• CN:11-2187/TH
  • ISSN:0577-6686

机械工程学报 ›› 2025, Vol. 61 ›› Issue (21): 152-167.doi: 10.3901/JME.2025.21.152

• 特邀专栏:纪念张启先院士诞辰 100 周年 • 上一篇    

扫码分享

面向空间多任务的自重构模块化机器人形态设计与性态分析

安小康1, 贾庆轩1, 陈钢1, 刘育强2, 刘华伟2   

  1. 1. 北京邮电大学智能工程与自动化学院 北京 100876;
    2. 北京空间飞行器总体设计部 北京 100094
  • 收稿日期:2025-03-31 修回日期:2025-07-22 发布日期:2025-12-27
  • 作者简介:安小康,男,1995年出生,博士研究生。主要研究方向为空间机器人、模块化机器人设计与控制。E-mail:xiaokangan@bupt.edu.cn
    贾庆轩,男,1964年出生,博士,教授,博士研究生导师。主要研究方向为机器人感知与控制、多机协同与特种机器人技术。E-mail:qingxuan@bupt.edu.cn
    陈钢(通信作者),男,1982年出生,博士,教授,博士研究生导师。主要研究方向为空间机器人、运动规划与控制。E-mail:buptcg@163.com
    刘育强,男,1985年出生,博士,研究员。主要研究方向为航天器总体技术。E-mail:echo33151223@sina.com.cn
    刘华伟,女,1987年出生,博士,高级工程师。主要研究方向为航天器总体技术。E-mail:lhw5646653@126.com
  • 基金资助:
    国家自然科学基金(62173044)和北京邮电大学优秀博士生创新基金(CX20243079)资助项目。

Morphological Design and Performance Analysis of a Self-reconfigurable Modular Robot for Multi-task Space Applications

AN Xiaokang1, JIA Qingxuan1, CHEN Gang1, LIU Yuqiang2, LIU Huawei2   

  1. 1. School of Intelligent Engineering and Automation, Beijing University of Posts and Telecommunication, Beijing 100876;
    2. Beijing Institute of Spacecraft System Engineering, Beijing 100094
  • Received:2025-03-31 Revised:2025-07-22 Published:2025-12-27

摘要: 模块化机器人的设计与应用是应对未来空间多任务的重要手段之一,而基础模块单元的形态又直接影响着模块化机器人任务适应性。面向从空间多任务需求到模块单元形态间的非线性映射难题,提出一种面向空间多任务的自重构模块化机器人形态设计与性态分析方法。首先,通过图论逆向映射并制定映射规则,构建了两类基础模块单元基础形态,引入体自由度以增强接口间姿态调节能力,并表征了形态参量。其次采用改进的自适应差分进化算法,以外形和体自由度夹角为优化变量,构建包含应力分布均匀性、运动灵活性及工作空间等性能的评价指标,优化评估并获得最优形态参数,即外观为球形、体自由度夹角为45°。然后分析两类模块单元的工作空间、运动灵活性及各向同性、外壳应力应变等性态在不同运动传递链路中的表达效应,证明了形态设计结果的优越性。最后,以桁架装配任务为例开展仿真分析,结果表明由模块单元组成的模块化机器人可自主重构为单臂到多分支等构型,可连续完成桁架抓取、搬运与装配等任务。该理论成果可为模块化机器人设计提供方案借鉴。

关键词: 自重构模块化机器人, 形态设计, 性态分析, 空间多任务, 桁架装配

Abstract: The design and application of modular robots represent a crucial approach to addressing future space multitasking challenges. The morphology of fundamental modular units directly influences the task adaptability of modular robots. To tackle the nonlinear mapping problem between space multitask requirements and modular unit morphology, this study proposes a morphology design and performance analysis method for self-reconfigurable modular robots tailored for space multitasking. First, through graph theory-based inverse mapping and the formulation of mapping rules, two types of basic modular unit morphologies are constructed. Body degrees of freedom (B-DOF) are introduced to enhance posture adjustment capability between interfaces, and morphological parameters are characterized. Next, an improved adaptive differential evolution algorithm is employed, with the unit shape and B-DOF angle as optimization variables. A performance evaluation index encompassing stress distribution uniformity, motion flexibility, and workspace is established to optimize and assess the morphology, yielding optimal parameters: a spherical exterior and a B-DOF angle of 45°. Subsequently, the performance expressions of the two modular units are analyzed, including workspace, motion flexibility, isotropy, and shell stress-strain distribution across different motion transmission chains, demonstrating the superiority of the proposed morphological design. Finally, a simulation of the entire truss assembly process is conducted as a multitask case study. The results show that the modular robot, composed of these units, can autonomously reconfigure into single-arm, dual-arm, and multi-branch configurations, successively accomplishing tasks such as truss grasping, transportation, and assembly. The theoretical findings provide valuable insights for the design of modular robots.

Key words: self-reconfiguring modular robot, morphological design, performance analysis, spatial multitasking, truss assembly

中图分类号: