基于智能路侧系统的高精度地图快速更新研究
Research on rapid update of highly automated driving map based on intelligent roadside system
王艳 1高凤飞 1孙宁 1宋娟 1韩增杨 2雷宇 1时慧1
作者信息
- 1. 北京车网科技发展有限公司,北京 102600
- 2. 北京汽车研究总院有限公司,北京 101300
- 折叠
摘要
针对高精度地图及其快速更新能力问题,提出一种实现高精度地图变化检测与更新的方法,旨在解决当前采用专业采集车方式,更新成本过高、周期长(季度更新)的问题.即在车路云一体化模式下,通过智能路侧系统输出检测范围内实时、稳定、高精度的传感器原始数据与感知结果数据,可实现在复杂场景城市路口提供实时的地图变化信息,结合云平台可为车辆提供高精度地图众源更新服务.有望实现以天为单位更新,甚至是小时、分钟级更新,缩减因未及时更新可能引起自动驾驶车辆的需要人工接管的概率、减少违反交规的行为、从而降低安全风险.
Abstract
This article proposes a method for achieving high-precision map change detection and update,aimed at addressing the key issue of achieving autonomous driving-high-precision maps and their fast update capabilities.The aim is to solve the problems of high update costs and long cycles(quarterly updates)currently using professional collection vehicles.In the vehicle road cloud integration mode,the intelligent roadside system outputs real-time,stable,and high-precision sensor raw data and perception result data within the detection range,which can provide real-time map change information for complex urban intersections.Combined with the cloud platform,it can provide high-precision map source update services for vehicles.It is expected to achieve daily updates,even hourly and minute level updates,reducing the probability of manual takeover of autonomous vehicles caused by delayed updates,reducing violations of traffic regulations,and thus reducing safety risks.
关键词
高精度地图/车路云一体化/智能路侧系统/地图变化信息/众源更新Key words
highly automated driving map/vehicle-road-cloud integration/intelligent roadside system/map change information/crowdsourcing update引用本文复制引用
基金项目
北京市科技计划项目(Z211100004221015)
科技创新2030-重大项目(2022ZD0116305)
出版年
2024
NETL
NSTL
万方数据