当前位置: 首页> 专利交易> 详情页
    待售中

    自动驾驶高精度地图的车道拾取算法[ZH]

    专利编号: ZL202606080601

    收藏

    拟转化方式: 转让;普通许可;独占许可;排他许可;作价投资;开放许可

    交易价格:面议

    专利类型:发明专利

    法律状态:授权

    技术领域:智能网联汽车

    发布日期:2026-06-08

    发布有效期: 2026-06-08 至 2038-11-19

    专利顾问 — 王老师

    电话咨询

    咨询电话

    13760886304

    专利基本信息
    >
    申请号 CN201811373474.5 公开号 CN109542994A
    申请日 2018-11-19 公开日 2019-03-29
    申请人 禾多科技(北京)有限公司 专利授权日期 2021-07-06
    发明人 郝文天;戴震 专利权期限届满日 2038-11-19
    申请人地址 100089 北京市海淀区阜外亮甲店1号恩济西园产业园21号楼一层21-14 最新法律状态 授权
    技术领域 智能网联汽车 分类号 G06F16/29
    技术效果 高效率 有效性 有效(授权、部分无效)
    专利代理机构 北京远大卓悦知识产权代理事务所(普通合伙) 11369 代理人 汤小东
    专利技术详情
    >
    01

    专利摘要

    本发明公开了一种自动驾驶高精度地图的车道拾取算法,包括:步骤1、若定位模块的定位精度下降,则进入应急检索状态;若定位模块为高精度定位,则进入步骤2;步骤2、判断是否进入局部搜索,若为局部搜索,则进入步骤6;若并非局部搜索,则进入步骤3瓦片搜索,得到车辆所处的位置范围以及位置范围内的街道ID后进入步骤4范围搜索,检索出车辆可能所处的街道的集合后进入步骤5粗糙搜索,找到车辆当前可能处于的街道集合;若未找到则进入步骤4扩大搜索范围搜索;若找到则进入步骤6精细搜索,得到车辆可能处于的车道集合中的首个元素即为最可能处于的车道。本发明可快速高效得从高精地图数据库中检索出当前车辆所处车道相关信息。
    展开 >
    02

    专利详情

    技术领域

    本发明涉及自动驾驶技术领域,尤其涉及一种自动驾驶高精度地图的车道拾取算
    法。

    背景技术

    随着自动驾驶技术的不断发展,以及商业落地场景需求的不断明晰,高精地图及
    其应用也扮演着越来越重要的角色,并服务于感知、定位、规划控制、决策以及人机交互等
    自动驾驶系统的各个模块。从技术角度来说,相对于传统地图,在提供路网拓扑关系的同
    时,高精地图具有更加丰富的车道级别的道路信息,不但为自动驾驶系统提供了重要的先
    验数据支持,也为保障行驶安全提供了强大的系统冗余。从商业落地角度来说,自动驾驶系
    统所依赖的各种传感器,总体性能不一,成本高昂,维护难度高,且易受到各种不同工况的
    影响,而高精地图可以为系统提供成本可控,易于更新且工作稳定的数据来源。

    因此,在自动驾驶技术逐渐由实验室研发过度到商业落地的今天,高精地图已经
    成为自动驾驶系统必不可少的一部分,并受到越来越多的关注。然而高精地图数据库丰富
    且巨大的信息,也为其数据检索带来了很大挑战。如何高效精准地检索出有效信息,对于实
    时性和安全性都要求极高的自动驾驶系统意义重大,也成为高精地图在实际应用中的重要
    技术瓶颈之一。具体来说,如何根据空间位置信息,快速高效地在高精地图数据库中定位到
    当前车辆行驶车道所对应的相关信息,实现高精度地图的车道拾取,是高精地图丰富数据
    信息得以充分利用的重要基础技术之一。相对于精确到街道级别的传统导航地图,自动驾
    驶高精地图能够精确到车道级别,有着更加丰富的信息,但也增大了数据检索的难度。

    发明内容

    本发明的一个目的是解决至少上述问题,并提供至少后面将说明的优点。

    本发明还有一个目的是提供一种自动驾驶高精度地图的车道拾取算法,可快速高
    效得从高精地图数据库中检索出当前车辆所处车道相关信息;并在定位模块工作异常时能
    继续提供车道级信息,直至定位模块恢复正常状态。

