Obstacle trajectory prediction method and device, storage medium and electronic equipment

Abstract

The invention discloses an obstacle trajectory prediction method and device, a storage medium and electronic equipment. According to the embodiment of the invention, the historical motion tracks of obstacles are input into a first model to obtain global interaction features; the map data of the more than two scales and the historical motion tracks of a to-be-predicted obstacle are input into a second model; map aggregation features corresponding to the to-be-predicted obstacle are obtained; the global interaction feature, the map aggregation feature and the historical motion track of the to-be-predicted obstacle are input into a third model; the third model can express the movement speed of the to-be-predicted obstacle through the historical movement track of the to-be-predicted obstacle,the map features adaptive to the motion speed of the to-be-predicted obstacle are adaptively selected from the map aggregation features of the maps aggregated with various scales to assist in predicting the motion track of the to-be-predicted obstacle, so that the finally obtained predicted motion track of the to-be-predicted obstacle is more accurate.

Description

technical field [0001] This specification relates to the field of unmanned driving technology, and in particular to a method, device, storage medium and electronic equipment for trajectory prediction of obstacles. Background technique [0002] At present, in the field of unmanned driving technology, the trajectory prediction of obstacles is an important technology. With the development of sensor technology in recent years, more accurate maps (such as radar maps, high-precision maps, etc.) are used to assist in predicting the trajectory of obstacles. [0003] In the prior art, when the trajectory prediction of obstacles is assisted by maps, for obstacles with different speeds, maps of the same scale are used to assist the trajectory prediction of obstacles. [0004] But in fact, obstacles at different speeds require different map scales. For example, if the speed is faster, you generally need to see farther, while at low speeds you only need to observe the closer scene. [...

AI Technical Summary

Problems solved by technology

[0005] Therefore, the method in the prior art that uses maps of the same scale to assist obstacle trajectory prediction will lead to inaccurate prediction results

Images