Design And Control Of Upper Limb Rehabilitation Robot System For Trunk Compensation Corrective

Trunk compensation is commonly used by patients with stroke during rehabilitation and negatively affects recovery outcomes.Previous studies have demonstrated that correcting compensation may be beneficial for functional recovery.Therefore,it is necessary to correct the trunk compensation in rehabilitation.Howerver,most researches focus on the automatic detection of TC,without involving the issue of trunk compensation reduction during training.Current therapeutic approach to reduce compensation relies on therapists or the physical restraints.,which are inefficient,small-scale and potential safety problems.Moreover,the methods by using visual-based or auditory-based feedback exist limtiations and little attention has been given to the role of robot in the correction of compensation.In view of the above limitations,this paper developed a new upper-limb rehabilitation robot that takes advantage of both end-effector and exoskeleton robots,and it is capable of assisting upper limb movements and correcting compensatory postures simultaneously.Firstly,based on the characteristics of human upper limb and hemiplegic trunk compensation,the structural design requirements are proposed.The system structure and hardware system composition are introduced in detail.Then the corresponding kinematics model is established to analyze the workspace and singularity of the upper limb and the system,which is the theoretical basis of the control system design.Based the idea of hierarchical control,the whole control system is constructed by including the bottom layer of motion control,the middle human-computer compliance interactive control and the upper layer control strategy to correct the trunk compensation.The upper control strategy includes position-based control and learning-based control.Finally,experiments are conducted to explore the effects of the designed system and the proposed strategies on trunk compensation correction and upper limb motor performance.The results show that the two control strategies have achieved good trunk compensation correction effect,but the learning-based strategy not only effectively reduces the motion error of the upper limb,but also significantly improves the fluency,enabling the subjects to complete the task training in a more correct limb movement mode.In general,the upper limb rehabilitation robot system proposed in this paper can effectively correct the trunk compensatory movement,and help to improve the motor ability of patients,which has important clinical application value and social significance.