Solution And Application Of Motion Intention Based On Disturbance Observer

Neural-rehabilitation robots have gradually become a hot topic at home and abroad,real-time intention detection is one of the key techniques for the rehabilitation robot to comprehend the patient's intention and the adaptive interaction with the patient.It is the basis for the realization of intelligent rehabilitation.The common intention detection techniques are based on the intention detection of electroencephalogram(EEG),electrooculography(EOG)and electromyography(EMG),as well as the detection of intention based on human behavior such as gestures,postures and facial expressions.The former is considered to be an intention recognition technology based on bioelectrical signals.Especially one of the most valuable frontier technologies associated with brain electroencephalogram(EEG),which is considered to be the most valuable of the cognitive activities of brain nerve.Both EEG and EMG have attracted scientists' attention.However,considering the complexity of the signal processing and the cost of the system,the author have achieved some achievements by using the biomechanical characteristics of the extremities to capture the user's motion intention.The researchers separate human movement intention through the "force" information detected by the multi-dimensional force sensors at the extremities.The adaptive control technology is used to adjust the output current of the driving motor to achieve the flexible interaction effect compatible with the motion intention.The sensorless estimation method is of great practical significance to reducing the cost of the equipment and improving the response of the system and the precision of the control.In this study,the friction force in the motion process of the equipment is estimated by theoretical modeling and combined with the disturbance observer(DOB),and the detection of the motion friction in the engineering practice is effectively solved,and the test is checked.When the friction is modeled and compensated to DOB,the external force exerted on the equipment in the human-computer interaction system can be stripped to determine the intention of the stroke patients and to obtain the motion intention.The main contents of the paper are as follows:(1)Dynamic modeling of the upper limb rehabilitation robot,the dynamic modeling of the upper limb rehabilitation robot system and the nominal model in the disturbance observer through the system dynamic model.(2)The design and Simulation of the disturbance observer,the basic principle of the disturbance observer is introduced,and the simulation experiments are carried out for the designed disturbance observer and the interaction force observer based on friction compensation and gravity compensation.(3)Parameter identification of friction model,the mathematical model of common friction is introduced,the friction force of the system is estimated with the interference observer(DOB),and the friction force is modeled,which effectively solves the problem of detecting the friction force in the engineering practice.(4)Motion intention solving in upper limb rehabilitation robot.The motion intention of the extremities is calculated based on the estimated friction force,and the effect of friction on the motion intention is eliminated.By analyzing the "force" information of the intention,the motion intention of the rehabilitation robot in the upper limb is solved.The results are of the same reference value for the research and application of force control in robotics.