Tracking Control Of The Omnidirectional Rehabilitative Training Walker With Interaction Forces

With the growing trend of the aging population,the number of patients with lower limb motor dysfunction is increasing caused by the factors covering brain or motor nerve diseases and accidents.And the lack of medical resources leads that some patients can not obtain effective rehabilitation training,which promotes rehabilitation walking training robot developed rapidly.In the process of rehabilitation training,the rehabilitation walking training robot will drive the patients to track the specified trajectory tracking set by the doctor.However,during the process,the trajectory tracking will be affected by the center of gravity shift,the change of the patient quality,the internal interference of the interaction forces between human and robot leading to reduce the tracking accuracy.In order to obtain more accurate tracking trajectory,this thesis takes an omnidirectional walking rehabilitation training robot as the research object.The problem of the trajectory tracking control with the interaction force is researched.The main research contents are as follows:Based on the dynamic model with the center of gravity shift of the omnidirectional walking rehabilitation training robot,an adaptive backstepping controller was designed to solve the the adverse effects of quality change caused by different patients.By adjusting the control parameters and adaptive parameters,the adaptive law was utilized to adapt to the change of quality parameter of the patient,which can ensure the asymptotic stability of the system tracking errors.The simulation results show that the designed controller can effectively improve the control accuracy.When the patients follow the omnidirectional walking robot for rehabilitation training,the interaction forces generated between human and robot can bring adverse effects to the control system.In this dissertation,the concept of man-robot interaction was defined at first.Then,a fuzzy identification method was proposed to construct the interaction model using the reduced value of tracking performance.On the basis of this model,the adaptive backstepping controller was designed to compensate the interaction forces and adapt the change of quality.The control parameters were adjusted so that the system tracking errors were asymptotically stable.The simulation results show that the designed controller can effectively resist the interaction forces.Further,the experiment results are provided to demonstrate the effectiveness of the proposed approach.Considering the real-time effect of the interaction forces on tracking accuracy of the robot,from the perspective of the safety of patient,this thesis proposed a new extended state observer to accomplish the online observation of the interaction forces.The extended state observer was designed utilizing the output location and speed information of the system,and the speed variation range of the interaction forces.On this basis,the backstepping controller was designed to resist the effects of the interaction forces.The system tracking errors and observation errors were asymptotically stable by adjusting control parameters and observation parameters.The feasibility of this method is proved by the simulation results.