Self-Reconfigurable Modular Robotic Manipulators for Space

本研究旨在开发新一代适应性强的机械手，该机械手将能够通过其模块臂和手的自重构来改变其形态。这涉及到开发一个具有可互换和可重构机器人柔顺手的臂关节对的模块化设计。深度学习技术将用于训练机器人机械手以半自主的方式承担日常任务。该设计的可行性将在模拟环境中进行测试，用于两个用例(i)大口径空间望远镜的轨道组装和维护（ii）小行星机器人采矿，其中操纵器可用于捕获小行星或更深地挖掘其他行星体。

This study aims at developing a new generation ofadaptable robotic manipulators that will be capable for changing its morphology through self-reconfiguration of its modulararm/s and hand/s. This involves developing a modular design for the link-joint pairs for the arm with an interchangeableand reconfigurable robotic compliant hand. Deep learning techniques will be used to train the robot manipulator toundertake routine tasks in a semi/autonomous manner. The feasibility of the design will be tested, in a simulationenvironment, for two use cases (i) for orbital assembly and maintenance of a large aperture space telescope (ii) roboticmining of asteroids where the manipulator could be used for capturing a small asteroid or for deeper excavation of otherplanetary bodies.