url = {http://resolver.tudelft.nl/uuid:bd671c3b-afc3-4215-a724-dd69512f4715},

year  = {2024},

date = {2024-01-01},

school = {TU Delft Aerospace Engineering},

abstract = {Neglecting actuator dynamics in nonlinear control and control allocation can lead to performance degradation, especially when considering fast dynamic systems. This thesis provides a novel method to account for actuator dynamics in the control allocation solution, dynamic incremental nonlinear control allocation, or D-INCA. The incremental approach allows for the implementation of a first order discrete-time actuator dynamics model in the quadratic programming (QP) solver. This model is used to find the optimal command inputs in addition to the desired physical actuator deflections, hereby compensating for actuator dynamics delays. Whereas, the baseline incremental nonlinear control allocation (INCA) approach requires pseudo-control hedging of the outer loop reference to increase closed loop stability margins under actuator dynamics delays. To its advantage, D-INCA does not require feedback of higher order output derivatives than INCA and can be used with nonlinear non-control affine systems. Furthermore, with adaptive D-INCA, or AD-INCA, an actuator dynamics parameter estimator is introduced to adapt the actuator model online, minimizing actuator tracking errors after actuator failures. The proposed methods are applied to a fighter aircraft model with an over-actuated innovative control effectors suite and results are compared to the baseline INCA controller.},

note = {de Visser, C.C. (mentor); Smeur, E.J.J. (graduation committee); Mooij, E. (graduation committee); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:b4f643b2-a64f-4fb4-a18f-5012364f7b0f,

title = {Optical Flow Determination using Neuromorphic Hardware with Integrate & Fire Neurons},

author = {Francesco Branca},

url = {http://resolver.tudelft.nl/uuid:b4f643b2-a64f-4fb4-a18f-5012364f7b0f},

year  = {2024},

date = {2024-01-01},

school = {TU Delft Aerospace Engineering},

abstract = {Spiking neural networks implemented for sensing and control of robots have the potential to achieve lower latency and power consumption by processing information sparsely and asynchronously. They have been used on neuromorphic devices to estimate optical flow for micro air vehicles navigation, however robotic implementations have been limited to hardware setups with sensing and processing as separate systems. This article investigates a new approach for training a spiking neural network for optical flow to be deployed on the speck2e device from Synsense. The method takes into account the restrictions of the speck2e in terms of network architecture, neuron model, and number of synaptic operations and it involves training a recurrent neural network with ReLU activation functions, which is subsequently converted into a spiking network. A system of weight rescaling is applied after conversion, to ensure optimal information flow between the layers. Our study shows that it is possible to estimate optical flow with Integrate-and-Fire neurons. However, currently, the optical flow estimation performance is still hampered by the number of synaptic operations. As a result, the network presented in this work is able to estimate optical flow in a range of [-4, 1] pixel/s.},

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

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:6a74fb80-425c-4366-8110-fecfb4a1a5fc,

title = {Novel Neuromorphic Hardware Inspired by the Olfactory Pathway Model of the \textit{Drosophila}: Leveraging bio-plausible computational primitives in digital circuits for spatio-temporal processing},

author = {Lyana Usa},

url = {http://resolver.tudelft.nl/uuid:6a74fb80-425c-4366-8110-fecfb4a1a5fc},

year  = {2024},

date = {2024-01-01},

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

abstract = {Olfactory learning in \textit{Drosophila }larvae exemplifies efficient neural processing in a small-scale network with minimal power consumption. This system enables larvae to anticipate important outcomes based on new and familiar odor stimuli, a process crucial for survival and adaptation. Central to this learning mechanism is the olfactory pathway model, which embodies the principles of synaptic plasticity and associative learning through prediction error coding mediated by specific neuromodulating neurons in the mushroom body, like dopaminergic neurons. There is a pressing need to develop novel computational frameworks that capture the spatio-temporal processes while remaining compatible with the constraints of small-scale neural networks. These frameworks should draw inspiration from the biophysical properties of neurons within the olfactory pathway model, enabling accurate emulation of neural dynamics and efficient learning processes using spiking neural networks. This thesis proposes a framework based on a phenomenological conductance-based leaky integrate-and-fire (COBALIF) neuron model, inspired by the olfactory pathway model of \textit{Drosophila} larvae. By first prototyping the spiking neural network in Intel's Lava Python-based framework, we validated the design on a neuron and system level for a neuromorphic hardware implementation. This was the foundation of a programmable, neuromorphic FPGA architecture capable of adaptive optimization, employed on a Zynq 7000 SoC FPGA. By implementing this architecture in a single-precision floating-point format, we model the real-time neural dynamics of the COBALIF neuron in one-tenth of a millisecond precision. Moreover, our FPGA implementation serves as a feasible prototype for deploying such biologically inspired neurons and their spatio-temporal dependencies in digital design, paving the way for scaling up to small-scale networks.},

