节点文献

面向人机协作的异构双臂机器人柔顺控制与同步示教研究

Compliance Control and Synchronous Demonstration of Heterogeneous Dual Arm Robot for Human-robot Collaboration

【作者】 陈涛

【导师】 徐文福;

【作者基本信息】 哈尔滨工业大学 , 机械工程, 2023, 硕士

【摘要】 由具有不同结构和功能的机器人组成的异构双臂机器人系统人可以充分发挥各自的优点、弥补各自的不足,实现1+1>2的操作能力,因而在工业、服务业、航空航天等领域具有重要的应用前景。然而,由于运动学、动力学方程变得更加复杂,响应特性存在较大差异,为整个系统的控制带来巨大挑战。本文以由绳驱机械臂及关节式协作臂组成的异构双臂机器人系统为对象,开展其复合体系统建模和协同柔顺控制方法的研究,并开展实验验证。根据典型任务场景和系统构型特点,建立了刚柔混合异构双臂机器人系统的运动学模型,并分析了双臂协同工作空间。采用D-H法分别推导了绳驱机械臂及关节式协作臂的正、逆运动学,进而根据松耦合和紧耦合两种不同状态推导了双臂协调下的运动学约束关系并进行求解。结合异构双臂的结构特点,基于包围盒碰撞检测构建了包含球体、子弹体以及圆柱体的混合包围盒,简化了双臂机器人的碰撞模型,并分析了运动过程中不同的碰撞情形。针对人机交互中难以区分有意接触和无意接触的问题,提出了一种人机接触状态识别及自适应导纳控制方法,实现了人机交互中的安全性。将人机协作控制框架分为安全层和协作层。在安全层,使用机器人的本体特性以及末端的力传感器,通过关节电流频率和末端力来判断机器人与人是主动接触还是被动碰撞;在协作层,将机器人分为领导者和顺从者两个角色,并设计了角色切换软开关,对末端力进行检测并根据检测结果实时确定机器人的角色。进而设计了模糊变导纳控制律,通过机器人的速度和加速度,对阻尼系数进行调整,保证系统的稳定性。为提高双臂协同作业轨迹规划的高效性并保证安全性,建立人-机-机之间的多重约束关系,并提出了一种异构双臂机器人同步示教控制方法,实现了双臂松耦合及紧耦合下的协同轨迹规划。对于双臂松耦合的情况,分别在绝对空间和增量空间进行示教;同于双臂紧耦合的情况,通过拖曳共同体同时对双臂进行示教。为了保证示教轨迹不超过关节极限和工作空间,在导纳控制器中增加了与关节极限和工作空间相关的项,在接近极限位置时阻尼系数迅速增大,阻止示教操作。最后,开发了协同控制器,并与异构双臂机器人系统进行集成,开展了实验验证。

【Abstract】 A heterogeneous dual-arm robot system composed of robots with different structures and functions can give full play to their respective advantages and compensate for their respective shortcomings to achieve a 1+1>2 operation capability,and thus has important application prospects in industry,service industry,aerospace,and other fields.However,as the kinematic and dynamical equations become more complex and the response characteristics differ greatly,it poses a great challenge for the control of the whole system.In this paper,a heterogeneous two-arm robot system consisting of a rope-driven robotic arm and an articulated collaborative arm is used to carry out research on its complex system modeling and collaborative supple control methods,and to conduct experimental validation.According to the typical task scenarios and system configuration characteristics,the kinematic model of the rigid-flexible hybrid heterogeneous dual-arm robot system is established and the dual-arm collaborative workspace is analyzed.The forward and reverse kinematics of the rope-driven robot arm and the articulated collaborative arm are derived using the D-H method,and then the kinematic constraints under the two different states of loose and tight coordination are derived and solved.Based on the structural characteristics of the heterogeneous dual-arm,a hybrid envelope containing a sphere,a bullet and a cylinder is constructed based on the envelope collision detection,which simplifies the collision model of the dual-arm robot and analyzes the different collision scenarios during the motion.For the problem that it is difficult to distinguish intentional contact and unintentional contact in human-robot interaction,a human-robot contact state recognition and adaptive conductive control method is proposed to achieve safety in human-robot interaction.The human-robot collaborative control framework is divided into a safety layer and a collaboration layer.In the safety layer,the robot’s body characteristics and the force sensors at the end are used to determine whether the robot is in active contact or passive collision with a human through the joint current frequency and the end force;in the collaboration layer,the robot is divided into two roles of leader and submissive,and a role switching soft switch is combined to detect the end force and determine the robot’s role in real time based on the detection results.Further,the fuzzy variable conductance control law is designed to adjust the damping coefficient by the robot’s speed and acceleration to ensure the stability of the system.In order to improve the efficiency and ensure the safety of cooperative trajectory planning of two-arm operation,multiple constraint relations between human-machine-machine are established,and a synchronous demonstration control method of heterogeneous two-arm robot is proposed to realize the cooperative trajectory planning under two-arm loose coupling and tight coupling.In the case of loosely coupled dual-arm robot,the trajectory is taught in absolute space and incremental space respectively,while in the case of tightly coupled dual-arm robot,the trajectory is taught simultaneously by the dragging community.In order to ensure that the teaching trajectory does not exceed the joint limit and the working space,terms related to the joint limit and the working space are added in the conductance controller,and the damping coefficient increases rapidly when the limit position is approached to prevent the teaching operation.Finally,a co-operative controller was developed and based on a heterogeneous two-arm robot system,and experimental validation was carried out.

  • 【分类号】TP242
节点文献中: 

本文链接的文献网络图示:

本文的引文网络