Road-Condition-Aware Dynamic Path Planning for Intelligent Vehicles

Abstract

Abstract Intelligent vehicles face a more complex driving environment, which requires their path planning algorithms to have the ability to respond quickly to complex environmental changes. Therefore, it is very necessary to plan the optimal path for smart vehicles in real-time. Recently, the convergence of graph dynamics and its equivalence with the shortest path solution have been proved, and the biased minimum consensus algorithm has been successfully applied to the shortest path planning problem. In this paper, considering the road conditions (degree of congestion, whether an accident occurred, etc.) and the Euclidean distance of the route, we improve biased minimum consensus algorithm to achieve optimal path planning in the complex situation of real-time updating of road conditions. In the simulation, we used the algorithm for path planning in the road topological maps of Beijing City, and compared with the RRT (Rapidly-exploring Random Trees Algorithm) algorithm and the BRRT (Bidirectional Rapidly-exploring Random Trees Algorithm) algorithm. The improved biased minimum consensus algorithm has better performance in real-time path planning problems. The research results shed new light on the real-time dynamic path planning in real traffic, which has practical significance.