note = {Frenkel, C. (mentor); Makinwa, K.A.A. (graduation committee); de Croon, G.C.H.E. (graduation committee); Nawrot, M. P. (graduation committee); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:491a7744-dc49-405c-ba38-5198b3e839a8,

title = {Towards a Robust Wireless Real-Time Ecological Monitoring System},

author = {Luke Waal},

url = {http://resolver.tudelft.nl/uuid:491a7744-dc49-405c-ba38-5198b3e839a8},

year  = {2024},

date = {2024-01-01},

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

abstract = {Climate change poses a serious threat to ecosystems and increases the need for accurate and rigorous monitoring of ecosystems. Current monitoring solutions are often bulky, expensive, and lack critical functionalities such as on-board inference capabilities, robust wireless connections, and a diverse sensor suite. Ecological monitoring projects often suffer from inefficiencies caused by the large time delays between collecting data and analyzing said data, as well as having to spend large amounts of time in the field setting up the sensors manually. This thesis addresses many of these issues by designing a sensor with an extensive sensor suite, robust wireless capabilities and an on-board audio classifier able to perform real-time inference. Furthermore, attention is paid to making the system extendable in the future and allow for potentially integrating the sensors with a drone delivery- and retrieval system. The system tests performed indicate that the system has great potential given more time to tweak some of its identified shortcomings.},

note = {Hamaza, S. (mentor); Rajan, R.T. (mentor); Smeur, E.J.J. (graduation committee); Hendriks, R.C. (graduation committee); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:3193131c-6b68-46a2-afe1-964a044dd6f9,

title = {Implications of Propeller-Wing Interactions on the Control of Aerodynamic-Surface-Free Tilt-Rotor Quad-Planes},

author = {Noah Wechtler},

url = {http://resolver.tudelft.nl/uuid:3193131c-6b68-46a2-afe1-964a044dd6f9},

year  = {2024},

date = {2024-01-01},

school = {TU Delft Aerospace Engineering},

abstract = {Quad-planes are a type of vehicle which combine the hovering capability of quadcopters and the forward flight efficiency of winged aircraft. Flight tests conducted on a dual-axis tilting-rotor quad-plane, designed to fly without aerodynamic surfaces, observed that the quad-plane suffered from insufficient roll authority during fast, forward flight. Subsequent wind tunnel testing confirmed a two- to fourfold reduction in roll moment generation from propellers mounted in front of the wing at similar levels of tilt as their rear counterparts, caused by propeller-wing interactions. To address the mismatch in actuator effectiveness shown by the wind tunnel experiment, the effect of the propeller-wing interactions was incorporated into the aero-propulsive model by means of a global polynomial, the structure of which was found using multivariate orthogonal function modelling. New flight tests demonstrated that, by including the propeller-wing interactions in the control allocation, the vehicle is capable of tracking a figure 8 maneuver without aerodynamic surfaces, and without compromising tracking performance.},

note = {Smeur, E.J.J. (mentor); Mancinelli, A. (graduation committee); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:eb522c6b-1b1d-4988-8a7a-e2846dc697c5,

title = {Self-Supervised Learning of Event-Based Optical Flow via Deep Equilibrium Models},

author = {Aleksandar Shokolarov},

url = {http://resolver.tudelft.nl/uuid:eb522c6b-1b1d-4988-8a7a-e2846dc697c5},

year  = {2024},

date = {2024-01-01},

school = {TU Delft Aerospace Engineering},

abstract = {The estimation of optical flow, which determines the movement of objects in a visual scene, is a crucial problem in computer vision. It is essential for applications such as autonomous navigation, where precise motion estimation is critical for performance and safety.<br/><br/>Frame-based cameras capture sequences of still images at regular intervals, from which optical flow is traditionally extracted using optimization-based or learning-based methods. Recently, event-based cameras, which detect changes in pixel brightness asynchronously, have gained traction due to their high temporal resolution and robustness to motion blur, and many algorithms have been developed to estimate optical flow from this data. IDNet is a learning-based approach that achieves state-of-the-art performance. However, IDNet and similar models face two major challenges: they require labeled ground-truth data for training, which is scarce and difficult to collect, and they rely on recurrent neural networks (RNNs) with a fixed number of refinement iterations. This fixed iteration scheme does not adapt to scene complexity, limiting accuracy for complex flows and increasing computational effort for simpler patterns.<br/><br/>The aim of this project is to explore, implement, and evaluate potential methods to address these two mentioned limitations and enhance the capabilities of models like IDNet.<br/><br/>To remove the need for ground-truth data, a self-supervised learning paradigm was implemented by introducing a novel contrast maximization loss that assesses the blur present when accumulating raw events for a certain time interval and compensating it with the predicted flow. To assess the effectiveness of this method, models were trained on the benchmark MVSEC dataset, showing improved results over previous methods with up to 15% on some sequences and an 8% improvement on average. Based on these experiments and results, further research directions were proposed.<br/><br/>As for the problem of the current fixed iteration scheme, Deep Equilibrium Models were found to provide a promising pathway to solving it. These novel models reformulate their iterative structure into a root-finding problem and utilize traditional solvers to find a solution based on some tolerance, providing a trade-off between speed and accuracy. Moreover, they allow for direct differentiation through the network using only their final estimate, compared to previous methods that keep track of their state through all iterations, leading to an O(1) memory consumption. Implementing these and some additional ideas, the trained DEQ IDNet model reached competitive performance on the DSEC dataset while consuming 15% less memory. Yet, further work is needed to close the gap and achieve state-of-the-art performance.},