    为了实现根据本发明的这些目的和其它优点,提供了一种自动驾驶高精度地图的
    车道拾取算法,包括:

    步骤1、通过判断对车辆进行定位的定位模块的定位精度的高低,确定车道拾取的
    状态,若所述定位模块的定位精度下降,则所述车道拾取进入应急检索状态;若所述定位模
    块为高精度定位,则进入步骤2。

    步骤2、判断是否进入局部搜索,若为局部搜索,则进入步骤6;若并非局部搜索,则
    进入步骤3。

    步骤3、进入瓦片搜索,根据所述瓦片ID特点以及所述定位模块定位的空间坐标,
    得到所述车辆所处的位置范围以及所述位置范围内的街道ID后进入步骤4。

    步骤4、进入范围搜索,以所述车辆的当前位置为中心,结合所述街道ID检索出所
    述车辆可能所处的所述街道的集合后进入步骤5。

    步骤5、进入粗糙搜索,以所述车辆的所述定位模块定位的坐标是否位于所述街道
    集合的范围内为依据,找到所述车辆当前可能处于的街道集合;若未找到所述可能处于的
    街道集合,则进入步骤4扩大搜索范围继续搜索;若检索到所述可能处于的街道集合,则进
    入步骤6。

    步骤6、进入精细搜索,根据所述可能处于的街道集合,得到所述车辆当前可能处
    于的车道集合,结合所述定位模块定位的坐标至所述车道的中心线距离,以及所述坐标是
    否处于所述车道内,得到所述车辆可能处于的车道集合,所述车道集合中的首个元素即为
    最可能处于的所述车道。

    优选的是,步骤1中所述应急检索状态下的具体算法为:

    根据车道线检测结果实时计算出所述车辆距离两侧所述车道线的距离,结合车道
    线检测惯性导航数据信息计算出所述车辆偏离上一时刻的位置信息,分析得出所述车辆是
    否在发生变道,以及变道的方向,以得出所述车辆当前所处所述车道。

    优选的是,步骤2中所述局部检索状态下的具体算法为:

    根据所述车辆上一时刻得到所处的街道,结合所述街道前后连接关系,找出当前
    所述车辆可能处于的街道集合,进而得到所述车辆可能所处的车道集合,并进入步骤6。

    优选的是,步骤1中若在一定时间窗口范围内所述定位精度下降的所述定位模块
    恢复所述高精度定位,则退出所述应急检索状态进入步骤2;若超过所述时间窗口,仍未能
    接收到有效的所述高精度定位的信息,则通知控制系统所述定位模块定位出现异常,并进
    一步提示驾驶员接管。

    优选的是,步骤6中所得结果还包括对所述最可能处于的车道的评价,所述评价为
    结果可信、不确定或者不可信。

    优选的是,步骤1中若所述定位模块为所述高精度定位,则进入全局检索,在确定
    所述车辆正常行驶在高精度地图区域的时候,可切换至所述局部检索模块。

    优选的是,步骤3中所述定位模块定位的空间坐标和步骤5、步骤6中的所述定位模
    块定位的坐标,均为所述车辆当前的坐标。

    本发明至少包括以下有益效果:

    本发明根据高精地图数据结构特点,结合自动驾驶系统实际应用场景,以及对系
    统安全性的要求,通过提出一套结合全局自顶向下的分级检索,局部先验信息快速检索以
    及定位精度下降工况下应急检索的综合解决策略,保证车辆行驶过程中,一定时间范围内
    出现定位精度下降的异常情况时,通过进入应急检索状态,仍能使系统在一定的时间内正
    常工作,以降低或避免定位精度下降造成的麻烦;所述算法是基于地图数据结构的全局检
    索算法,其中,针对自动驾驶优化的局部快速检索,保障系统冗余的应急检索,以及保障系
    统安全性的异常退出机制,充分考虑到自动驾驶系统对于检索速度、系统负荷以及安全冗
    余的要求,是一套综合完整的专为自动驾驶定制的高精地图车道级数据检索策略。并充分
    考虑到其他模块可能的应用需求,相对于仅返回一个简单唯一的结果而言,通过返回一个
    当前车辆可能所处车道的集合,按照可能性大小排列并给出相应评价更为准确。所述算法
    根据车辆高精度车道级定位信息,快速高效得从高精地图数据库中检索出当前车辆所处车
    道相关信息。

