Path planning method and device

Abstract

The invention provides a path planning method and device, and the method comprises the steps: firstly determining a target driving region comprising a plurality of sub-regions, determining candidate sub-regions according to the predicted congestion values of all sub-regions at a preset time point, then determining a navigation path of the target region according to the travel starting point, travel terminal point and candidate sub-regions of a to-be-planned vehicle, and when the to-be-planned vehicle enters the current sub-region along the target region navigation path, determining a target lane navigation path according to the target region navigation path, the current position of the to-be-planned vehicle and the traffic flow load of each road section in the current sub-region. Accordingto the invention, a vehicle planning problem in a large range is divided into a plurality of vehicle planning problems in a small range, a large number of vehicle planning scenes can be satisfied, the traffic flow loads of each sub-region and each road section are integrated during vehicle planning, the effects of avoiding congestion sub-regions and realizing load balancing of the whole traffic network can be achieved at the same time. and therefore, the applicability is high.

Description

technical field [0001] The present application relates to the technical field of automatic driving, and in particular to a path planning method and device. Background technique [0002] When the self-driving vehicle is driving, it is necessary to plan the driving path for the self-driving vehicle, which involves the planning of each self-driving vehicle in a certain area. Existing planning methods can only satisfy small-scale path planning and single-scenario planning, and are not suitable for large-scale vehicle planning, and cannot adapt to changing application scenarios. In path planning, it can only achieve the shortest planning time for vehicles to be planned. The path or the shortest distance path cannot simultaneously avoid congestion areas in the traffic network and achieve load balancing from the perspective of the entire traffic network. [0003] Therefore, the existing path planning methods have technical problems of weak applicability and need to be improved. ...

AI Technical Summary

Problems solved by technology

Existing planning methods can only satisfy small-scale path planning and single-scenario planning, and are not suitable for large-scale vehicle planning, and cannot adapt to changing application scenarios. In path planning, it can only achieve the shortest planning time for vehicles to be planned. The path or the shortest distance path cannot simultaneously avoid congestion areas in the traffic network and achieve load balancing from the perspective of the entire traffic network

[0003] Therefore, the existing path planning methods have technical problems of weak applicability and need to be improved

Images