Research on Control Strategy of the6-DOF Parallel Robot for Rehabilitation Purpose
|School||Wuhan University of Technology|
|Course||Communication and Information System|
|Keywords||Lower limb rehabilitation parallel robot fuzzy PID control sliding modecontrol force/position control|
Medical rehabilitation robot, which is an interdisciplinary research area where medicine technologies combined with mechanics, electronics, computer science, and robotics, has become an outstanding field that attracts more and more attentions in both industrial and academic domains. Applying robotic technology to rehabilitation can not only release rehabilitation physicians from the heavy burden of training, but also provide a platform to evaluate the convalescence results by means of analyzing the data recorded during training process. Due to advantages of its superior stiffness, compact structure, high load capacity and operation accuracy,6-DOF parallel platform has been widely used as carrier of flight simulators, telescopes and medical devices. While applications of6-DOF parallel robots in rehabilitation and related technologies in health monitoring and evaluation are able to effectively improve the recovery performance of patients after rehabilitation and training.The Stewart platform is a parallel robot whose operation is based on the simultaneous movement of its six-axis. In this paper, a multi-DOF parallel robot based on Stewart platform is designed for lower limb rehabilitation, after studying its kinematics and dynamic models, a high-precision velocity control strategy based on fuzzy adaptive algorithm is proposed and verified. To guarantee the control precision of robots that applied for lower limb rehabilitation, an actual force control system was established and a closed-loop control strategy based on fuzzy sliding mode algorithm and force control strategy based on hybrid and impedance control is proposed.The research content in this thesis mainly contains the following parts:(1) Firstly the kinematics and dynamic model of the parallel robot, especially the inverse kinematics solution are studied. Thus, the actual length of each joint can be determined according to parallel robot’s position in task space based on MATLAB. Furthermore, an actual parallel robot platform in real environment is established based on dynamic relationship between the task space and each joint.(2) With the combination of simulation model and actual platform of the6-DOF robot developed, its closed-loop control of position and orientation are studied, and several classical control strategies such as PID, fuzzy PID control algorithms are designed and implemented; while the software and DSP-based motion controller are proposed, and in this system the parallel robot platform can be controlled accurately.(3) A sliding mode controller (SMC) for the6-DOF parallel robot is proposed, and the applicable sliding mode control system is established based on the robot model; with the combination of fuzzy logic, a fuzzy sliding mode controller is designed in order to improve the precision of position control. The high-precision control of parallel robot is achieved in the simulation and physical environment.(4) In order to achieve different rehabilitation robot training mode, the force sensor feedback information is considered to propose the parallel robot supple control strategy (hybrid force/position control, impedance control algorithm); a model of active force control system is established and the actual test platform is implemented to realize the active rehabilitation robot control strategies.