    本发明的其它优点、目标和特征将部分通过下面的说明体现,部分还将通过对本
    发明的研究和实践而为本领域的技术人员所理解。

    附图说明

    图1为本发明所述自动驾驶高精度地图的车道拾取算法的流程图。

    具体实施方式

    下面结合附图对本发明做进一步的详细说明,以令本领域技术人员参照说明书文
    字能够据以实施。

    应当理解,本文所使用的诸如“具有”、“包含”以及“包括”术语并不排除一个或多
    个其它元件或其组合的存在或添加。

    如图1所示,本发明提供一种自动驾驶高精度地图的车道拾取算法,包括:

    步骤1、通过判断对车辆进行定位的定位模块的定位精度的高低,确定车道拾取的
    状态,若所述定位模块的定位精度下降,则所述车道拾取进入应急检索状态;若所述定位模
    块为高精度定位,则进入步骤2。

    步骤2、判断是否进入局部搜索,若为局部搜索,则进入步骤6;若并非局部搜索,则
    进入步骤3。

    步骤3、进入瓦片搜索,根据所述瓦片ID特点以及所述定位模块定位的空间坐标,
    得到所述车辆所处的位置范围以及所述位置范围内的街道ID后进入步骤4。

    步骤4、进入范围搜索,以所述车辆的当前位置为中心,结合所述街道ID检索出所
    述车辆可能所处的所述街道的集合后进入步骤5。

    步骤5、进入粗糙搜索,以所述车辆的所述定位模块定位的坐标是否位于所述街道
    集合的范围内为依据,找到所述车辆当前可能处于的街道集合;若未找到所述可能处于的
    街道集合,则进入步骤4扩大搜索范围继续搜索;若检索到所述可能处于的街道集合,则进
    入步骤6。

    步骤6、进入精细搜索,根据所述可能处于的街道集合,得到所述车辆当前可能处
    于的车道集合,结合所述定位模块定位的坐标至所述车道的中心线距离,以及所述坐标是
    否处于所述车道内,得到所述车辆可能处于的车道集合,所述车道集合中的首个元素即为
    最可能处于的所述车道。

    在上述方案中,根据高精地图数据结构特点,结合自动驾驶系统实际应用场景,以
    及对系统安全性的要求,通过提出一套结合全局自顶向下的分级检索,局部先验信息快速
    检索以及定位精度下降工况下应急检索的综合解决策略,保证车辆行驶过程中,一定时间
    范围内出现定位精度下降的异常情况时,通过进入应急检索状态,仍能使系统在一定的时
    间内正常工作,以降低或避免定位精度下降造成的麻烦;所述算法是基于地图数据结构的
    全局检索算法,其中,针对自动驾驶优化的局部快速检索,保障系统冗余的应急检索,以及
    保障系统安全性的异常退出机制,充分考虑到自动驾驶系统对于检索速度、系统负荷以及
    安全冗余的要求,是一套综合完整的专为自动驾驶定制的高精地图车道级数据检索策略。
    并充分考虑到其他模块可能的应用需求,相对于仅返回一个简单唯一的结果而言,通过返
    回一个当前车辆可能所处车道的集合,按照可能性大小排列并给出相应评价更为准确。所
    述算法根据车辆高精度车道级定位信息,快速高效得从高精地图数据库中检索出当前车辆
    所处车道相关信息。

    相较于单一策略检索或者针对其他应用场景的地图检索策略,充分考虑到了自动
    驾驶系统对于检索速度,系统负荷以及安全冗余的要求,是一套综合完整的专为自动驾驶
    定制的高精地图车道级数据检索算法,其中,高精地图数据一般按照地理空间范围分级切
    分存储,按照空间范围由大到小的包含关系,依次为Tile(瓦片)、Link(街道)、Lane(车道)、
    以及Lane Mark(车道线)。每一组数据都有一个全局唯一的ID,根据该ID可以在数据库中检
    索出其对应的信息。具体来说,高精地图数据由若干分级的瓦片组成,每个瓦片包含其对应
    属性以及若干街道ID,每个街道包含其对应属性以及若干车道ID,每个车道包含其对应属
    性以及其对应的车道线ID,每个车道线包含其对应属性。

    一个优选方案中,步骤1中所述应急检索状态下的具体算法为:

