Rapid humanoid motion learning through coordinated, parallel evolution

Marijn Stollenga, Jürgen Schmidhuber, Faustino Gomez

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Planning movements for humanoid robots is still a major challenge due to the very high degrees-of-freedom involved. Most humanoid control frameworks incorporate dynamical constraints related to a task that require detailed knowledge of the robot's dynamics, making them impractical as efficient planning. In previous work, we introduced a novel planning method that uses an inverse kinematics solver called Natural Gradient Inverse Kinematics (NGIK) to build task-relevant roadmaps (graphs in task space representing robot configurations that satisfy task constraints) by searching the configuration space via the Natural Evolution Strategies (NES) algorithm. The approach places minimal requirements on the constraints, allowing for complex planning in the task space. However, building a roadmap via NGIK is too slow for dynamic environments. In this paper, the approach is scaled-up to a fully-parallelized implementation where additional constraints coordinate the interaction between independent NES searches running on separate threads. Parallelization yields a 12× speedup that moves this promising planning method a major step closer to working in dynamic environments. © 2014 Springer International Publishing Switzerland.
Original languageEnglish (US)
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
PublisherSpringer Verlagservice@springer.de
Pages210-219
Number of pages10
ISBN (Print)9783319088631
DOIs
StatePublished - Jan 1 2014
Externally publishedYes

Bibliographical note

Generated from Scopus record by KAUST IRTS on 2022-09-14

Fingerprint

Dive into the research topics of 'Rapid humanoid motion learning through coordinated, parallel evolution'. Together they form a unique fingerprint.

Cite this