Differentiable Motion Manifold Primitives for Reactive Motion Generation under Kinodynamic Constraints

Fuente: arXiv
Saved in:
Bibliographic Details
Main Author: Lee, Yonghyeon
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908463040299008
author Lee, Yonghyeon
author_facet Lee, Yonghyeon
contents Real-time motion generation -- which is essential for achieving reactive and adaptive behavior -- under kinodynamic constraints for high-dimensional systems is a crucial yet challenging problem. We address this with a two-step approach: offline learning of a lower-dimensional trajectory manifold of task-relevant, constraint-satisfying trajectories, followed by rapid online search within this manifold. Extending the discrete-time Motion Manifold Primitives (MMP) framework, we propose Differentiable Motion Manifold Primitives (DMMP), a novel neural network architecture that encodes and generates continuous-time, differentiable trajectories, trained using data collected offline through trajectory optimizations, with a strategy that ensures constraint satisfaction -- absent in existing methods. Experiments on dynamic throwing with a 7-DoF robot arm demonstrate that DMMP outperforms prior methods in planning speed, task success, and constraint satisfaction.
format Preprint
id arxiv_https___arxiv_org_abs_2410_12193
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Differentiable Motion Manifold Primitives for Reactive Motion Generation under Kinodynamic Constraints
Lee, Yonghyeon
Robotics
Artificial Intelligence
Real-time motion generation -- which is essential for achieving reactive and adaptive behavior -- under kinodynamic constraints for high-dimensional systems is a crucial yet challenging problem. We address this with a two-step approach: offline learning of a lower-dimensional trajectory manifold of task-relevant, constraint-satisfying trajectories, followed by rapid online search within this manifold. Extending the discrete-time Motion Manifold Primitives (MMP) framework, we propose Differentiable Motion Manifold Primitives (DMMP), a novel neural network architecture that encodes and generates continuous-time, differentiable trajectories, trained using data collected offline through trajectory optimizations, with a strategy that ensures constraint satisfaction -- absent in existing methods. Experiments on dynamic throwing with a 7-DoF robot arm demonstrate that DMMP outperforms prior methods in planning speed, task success, and constraint satisfaction.
title Differentiable Motion Manifold Primitives for Reactive Motion Generation under Kinodynamic Constraints
topic Robotics
Artificial Intelligence
url https://arxiv.org/abs/2410.12193