Research On Object Location And Trajectory Planning Of Upper Limb Rehabilitation Robot Based On Eye Movement Information

Spinal cord injury and stroke,as the typical representatives of multiple trend neurological injury diseases,have become a serious threat to the health of Chinese residents.70%～80% of the survivors have some degree of limb movement disorders.The upper limb activities of the human body occupy a greater proportion in daily activities than the lower limb,so the rehabilitation training of patients with upper limb movement disorders is particularly important.Aiming at the active rehabilitation training of patients with upper limb motor dysfunction after surgery,this paper proposes a rehabilitation strategy scheme of upper limb rehabilitation robot based on eye movement information,mainly in the aspects of eye movement information,target object positioning,kinematics and trajectory planning.Firstly,according to the mechanism of eye movement rehabilitation,the strategy of eye movement rehabilitation is proposed.In view of the extraction of target objects of interest to patients,this paper adopts eye tracker to extract the corresponding eye movement information,and processes and analyzes the acquired eye movement information to obtain eye movement trajectory diagram and hot spot map,and compares and determines the target objects of interest to patients.Secondly,the target object is located and the system coordinates are unified.For target object positioning,this paper uses the binocular 3D reconstruction algorithm based on eye tracker to estimate the position of the target object in the eye tracker coordinate system,then uses the six-axis attitude Angle sensor to conduct real-time tracking of the head attitude,establishes the conversion relationship of each module,and finally converts the coordinates of the target object to the coordinate system of the upper limb rehabilitation robot to achieve target object positioning.Furthermore,the D-H method was used for kinematics analysis of the robot,and the quintic B-spline method was used for trajectory planning of the robot.After the planning was completed,aiming at the two optimization objectives of energy-impact,an improved NSGA-II algorithm was proposed.The Circle function of chaotic strategy was added to the initialization of its population to increase the randomness of its search.At the same time,the concept of dynamic crowding degree is added to replace the original fixed crowding degree during evolution,variation and selection,so as to increase the diversity of optimal solutions.Finally,on the basis of building the electrical system of the upper limb rehabilitation robot,the target object confirmation and positioning experiments were carried out,which provided the data basis for the target object grasping experiment.At the same time,the track of the target object grasping experiment was tracked,which proved that the trajectory tracking accuracy of the rehabilitation robot was high,and the correctness of the trajectory planning was verified.