Safe driving decision-making method for highly human-like automatic driving commercial vehicle

Abstract

The invention discloses a safe driving decision-making method for a highly human-like automatic driving commercial vehicle. The method comprises the following steps: firstly, synchronously collecting multi-source information of driving behaviors in a typical traffic scene, and constructing an expert track data set representing the driving behaviors of excellent drivers; secondly, comprehensively considering the influence of factors such as forward collision, backward collision, lateral collision, vehicle roll stability and driving smoothness on driving safety, simulating the driving behavior of an excellent driver by utilizing a generative adversarial imitation learning algorithm, and constructing a generator and a discriminator by respectively adopting a near-section strategy optimization algorithm and a deep neural network; and a safe driving decision model with a high human-like level is established. And finally, training the safe driving decision model to obtain safe driving strategies under different driving conditions. The driving intention of excellent human drivers can be simulated, a more reasonable and safer driving strategy is provided for the automatic driving operation vehicle, and the driving safety of the automatic driving operation vehicle can be effectively guaranteed.

Description

technical field [0001] The invention relates to a driving decision-making method for commercial vehicles, in particular to a highly human-like automatic driving commercial vehicle safe driving decision-making method, which belongs to the technical field of automobile safety. Background technique [0002] Commercial vehicles are the main undertakers of road transportation in my country, and they are also the main body of accidents involving mass casualties and casualties. In order to significantly improve traffic safety and transportation efficiency, advanced driving assistance and even fully unmanned commercial vehicle automatic driving technology have received high attention and development in recent years. [0003] Human-machine co-driving is the only way for the development of smart vehicles. As a key part of realizing high-quality automatic driving, driving decisions determine the safety and rationality of autonomous driving of operating vehicles in the process of human-...

AI Technical Summary

Problems solved by technology

[0007] Overall, existing "human-like" driving decision-making methods for small passenger vehicles cannot be directly applied to commercial vehicles

Research on safe driving decision-making for autonomous driving commercial vehicles is relatively scarce, especially in the research on safe driving decision-making of vehicles with a high human-like level, which is still blank

Images