Automatic driving autonomous lane changing decision method and device and autonomous vehicle

Abstract

The invention discloses an automatic driving autonomous lane changing decision method and device and an autonomous vehicle. The automatic driving autonomous lane changing decision method comprises thefollowing steps of step 100, updating vehicle parameters; step 200, judging whether a lane changing intention exists or not; and if the lane changing intention exists, entering a step 300; and if thelane changing intention does not exist, entering a step 400; step 300, calculating lane changing benefit value; and step 400, outputting a lane changing decision result, wherein the lane changing decision result comprises not changing the lane, changing to a left lane and changing to a right lane. The automatic driving autonomous lane changing decision method has the advantages of simple rules, clear logic, good adaptability and easily adjusted parameters. Under complicated road environments, the autonomous vehicle sufficiently uses lane changing experiences of human drivers, so that the lanechanging efficiency and benefit are furthest increased on the premise of guaranteeing the safety.

Description

technical field [0001] The present invention relates to the technical field of automatic driving, in particular to an automatic driving autonomous lane change decision method and device and an automatic driving vehicle. Background technique [0002] Autonomous driving has gradually become a hot topic in recent years. Many universities, traditional car companies and Internet companies have started research, and it has developed to a certain level. Research results at home and abroad have predicted that autonomous driving will bring subversive changes in areas such as alleviating traffic congestion, improving road safety, and reducing air pollution. Autonomous driving comprehensively uses advanced technologies such as visual analysis, artificial intelligence, and distributed computing to enable vehicles to have functions such as autonomous cruise, traffic sign recognition, and collision warning. Among them, autonomous vehicle lane changing is one of its important research cont...

AI Technical Summary

Problems solved by technology

[0007] The rule-based lane-changing model has complex judgment rules and numerous constraints, resulting in poor portability of the algorithm, and thus cannot satisfy the lane-changing logic in various scenarios.

And the existing algorithm only considers the state of the self-vehicle and surrounding obstacles, but does not take the geometric structure of the road as a decision-making factor, which obviously does not conform to the driving habits of human drivers.

Images