Multi-task continuous manipulation for mobile manipulator inspired by mechanism theory: a Path Kinematic Chain (PKC) method

DOI number:

10.1016/j.cjme.2026.100279

Journal:

Chinese Journal of Mechanical Engineering

Key Words:

Mobile manipulator; multi-task manipulation; workspace analysis

Abstract:

The capability of robots to consecutively execute multi-task manipulation is essential for task adaptability. The challenge is that the configuration required for one task may hinder the execution of the next, causing interruptions. This work brings the concept of kinematic chains from mechanism theory into robot manipulation and proposes a novel method for multi-task continuous execution. Owing to the degrees of freedom of the mobile base and torso, each task path is treated as a rigid link and connected with others in the robot’s root coordinate frame. Therefore, the interdependence among tasks can be modeled as a Path Kinematic Chain (PKC). Then, the PKC can adaptively move to feasible workspace regions to obtain valid configurations, i.e., base sequences for tasks. To construct such regions, a Point Family Map is designed to convert configuration feasibility evaluation into a geometric relationship judgment, enabling integrated trajectory generation. This method can accommodate diverse task paths in kinematics, manipulator structures, and motion forms of mobile base and torso. Benchmarks demonstrate up to 96% reliability, outperforming baseline approaches. Physical experiments on service tasks validate its applicability.

Co-author:

Sipu Ruan,Yazhe Luo

First Author:

Yifei Li

Indexed by:

Journal paper

Correspondence Author:

Yong Tao,Diansheng Chen

Document Type:

J

Translation or Not:

no

Date of Publication:

2026-03-27

Included Journals:

SCI

Links to published journals:

https://www.sciencedirect.com/science/article/pii/S1000934526000684