Unified automatic driving transverse planning method and system

Abstract

Disclosed is a unified automatic driving transverse planning method. The method comprises the following steps: describing vehicle surrounding environment information in a unified mode; based on the unifiedly described vehicle surrounding environment information, planning an executable locus in real time in a driving process of a vehicle; executing the executable locus to guide the vehicle to run transversely; and through the previous three steps, realizing optimal local locus planning based on various constraints, performance indexes and driving behaviors. Disclosed is also a unified automatic driving transverse planning system. The system comprises an environment model sub-module, a planning algorithm sub-module and a vehicle control sub-module, wherein the environment model sub-module, according to environment perceiving information, in accordance with traffic rules and properties of traffic participants, constructs a unified local environment model, describes the vehicle surrounding environment information in a unified mode and reflects a potential dangerous degree around the vehicle; and the planning algorithm sub-module takes vehicle nonholonomic constraints and safety and comfortableness requirements into consideration, and based on the environment model, planning the executable locus in real time in the driving process of the vehicle.

Description

technical field [0001] The invention belongs to the technical field of automobiles and relates to an automobile automatic driving system, in particular to a horizontal planning technology of the automobile automatic driving system. Background technique [0002] Smart car technology has great potential in terms of travel safety, energy saving and environmental protection, and is considered to be an effective way to solve traffic congestion, reduce traffic accidents and improve environmental pollution. In the recent period, self-driving car technology has received widespread attention and has become one of the main directions of future car development. [0003] Generally speaking, an automatic driving system includes the following modules: environmental perception module, motion planning module, control execution module and human-machine interface module. The above four modules are crucial to the entire automatic driving system and directly affect the intelligence of the syste...

AI Technical Summary

Problems solved by technology

At present, there are many researches on lateral motion planning for autonomous driving at home and abroad, including traditional artificial potential field method, grid-based search method, sampling search method and discrete optimal method, etc. The above schemes have their own advantages, but there are still shortcomings. There is an urgent need for further improvement

Images