Target tracking system and tracking method applied to driverless car

Abstract

The invention discloses a target tracking system and tracking method applied to a driverless car. The tracking system comprises a vehicle-borne radar system and a robot operating system (ROS), whereinthe vehicle-borne radar system is composed of a laser radar and a millimeter wave radar. The laser radar is mounted at the top of the car and has relatively high measurement accuracy, and but is susceptible to environmental factors such as illumination, the millimeter wave radar is installed at a license plate at the front of the car, has relatively low measurement accuracy, but is just slightlyaffected by external environment, and therefore, the vehicle-borne radar system integrating the laser radar and the millimeter wave radar combined can improve measurement accuracy; and the node set ofthe robot operating system (ROS) includes a radar sensor node and an unscented Kalman filter estimator node. According to the unscented Kalman filter algorithm, a series of sigma points are adopted to generate states through nonlinear transformation, and the estimated sigma points are used to cover state estimation points and covariances, and therefore, the fusion of the data received by the radar system can be realized, and target tracking can be achieved.

Description

technical field [0001] The invention relates to a vehicle target tracking system and method, in particular to a target tracking system and method based on a lossless Kalman filter for an unmanned vehicle equipped with a vehicle radar system. Background technique [0002] Unmanned vehicles need to sense the environment during unmanned driving, so as to realize the control of the vehicle and the planning of the path according to the environmental information, and this process depends on the corresponding on-board sensors. These sensors mainly include cameras, lidar, millimeter-wave radar, infrared cameras, GPS and inertial navigation systems, etc. Millimeter-wave radar has a strong ability to penetrate fog, smoke, and dust, and has the advantages of all-weather and all-weather, but its detection accuracy is low; while lidar has high detection accuracy but is easily disturbed by the external environment. Therefore, compared with a single sensor, the vehicle radar system compos...

AI Technical Summary

Problems solved by technology

[0005] In the current technology, the application number "201210555170.7", the patent application publication number "CN103889047A" "a target tracking algorithm based on Kalman filter" and the application number "201610809777.1", the patent application publication number "106443661A" "based on None of the "maneuvering extended target tracking methods with trace Kalman filter" combined the application of lossless Kalman filter with the vehicle-mounted radar system of unmanned vehicles, and did not form a complete target tracking and information processing system.

Images