欠驱动连续型刚柔耦合并联指机构的综合方法与设计理论研究

As the intelligent equipment providing service for human beings in the unstructured environment, service robot has become a strategic emerging industry in our country. The dexterous grasping mechanism with underactuation plays an important role in improving the intelligence and enlarging the application of the service robot, since it is an essential component for the interaction between the service robot and the unstructured environment. Aiming at the academic frontier in robot mechanism and the requirements from the service robot, the design of a novel parallel finger mechanism with underactuated, continuous, rigid-flexible coupling features is implemented. First of all, structural innovation and topology synthesis are carried out. Then the motion/force mapping models between limited inputs and multiple outputs are formulated considering the mechanics characteristic of the flexible material, on the basis of which, the performance evaluation index is proposed for the optimal design. The parameterized optimization in terms of kinematic and static performance is investigated, as well as the assessment relating to the stability of the mechanism and the grasping operation. Moreover, the analysis of the driving force optimization in the force closed field and the material selections based on various grasping tasks are conducted. Finally, the prototype is built and the experimental analysis is performed. The proposed procedure lays a foundation for the innovation and design of the dexterous grasping mechanism with underactuation.

服务机器人，作为非结构环境下为人类提供服务的智能化装备，是我国着力培育的战略性新兴产业。欠驱动灵巧型抓持机构作为服务机器人与非结构环境交互的一类关键使能部件，对提高和扩展服务机器人的智能化程度和应用范围具有重要作用。面向机器人机构学国际学术前沿和服务机器人对欠驱动灵巧型抓持机构的重要需求，本项目拟研究一类新型欠驱动连续型刚柔耦合并联指机构的构型创新和综合方法，考虑柔性材料的力学特性，构建运动/力在机构输入端和多个输出端间的映射模型，提出可指导优化设计的性能评价指标，掌握计及运动学和静力学性能的参数优化设计理论，探索抓持操作稳定性和机构自身稳定性的评价方法以及满足力封闭域的驱动力优化策略，完成可实现不同抓持操作任务的材料选择，据此搭建原理样机并开展实验研究，为欠驱动灵巧型抓持机构的自主创新和工程设计奠定重要的理论与技术基础。

本项目面向机器人机构学国际学术前沿和服务机器人对欠驱动灵巧型抓持机构的重要需求，开展了一类新型欠驱动连续型刚柔耦合抓持机构以及刚-柔-软变刚度机械手的构型综合、性能设计、感知识别、仿生制造与综合实验等研究工作。融合柔性抓持机构高度适应性与刚性抓持机构稳定性好的优势，提出了一种柔顺与刚度兼顾的刚柔耦合抓持机构。受鳞甲结构启发，提出了一种基于软体机械手的鳞甲仿生变刚度结构，可显著提升软体机械手的负载能力。同时针对上述设计的软体机械手，提出了一种基于传感器数据反馈的自主抓持控制策略，并开发了一套自主抓持的控制系统，改善了软体机械手的抓持稳定性。借助视觉与触觉传感器，提出了一种基于机器学习的视触融合识别算法，可显著提升物体识别的效率与准确性。利用3D打印机技术，开展了软体机械手的制造方法研究，构建了“模具筑造+3D打印”的多材料一体化制造体系。最后，基于前期理论方面的研究成果，研制出刚柔耦合抓持机构、变刚度软体机械手的原型样机，并开展相应的抓持测试、感知识别等性能实验，实验结果充分验证了上述理论和方法的有效性，为欠驱动灵巧型抓持机构的自主创新和工程设计及进一步开发非结构化环境下刚柔耦合软体机械手奠定重要的理论与技术基础。项目组在IEEE TRO、TIE等期刊发表SCI论文11篇，国际会议论文2篇，申请国家发明专利7项，已授权4项；培养硕博士研究生8人；培养青年教师2人，其中项目负责人入选教育部青年长江学者和国家万人计划青年拔尖人才。