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

Abstract

This specification discloses a trajectory prediction method, device, storage medium, and electronic equipment. The embodiment of this specification determines the historical interaction characteristics between the specified target and each obstacle according to the historical trajectory of the specified target and each obstacle, and the specified target The trajectory of the motion is predicted to obtain the initial predicted trajectory. Then, based on the initial predicted trajectory and the planned trajectory of each obstacle, the future interaction characteristics between the specified target and each obstacle are determined. According to the future interaction features, the final predicted trajectory of the specified target is obtained. In this method, the time dimension of the trajectory based on the prediction of the initial prediction trajectory of the specified target is the historical time dimension, and the time dimension of the trajectory based on the prediction of the final prediction trajectory of the specified target is the future time dimension. In this way, the problem of different time dimensions can be avoided, thereby improving the prediction accuracy of the motion trajectory of the specified target.

Description

technical field [0001] This specification relates to the field of unmanned driving, and in particular to a trajectory prediction method, device, storage medium and electronic equipment. Background technique [0002] In the field of unmanned driving, unmanned equipment needs to plan the corresponding trajectory for itself in real time according to the changes in its surrounding environment. Whether the motion trajectory planned by the unmanned device for itself is reasonable or not depends on the prediction of the motion trajectory of the obstacles around itself by the unmanned device. [0003] In the prior art, obstacles that the unmanned equipment needs to predict can be used as specified targets. When predicting the specified target, the obtained historical trajectory of the specified target, the historical trajectory of other obstacles around the specified target, and the initial planning trajectory of the unmanned device itself can be input into the prediction model. Th...

AI Technical Summary

Problems solved by technology

[0004] Since the prediction model in the prior art directly fuses the characteristics of the initial planning trajectory with the characteristics of the historical motion trajectory, and does not consider the difference in the time dimension between the initial planning trajectory and the historical motion trajectory, the prediction model is not accurate enough for the movement of the specified target. Trajectory prediction is inaccurate

Images