    根据车道线检测结果实时计算出所述车辆距离两侧所述车道线的距离,结合车道
    线检测惯性导航数据信息计算出所述车辆偏离上一时刻的位置信息,分析得出所述车辆是
    否在发生变道,以及变道的方向,以得出所述车辆当前所处所述车道。

    在上述方案中,惯性导航的基本工作原理是以牛顿力学定律为基础,通过测量载
    体在惯性参考系的加速度,将它对时间进行积分,且把它变换到导航坐标系中,就能够得到
    在导航坐标系中的速度、偏航角和位置等信息;通过实时计算车辆距离两侧车道线的距离,
    结合惯性导航系统给出的车辆对比当前的上一时刻的位置信息,可得出车辆有没有脱离上
    一时刻行驶的车道,以及变车道的方向,再结合定位模块的定位坐标得到车辆当前所处车
    道,将车辆惯性导航数据与实时算法结合,保证定位模块定位精度下降的情况下的定位的
    准确度。

    一个优选方案中,步骤2中所述局部检索状态下的具体算法为:

    根据所述车辆上一时刻得到所处的街道,结合所述街道前后连接关系,找出当前
    所述车辆可能处于的街道集合,进而得到所述车辆可能所处的车道集合,并进入步骤6。

    在上述方案中,通过结合上一时刻的车辆定位的数据信息,在结合已知的前后街
    道的连接关系,可快速找出当前车辆可能处于的街道的集合,从而缩小了街道集合的范围,
    便于更块定位可能所处的车道集合。

    一个优选方案中,步骤1中若在一定时间窗口范围内所述定位精度下降的所述定
    位模块恢复所述高精度定位,则退出所述应急检索状态进入步骤2;若超过所述时间窗口,
    仍未能接收到有效的所述高精度定位的信息,则通知控制系统所述定位模块定位出现异
    常,并进一步提示驾驶员接管。

    在上述方案中,应急检索状态适用于定位模块一定时间内的定位精度的下降,并
    不能一直处于应急搜索状态,超出设定的时间结点,定位模块的定位精度仍为恢复的话,控
    制系统无法接收到准确有效的定位信息,控制系统即确定定位模块出现功能异常,则退出
    定位机制,由驾驶员自己通过其他办法确认车道或再次重启定位机制,避免了定位偏差带
    来的不必要的麻烦。

    一个优选方案中,步骤6中所得结果还包括对所述最可能处于的车道的评价,所述
    评价为结果可信、不确定或者不可信。

    在上述方案中,通过在给出最可能车道的同时,并对最可能的车道的这种可能性
    给予评价,以供驾驶员参考,定位的最终结果是否可信、不确定或者是不可信,以便驾驶员
    最最终判断,或是否进行二次搜索定位。

    一个优选方案中,步骤1中若所述定位模块为所述高精度定位,则进入全局检索,
    在确定所述车辆正常行驶在高精度地图区域的时候,可切换至所述局部检索模块。

    在上述方案中,步骤1中根据高精度定位信息,全局检索可以精确检索出当前车所
    处的车道,但计算量较大,在完成全局检索,并且确定车辆正常行驶在高精地图区域的时
    候,可以切换到局部检索模式,以减少系统运算负荷,保证检索速度。

    一个优选方案中,步骤3中所述定位模块定位的空间坐标和步骤5、步骤6中的所述
    定位模块定位的坐标,均为所述车辆当前的坐标

    在上述方案中,步骤3中所述定位模块定位的空间坐标为瓦片搜索状态下所述车
    辆所处的空间坐标;步骤5中所述定位模块定位的坐标为粗糙搜索状态下的所述车辆的所
    处的坐标;步骤6中所述定位模块定位的坐标为精细搜索状态下的所述车辆的所处的坐标;
    每个搜索状态下均以当前的车辆的坐标为参考,以保证最终搜索结果的相对准确度。

    尽管本发明的实施方案已公开如上,但其并不仅仅限于说明书和实施方式中所列
    运用,它完全可以被适用于各种适合本发明的领域,对于熟悉本领域的人员而言,可容易地
    实现另外的修改,因此在不背离权利要求及等同范围所限定的一般概念下,本发明并不限
    于特定的细节和这里示出与描述的图例。

    自动驾驶高精度地图的车道拾取算法

    Technical Field

    The invention relates to an automatic driving technical field, in particular to a kind of automatic drive high precision map lane pick-up algorithm.

