具体实施方式

下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述。显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有作出创造性劳动的前提下所获得的所有其他实施例，都属于本发明保护的范围。

此处所称的“一个实施例”或“实施例”是指可包含于本发明至少一个实现方式中的特定特征、结构或特性。在本发明的描述中，需要理解的是，术语“上”、“下”、“顶”、“底”等指示的方位或位置关系为基于附图所示的方位或位置关系，仅是为了便于描述本发明和简化描述，而不是指示或暗示所指的装置或元件必须具有特定的方位、以特定的方位构造和操作，因此不能理解为对本发明的限制。此外，术语“第一”、“第二”仅用于描述目的，而不能理解为指示或暗示相对重要性或者隐含指明所指示的技术特征的数量。由此，限定有“第一”、“第二”的特征可以明示或者隐含的包括一个或者更多个该特征。而且，术语“第一”、“第二”等是用于区别类似的对象，而不必用于描述特定的顺序或先后次序。应该理解这样使用的数据在适当情况下可以互换，以便这里描述的本发明的实施例能够以除了在这里图示或描述的那些以外的顺序实施。

需要说明的是，参考图1，本发明依据的高精地图道路数据模型是结合导航电子地图和高精地图生产规范制定的满足自动驾驶需求的数据模型，包括道路参考线、车道边界线、车道中心线、地面和地上道路附属设施、停车区域和限速标识等。

本发明后述中所提到的拓扑关系主要指参考线、车道线和车道中心线的连接关系；关联关系主要指交通附属设施与道路或车道的关系，如停止线与车道间的逻辑关联关系，道路边界与参考线的逻辑关联关系等；待处理路口为规则路口，是指能够参数化的路口。

参阅图2，其所示为本发明实施例提供的一种路口高精地图生成方法的流程示意图。本说明书提供了如实施例或流程图所述的方法操作步骤，但基于常规或者无造性的劳动可以包括更多或者更少的操作步骤。实施例中列举的步骤顺序仅仅为众多步骤执行顺序中的一种方式，不代表唯一的执行顺序。在实际中的系统装置或产品执行时，可以按照实施例或者附图所示的方法顺序执行或者并行执行（例如并行处理器或者多线程处理的环境）。本发明实施例提供的路口高精地图生成方法包括：

S201，基于待处理路口的参数和路口外道路参考线拓扑关系，确定所述待处理路口的路口内参考线起始节点；

具体的，先基于路段先验信息和人工目视解译参数化配置规则的待处理路口，所述待处理路口的参数包括路口类型、驶入和驶出道路数量和驶入路口道路转向关系，在一种可能的实施方式中，根据预先采集的所述待处理路口的图片数据，确定路口类型，将所述路口类型作为第一参数P₁；基于与所述待处理路口相交道路的道路模型数据，生成驶入和驶出道路数量，将所述驶入和驶出道路数量作为第二参数P₂；根据驶入路口道路与路标方向箭头的关联关系，确定驶入路口道路转向关系，将所述驶入路口道路转向关系作为第三参数P₃；所述驶入路口道路转向关系包括直行、左转和右转；所述路标方向箭头指示车辆在所述驶入路口道路执行不同方向上的转向；基于所述第一参数、所述第二参数和所述第三参数，生成所述待处理路口的所述参数，以{P₁,P₂, P₃}作为参数标记所述待处理路口，用于后续的路口高精地图的自动化生成

参考图3，在一种可能的实施方式中，步骤S201还包括：

S301，获取与所述待处理路口相交道路的路口外道路参考线和停止线，并基于所述停止线生成所述待处理路口的四至范围路段；所述四至范围路段包括与所述待处理路口相交的路段；

S302，基于所述待处理路口的参数和所述四至范围路段内的路口外道路参考线拓扑关系将所述路口外道路参考线划分为满足预设条件的驶入路口道路参考线和驶出路口道路参考线；

具体的，根据道路空间几何相交关系，查询与所述待处理路口相交的所有路口外道路参考线和停止线，并基于停止线自动生成所述待处理路口的四至范围路段，后续所提及的路口外道路参考线等道路数据模型均指所述待处理路口的四至范围路段的道路数据模型;基于路口类型、驶入和驶出道路数量以及路口外道路参考线拓扑关系将路口外道路参考线分组，在一种可能的实施方式中，步骤S302还包括：

（1）基于所述路口外道路参考线拓扑关系模型、所述路口类型及所述驶入和驶出道路数量，确定驶入路口道路参考线集合和驶出路口道路参考线集合；

（2）计算所述路口外道路参考线中各道路参考线与所述待处理路口的中心的平均距离；

（3）将所述平均距离最小的所述道路参考线确定为所述满足预设条件的驶入路口道路参考线和驶出路口道路参考线。

具体的，参考图5，驶入路口道路参考线为箭头指向所述待处理路口的箭头线，确定驶入路口道路参考线集合为{L₁,L₂,L₃,L₄}；驶出路口道路参考线为箭头远离所述待处理路口的箭头线，确定驶出路口道路参考线集合为{K₁,K₂,K₃,K₄}。其中L_i或K_i包含一条或多条参考线，i=1,2,3,4。若L_i或K_i中包含一条以上参考线时，根据公式（1）分别计算与待处理路口的中心的平均距离，保留最小平均距离对应的参考线，实现集合L_i或K_i中仅包含一条满足预设条件的驶入路口道路参考线（即箭头指向所述待处理路口的箭头实线）、驶出路口道路参考线（即箭头远离所述待处理路口的箭头实线）和对应挂接的多条虚参考线，其中预设条件即为与所述待处理路口的中心的平均距离最小的道路参考线。

