DK-RRT: Deep Koopman RRT for Collision-Aware Motion Planning of Space Manipulators in Dynamic Debris Environments

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Chen, Qi, Liu, Rui, Mo, Kangtong, Zhang, Boli, Yu, Dezhi
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866913927580876800
author Chen, Qi
Liu, Rui
Mo, Kangtong
Zhang, Boli
Yu, Dezhi
author_facet Chen, Qi
Liu, Rui
Mo, Kangtong
Zhang, Boli
Yu, Dezhi
contents Trajectory planning for robotic manipulators operating in dynamic orbital debris environments poses significant challenges due to complex obstacle movements and uncertainties. This paper presents Deep Koopman RRT (DK-RRT), an advanced collision-aware motion planning framework integrating deep learning with Koopman operator theory and Rapidly-exploring Random Trees (RRT). DK-RRT leverages deep neural networks to identify efficient nonlinear embeddings of debris dynamics, enhancing Koopman-based predictions and enabling accurate, proactive planning in real-time. By continuously refining predictive models through online sensor feedback, DK-RRT effectively navigates the manipulator through evolving obstacle fields. Simulation studies demonstrate DK-RRT's superior performance in terms of adaptability, robustness, and computational efficiency compared to traditional RRT and conventional Koopman-based planning, highlighting its potential for autonomous space manipulation tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2507_03878
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DK-RRT: Deep Koopman RRT for Collision-Aware Motion Planning of Space Manipulators in Dynamic Debris Environments
Chen, Qi
Liu, Rui
Mo, Kangtong
Zhang, Boli
Yu, Dezhi
Robotics
Trajectory planning for robotic manipulators operating in dynamic orbital debris environments poses significant challenges due to complex obstacle movements and uncertainties. This paper presents Deep Koopman RRT (DK-RRT), an advanced collision-aware motion planning framework integrating deep learning with Koopman operator theory and Rapidly-exploring Random Trees (RRT). DK-RRT leverages deep neural networks to identify efficient nonlinear embeddings of debris dynamics, enhancing Koopman-based predictions and enabling accurate, proactive planning in real-time. By continuously refining predictive models through online sensor feedback, DK-RRT effectively navigates the manipulator through evolving obstacle fields. Simulation studies demonstrate DK-RRT's superior performance in terms of adaptability, robustness, and computational efficiency compared to traditional RRT and conventional Koopman-based planning, highlighting its potential for autonomous space manipulation tasks.
title DK-RRT: Deep Koopman RRT for Collision-Aware Motion Planning of Space Manipulators in Dynamic Debris Environments
topic Robotics
url https://arxiv.org/abs/2507.03878