Research On Human-robot Coupling Multi-modal Sensing System For Lower Limb Rehabilitation Training Robot

According to the current population structure and development trend,aging will be a huge challenge for China in the future.Due to the aging of the aging,the disease spectrum of our society is also undergoing irreversible changes.Strokes(also known as strokes)caused by cardiovascular and cerebrovascular diseases are gradually becoming a huge threat to people's health.Medical research has found that the timely treatment of nerve damage caused by stroke can be effectively assisted in restoring healthy walking ability with scientific exercise-assisted training.However,although the existing lower limb rehabilitation training robot can perform passive rehabilitation training well,it lacks effective sensing and recognition of the patient's lower limb movement intention in active training.In this study,the human-robot coupled multimodality sensing system of the lower limb rehabilitation training robot is studied to realize the judgment of the patient's lower limb movement intention,thereby improving on-demand assistance and safe and supple rehabilitation training.Based on the retrieval and analysis of the research results of lower limb rehabilitation training,the multi-modal information in the human-robot coupling process of lower limb gait rehabilitation training is proposed according to the gait phase of the lower limb gait rehabilitation training and the judgment of the patient's movement intention.Firstly,based on the mechanism of human lower limb movement,the patient's gait parameter sensing system is constructed to obtain the relevant feature information(hip/knee joint angle,plantar pressure)in the lower limb walking movement.Secondly,the robot motion parameter sensing system is designed to realize the state monitoring of the robot assisted patients during rehabilitation training.Then,according to the judgment requirements of the human body's lower limbs motion intention in the active rehabilitation training,the human-robot coupling force sensing system was developed.In order to properly assist the patient and stimulate the patient to actively participate in the rehabilitation training,the gait of the patient's lower limb walking movement must be distinguished and identified.Therefore,the multi-modal information fuzzy fusion algorithm is designed.The multi-modality sensing system is used to obtain the human and robot motion parameters and force parameters,and the gait phase is identified based on the fuzzy fusion algorithm.In order to verify the multi-modal information sensing system and fuzzy fusion algorithm,the real-time detection of human-robot coupling force and real-time detection of robot motion parameters are completed on the prototype and test platform through different walking tests of different height test objects at different speeds.Realtime detection of hip and knee joint activity angle and foot pressure information in human lower limbs.The test results show that the multi-modality sensing system is designed to satisfy the requirements of human-robot system motion and force parameters.Based on the acquired human motion parameters,the gait phase recognition is completed,and the feasibility of the fuzzy fusion algorithm is verified,which lays a foundation for the research of the control strategy of the gait rehabilitation training robot.