Modular Robots

Lamarckian Evolution of Simulated Modular Robots

We study evolutionary robot systems where not only the robot brains but also the robot bodies are evolvable. Such systems need to include a learning period right after ‘birth' to acquire a controller that fits the newly created body. In this paper we …

Revolve: A Versatile Simulator for Online Robot Evolution

Developing robotic systems that can evolve in real-time and real-space is a long term objective with technological as well as algorithmic milestones on the road. Technological prerequisites include advanced 3D-printing, automated assembly, and robust …

Analysing the Relative Importance of Robot Brains and Bodies

The evolution of robots, when applied to both the morphologies and the controllers, is not only a means to obtain high-quality robot designs, but also a process that results in many body-brain-fitness data points. Inspired by this perspective, in …

Directed Locomotion for Modular Robots with Evolvable Morphologies

Morphologically evolving robot systems need to include a learning period right after ‘birth' to acquire a controller that fits the newly created body. In this paper, we investigate learning one skill in particular: walking in a given direction. To …

Morphological Attractors in Darwinian and Lamarckian Evolutionary Robot Systems

Morphological evolution in a robotic system produces novel robot bodies after each reproduction event. This implies the necessity for life-time learning so that newborn robots can acquire a controller that fits their body. Thus, we obtain a system …

Real-World Evolution of Robot Morphologies: A Proof of Concept

Evolutionary robotics using real hardware has been almost exclusively restricted to evolving robot controllers, but the technology for evolvable morphologies is advancing quickly. We discuss a proof-of-concept study to demonstrate real robots that …

Improving RL Power for On-Line Evolution of Gaits in Modular Robots

This paper addresses the problem of on-line gait learning in modular robots whose shape is not known in advance. The best algorithm for this problem known to us is a reinforcement learning method, called RL PoWER. In this study we revisit the …

Revolve

An open source software framework for robot evolution

The Robot Baby Project

The first implementation of the life cycle of self-reproducing robots