    Background Art

    With the continuous development of automatic drive technology, and the needs of commercial floor scene continuously clear, high-precision map and its application also are playing an increasingly important role, and services the perception, positioning, planning control, decision-making and human-computer interaction and the like of the automatic driving system of each module. From a technical standpoint, relative to the traditional map, to provide network topological relation of at the same time, the high-precision map has more abundant lane level road information, not only for the automatic drive system provides important support of a priori data, also for the protection of the driving safety and provides a powerful system redundancy. The ground from a commercial perspective, the automatic drive system relied on various sensor, the overall performance of the different, high cost, high maintenance difficulty, and vulnerable to the effects of various different working conditions, while the high-precision map provides the system cost can be controlled, easy to work and renewal of stable data sources.

    Therefore, in the automatic drive technology gradually by the laboratory research over to the commercial floor today, high-precision map has become automatic drive system an essential part of, and the subject of increasing concern. However the high-precision map database is abundant and the immense information, also for its data retrieval brings great challenges. How to efficiently and accurately the effective retrieval of information, real time and safety are all extremely high requirements for an automatic driving system is of great significance, as well as the high-precision map in the practical application of one of the important technical bottleneck. In particular, how according to the space position information, fast and efficient in high-precision map database positioning to the current vehicle traveling lane corresponding to the relevant information, to realize high precision map lane pick-up, high-precision map is rich data information can be fully utilized an important basis of one of the technology. With respect to the accurate to the street level of the traditional navigation map, the automatic drive high-precision map can be accurate to the lane level, there are more abundant information, but also increases the difficulty of data retrieval.

    Content of the invention

    One of the purposes of the present invention is to solve the at least the above-mentioned problems, and to provide at least the rear and the advantages of the note.

    The invention has another purpose is to provide an automatic driving high precision map lane pick-up algorithm, can be quickly and efficiently obtained from high-precision map database to search out the current vehicle lane at the relevant information; and the positioning module abnormal working state can continue to provide lane level information, until the positioning module resumes the normal state.

    In order to achieve these purposes according to the present invention and other advantages, provides an automatic driving high precision map lane pick-up algorithm, including:

    Step 1, by judging the vehicle for positioning of the positioning accuracy of the height of the module, determining the state of the lane pick-up, if the locating module of positioning accuracy is lowered, then the states the traffic lane pick-up into the emergency retrieval state; if the positioning module is a high-precision positioning, then enter the step 2.

    Step 2, judges whether to enter the local search, if it is a local search, enter step 6; and if the non-local search, then enter the step 3.

    Step 3, into the tile search, according to the ID [...] characteristics of said positioning module and positioning space coordinate, to obtain the position of the range of the vehicle and the position of the ID within the streets after entering step 4.

    Step 4, enter the range search, to the current position of a vehicle as the center, of a combination of the street ID search of the vehicle may be at the position of the street after the set in the step 5.

    Step 5, into the rough search, to the position of the vehicle are the coordinates of the positioning module located within the range of the street is set according to, find the vehicle present may be at street set; and if not found the may be in street set, then enter the step 4 to expand the scope of search to continue to search; if the search to the may be in street set, then enter the step 6.

    Step 6, the fine search, according to the stated may be in street set, to obtain the current vehicle may be in the lane of the set, a combination of the positioning module positioning coordinate to the lane of the distance between the center lines, and the coordinate is in the lane, to obtain the vehicle may be in the lane of the set, the lane in the set of the first element is the most may be in said lane.

    Preferably, the step 1 in the emergency search the status of the specific algorithm is:

    The lane detection result according to the real-time calculated vehicle distance on both sides of the lane of the distance, combined with the lane detecting inertial navigation data information calculated at a last time the vehicle deviates from the position information, the vehicle whether the analysis in the event of a lane change, and the direction of the lane change, in order to obtain said vehicle that said current the lane.

    Preferably, step 2 in the local search the status under the specific algorithm is:

    According to the stated vehicle at a last time is obtained at the street, a combination of the street front and rear connection relationship, find the present the vehicle may be in street set, and then to obtain the vehicles may be at the position of the lane set, and proceed to step 6.

    Preferably, the step 1 in a certain time window if within the scope of the decline in the positioning accuracy of the positioning module to restore the high-precision positioning, is exiting the emergency search the status enter step 2; if more than the time window, still fails to be received effectively the high precision positioning information, notify the control system positioning said positioning module is abnormal, and further prompt the driver to take over.

