Design of Active Tracking Algorithm for Motion Training Assisted Robots Based on Multiple Sensors
A multi-sensor based active tracking algorithm for motion training assistance robots is proposed to address the accuracy issue of detecting and tracking aerobics athletes using motion training assistance robots.Firstly,the overall scheme of the active track-ing algorithm is designed,and a robot active tracking method based on visual sensors and a robot active tracking method based on so-nar sensors are proposed;Then,a decision layer based data fusion method is used to fuse the tracking schemes obtained by the two tracking methods,compensating each other to obtain the best tracking scheme;Finally,through experimental testing,whether the al-gorithm proposed in this article can achieve active tracking of aerobics athletes by robots.The experimental results show that the multi-sensor based motion training assisted robot active tracking algorithm designed in this article can enable the robot to actively track aer-obics athletes,and the entire tracking process can quickly detect and approach the target.At the same time,by adjusting its own speed to safely and smoothly track the target,the goal of assisting athletes in training can be achieved.
万方数据
维普
