Obstacle backing track prediction method, device and equipment and readable storage medium

Abstract

The invention discloses an obstacle backing track prediction method, device and equipment and a readable storage medium. The method comprises the steps: obtaining current driving information of an obstacle, screening the current driving information, obtaining an obstacle in a backing state as a first obstacle, determining the limit curvature of the backing track of the first obstacle, determining limit position information of rotation centers on two sides of a first obstacle by utilizing a limit curvature of a backing track of the first obstacle, and for each rotation center, determining a limit backing track of the first obstacle according to the corresponding limit curvature and the limit position information, and uniformly generating a plurality of reference backing tracks between the two determined limit backing tracks. According to the method, the obstacle in the backing state is screened out by using the current driving information of the obstacle, the limit curvature of the obstacle is determined by using the current driving information of the obstacle obtained by screening, and the reference backing track of the obstacle is finally determined based on the limit curvature, so that prediction of the backing track of the obstacle is realized.

Description

technical field [0001] The present application relates to the technical field of automatic driving, and in particular to a method, device, equipment and readable storage medium for predicting an obstacle reversing trajectory. Background technique [0002] In recent years, with the rapid development of science and technology, research on autonomous driving technology has become a hot spot in various industries. When the self-driving vehicle is driving, it will recognize the surrounding obstacles and predict the trajectory of the obstacles, so as to generate corresponding decisions to control the driving of the vehicle and avoid collisions with obstacles. At present, the common way to predict the trajectory of obstacles is to use the neural network model to analyze the obstacles, so as to predict the trajectory of obstacles. Among them, when predicting the trajectory of the obstacle, the prediction of the reversing trajectory of the obstacle is also included, but the reversin...

AI Technical Summary

Problems solved by technology

Among them, when predicting the trajectory of the obstacle, the prediction of the reversing trajectory of the obstacle is also included, but the reversing of the obstacle occurs less, so when the neural network model is trained, the training samples involved are relatively small, which means It will make the neural network model trained to predict the reversing trajectory of obstacles may be less accurate. Therefore, how to predict the reversing trajectory of obstacles is a problem that people have been paying attention to.

Images