    Preferably, step 6 in the results also includes [...] may be in the lane of the evaluation, the evaluation results are credible, uncertain or untrusted.

    Preferably, the step 1 if states positioning module for the high-precision positioning, then enter the full board retrieval, in the determination of the normal running of the vehicle in the high precision of the map area, can be switched to the local search module.

    Preferably, step 3 in the positioning of the locating module space coordinate and step 5, step 6 in the positioning module positioning coordinate, are the coordinates of the current vehicle.

    The invention comprises at least the following advantages:

    The invention according to the characteristics of high-precision map data structure, combination automatic driving system actual application scene, and to the safety of the system requirements, through a set of combined overall from top hierarchical search, some a priori information fast retrieval and positioning accuracy is lowered under the operating condition of the emergency retrieval of a comprehensive solution to the strategy, ensure that during running of the vehicle, a certain period of time in the range of of positioning accuracy is lowered when the abnormal condition, by entering into a emergency search the status, can still make the system work normally in a certain period of time, in order to reduce or avoid the trouble caused by lowering of a position precision; said algorithm is based on the map data structure of the overall search algorithm, wherein in response to the automatic drive optimized partial fast retrieval, security system redundant emergency retrieval, and to protect the safety of the system abnormal withdrawal mechanism, fully taking into account the automatic driving system for the retrieval speed, system load requirements and safety redundancy, is a comprehensive and complete set of designed specifically for automatic drive customized high-precision map lane level data search strategies. To take fully into account other module possible application requirements, only a simple relative to the return in the case of the only result, by returning a current vehicle may be set at the lane, sized and arranged in accordance with the possibility and to provide the corresponding evaluation is more accurate. The algorithm according to the vehicle lane level positioning information with high precision, high speed and high efficiency obtained from the high-precision map database to search out the current vehicle lane at the relevant information.

    Other advantages of the present invention, objectives and features will reflect through the lower part of the note, segment will also be through to the study and practice of this invention is in the field of the technical understood.

    Description of drawings

    Figure 1 is the flow chart of the present invention the automatic driving high precision map lane pick-up algorithm.

    Mode of execution

    The Figure below to the further detailed description of this invention, in order to make the technical personnel in the field specification can be on the basis of the implementation of the reference characters.

    It should be understood, used herein such as "has", "comprising" and "including" terminology does not exclude one or a plurality of other components or a combination thereof the presence or added.

    As shown in Figure 1, the invention provides an automatic driving high precision map lane pick-up algorithm, including:

    Step 1, by judging the vehicle for positioning of the positioning accuracy of the height of the module, determining the state of the lane pick-up, if the locating module of positioning accuracy is lowered, then the states the traffic lane pick-up into the emergency retrieval state; if the positioning module is a high-precision positioning, then enter the step 2.

    Step 2, judges whether to enter the local search, if it is a local search, enter step 6; and if the non-local search, then enter the step 3.

    Step 3, into the tile search, according to the ID [...] characteristics of said positioning module and positioning space coordinate, to obtain the position of the range of the vehicle and the position of the ID within the streets after entering step 4.

    Step 4, enter the range search, to the current position of a vehicle as the center, of a combination of the street ID search of the vehicle may be at the position of the street after the set in the step 5.

    Step 5, into the rough search, to the position of the vehicle are the coordinates of the positioning module located within the range of the street is set according to, find the vehicle present may be at street set; and if not found the may be in street set, then enter the step 4 to expand the scope of search to continue to search; if the search to the may be in street set, then enter the step 6.

    Step 6, the fine search, according to the stated may be in street set, to obtain the current vehicle may be in the lane of the set, a combination of the positioning module positioning coordinate to the lane of the distance between the center lines, and the coordinate is in the lane, to obtain the vehicle may be in the lane of the set, the lane in the set of the first element is the most may be in said lane.

