A method for determining key conflict points to ensure the safety of automatic driving intersections

Abstract

The invention discloses a method for determining a key conflict point for ensuring traffic safety at an automatic driving intersection, and belongs to the field of intelligent traffic control. In the case of no signal control at the intersection of automatic driving, the automatic driving vehicle avoids the conflict by calculating the time to reach the conflict point on the route, and determines the key conflict point for the conflict point on the route, and the vehicle is controlled according to the key conflict point; first obtain the intersection information, Establish a Cartesian coordinate system, calculate the equation and conflict point coordinates of each path inside the intersection; update the conflict point time after the vehicle passes through the conflict point, perform iterative screening according to the updated time and conflict point position, and obtain the filtered conflict point set; Select a reference point in the set, judge whether to pass through each conflict point at the fastest speed from the reference point time, update the reference point time according to the passing situation, and determine the point at which the last update reference point time is the key conflict point. Improve control efficiency under the premise of safe passage.

Description

technical field [0001] The invention belongs to the field of intelligent traffic control, relates to the technical field of urban road traffic control for automatic driving vehicles passing through intersections, and more specifically relates to a method for determining key conflict points for ensuring the safety of automatic driving intersections. Background technique [0002] Many studies have shown that in the autonomous driving environment, compared with the signal-controlled intersection, the traffic efficiency of vehicles passing through the unsignalized intersection is higher. development trend. [0003] At a no-signal control intersection, multiple self-driving vehicles in different directions pass through the intersection, and there are bound to be conflicts. There are two main ways to avoid conflicts. One is to combine all vehicles that are about to enter the intersection in pairs, and judge whether the two How do conflicting vehicles safely pass through a conflic...

AI Technical Summary

Problems solved by technology

[0003] At a no-signal control intersection, multiple self-driving vehicles in different directions pass through the intersection, and there are bound to be conflicts. There are two main ways to avoid conflicts. One is to combine all vehicles that are about to enter the intersection in pairs, and judge whether the two How do conflicting vehicles safely pass through a conflict point? This method is effective when the traffic volume is small. When the traffic volume increases and there are many conflict points, the calculation times of the control model increase in a factorial order, and the calculation is difficult to meet the timeliness requirements. Even No feasible solution can be obtained. Under this control framework, many studies are based on the premise of constant vehicle speed. If the vehicle speed is considered to be optimized, the driving speed and the time to reach the conflict point are mutually constrained, and no feasible solution can be obtained; another control The method is to consider only one vehicle passing through all the conflict points on the path at a time. Before that, several vehicles have passed the intersection, and the time of each conflict point is updated. The vehicle only needs to reach the conflict after the time of each conflict point It can ensure that there will be no conflicts with other vehicles, but safely passing through a certain conflict point cannot guarantee safe passing through other conflict points on the path. Safety, there are two main problems: firstly, it causes the vehicle to adjust the speed frequently, which seriously affects the comfort; secondly, when actually solving the model, it is found that when there are many conflict points, the difficulty of model solution increases

[0005] After searching the literature of the existing technology, it is found that there are already literatures on "automatic driving intersections", which mainly study the traffic control methods of intersections in the autonomous driving environment, and there is no research specifically on the method of determining key conflict points

Images