变负载双臂机器人阻抗自适应控制系统研究

发布时间：2018-08-27 06:42

【摘要】：人类双臂协作完成各种复杂的任务具有独特优势,如在变化的非结构环境下双手抓取物体进行协作安装或打磨：传统的单臂工业机器人在实际的应用中已经非常成熟,但无法完成类似的动作,模仿人类双臂协作的生物学机理,研发双臂协作机器人将是下一代机器人研究的方向和重点。在非结构环境下研究和解决协作机器人的变负载问题是解决双臂协作机器人实际应用的关键,具有重要的理论意义和实际应用价值。本文结合江苏省科技支撑重点项目“多功能冗余自由度双臂工业机器人开发及示范应用”(编号：BE2013003),对双臂协作机器人进行了深入研究,研制出冗余自由度双臂协作机器人系统,开展了变负载的双臂协作机器人双环自适应阻抗控制研究,设计了基于模型的非线性控制系统,通过系统仿真与实验,验证了提出的控制策略与算法。主要的研究内容和创新点如下：(1)提出了基于外部扰动观测的变负载模糊自适应滑模控制算法。设计了基于迭代算法的非线性观测器,精确估算变外力负载产生的扰动。引入自适应模糊逻辑算法,实时更新滑模切换增益,有效地消除滑模控制中的抖动问题。根据李亚普诺夫函数,设计滑模控制器控制率,保证了系统的稳定性。解决了由于变负载以及机器人动力学模型不确定性对机器人系统跟踪精度影响的问题。(2)提出了基于系统能耗最小的内力动态优化分配方法。通过建立双臂机器人抓取物体的运动学动力学模型,设计了抓取矩阵,分析了运动约束关系。采用最小范数法和零空间解法,将双臂夹持物体的合力分解为内力和外力,以系统的能量消耗最小为优化目标,采用能量代价函数对内力进行二次优化分配。解决了目标物体的合力分解到双臂机器人操作空间的力映射问题,为下章的力控制提供理论模型。(3)提出了双环阻抗变刚度力跟踪控制策略。针对目标物体合力分解为内力和外力的方法,分别采用内环阻抗和外环阻抗的控制策略,外环阻抗对理想的外力进行跟踪控制,可实现物体与环境良好接触,稳定交互。以手臂的阻抗特性分析,提出了基于PD变刚度调整阻抗控制策略,提高了系统的鲁棒性。为了防止内力过大而损坏物体,设计了内环阻抗控制模型,通过李亚普诺夫函数证明全局控制系统的稳定性。解决了双臂协作抓取及物体与环境交换下力的稳定性控制的问题。(4)开发双臂协作机器人软硬件控制系统。针对双臂协作机器人不同任务作业的复杂控制算法,采用开源的Linux+RTAI系统,构建了视觉、力觉、电流检测等感知能力系统,开发出能快速计算和数据处理能力的总控系统。采用基于总线通讯模式,实现各关节控制节点与总控系统的信息传递速度与准确性的大幅度提高。(5)搭建了实验平台,对控制算法进行了验证。针对双臂协调变负载问题,开展了机械臂末端变负载实验、双臂机器人的变外力交互实验以及基于任务分解的双臂机器人轴孔装配实验；通过两7-DOF机械臂仿真平台和实验手段验证了提出的控制算法的有效性。
[Abstract]:Human double-arm cooperation has unique advantages in completing various complex tasks, such as cooperative installation or polishing of objects grasped by both hands in a changing unstructured environment: the traditional single-arm industrial robot has been very mature in practical applications, but it can not complete similar movements, imitate the biological mechanism of human double-arm cooperation, and develop double-arm cooperation. The research of arm cooperative robot will be the direction and focus of the next generation robot. It is the key to solve the variable load problem of the cooperative robot in unstructured environment. It has important theoretical significance and practical application value. The development and demonstration application of dual-arm industrial robot with redundant degree of freedom (No. BE2013003) are introduced. The dual-arm cooperative robot system with redundant degree of freedom is developed. The dual-loop adaptive impedance control of dual-arm cooperative robot with variable load is studied. A model-based nonlinear control system is designed. The main research contents and innovations are as follows: (1) A variable load fuzzy adaptive sliding mode control algorithm based on external disturbance observation is proposed. A nonlinear observer based on iterative algorithm is designed to accurately estimate the disturbance caused by variable external force load. Fuzzy logic algorithm updates the sliding mode switching gain in real time and effectively eliminates the jitter problem in sliding mode control. According to Lyapunov function, the control rate of sliding mode controller is designed to ensure the stability of the system. A dynamic optimal allocation method of internal force based on the minimum energy consumption of the system is proposed.The grabbing matrix is designed by establishing the kinematics dynamic model of the two-arm robot grabbing the object,and the motion constraints are analyzed.The resultant force of the two-arm clamping object is decomposed into internal force and external force by using the minimum norm method and the zero-space method. The problem of force mapping from the resultant force of the target object to the manipulation space of the dual-arm robot is solved, and a theoretical model is provided for the force control of the next chapter. (3) A dual-loop impedance variable stiffness force tracking control strategy is proposed. The method of internal force and external force respectively adopts the control strategy of inner loop impedance and outer loop impedance, and the outer loop impedance can track and control the ideal external force, which can realize good contact and stable interaction between object and environment. The inner loop impedance control model is designed to prove the stability of the global control system by Lyapunov function. The problems of cooperative grasping with two arms and the stability control of force exchange between object and environment are solved. (4) The hardware and software control system of double arm cooperative robot is developed. Complex control algorithm of task task task, using open source Linux+RTAI system, constructs perceptual system of vision, force sense, current detection and so on, develops the master control system which can calculate and process data quickly. Adopts bus communication mode, realizes the speed and accuracy of information transmission between each joint control node and master control system. (5) Set up the experimental platform and verify the control algorithm. Aiming at the problem of the coordinated variable load of the two arms, the experiment of variable load at the end of the manipulator, the experiment of variable external force interaction of the two-arm robot and the experiment of the assembly of the two-arm robot shaft hole based on task decomposition are carried out. The effectiveness of the proposed control algorithm is also discussed.
【学位授予单位】：中国科学技术大学
【学位级别】：博士
【学位授予年份】：2016
【分类号】：TP242

【参考文献】