    In the above-mentioned scheme, according to the characteristics of high-precision map data structure, combination automatic driving system actual application scene, and to the safety of the system requirements, through a set of combined overall from top hierarchical search, some a priori information fast retrieval and positioning accuracy is lowered under the operating condition of the emergency retrieval of a comprehensive solution to the strategy, ensure that during running of the vehicle, a certain period of time in the range of of positioning accuracy is lowered when the abnormal condition, by entering into a emergency search the status, can still make the system work normally in a certain period of time, in order to reduce or avoid the trouble caused by lowering of a position precision; said algorithm is based on the map data structure of the overall search algorithm, wherein in response to the automatic drive optimized partial fast retrieval, security system redundant emergency retrieval, and to protect the safety of the system abnormal withdrawal mechanism, fully taking into account the automatic driving system for the retrieval speed, system load requirements and safety redundancy, is a comprehensive and complete set of designed specifically for automatic drive customized high-precision map lane level data search strategies. To take fully into account other module possible application requirements, only a simple relative to the return in the case of the only result, by returning a current vehicle may be set at the lane, sized and arranged in accordance with the possibility and to provide the corresponding evaluation is more accurate. The algorithm according to the vehicle lane level positioning information with high precision, high speed and high efficiency obtained from the high-precision map database to search out the current vehicle lane at the relevant information.

    Compared with the single strategy search or to other application scene map search strategies, fully taking into account the automatic driving system for the retrieval speed, system load requirements and safety redundancy, is a comprehensive and complete set of designed specifically for automatic drive customized high-precision map lane level data retrieval algorithm, wherein high-precision map data generally in accordance with the geographical space range hierarchical segmentation memory, in accordance with the spatial range from large to small relation of inclusion, is successively Tile (tile), Link (street), Lane (lane), and Lane Mark (lane). Each group of data have a globally unique ID, according to the ID in the database can be retrieved in the corresponding information. Specifically, the high-precision map data of a by a plurality of tiles, each tile comprising a plurality of streets and its corresponding attribute ID, each containing the same and they are advised to attribute ID corresponding to a plurality of lanes, each lane containing the same and corresponding to the attribute ID of the corresponding lane, each lane line includes its corresponding attribute.

    In one preferred embodiment, step 1 in the emergency search the status of the specific algorithm is:

    The lane detection result according to the real-time calculated vehicle distance on both sides of the lane of the distance, combined with the lane detecting inertial navigation data information calculated at a last time the vehicle deviates from the position information, the vehicle whether the analysis in the event of a lane change, and the direction of the lane change, in order to obtain said vehicle that said current the lane.

    In the above-mentioned scheme, inertial navigation of the basic operating principle is based on the law of Newton's mechanics, by measuring the carrier in the inertial frame of reference acceleration, it to the time integral, and converts these to the navigation coordinate system, can be obtained in the speed of the navigation coordinate system, information such as the yaw angle and the position of the; vehicle distance through the real-time calculation of the distance between the two sides of the lane, combined inertial navigation system provided on the vehicle of a time compared with the current position information of the, can obtain the vehicle there from at a last time the running lane, and change the direction of the lane, a combination with the positioning module of the vehicle and the positioning coordinate on the present lane, the vehicle inertial navigation data and real-time algorithm combined, ensure the positioning accuracy of the positioning module of the decline in the accuracy of location.

    In one preferred embodiment, step 2 in the local search the status under the specific algorithm is:

    According to the stated vehicle at a last time is obtained at the street, a combination of the street front and rear connection relationship, find the present the vehicle may be in street set, and then to obtain the vehicles may be at the position of the lane set, and proceed to step 6.

    In the above-mentioned scheme, by combining the last time vehicle positioning data information, in conjunction with known connection relationship of the front and rear of the street, can quickly find the current vehicle may be in of a street set, thereby narrowing the scope of the street set, block positioning may be more convenient to the lane at the set.

    In one preferred embodiment, step 1 in a certain time window if within the scope of the decline in the positioning accuracy of the positioning module to restore the high-precision positioning, is exiting the emergency search the status enter step 2; if more than the time window, still fails to be received effectively the high precision positioning information, notify the control system positioning said positioning module is abnormal, and further prompt the driver to take over.

    In the above-mentioned scheme, emergency search the status suitable for positioning module within a certain time of the decline of the positioning accuracy, did not always is in emergent search state, exceeds the setting time node, the positioning accuracy of the positioning module to restore it still, control system cannot receive the accurate and effective positioning information, control system determine the positioning module appear abnormal function, withdrawn from the positioning mechanism, by the driver through the other recognize the own lane or once again restart the positioning mechanism, avoiding the positioning deviation of the trouble is not necessary.

    In one preferred embodiment, step 6 in the results also includes [...] may be in the lane of the evaluation, the evaluation results are credible, uncertain or untrusted.

