An Investigation of the Factors Influencing Evolutionary Dynamics in the Joint Evolution of Robot Body and Control

TL;DR

An in-depth study of the factors that influence the evolution of robots that can be physically created rather than just simulated, in a rich morphological space that includes a voxel-based chassis, wheels, legs and sensors.

Abstract

In evolutionary robotics, jointly optimising the design and the controller of robots is a challenging task due to the huge complexity of the solution space formed by the possible combinations of body and controller. We focus on the evolution of robots that can be physically created rather than just simulated, in a rich morphological space that includes a voxel-based chassis, wheels, legs and sensors. On the one hand, this space offers a high degree of liberty in the range of robots that can be produced, while on the other hand introduces a complexity rarely dealt with in previous works relating to matching controllers to designs and in evolving closed-loop control. This is usually addressed by augmenting evolution with a learning algorithm to refine controllers. Although several frameworks exist, few have studied the role of the \textit{evolutionary dynamics} of the intertwined `evolution+learning' processes in realising high-performing robots. We conduct an in-depth study of the factors that influence these dynamics, specifically: synchronous vs asynchronous evolution; the mechanism for replacing parents with offspring, and rewarding goal-based fitness vs novelty via selection. Results show that asynchronicity combined with goal-based selection and a `replace worst' strategy results in the highest performance.

Figures (7)

• Figure 1: An example of evolved body-plan with a the components labelled. The components are attached on the surface of chassis with the white connectors and the head is attached in the centre of the chassis.
• Figure 2: The arena used for the exploration task.
• Figure 3: The average number of tiles visited by robots in the parent pool. The robot's index corresponds to the number of robots evaluated. The dots indicate when the parents' pool was updated.
• Figure 4: The morphological variance over the robot's index. The robot's index corresponds to the number of robots evaluated. The morphological variance is computed based on a descriptor of the body-plan design comprising: the number of wheels, legs, sensors, and the height, width and depth of the chassis. The higher the value, the higher the variance.
• Figure 5: The morphological and behavioural variance for each variant of the 20 best robots (selected according to their task performance). The morphological variance is computed based on a descriptor of the body-plan design comprising: the number of wheels, legs, sensors, and the height, width and depth of the chassis. The behavioural variance is computed based on the trajectory of the robots of their best controller.