Learning-based Initialization of Trajectory Optimization for Path-following Problems of Redundant Manipulators

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Yoon, Minsung, Kang, Mincheul, Park, Daehyung, Yoon, Sung-Eui
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866918328508874752
author Yoon, Minsung
Kang, Mincheul
Park, Daehyung
Yoon, Sung-Eui
author_facet Yoon, Minsung
Kang, Mincheul
Park, Daehyung
Yoon, Sung-Eui
contents Trajectory optimization (TO) is an efficient tool to generate a redundant manipulator's joint trajectory following a 6-dimensional Cartesian path. The optimization performance largely depends on the quality of initial trajectories. However, the selection of a high-quality initial trajectory is non-trivial and requires a considerable time budget due to the extremely large space of the solution trajectories and the lack of prior knowledge about task constraints in configuration space. To alleviate the issue, we present a learning-based initial trajectory generation method that generates high-quality initial trajectories in a short time budget by adopting example-guided reinforcement learning. In addition, we suggest a null-space projected imitation reward to consider null-space constraints by efficiently learning kinematically feasible motion captured in expert demonstrations. Our statistical evaluation in simulation shows the improved optimality, efficiency, and applicability of TO when we plug in our method's output, compared with three other baselines. We also show the performance improvement and feasibility via real-world experiments with a seven-degree-of-freedom manipulator.
format Preprint
id arxiv_https___arxiv_org_abs_2602_03418
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Learning-based Initialization of Trajectory Optimization for Path-following Problems of Redundant Manipulators
Yoon, Minsung
Kang, Mincheul
Park, Daehyung
Yoon, Sung-Eui
Robotics
Trajectory optimization (TO) is an efficient tool to generate a redundant manipulator's joint trajectory following a 6-dimensional Cartesian path. The optimization performance largely depends on the quality of initial trajectories. However, the selection of a high-quality initial trajectory is non-trivial and requires a considerable time budget due to the extremely large space of the solution trajectories and the lack of prior knowledge about task constraints in configuration space. To alleviate the issue, we present a learning-based initial trajectory generation method that generates high-quality initial trajectories in a short time budget by adopting example-guided reinforcement learning. In addition, we suggest a null-space projected imitation reward to consider null-space constraints by efficiently learning kinematically feasible motion captured in expert demonstrations. Our statistical evaluation in simulation shows the improved optimality, efficiency, and applicability of TO when we plug in our method's output, compared with three other baselines. We also show the performance improvement and feasibility via real-world experiments with a seven-degree-of-freedom manipulator.
title Learning-based Initialization of Trajectory Optimization for Path-following Problems of Redundant Manipulators
topic Robotics
url https://arxiv.org/abs/2602.03418