Vehicle Trajectory Tracking Method Based on Kalman Filter and χ2 Detection Smoothing

Abstract

The invention discloses a method based on Kalman filtering and χ 2 The vehicle trajectory tracking method with detection and smoothing processing aims to find a more effective vehicle tracking and detection method, which includes the following steps: constructing a kinematic model and a measurement model of the target vehicle driving process; The model and measurement model are substituted into the time update equation and state update equation in the Kalman filter to iteratively update the motion state of the target vehicle and predict the motion state at the next moment; construct χ 2 The fault detector performs fault detection on the prediction result. The present invention applies χ to problems such as identification error or loss that may occur during tracking in the prior art 2 The fault detector can effectively reduce the deviation of trajectory prediction and improve the accuracy of trajectory recognition.

Description

technical field [0001] The present invention relates to the field of automatic driving, and in particular, the present invention relates to a method based on Kalman filtering and χ 2 A vehicle trajectory tracking method for detection smoothing. Background technique [0002] With the continuous development and progress of science and technology, driverless vehicles are the main trend in the development of modern vehicles. Major car companies and universities have carried out various research applications and experimental tests around their various technologies and functions. Trajectory tracking control is the basis for realizing unmanned driving, and it is also an important technology for vehicle control of intelligent vehicles. To obtain the driving trajectory of the vehicle in the current scene, it is necessary to track the vehicle for a long time. [0003] For the vehicle tracking detection method through video images, the original vehicle trajectory obtained has the disa...

AI Technical Summary

Problems solved by technology

[0003] For the method of vehicle tracking and detection through video images, the original vehicle trajectory obtained by it has the disadvantages of unstable detection frames, many jumps in the trajectory, and poor smoothness. Noise problems such as clutter and dark changes, the processing results obtained by applying traditional tracking algorithms are inaccurate, and it is difficult to accurately obtain the position information and speed information of the target vehicle

Images