（1）

其中，[X, Y]表示待处理路口的中心的平面坐标；[X_i,Y_i]为L_i或K_i中包含的道路参考线上节点的平面坐标，n为对应道路参考线上节点数，为平均距离。

S303，确定所述驶入路口道路参考线和所述驶出路口道路参考线与所述待处理路口连接的节点集合，所述节点集合包括驶入节点集合和驶出节点集合；

具体的，根据步骤S302中获取的驶入路口道路参考线和驶出路口道路参考线，找到参考线上与待处理路口连接的节点，包括虚参考线节点，如图5所示。满足预设条件的路口外道路参考线中，若其节点位于参考线的起点位置，将节点加入到驶出节点V_i集中；若其节点位于参考线的终点位置，将节点加入到驶入节点U_i集中，依据参考线在驶入或驶出路口道路的位置，参照S302，i值可设定为1,2,3,4。待处理路口内不存在车道线地面标识，难以基于图像和点云数据源自动化提取车道模型，通过基于路口外道路参考线和待处理路口参数确定驶入节点和驶出节点，用于后续确定路口内参考线的节点，进一步确定路口内参考线。

S304，根据所述驶入路口道路转向关系，确定所述节点集合中各驶入节点在不同道路转向关系中对应的各驶出节点；

在一种可能的实施方式中，步骤S304还包括：

（1）根据所述驶入路口道路转向关系确定与所述各驶入节点相交的道路车道线；

（2）计算所述各驶出节点与待匹配驶入节点相交的所述道路车道线的距离值和方向值；所述待匹配驶入节点为所述驶入节点集合中的任一驶入节点；

（3）将所述距离值最大和所述方向值为正的对应驶出节点作为所述待匹配驶入节点在左转方向上的驶出节点；将所述距离值最大和所述方向值为负的对应驶出节点作为所述待匹配驶入节点在右转方向上的驶出节点；

（4）计算所述各驶出节点与待匹配驶入节点形成的方向线与所述道路车道线的夹角值；

（5）将所述夹角值最小的对应驶出节点作为所述待匹配驶入节点在直行方向上的驶出节点。

具体的，根据公式（2）计算各驶出节点V_i与待匹配驶入节点相交的所述道路车道线的距离值Dist和方向值Dir，若匹配驶入节点在左转方向上的驶出节点，则寻找Dist最大和Dir为正的点；若匹配驶入节点在右转方向上的驶出节点，则寻找Dist最大和Dir为负的点；若匹配驶入节点在直行方向上的驶出节点，则采用公式（2）分别计算待匹配驶入节点与驶出节点V_i形成的方向线与所述道路车道线的夹角值Ang，Ang最小值对应的驶出节点即为所求。

（2）

其中，{ A,B, C }表示道路车道线的直线方程表达系数;[X₀,Y₀]为驶出节点V_i的平面坐标；[X₁,Y₁]和[X₂,Y₂]为道路车道线的两节点Node1和Node2坐标，Node2位于Node1的下一个节点，且Node1和Node2是道路车道线上离驶出节点V_i最近的两节点，Dir大于0表示驶出节点V_i位于道路车道线的左侧，小于0位于右侧，等于0位于道路车道线上。

