The Design Of Control System Platform And Control Algorithm Analysis For Lower Limb Exoskeleton Robot

As a newly developed research field, the technology of lower limb exoskeleton robot involves robotics, Bionics, artificial intelligence and many other subjects. The development of exoskeleton robots draws on the experience and technology of humanoid robots. It can be worn in the human body and realizes the human-machine integration. It has a great development in the field of the medical rehabilitation and military area. The lower limb exoskeleton rehabilitation robots help patients free from the traditional "one therapist to one patient" treatment of the rehabilitation training. It has changed people’s cognition to the traditional medical rehabilitation. In the military field, exoskeleton system can maximize the ability and potential of soldiers and improve their overall quality. This thesis based on the previous research mainly focuses on the design of the control system platform and the implementation of the control algorithms of the lower limb exoskeleton robots.First, the research status of lower limb exoskeleton robots at home and abroad is summarized in this thesis. Referred to the previous successful achievement, the system is designed. The control master is designed based on the software TwinCAT and the Ethernet is employed as the system bus of the control system. The EtherCAT is used as the communication protocol which improves the data propagation speed and control accuracy. The coordinated motion of multi-motors is achieved in the end.In order to analyze the feasibility of the system, the exoskeleton robot is simplified into a 5-DOF link model when it is simulated. The models of dynamics and kinematics are established based on Lagrange Equation. The PD control algorithm based on the gravity compensation and the fuzzy adaptive PID control algorithm are respectively applied for the simulation in MATLAB/Simulink. Comparing the results, it is concluded that fuzzy PID control has a better control effect.Hardware platform have been built and the whole control system is divided into four layers, including the network layer、the master layer、the slave layer and the execution layer. The actuators are controlled by double closed-loops. The inner closed-loop is consisted of the master controller、drive and motor. The motion state of the motor is feedback by the motor encoder. The outer closed-loop is consisted of the sensors and PC. The sensors collect the actual trajectory of exoskeleton and input them into the PC. In the PC, the accurate trajectory is obtained by the fuzzy PID control algorithm and then sent to the controller. In the controller, the control command is sent to the drive by motion control functions. All of these make the exoskeleton system simulate the human gait well. The validity of the control system is verified by the experiment.