    In the above-mentioned scheme, most likely through lane of given at the same time, and the lane of the most likely possibility of given to the evaluation, in order to for the driver reference, whether positioning the final outcome of the trusted, uncertain or not credible, the driver seat to the most final judgment, or whether the secondary search positioning.

    In one preferred embodiment, step 1 if states positioning module for the high-precision positioning, then enter the full board retrieval, in the determination of the normal running of the vehicle in the high precision of the map area, can be switched to the local search module.

    In the above-mentioned scheme in, step 1 in accordance with the high-precision positioning information, the whole board retrieval can be accurately retrieved at the present vehicle lane, but the computational complexity to be relatively large, the overall situation in the completion of the search, and the determination of the normal running of the vehicle at the time of high-precision map region, can switch to the local search mode, in order to reduce the operation load of the system, to ensure that the retrieval speed.

    In one preferred embodiment, step 3 in the positioning of the locating module space coordinate and step 5, step 6 in the positioning module positioning coordinate, are the coordinates of the current vehicle

    In the above-mentioned scheme in, step 3 in the positioning of the locating module space coordinate tile under the search for the state of the vehicle at the space coordinate; step 5 in the locating module positioning coordinate is rough search for the state of the vehicle under the coordinates of the of location; step 6 in the positioning of the locating module coordinates the fine search for the state of the vehicle under the coordinates of location; each of the search condition to the current of the vehicle are coordinate is the reference, in order to guarantee the relative accuracy of the final search result.

    Although the embodiments of the invention have been disclosed above, but not limited to the specification and embodiments set out in the application, it is fully can be suitable for various suitable for the field of this invention, familiar in the case of the field personnel, can be easily achieved in addition changes, therefore without departing from the claims and the equivalent of the scope of the general concept of defined, the invention is not limited to the specific details shown and described here of the legend.

    Lane pickup algorithm for autopilot high-precision map
    展开 >
    说明书附图
    >
    交易服务流程
    >

    挑选中意的板块

    ----

    客服确认选择专利的交易信息和价格并支付相应款项

    办理转让材料

    ----

    协助双方准备相应的材料

    签订协议

    ----

    协助卖家签订协议

    办理备案手续

    ----

    买卖双方达成一致后

    交易完成

    ----

    交易完成可投入使用

    过户资料 & 安全保障 & 承诺信息
    >

    过户资料

    买卖双方需提供的资料
    公司 个人
    买家 企业营业执照
    企业组织机构代码证
    身份证
    卖家 企业营业执照
    专利证书原件
    身份证
    专利证书原件
    网站提供 过户后您将获得
    专利代理委托书
    专利权转让协议
    办理文件副本请求书
    发明人变更声明
    专利证书
    手续合格通知书
    专利登记薄副本

    安全保障

    承诺信息

    我方拟转让所持标的项目,通过中国汽车知识产权应用促进中心公开披露项目信息和组织交易活动,依照公开、公平、公正和诚信的原则作如下承诺:

    1、本次项目交易是我方真实意思表示,项目标的权属清晰,除已披露的事项外,我方对该项目拥有完全的处置权且不存在法律法规禁止或限制交易的情形;
    2、本项目标的中所涉及的处置行为已履行了相应程序,经过有效的内部决策,并获得相应批准;交易标的涉及共有或交易标的上设置有他项权利,已获得相关权利 人同意的有效文件。
    3、我方所提交的信息发布申请及相关材料真实、完整、准确、合法、有效,不存在虚假记载、误导性陈述或重大遗漏;我方同意平台按上述材料内容发布披露信息, 并对披露内容和上述的真实性、完整性、准确性、合法性、有效性承担法律责任;
    4、我方在交易过程中自愿遵守有关法律法规和平台相关交易规则及规定,恪守信息发布公告约定,按照相关要求履行我方义务;
    5、我方已认真考虑本次项目交易行为可能导致的企业经营、行业、市场、政策以及其他不可预计的各项风险因素,愿意自行承担可能存在的一切交易风险;
    6、我方在平台所组织交易期间将不通过其他渠道对标的项目进行交易;
    7、我方将按照平台收费办法及相关交易文件的约定及时、足额支付相关费用,不因与受让方争议或合同解除、终止等原因拒绝、拖延、减少交纳或主张退还相关费用。