Analysing the Relative Importance of Robot Brains and Bodies

Abstract

The evolution of robots, when applied to both the morphologies and the controllers, is not only a means to obtain highquality robot designs, but also a process that results in many body-brain-fitness data points. Inspired by this perspective, in this paper we investigate the relative importance of robot bodies and brains for a good fitness. We introduce a method to isolate and quantify the effect of the bodies and brains on the quality of the robots and perform a case study. The method is general in that it is not restricted to evolutionary systems. For the case study, we use a system of modular robots, where the bodies are evolvable and the brains are evolvable and learnable. These case studies validate the usefulness of our method and deliver interesting insights into the interplay between bodies and brains in evolutionary robotics.

Morphological Attractors in Darwinian and Lamarckian Evolutionary Robot Systems

“…Recent research explores the influence of fitness functions on the outcome of evolution [21]. The basis for this research is the morphological descriptors defined in [22] along with [16].…”

“…Lamarckian regime) can provide a benefit to a newly-born robot [15]. The same set-up has been tested in another investigation that shows the greater influence of the body structure against the brain throughout the robot's lifetime [16]. In this paper, we investigate how does the inherited knowledge influences the evolutionary development over several generations.…”

The Limits of Lexicase Selection in an Evolutionary Robotics Task

“…Evolutionary robotics (ER) optimizes robotic systems by employing concepts from biological evolution in a digital system (Nolfi and Floreano, 2000). ER has been employed across robotic systems to optimize both body structure alone (Auerbach and Bongard, 2010;Cheney et al, 2013;Collins et al, 2018) or together with control (Jelisavcic et al, 2018;Kriegman et al, 2018). Evolved robots have been transferred to reality (Ruud et al, 2016) although the reality gap remains a persistent issue (Stanton, 2018;Koos et al, 2010;Jakobi, 1998).…”