Learning Locomotion Skills in Evolvable Robots
It addresses a key challenge in robotic reproduction for practical applications, though it is incremental as it builds on prior work in evolving robot systems.
The paper tackles the problem of generating controllers for newborn robots in physically evolving systems, introducing a method that enables modular robots of arbitrary shapes to learn targeted locomotion, achieving successful validation on three robots in real-world scenarios.
The challenge of robotic reproduction -- making of new robots by recombining two existing ones -- has been recently cracked and physically evolving robot systems have come within reach. Here we address the next big hurdle: producing an adequate brain for a newborn robot. In particular, we address the task of targeted locomotion which is arguably a fundamental skill in any practical implementation. We introduce a controller architecture and a generic learning method to allow a modular robot with an arbitrary shape to learn to walk towards a target and follow this target if it moves. Our approach is validated on three robots, a spider, a gecko, and their offspring, in three real-world scenarios.