Tracking algorithm for swarm robots based on improved Bytetrack
In practical tracking scenarios,swarm robots are often affected by factors such as occlusion,target density,and scale transformation,resulting in missed detection,trajectory interruption,and frequent ID jumps.Based on the Bytetrack tracking algorithm,the state variables of the Kalman filter was improved,a noise scale adaptive Kalman algorithm(NASA-Kalman)was proposed,and acceleration parameters(AMA)were introduced into Kalman filter to improve the tracking accuracy.Compared to the original algorithm,MOTA and MOTP were improved.In order to further verify the effectiveness of the tracking algorithm,the algorithm was evaluated on the MOT20 dataset,and improved by 0.65% and 1.26% in MOTA and MOTP respectively.
NETL
NSTL
万方数据
维普