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

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:7e9dd0aa-4d24-4100-a224-14e71f86cdda,

title = {IUVO: An Emergency Response Flyer},

author = {Gabriel Gervas Montoya},

url = {http://resolver.tudelft.nl/uuid:7e9dd0aa-4d24-4100-a224-14e71f86cdda},

year  = {2024},

date = {2024-01-01},

school = {TU Delft Aerospace Engineering; Grochowski, Bartłomiej},

note = {Smeur, E.J.J. (mentor); Varriale, Carmine (graduation committee); Georgopoulos, P. (graduation committee); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:8649d62f-6266-44a4-89de-5b5805d83ae5,

title = {Deep Learning Fusion of Monocular and Stereo Depth Maps Using Convolutional Neural Networks},

author = {Dani Tóth},

url = {http://resolver.tudelft.nl/uuid:8649d62f-6266-44a4-89de-5b5805d83ae5},

year  = {2024},

date = {2024-01-01},

school = {TU Delft Aerospace Engineering},

abstract = {This paper presents an encoder-decoder-style convolutional neural network (CNN) for the purpose of improving monocular and stereo depth estimation (SDE) estimates, by combining them with the corresponding monocular estimates through a fusion network, assisted by prior information to provide context for the fusion. Video cameras are commonly used for depth perception in robotics, especially weight-sensitive applications, such as on Micro Aerial Vehicles (MAV). The two primary paradigms for vision-based depth perception are monocular and stereo depth or disparity estimation, each having their own strengths and weaknesses. These strengths and weaknesses seem to be complementary, and thus a fusion of the two may result in more accurate predictions. In this paper, we investigate this fusion by training a CNN that combines stereo and monocular depth or disparity estimates. The fusion network is agnostic to the choice of the input networks, providing great flexibility. It was found that such a fusion network, while increasing the computational complexity of the depth perception pipeline, indeed improves the accuracy of the estimates. The number of outlier predictions has been significantly decreased, while also limiting some fundamental limitations of both stereo and monocular methods, such as errors arising from occluded regions.},

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

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:5aeb0475-b3d5-45a4-82c8-7b92fabbb683,

title = {Fault Tolerant Control in Over-Actuated Hybrid Tilt-Rotor Unmanned Aerial Vehicles},

author = {Nico Voß},

url = {http://resolver.tudelft.nl/uuid:5aeb0475-b3d5-45a4-82c8-7b92fabbb683},

year  = {2024},

date = {2024-01-01},

school = {TU Delft Aerospace Engineering},

abstract = {Quad-planes combine hovering and vertical takeoff and landing capability with fast and efficient forward flight. Regular Quad-planes with dedicated pusher motor can be subject to gust disturbances, and are not well-equipped to deal with actuator faults. Dual-axis Tilt-Rotor quad-planes are more maneuverable due to their overactuation. This also increases their gust resilience and allows them to hover statically after actuator failures. The vehicle in this paper uses an Incremental Nonlinear Dynamic Inversion (INDI ) controller, combined with a nonlinear Sequential Quadratic Programming (SQP) Control Allocation (CA ) algorithm, which can also find hover solutions in the case of actuator failures. We investigate both a combined allocation of linear and angular accelerations, as well as a cascaded allocation scheme. Due to large required changes in roll and pitch angles, the cascaded approach is selected in this research. Introduction of a tertiary control effort term, separation of attitude and actuator command optimization and a simulated Fault Detection and Identification ( FDI) mechanism led to repeated successful recovery from a motor failure in hover. Position tracking was demonstrated under failure in the recon- figured flight condition. Index Terms- Tilt-rotor, dual-axis tilt, quad-plane, FTC, over- actuated, control allocation},

note = {Smeur, E.J.J. (mentor); Mancinelli, A. (mentor); Bombelli, A. (graduation committee); de Visser, C.C. (graduation committee); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:9ab2b4ba-8f91-4891-8190-4a96f77c471e,

