Research on Autonomous Driving Algorithm Based on Multi-Agent
Download as PDF
The research of driverless cars is mainly about the interaction of the agent of a single car. However, there are few studies on the traffic congestion of multiple intelligent vehicles in ITS. Therefore, the advantages of driverless cars are used to solve specific traffic congestion problems. First of all, the state of the car driving in ITS can be divided into straight-line driving and when driving at an intersection, there are relatively few cooperative researches between car agents at the intersection. To a certain extent, the realization of driverless car technology can get rid of the traditional traffic light control technology at intersections. The control of traffic lights will cause frequent braking of the car and cause loss of car performance. At the same time, it also promotes irregular traffic post switching times and leads to serial congestion. Therefore, this article focuses on the intelligent body of the driverless car as the research object, and uses the car laser rangefinder to perceive the vehicle's position information and speed information within 200m. In order to avoid the traffic congestion problem caused by the use of traffic lights on the traditional traffic network, this paper adopts the principle of not stopping cars on the highway as much as possible, and researches the distance between the intelligent car at the intersection and the car on the straight road.
Autonomous driving algorithm, Multi-agent study, Calculation method