Automatic overtaking trajectory planning method for unmanned vehicle

Abstract

The invention provides an automatic overtaking trajectory planning method for an unmanned vehicle. The automatic overtaking trajectory planning method comprises the following steps: historical overtaking data under a manual driving state are selected to establish an overtaking intent model according with human driver habits; under the unmanned driving state, whether the overtaking intent is generated or not is determined according to information collected by a vehicle-borne three-dimensional laser radar and a wheel speed sensor in real time; according to the collision time, whether the overtaking condition is met or not is judged; whether overtaking operation is executed or not is manually determined; overtaking trajectories meeting safety and comfort constraints are automatically planned;and the overtaking trajectories are dynamically updated and displayed on a driving cab display screen. Compared with the prior art, based on overtaking habits of a human driver, influences of the physical dimension and motion state of a front vehicle on the overtaking intent and trajectory planning of the vehicle are taken into full consideration, the optimal overtaking trajectory is selected, the overtaking accident is effectively avoided, and thus the automatic overtaking trajectory planning method is suitable for being used and popularized.

Description

technical field [0001] The invention relates to the field of intelligent driving and active safety, in particular to a method for planning an automatic overtaking trajectory of an unmanned vehicle. Background technique [0002] Overtaking is a common driving behavior in the process of vehicle driving. Relevant statistics show that road traffic accidents caused by overtaking account for about 20% of the total number of accidents every year, most of which are caused by human factors. During the overtaking process, due to the limitations of their own perception ability, it is difficult for human drivers to accurately obtain the surrounding driving environment information, and fail to plan the overtaking trajectory reasonably, thus causing traffic accidents. Unmanned vehicles can obtain rich environmental information through on-board perception sensors, and plan safe driving trajectories in real time in combination with changes in the driving conditions of the vehicle and surrou...

AI Technical Summary

Problems solved by technology

[0004] However, although the above two invention patents have established overtaking conditions and can dynamically generate overtaking trajectories, they ignore the overtaking habits and driving intentions of human drivers, and do not consider the influence of the geometry of the preceding vehicle on overtaking trajectory planning.

Images