title = {Adaptive Visual Servoing Control for Quadrotors: A Bio-inspired Strategy Using Active Vision},

author = {Sander Hazelaar},

url = {http://resolver.tudelft.nl/uuid:9ab2b4ba-8f91-4891-8190-4a96f77c471e},

year  = {2024},

date = {2024-01-01},

school = {TU Delft Aerospace Engineering},

abstract = {New insights into the landing behavior of bumblebees show an adaptive strategy where the optical flow expansion of the landing target is step-wise regulated. In this article, the potential benefits of this approach are studied by replicating the landing experiment with a quadrotor. To this end, an open-loop switching method is developed, enabling fast steps in divergence. An adaptive control law is used to deal with non-linear system dynamics, where the control gain is scheduled based on the control effectiveness of the actuator inputs during the steps. It is demonstrated that the quadrotor can reliably land on the target from varying initial positions, and the switching strategy shows a slight reduction in landing time compared to a constant divergence strategy with the same average divergence over distance. This strategy also reduces the maximum velocity during the landing.},

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

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:9a295d44-1e95-4911-a4a2-4a96c498fe79,

title = {Design of an inherently fully dynamically balanced aerial manipulator with omnidirectional workspace},

author = {Alexander Bom},

url = {http://resolver.tudelft.nl/uuid:9a295d44-1e95-4911-a4a2-4a96c498fe79},

year  = {2024},

date = {2024-01-01},

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

abstract = {Drones are increasingly used nowadays, primarily for visual inspection tasks facilitated by onboard cameras. The field of aerial manipulation tries to expand the capabilities of drones by attaching a manipulator, enabling physical interaction. Unfortunately, the usability of aerial manipulators is hindered by disturbances resulting from the movements of the manipulator. These disturbances, including reaction forces and a shifting centre of mass, not only affect manipulation accuracy but also pose safety risks by potentially destabilizing the drone. In this thesis, a design is presented that addresses this challenge by leveraging the theory of dynamic balance. <br/>A new design approach of making a manipulator fly, instead of the common approach of mounting a manipulator arm to a drone was used. This new approach avoids interference with the drone's components, allowing to focus on the design of the manipulator arm. Furthermore, it made it possible to create a manipulator which can manipulate above, to the side and underneath itself. This makes the presented manipulator arm more versatile than common aerial manipulators whose workspace is mostly located only above or below the drone. The kinematics, workspace and balance conditions of the manipulator arm are presented. Furthermore, the design's workspace is optimised while the mass of the manipulator is minimized in a bilevel optimisation. Finally, the design is validated both by simulation and measurements performed with the built prototype.<br/>The design presented is the first inherently fully dynamically balanced manipulator with omnidirectional workspace which can be used for aerial manipulation.<br},

note = {van der Wijk, V. (mentor); Hamaza, S. (mentor); Herder, J.L. (graduation committee); Goosen, J.F.L. (graduation committee); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}

@mastersthesis{uuid:a164abc6-0103-4fa0-b7ac-c15bce2bce64,

title = {A lightweight quadrotor autonomy system: To navigate in densely cluttered forest environments},

author = {Andreas Zwanenburg},

url = {http://resolver.tudelft.nl/uuid:a164abc6-0103-4fa0-b7ac-c15bce2bce64},

year  = {2024},

date = {2024-01-01},

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

abstract = {These days, people see more and more applications for drones, including monitoring rainforests to protect plant and animal species. However, drones face challenges when navigating through the dense and cluttered vegetation of the forest. These environments necessitate advanced autonomous detection and navigation to make the drone traverse robustly and fly safely. In addition, the forest brings extra challenges, such as blocked signals for GPS localisation, remote control, and remote supervising.<br/><br/>In this thesis project, a drone is designed, built, and programmed to navigate autonomously in the rainforest with complete onboard computing and no GPS localisation. This 500-gram drone is being extensively tested and optimized in real forest conditions, and a dataset is being created from its autonomous flights to simulate various configurations of the path-planning algorithm. The results of these simulations on this dataset are then used for thorough research on how the algorithm can downscale to smaller systems and how this affects performance.<br/><br/>By using the results of this research on downscaling, a 100-gram drone is built and programmed to fly in forest conditions with complete onboard computation. Challenging on this small-size drone is the use of low-quality lightweight sensors and processor. The processor only weighs 10 grams, and the depth camera weighs 8 grams. Unique on this small drone is the 3D path planning fully computed onboard and the implementation of a new type of depth camera.},

note = {Wisse, M. (mentor); Hamaza, S. (graduation committee); Benders, D. (graduation committee); Delft University of Technology (degree granting institution)},

keywords = {},

pubstate = {published},

tppubtype = {mastersthesis}

}