Learning Task-Agnostic Action Spaces for Movement Optimization

Amin Babadi, Michiel Van de Panne, Caren Liu, Perttu Hamalainen

Research output: Contribution to journalArticleScientificpeer-review

2 Citations (Scopus)
46 Downloads (Pure)

Abstract

We propose a novel method for exploring the dynamics of physically based animated characters, and learning a task-agnostic action space that makes movement optimization easier. Like several previous papers, we parameterize actions as target states, and learn a short-horizon goal-conditioned low-level control policy that drives the agent's state towards the targets. Our novel contribution is that with our exploration data, we are able to learn the low-level policy in a generic manner and without any reference movement data. Trained once for each agent or simulation environment, the policy improves the efficiency of optimizing both trajectories and high-level policies across multiple tasks and optimization algorithms. We also contribute novel visualizations that show how using target states as actions makes optimized trajectories more robust to disturbances; this manifests as wider optima that are easy to find. Due to its simplicity and generality, our proposed approach should provide a building block that can improve a large variety of movement optimization methods and applications.

Original languageEnglish
Pages (from-to)4700-4712
JournalIEEE Transactions on Visualization and Computer Graphics
Volume28
Issue number12
Early online date2021
DOIs
Publication statusPublished - Dec 2022
MoE publication typeA1 Journal article-refereed

Keywords

  • action space
  • Aerospace electronics
  • hierarchical reinforcement learning
  • movement optimization
  • Optimization
  • policy optimization
  • Reinforcement learning
  • Splines (mathematics)
  • Task analysis
  • Training
  • trajectory optimization
  • Trajectory optimization

Fingerprint

Dive into the research topics of 'Learning Task-Agnostic Action Spaces for Movement Optimization'. Together they form a unique fingerprint.

Cite this