Knee joint exoskeleton man-machine posture deviation identification method and device based on subspace

Abstract

The invention provides a sub-space-based knee joint exoskeleton man-machine attitude deviation identification method and device. The method comprises the following steps: acquiring a torque tau applied by a knee joint exoskeleton; inputting the moment tau into a state space model, and acquiring angle information capable of being used for human body posture correction, wherein the state space model conducts deviation identification on the knee joint exoskeleton joint angle and the actual human body knee joint angle. According to the method and device, the state space model of the exoskeleton joint torque and the knee joint posture deviation is established, the model is identified by using the subspace algorithm, and the corrected human body posture data is obtained, so that the accurate calculation of the human body posture in the state of wearing the exoskeleton is realized.

Description

technical field [0001] The invention relates to the field of exoskeletons, in particular to a subspace-based human-machine attitude deviation identification method and device for knee joint exoskeletons. Background technique [0002] Wearable exoskeleton robots assist the human body to complete actions through precision mechanical devices. It combines exoskeleton bionic technology and information control technology, and involves interdisciplinary knowledge such as biokinematics, robotics, information science, and artificial intelligence. At present, there are two ways to estimate the human body pose in the wearable exoskeleton state. One is to directly use the angle encoder on the exoskeleton or the inertial sensor (IMU) for calculation. In fact, the exoskeleton is calculated in this way. posture, ignoring the fact that there is a certain deviation between the posture of the human body and the posture of the exoskeleton; another method is to estimate the posture of the human...

AI Technical Summary

Problems solved by technology

At present, there are two ways to estimate the human body pose in the wearable exoskeleton state. One is to directly use the angle encoder on the exoskeleton or the inertial sensor (IMU) for calculation. In fact, the exoskeleton is calculated in this way. posture, ignoring the fact that there is a certain deviation between the posture of the human body and the posture of the exoskeleton; another method is to estimate the posture of the human body through methods such as Kalman filtering through an inertial sensor (IMU) bound to the human body. More accurate measurements can be obtained without wearing an exoskeleton, but in the case of wearing an exoskeleton, its installation position and measurement accuracy will be affected by the exoskeleton assisting process

Images