Target tracking post-processing method based on vehicle physical kinematics model and vehicle

Abstract

The invention discloses a target tracking post-processing method based on a vehicle physical kinematics model and an automobile. The method comprises the following steps: acquiring data information of a vehicle target; eliminating outlier points in the data information; selecting a target vehicle for tracking; continuously and stably tracking the selected target vehicle; performing parameter calculation on the tracked target vehicle; and judging whether the tracked target vehicle disappears or not, if so, ending, and if not, continuously tracking. According to the invention, some detected errors and inaccurate results can be corrected according to physical characteristics of a moving vehicle, missing detection and false detection conditions of detection are processed, target tracking is made to be continuous and stable, and meanwhile, kinematics characteristics based on target vehicle motion are combined with vehicle motion information and detection information, so that target tracking accuracy is improved. And the pure vision scheme information is used for tracking the vehicle related target, so that the inspection and data acquisition cost is reduced.

Description

technical field [0001] The invention relates to the technical field of intelligent driving, in particular to a target tracking post-processing method based on a physical kinematics model of a vehicle and a vehicle. Background technique [0002] Target detection and tracking is an important research direction in the field of autonomous driving. The accuracy of target detection and tracking is closely related to the safety of autonomous driving. Ensuring the accuracy of target tracking can greatly avoid the possibility of collision between autonomous vehicles and target vehicles. . [0003] There are many related technical solutions for target tracking at this stage. For example, the invention patent application with publication number CN111932580A discloses a road 3D vehicle tracking method and system based on Kalman filter and Hungarian algorithm. The Hungarian algorithm performs the best matching to predict the tracking trajectory of the target, but this method does not co...

AI Technical Summary

Problems solved by technology

[0003] There are many related technical solutions for target tracking at this stage. For example, the invention patent application with the publication number CN111932580A discloses a road 3D vehicle tracking method and system based on Kalman filter and Hungarian algorithm. The method uses Kalman filter for prediction and tracking. The Hungarian algorithm performs the best matching to predict the tracking trajectory of the target, but this method does not take into account the processing of errors that do not satisfy the Gaussian distribution due to other factors such as ranging, or the tracking generated when the detection network is abnormal Processing during abnormality; another example is the invention patent application with publication number CN112487919A, which discloses a 3D target detection and tracking method based on cameras and laser radars. For matching and tracking, this method needs to use lidar to obtain the true value, and does not make full use of the target's movement and visual spatial information, which greatly increases the cost of detection and data acquisition

Images