Method and system for predicting lane changing track of target vehicle on unstructured road

Abstract

The invention discloses a method and system for predicting a lane changing track of a target vehicle on an unstructured road. The method comprises the following steps: determining a virtual lane of an unstructured road and a vehicle lane changing behavior; obtaining driving data on the unstructured road; extracting vehicle track data according to the driving data; determining lane changing data based on the virtual lane, the vehicle lane changing behavior and the vehicle track data; training a Gaussian mixture-hidden Markov model through the lane changing data to obtain a driving intention recognition model, wherein the output of the driving intention recognition model is the lane changing intention of the vehicle; determining a possible lane changing terminal point according to the lane changing intention of the vehicle and the historical track of the vehicle; according to the current position of the vehicle and the possible lane changing end point, determining a possible vehicle lane changing track set based on the vehicle motion constraint and a polynomial curve model; and matching the possible vehicle lane changing track set with an actual track to obtain a vehicle prediction track. According to the method, the lane changing track with uncertain transverse and longitudinal displacement under the unstructured road can be predicted.

Description

technical field [0001] The invention relates to the technical field of vehicle trajectory prediction, in particular to a method and system for predicting a lane-changing trajectory of an unstructured road target vehicle. Background technique [0002] The unmanned driving system is mainly composed of environment perception module, behavior decision module, motion planning module and motion control module. With the continuous development of related technologies, the functions of each module are also increasing and perfecting. In the environmental perception module, one type of research focuses on predicting the trajectory of the target vehicle that affects the driving of the unmanned vehicle based on the information obtained by the environmental sensor or the vehicle-vehicle interaction system, so as to ensure that the unmanned vehicle can perform reasonable behavior. Make decisions and plan collision-free motion paths. [0003] Existing trajectory prediction methods are gen...

AI Technical Summary

Problems solved by technology

[0006] (2) Under unstructured road conditions, due to the lack of traffic rules and traffic sign references, the actual lane-changing trajectory of the vehicle is more uncertain than that of the urban scene

However, in unstructured road scenarios, since there are no lane lines on the road, the lateral displacement of lane-changing behavior is difficult to determine, which is affected by the road environment and the driver's personal habits.

Trajectory uncertainty brings new challenges to orbit prediction methods

In addition, most current trajectory prediction methods are based on historical trajectory data, and it is difficult to consider the constraints of the environment and vehicle kinematics on vehicle driving in the prediction.

Images