Comparing lifetime learning methods for morphologically evolving robots

March 08, 2022 · Entered Twilight · 🏛 GECCO Companion

"Last commit was 6.0 years ago (≥5 year threshold)"

Evidence collected by the PWNC Scanner

Repo contents: .circleci, .gitignore, .gitmodules, CMakeLists.txt, Dockerfile, README.md, RELEASE.md, cpprevolve, docker, docs, experiments, models, pyrevolve, requirements.txt, revolve.py, test_py, test_robots.py, thirdparty, tools, worlds

Authors Fuda van Diggelen, Eliseo Ferrante, A. E. Eiben arXiv ID 2203.03967 Category cs.RO: Robotics Cross-listed cs.NE Citations 10 Venue GECCO Companion Repository https://github.com/fudavd/revolve/tree/learning ⭐ 2 Last Checked 1 month ago

Abstract

Evolving morphologies and controllers of robots simultaneously leads to a problem: Even if the parents have well-matching bodies and brains, the stochastic recombination can break this match and cause a body-brain mismatch in their offspring. We argue that this can be mitigated by having newborn robots perform a learning process that optimizes their inherited brain quickly after birth. We compare three different algorithms for doing this. To this end, we consider three algorithmic properties, efficiency, efficacy, and the sensitivity to differences in the morphologies of the robots that run the learning process.