An intelligent vehicle driving behavior personification decision-making method based on a driving prediction field and a BP neural network

Abstract

The invention discloses an intelligent vehicle driving behavior personification decision-making method based on a driving prediction field and a BP neural network. The method comprises the following steps: 1, dividing surrounding vehicle behaviors into 9 typical behaviors bi according to combination of transverse and longitudinal aspects, fitting a similarity track corresponding to each vehicle behavior bi, and setting an area through which a target vehicle passes according to the similarity track; 2, taking front and rear vehicles of the current lane of the intelligent vehicle and front and rear vehicles of an adjacent lane as surrounding vehicles, and enabling the intelligent vehicle to use V2V communication to obtain the position, speed and acceleration of each vehicle at each moment inreal time; 3, establishing a safety prediction field, an efficiency prediction field and a driving comfort prediction field; And step 4, enabling the intelligent vehicle to obtain the field intensitysum of the sub-prediction fields of each behavior bi in the driving prediction field in real time, inputting the field intensity sum into a BP neural network driving behavior decision model after normalization processing, and outputting a y vector to decode to obtain a most reasonable driving behavior decision result.

Description

technical field [0001] The invention belongs to the field of intelligent driving decision-making, and in particular relates to an anthropomorphic decision-making method for intelligent vehicle driving behavior based on a driving prediction field and a BP neural network. Background technique [0002] An intelligent driving vehicle is a vehicle with the ability to drive autonomously. In addition to being able to complete conventional car driving actions, it also has human-like behavior capabilities such as environment perception, behavior decision-making, motion planning, vehicle control, and automatic obstacle avoidance for traffic scenes. As an important manifestation of the intelligence level of unmanned vehicles, driving behavior decision-making has become the focus and difficulty of research by experts and scholars. For the complex behavior of driving a vehicle, drivers often rely on external sensory organs such as eyes and ears to collect road environment information, an...

AI Technical Summary

Problems solved by technology

[0003] At this stage, the research on driving behavior decision-making models usually classifies the driving traffic scene first and then makes behavior decisions for the driving scene. Although better decision-making results can be obtained for some specific driving scenes, traffic elements in real traffic environments There are many complex, random and uncertain factors, and the decision-making system cannot completely cover all possible driving scenarios, so such methods generally cannot achieve better robustness and adaptability

In addition, the existing intelligent vehicle driving behavior decision-making methods often focus on the ability of the vehicle to avoid driving hazards, while the normal rational driver's behavior decision-making when driving can be abstracted as a process of constantly pursuing the maximum benefit, such methods ignore the driving occupants' or passengers' expected benefits on other vehicle performances, such as efficiency, driving comfort, etc.

Images