Vehicle trajectory prediction method and device, electronic equipment and computer readable medium

Abstract

The invention relates to a vehicle trajectory prediction method and device based on a high-precision map, electronic equipment and a computer readable medium. The method comprises the steps that information of other vehicles around the intelligent automobile, lane information and sidewalk information are acquired based on a high-precision map; an adjacent relation graph is generated based on the vehicle information, the other vehicle information, the lane information and the sidewalk information of the intelligent vehicle, and the intelligent vehicle, the other vehicles, the lanes and the sidewalks are all nodes of the adjacent relation graph; a local feature of each node in the adjacent relation graph is extracted; and the trajectories of the other vehicles is predicted based on a globalconvolutional network, the local features and the adjacent relation graph. According to the vehicle trajectory prediction method and device based on the high-precision map, the electronic equipment and the computer readable medium, an automatically-driven intelligent vehicle can obtain a more reasonable driving strategy, and a safer driving route is planned.

Description

technical field [0001] The present invention relates to the field of intelligent vehicle path planning, in particular to a high-precision map-based vehicle trajectory prediction method, device, electronic equipment, and computer-readable medium. Background technique [0002] Vehicle behavior prediction solves the problem of cooperative interaction between self-driving vehicles and other moving vehicles. Existing prediction methods generally include the following: 1. Prediction based on physical laws. It is assumed that the movement mode of the moving object satisfies certain rules. For example, the vehicle satisfies kinematic constraints, etc.; 2. Prediction based on motion behavior. It is assumed that the behavior of moving objects such as vehicles is a combination of basic behaviors such as cruising, turning left, turning right, following, overtaking, parking, etc.; 3. Behavior prediction based on high-precision maps. HD maps provide rich traffic information (such as la...

AI Technical Summary

Problems solved by technology

This approach is to convert the information of the high-precision map into a semantic image, and let the neural network learn the semantics, which is very dependent on the way of generating the semantic image. For a long time, the performance is greatly affected

Images