Intelligent vehicle route planning method based on threat assessment

Abstract

The invention discloses an intelligent vehicle route planning method based on threat assessment and relates to the technical field of intelligent vehicle decision making and control. The method comprises steps as follows: multiple factors such as external target features, environment parameters and the like influencing threat assessment of an intelligent vehicle are determined firstly, and a Bayesian network based intelligent vehicle threat assessment model is established; real-time data of the threat factors are collected in a running process of the intelligent vehicle, and intelligent vehicle threat indexes of targets in an outer environment are calculated according to the model; a repulsion potential field model of the targets on the intelligent vehicle is constructed according to the intelligent vehicle threat indexes of the targets and is fused with a gravitational potential field model of an intelligent vehicle traveling destination on the intelligent vehicle, a comprehensive potential field module of the intelligent vehicle is obtained, and route planning of the intelligent vehicle is obtained after solving. The multiple factors influencing threat assessment of the intelligent vehicle are taken into consideration, so that the driving environment of the intelligent vehicle is more sufficiently cognized, and a more reasonable and more effective traveling route is planned out.

Description

technical field [0001] The invention belongs to computer and automation technology, in particular to the field of intelligent vehicle decision-making and control technology, and specifically relates to a method for estimating intelligent vehicle threat by using Bayesian network and thereby realizing intelligent vehicle path planning. Background technique [0002] Path planning is an important part of intelligent vehicles and the basis for realizing automatic driving control of intelligent vehicles. Compared with mobile robots, intelligent vehicles face many factors such as high speed, dynamic complex environment, strict real-time and safety constraints, etc., so it is necessary to fully consider different types of road vehicles, pedestrians and pedestrians with different motion characteristics when planning the path. Obstacle details, planning a reasonable path to support intelligent vehicles to achieve safe and efficient automatic driving. [0003] Among the existing path ...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the problem that the existing vehicle path planning method does not consider the target type and motion characteristics when evaluating the threats of road vehicles, pedestrians, obstacles and other targets in the external environment, the invention proposes a method of using Bayesian network to detect threats to intelligent vehicles A method for estimating and implementing intelligent vehicle path planning. This method first constructs a Bayesian network model that includes target types, motion characteristics, and environmental factors to estimate the threats to targets such as road vehicles, pedestrians, and obstacles in the external environment. Index, and then establish the repulsion potential field model of the target in the external environment based on the threat index to the smart vehicle, combined with the gravitational potential field model of the smart vehicle’s driving end point to the smart vehicle, to obtain the comprehensive potential field model of the smart vehicle at the current position, In order to plan an intelligent vehicle path that can take into account both safety and efficiency to support autonomous driving

Images