Lower limb movement pose quick predicting system and method based on fusion of a plurality of sensors

Abstract

The invention relates to a lower limb movement pose quick predicting system based on fusion of a plurality of sensors. The lower limb movement pose quick predicting system comprises surface myoelectricity sensors, inertial sensors, a pose responding module, a pose resolving module and an external auxiliary device, wherein the external auxiliary device is worn to the waist and the lower limbs of awearer; the surface myoelectricity sensors are used for obtaining muscle surface myoelectricity signals of the lower limbs of the wearer in a real-time manner; the inertial sensors are fixed to the external auxiliary device; the pose responding module is used for determining the motion state corresponding to the current muscle surface myoelectricity signal of the lower limbs according to the collected sequencing of each muscle surface myoelectricity signal; and the pose resolving module is used for performing short time dynamic weighting data fusion on each joint movement angle collected by the inertial sensors and the joint movement angle estimated by resolving, so that the lower limb movement posture and the position coordinates of the joints of the wearer are obtained in a real-time manner. The invention further provides a lower limb movement pose quick predicting method. Through the adoption of the lower limb movement pose quick predicting system and method disclosed by the invention, the movement pose of the lower limbs of the human body can be quickly predicted, and besides, the estimating accuracy of the continuous movement amount is improved.

Description

technical field [0001] The invention relates to a system and method for fast prediction of lower limb motion pose based on multi-sensor fusion, and belongs to the field of exoskeleton robots and human-machine collaborative control. Background technique [0002] At present, in terms of rehabilitation training and walking assistance, exoskeleton robots are becoming more and more popular with patients as an important device that can replace manual assistance. However, the human-machine collaboration effect of existing exoskeleton robots is not good, mainly because the prediction method of human body motion information is mostly carried out by physical sensors. Physical sensors can only collect human body information when the wearer has a certain range of motion. Motion information, resulting in a large time delay, can not meet the requirements of fast prediction; at the same time, a single sensor has a certain accuracy in the prediction of a discrete action, but there is a big ...

AI Technical Summary

Problems solved by technology

However, the human-machine collaboration effect of existing exoskeleton robots is not good, mainly because the prediction method of human body motion information is mostly carried out by physical sensors, and physical sensors can only collect human body information when the wearer has a certain range of motion Motion information, resulting in a large time delay, can not meet the requirements of fast prediction; at the same time, a single sensor has a certain accuracy in the prediction of a discrete action, but there is a big shortage in the continuous motion estimation

Images