English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Evolution sculpts both the body plans and nervous systems of agents togetherover time. In contrast, in AI and robotics, a robot's body plan is usuallydesigned by hand, and control policies are then optimized for that fixeddesign. The task of simultaneously co-optimizing the morphology and controllerof an embodied robot has remained a challenge. In psychology, the theory ofembodied cognition posits that behavior arises from a close coupling betweenbody plan and sensorimotor control, which suggests why co-optimizing these twosubsystems is so difficult: most evolutionary changes to morphology tend toadversely impact sensorimotor control, leading to an overall decrease inbehavioral performance. Here, we further examine this hypothesis anddemonstrate a technique for "morphological innovation protection", whichtemporarily reduces selection pressure on recently morphologically-changedindividuals, thus enabling evolution some time to "readapt" to the newmorphology with subsequent control policy mutations. We show the potential forthis method to avoid local optima and converge to similar highly fitmorphologies across widely varying initial conditions, while sustaining fitnessimprovements further into optimization. While this technique is admittedly onlythe first of many steps that must be taken to achieve scalable optimization ofembodied machines, we hope that theoretical insight into the cause ofevolutionary stagnation in current methods will help to enable the automationof robot design and behavioral training -- while simultaneously providing atestbed to investigate the theory of embodied cognition.

Scalable Co-Optimization of Morphology and Control in Embodied Machines