Multi-source target fusion method based on track matching

Abstract

The invention discloses a multi-source target fusion method based on track matching. The multi-source target fusion method includes the steps: acquiring target information by the aid of various sensors; performing time alignment and space alignment on the target information acquired by the sensors; matching image target tracks with target tracks acquired by the other sensors, and supplementing information of the target tracks in the other sensors through information of the image target tracks if the image target tracks can be successfully matched with the target tracks acquired by the other sensors; selecting a target track generating result of a single sensor, a target matching result among different sensors and a track matching result among different sensors as three evidences of targetexistence, calculating probability of the three evidences of the target existence, and taking the probability as basic probability assignment of a DS (dempster-shafer) evidence theory; performing target existence fusion and category fusion by the aid of the DS evidence theory; outputting a fusion result. The method can solve the problem that a global target track is frequently interrupted as an image target position is inaccurate, and accuracy and reliability can be improved.

Description

Technical field [0001] The invention belongs to the technical field of automatic driving and advanced assisted driving, and particularly relates to a multi-source target fusion method based on trajectory matching. Background technique [0002] In the process of autonomous driving and advanced assisted driving, there are many dynamic obstacle targets in traffic scenes, such as vehicles, pedestrians, and bicycles. The speed, category, position, orientation and other information of the dynamic obstacle target are attribute information that affects the safety and comfort of automatic driving and assisted driving, especially the category information, which is related to the decision accuracy of the control planning module . [0003] At present, different types of target sensing sensors are mostly used to obtain various information of dynamic obstacle targets. There are many kinds of target sensing sensors, such as lidar, camera, millimeter wave radar and so on. At present, with the r...

AI Technical Summary

Problems solved by technology

The multi-source target fusion method of the present invention can provide strong evidence of the existence of the target, can solve the problem of frequent interruption of the target trajectory caused by inaccurate target position in the image, and can improve the accuracy and reliability of environment perception in the process of automatic driving

Images