Research on Human Motion Intention Prediction Technology Based on DTW

To enhance the compliance of the lower limb exoskeleton's follow-up control, the control system should accurately identify the human gait phase and perceive the human motion intention in real time. There is a delay in the process of obtaining gait information through sensors. Therefore, to ensure the real-time performance of the follow-up control, it is necessary to accurately predict the human motion intention. This paper proposes an algorithm based on Dynamic Time Warping (DTW) for long-term prediction of human motion intention. According to the current pose state of the exoskeleton, multi-sensor information fusion combined with DTW is used to predict the motion intention for a long time, which compensates for the poor long-term prediction effect and the poor handling of sudden changes during walking of the Kalman filter. It realizes the long-term and accurate prediction of the lower limb motion intention after the human wears the exoskeleton.