Small robots with limited computational and sensory capabilities are becoming more commonplace. Designing effective behavior for small robotic platforms to complete complex tasks is a major challenge. A promising methodology to address this problem is found in Evolutionary Robotics (ER), in which a robot’s controller, and possibly its body, is optimized using Evolutionary Learning (EL) techniques. This approach satisfies given computational constraints, while often resulting in unexpected solutions that exploit sensory-motor coordination to achieve complex tasks.
After more than 20 years of research on ER there are still many open questions. Most issues surround the question of how ER is implemented on a real robotic platform resulting in effective robust behavior. Additionally, most implementation of ER to date have been to robots completing relatively simple tasks in a constrained environment.
Can Evolutionary Robotics be used to enable small, computationally limited flying robotic platforms to complete goal-driven tasks in complex environments?
Publications