The compliance of robot is required during the manufacturing processing and human-robot cooperation. Moreover, the excellent motion performances, such as fast response, small tracking error, are also necessary in order to improve the productivity. Compliance and excellent motion performance are conflicting requirement. The variable stiffness joint is the key to solving the conflicting requirement of robot. Aiming at the requirement, a novel variable stiffness joint with two actuators is proposed in the project. The joint is based on cam theory. The topological structure and the cam shape are optimized. The relationship between joint torque and joint deflection is built by theoretical derivation and test method. The hysteresis model of the compliant joint is studied. The dynamic model, in which the potential energy of the spring is considered, is built, and its parameters are identified. The inverse dynamic of compliant joint is solved. The project will also design the stiffness observer and torque observer of the compliant joint. Taking the single compliant joint and the compliant robot with three degree of freedoms as the object of the research, the full state feedback controllers are designed. The problems of gravity compensation of vibration due to the low damping are addressed in the project. The momentum observer of the joint or robot is deduced by theoretical method. The signal feature of the momentum is analyzed and extracted during the collision between the robot and the environment. The mechanism of collision of the compliant robot is built. The project will solve the problem of the inverse dynamic, the compensation of the hysteresis and optimal control in order to gain the fast positioning and the low tracking error. The project will provide the basis for the application of the compliant robot in the manufacturing, such as assembling, grinding, polishing and forging.
智能制造工艺和人-机协作安全要求机器人具有柔顺性,同时制造效率要求机器人具有优越的运动性能,具有变刚度特性的机器人关节是解决机器人制造时柔顺性与高性能运动矛盾要求的关键。针对以上需求,提出一种基于凸轮原理的双伺服驱动被动柔顺关节新构型,优化关节拓扑布局及凸轮轮廓形状;建立柔顺关节力矩-挠度之间的映射关系及迟滞模型,包含关节弹性势能的柔顺关节动力学建模及参数辨识,柔顺关节逆动力学求解;以单自由度柔顺关节和三自由度柔顺机器人为对象,设计柔顺关节刚度和力矩观测器,研究基于观测值的柔顺关节多输入多输出全状态反馈控制算法、在线重力补偿算法和主动阻尼控制算法;建立柔顺关节动量观测器,提取柔顺关节碰撞时动量变化特征,建立柔顺关节碰撞检测机制和反应机制。解决柔顺关节逆动力学求解、迟滞补偿、优化控制等关键问题以实现柔顺关节的快速准确定位和精确轨迹跟踪,为柔顺机器人在装配、磨抛、锻造等制造工艺应用提供条件。
在非结构化环境下操作,要求机器人具有柔顺性,而具有柔顺性能的机器人关节是确保机器人柔顺性的关键部件。针对以上需求及存在的科学问题,本项目提出基于凸轮结构的变刚度机器人柔性关节新构型,具有双伺服驱动、刚柔转换特性,并优化了关节参数;针对柔性关节机器人动力学辨识不精确的问题,提出了基于能量法的改进柔性关节动力学模型,引入关节柔性特征的辨识方法;提出了一种基于指数坐标法的柔性机器人运动学参数标定方法,验证了方法的可行性,稳定性更高;提出了基于sigmoid函数改进的时滞控制算法及在线重力补偿算法,解决了柔性关节控制的抖动问题,降低了关节稳态误差;针对柔性关节存在的迟滞现象,基于GPI模型对柔性关节进行了迟滞建模,利用退火算法求得最优模型参数;设计制造了多柔性关节实验平台,对柔性关节动力学建模、参数辨识方法、运动学标定方法、控制算法、迟滞模型进行了验证。研究成果可应用于人机协作、空间非合作目标抓取等场景;获得上海市科技进步一等奖,撰写英文论文稿3篇,发表国际会议论文1篇,中文论文3篇,申请发明专利2项,培养在读博士1人,毕业硕士3人,毕业本科生4人。
{{i.achievement_title}}
数据更新时间:2023-05-31
基于LASSO-SVMR模型城市生活需水量的预测
基于SSVEP 直接脑控机器人方向和速度研究
基于分形维数和支持向量机的串联电弧故障诊断方法
滚动直线导轨副静刚度试验装置设计
物联网中区块链技术的应用与挑战
并行双驱动变刚度系统的设计理论与控制研究
变物理阻尼柔顺关节机构设计与驱动控制研究
变刚度弹性驱动伸缩腿双足机器人动力学与控制研究
自适应地形的变刚度仿蛇搜救机器人机构设计及其柔顺控制研究