Research On Intention Recognition And Motion Control Of Lower Limb Assisted Exoskeleton Robot

Nowadays,with the development of robot technology,exoskeleton robot technology has been widely used in the fields of medical rehabilitation,logistics transportation,disaster relief and rescue.In the process of disaster relief,because of the complex terrain,the relief workers need to carry on the relief equipment to rescue the wounded for a long time.In order to deal with the unexpected situation in the disaster relief environment,exoskeleton robots used for disaster relief personnel need to have accurate motion intention recognition system and appropriate motion control system to ensure the rescue efficiency and personal safety of disaster relief personnel.Therefore,this paper designs a kind of lower limb assisted exoskeleton,which is used to assist the wearer to walk with heavy load for a long time.In this paper,through the analysis of the characteristics of the main joint structure of the lower extremities and the gait of the human body,combined with the work tasks of the assisted lower exoskeleton robot,the classification of the finite state machine of the assisted extremity exoskeleton robot suitable for this system is proposed.Based on the finite state machine conversion of the exoskeleton robot,in-depth analysis of the data characteristics of the exoskeleton robot’s motion state and human-computer interaction detection,the sensor system of the lower extremity assisted exoskeleton robot is designed.Based on the data feature analysis of human motion state,an exoskeleton robot motion state recognition algorithm based on the combination of plantar pressure data detection and support vector machine model is designed.Based on the human-machine interactive detection system of the exoskeleton robot,according to the finite state machine conversion of the exoskeleton robot,the assistance and follow intervals of the exoskeleton robot are divided,and the impedance-based fuzzy control algorithm and the DMP-based steps are designed for the assistance and follow intervals Trajectory generation position control algorithm.In this paper,the lower limb assisted exoskeleton system mainly completes the design and verification of motion intention recognition algorithm and motion control algorithm.The gait recognition algorithm based on the rising edge and falling edge detection of plantar pressure.The state recognition algorithm based on the support vector machine of lower limb posture are adopted.After experimental analysis,the algorithm has a high recognition accuracy and can be used in this system and the accuracy rate is over 90%.In different motion states,adaptive impedance control algorithm based on fuzzy control and position control algorithm based on DMP are used.Through the simulation experiment analysis in MATLAB,The adaptive impedance control algorithm has good environmental adaptability and self-adjusting ability.Within the knee gait cycle,the tracking deviation is within 8%.The DMP-based gait trajectory generation algorithm has good periodic learning ability,and the generated gait trajectory curve inherits the motion characteristics of the previous gait cycle.The exoskeleton robot system can adapt to the complex and changeable environment,can accurately identify the wearer’s movement intention,improve the human-machine collaboration ability,and can assist to improve the disaster relief personnel’s ability to walk under a complex and changeable environment.