Vehicle trajectory prediction method based on vehicle behavior characteristics and environment interaction information

Abstract

The invention discloses a vehicle trajectory prediction method based on vehicle behavior characteristics and environment interaction information, and relates to the technical field of automatic driving vehicles, and the method comprises the steps: inputting a historical trajectory sequence of a target vehicle and a historical trajectory sequence of adjacent vehicles around the target vehicle into a vehicle trajectory prediction model, and outputting a predicted trajectory sequence of all vehicles in a scene, the vehicle track prediction model comprises a vehicle information dynamic coding module, a multi-information fusion module and a vehicle track prediction module. The multi-information fusion module is used for processing the dynamic sequence features of all vehicles in the scene to obtain environment interaction information among the vehicles in the scene, behavior features of each vehicle and a vehicle information coding result of a target vehicle, and then performing information fusion to obtain fusion features; according to the method, multi-factor fusion consideration is carried out on two factors which have great influence on the vehicle track, namely the vehicle behavior characteristics and the environment interaction information, so that the accuracy and the prediction time duration of the predicted vehicle track can be improved.

Description

technical field [0001] The invention relates to the technical field of self-driving vehicles, in particular to a vehicle trajectory prediction method based on vehicle behavior characteristics and environment interaction information. Background technique [0002] At present, self-driving vehicle technology and intelligent transportation systems have gradually entered a stage of vigorous development. This technology requires self-driving vehicles to have active behavior decision-making capabilities, and needs to complete driving and lane switching according to the traffic environment in a complex and dynamic traffic environment. , Accelerate to overtake, brake to decelerate and other driving behaviors. In the traffic environment where the vehicle is located, the dynamic driving behavior of the surrounding vehicles is an important factor affecting the decision-making of the driving vehicle behavior. If the future trajectory of the surrounding vehicles is predicted, the vehicle ...

AI Technical Summary

Problems solved by technology

However, as the number of traffic participants continues to increase, the road condition information becomes increasingly complex. In the entire traffic scene, motor vehicles, bicycles, pedestrians, and the road environment will all have an impact on autonomous driving vehicles. The existing prediction methods for these vehicle trajectories above Accuracy is still not ideal

Images