Intelligent vehicle laser radar maneuvering multi-target tracking method based on IMM-MHT algorithm

Abstract

The invention discloses an intelligent vehicle laser radar maneuvering multi-target tracking method based on an IMM-MHT algorithm, and the intelligent vehicle laser radar maneuvering multi-target tracking method belongs to the technical field of intelligent vehicles. The intelligent vehicle laser radar maneuvering multi-target tracking method is implemented by generating a tracking gate firstly, receiving a measurement value and performing detection and identification on the measurement value, making a joint hypothesis of a flight path and measurement, simplifying and managing the hypothesis, performing detection and identification on the flight path, and finally filtering and predicting the flight path. The intelligent vehicle laser radar maneuvering multi-target tracking method comprises the steps of: step (1), forming a flight path tracking gate; step (2), receiving a t-moment measurement value and identifying the t-moment measurement value by adopting an image processing method; step (3), generating an association hypothesis of measurement and a flight path in the tracking gate; step (4), managing the hypothesis; step (5), calculating information such as speed and angular speed of the flight path, and identifying the flight path; step (6), and filtering and predicting the flight path by adopting an interactive multi-model algorithm. Through integrating the interactive multi-model algorithm and a multi-hypothesis tracking algorithm, the intelligent vehicle laser radar maneuvering multi-target tracking method can better deal with the intelligent vehicle maneuvering multi-target tracking problem in a complex environment.

Description

technical field [0001] The invention belongs to the technical field of intelligent vehicles, and in particular relates to an intelligent vehicle lidar mobile multi-target tracking method based on the IMM-MHT algorithm. Background technique [0002] Because lidar has the advantages of high precision and little environmental impact, it is more and more widely used in intelligent driving technology. Smart cars use lidar to track obstacles on the road, which is a kind of mobile multi-target tracking. Most of the dynamic obstacles on the road are pedestrians and vehicles. Because of the unpredictable movement of motor vehicles, they behave as maneuvering characteristics, and single-model filtering is no longer applicable. The number of obstacles in the radar scanning range cannot be determined, and the tracking process includes the generation and elimination of tracking tracks. [0003] Interactive multi-model algorithm (IMM) is an algorithm to solve maneuvering target tracking...

AI Technical Summary

Problems solved by technology

Most of the dynamic obstacles on the road are pedestrians and vehicles. Motor vehicles behave as maneuvering characteristics due to their unpredictable motion, and single-model filtering is no longer applicable.

Images