基于力观测器的微创机器人力检测与力反馈控制研究

发布时间：2019-05-18 13:23

【摘要】：微创外科机器人具有灵活性高、定位准确、运动平稳和手眼协调等诸多优点,然而,由于微创手术的特殊性,使得现有微创机器人系统缺乏力检测和力反馈。针对上述问题,在分析现有力检测技术的基础上,构建了外力观测器对从机器人与环境间的作用力进行观测,设计了主从鲁棒滑模控制器和主从力反馈结构,实现了无力/力矩传感器力检测与力反馈控制,为无力/力矩传感器下的主从运动和力反馈控制奠定了研究基础。本论文的主要研究内容如下:首先,基于旋量理论分析了主从微创机器人的运动学,并结合拉格朗日动力学建模方法建立了主从机器人动力学模型。通过ADAMS和MATLAB联合仿真验证了从机器人动力学模型的正确性,并分析了主从机器人动力学方程中各项对总力矩的贡献,为主从微创机器人动力学模型的简化奠定基础。其次,在分析非线性干扰观测器的基础上,提出了从机器人非线性外力观测器,用于实现从机器人与环境间相互作用力的观测,同时证明了外力观测器的稳定性。鉴于关节摩擦对观测精度的影响,确定了 Stribeck模型为关节摩擦模型,采用遗传算法对摩擦参数进行了辨识,在从机器人外力观测器中进行了摩擦补偿,实现精准的外力观测。再次,设计了微创机器人主从鲁棒滑模控制器,构建了主从力反馈系统,实现了对主从机器人的精准的位置控制和力反馈控制。通过Simulink仿真验证了在有/无摩擦补偿的情况下力反馈控制系统的有效性和准确性,且具有良好的主从位置跟踪能力。最后,构建了主从微创机器人实验平台,对外力观测器和主从运动和力反馈控制进行实验研究,结果表明该控制系统具有较好的力反馈性能,实现了无力/力矩传感器下的外力观测和力反馈控制。
[Abstract]:Minimally invasive surgical robots have many advantages, such as high flexibility, accurate positioning, stable motion and hand-eye coordination. However, due to the particularity of minimally invasive surgery, the existing minimally invasive robot systems lack force detection and force feedback. In order to solve the above problems, based on the analysis of the current powerful detection technology, an external force observer is constructed to observe the force between the slave robot and the environment, and the master-slave robust sliding mode controller and the master-slave force feedback structure are designed. The force detection and force feedback control of powerless / torque sensor are realized, which lays a research foundation for master-slave motion and force feedback control under powerless / torque sensor. The main research contents of this paper are as follows: firstly, the kinematics of master-slave minimally invasive robot is analyzed based on spinor theory, and the dynamic model of master-slave robot is established by combining Lagrangian dynamic modeling method. The correctness of the slave robot dynamics model is verified by the joint simulation of ADAMS and MATLAB, and the contribution of the master-slave robot dynamics equation to the total torque is analyzed, which lays a foundation for the simplification of the master-slave minimally invasive robot dynamics model. Secondly, on the basis of analyzing the nonlinear disturbance observer, a slave robot nonlinear external force observer is proposed to observe the interaction between the robot and the environment, and the stability of the external force observer is proved at the same time. In view of the influence of joint friction on the observation accuracy, the Stribeck model is determined as the joint friction model. The friction parameters are identified by genetic algorithm, and the friction compensation is carried out from the robot external force observer to realize accurate external force observation. Thirdly, the master-slave robust sliding mode controller of minimally invasive robot is designed, and the master-slave force feedback system is constructed to realize the accurate position control and force feedback control of the master-slave robot. The effectiveness and accuracy of the force feedback control system with / without friction compensation are verified by Simulink simulation, and the force feedback control system has good master-slave position tracking ability. Finally, the experimental platform of master-slave minimally invasive robot is constructed, and the external force observer and master-slave motion and force feedback control are studied experimentally. the results show that the control system has good force feedback performance. The external force observation and force feedback control under powerless / torque sensor are realized.
【学位授予单位】：天津工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP242

【参考文献】