S305，基于所述各驶入节点和所述各驶出节点确定所述路口内参考线起始节点。

S202，基于所述路口内参考线起始节点，确定所述待处理路口在不同道路转向上的路口内参考线；所述路口内参考线指示车道行驶方向；

参考图4，在一种可能的实施方式中，步骤S202还包括：

S401，获取与驶入节点预设范围内相交的第一道路车道线以及与驶出节点预设范围内相交的第二道路车道线；

S402，对于所述第一道路车道线，确定所述驶入节点与所述第一道路车道线连接的最近两节点连接线段的第一方向向量；

S403，对于所述第二道路车道线，确定所述驶出节点与所述第二道路车道线连接的最近两节点连接线段的第二方向向量；

S404，基于所述驶入节点、所述驶出节点、所述第一方向向量和所述第二方向向量，确定所述待处理路口在不同道路转向上的路口内参考线。

具体的，本发明实施例中所述预设范围指各以驶入节点和驶出节点为中心，分别构建半径为0.5米的缓冲区，基于矢量空间相交方法，获取与缓冲区相交的第一道路车道线和第二道路车道线，计算第一道路车道线与驶入节点连接的最短线段的第一方向向量Vl₁(x,y)和第二道路车道线与驶出节点连接的最短线段的第二向向量Vr₂(x,y)，将驶入节点坐标、驶出节点节点坐标和第一方向向量Vl₁(x,y)、第二向向量Vr₂(x,y)带入曲线公式（3）中，求解参考线曲线方程的参数。其中[X(p),Y(p)]为求解的参数三次曲线函数；[aU, bU,cU, dU]和[aV, bV, cV, dV]为多项式参数；参数p定义曲线上的点，起点对应的p值为0，终点对应的p值为1，前述的驶入节点即为曲线的起点，驶出节点即为曲线的终点；公式（3）中和是对p的一次求导，代入曲线起点和终点的切向方向值Vl₁(x,y)和Vr₂(x,y)，求解8个多项式参数。求出曲线函数表达式后，再以1m间隔内插节点，基于内插的节点，搜索与该节点最近的点云，赋点云Z值到该节点，生成所述路口内参考线。通过求解曲线方程的方法确定路口内参考线，在存在多个行驶方向上路线交汇的交通复杂的路口内，能够独立地基于驶入节点和驶出节点构建路口内参考线拓扑关系模型。

（3）

S203，根据所述路口内参考线确定对应车道的车道线起始节点，并基于所述车道线起始节点生成路口内车道线；

具体的，依据驶入道路对应的车道行驶方向和驶出道路对应的车道行驶方向，确定路口内左转、直行和右转道路参考线对应的车道数，基于参考线左侧车道数始终为1的原则，确定参考线对应的车道线连接的驶入节点和驶出节点，进一步确定所述驶入节点和驶出节点和切线方向向量，同方法步骤S404，基于公式（3）求解车道线曲线方程，以1m间隔内插节点，然后基于内插的节点，搜索与该节点最近的点云，赋点云Z值到该节点，生成路口内车道线。

S204，基于所述路口内车道线确定车道中心线起始节点，并基于所述车道中心线起始节点生成所述路口内车道线对应的路口内车道中心线；

在一种可能的实施方式中，步骤S204还包括：

（1）针对所述路口内车道线，获取其中相对应的左车道线和右车道线，确定左车道线长度和右车道线长度；

（2）针对所述左车道线上每个点，基于所述点与所述左车道线第一端点的距离、所述左车道线长度和右车道线长度确定所述右车道线上的各点中与所述点相对应的对应点；

（3）确定所述左车道线上每个点和所述对应点连线的中点，得到多个所述车道中心线起始节点；

（4）基于所述车道中心线起始节点生成所述左车道线和右车道线对应的路口内车道中心线。

具体的，选取所述路口内车道线中两条相对应的左车道线和右车道线，遍历线左车道线上所有的点N_t，计算每个N_t与左车道线第一端点的距离D、确定左车道线的总长度Length1和右车道线的总长度Length2，基于公式（4），以D * Length2/Length1反算右车道线上的各点中与每个N_t相对应的对应点M_t，计算形状点N_t和M_t的中心点，作为路口内车道中心线的节点。

（4）

其中，D表示N_t与左车道线第一端点的距离；Length1和Length2表示左车道线的总长度和右车道线的总长度；[X₁,Y₁]为点N_t的坐标，[X₂,Y₂]为点M_t的坐标；[X,Y]为生成的路口内车道中心线的节点坐标；dCurMe表示右车道线上O_t距右车道线第一端点的平面距离值，所述第一端点和第二端点分别是左车道线和右车道线相对应的端点，O_t是右车道线上距离M_t最近的点，点O_t距右车道线第一端点的平面距离应小于M_t距右车道线第一端点的平面距离；[X_S,Y_S]为点O_t的坐标,dCurSegLength为RightLaneLine上与M_t最近前后两节点O_t和O_t+1的平面距离。采用等比例距离法，确定左右车道线上对应的点，基于对应点确定车道中心线起始节点，进一步生成的路口内车道中心线曲率、坡度平缓，更好地服务于自动驾驶车辆行驶。

S205，基于所述路口内参考线、路口内车道线和所述路口内车道中心线，以及所述路口内参考线、所述路口内车道中心线与所述待处理路口相关的交通设施对象建立的逻辑关联模型，生成所述待处理路口的路口高精地图。

在一种可能的实施方式中，步骤S205还包括：

（1）基于所述路口内参考线、路口内车道线和所述路口内车道中心线生成所述待处理路口的路口道路几何模型和拓扑关系；

（2）构建所述路口内参考线、所述路口内车道中心线与所述待处理路口内交通设施对象的逻辑关联关系；

（3）基于所述路口道路几何模型和拓扑关系以及所述逻辑关联关系生成所述待处理路口的路口高精地图。

具体的，根据停止线与路口内车道中心线间的空间位置相交关系，构建路口内车道中心线与停止线的逻辑关联关系；以路口内车道中心线结束的节点为基准，沿着道路行驶方向前方一定距离搜索最近的红绿灯组合对象，本发明实施例中所述一定距离设定为30m，基于查找到的红绿灯位置和路口内车道中心线的车辆行驶方向，完成路口内车道中心线与各红绿灯元素的逻辑关联关系，最后基于所述路口道路几何模型和拓扑关系以及所述逻辑关联关系生成所述待处理路口的路口高精地图。将几何模型、拓扑关系和关联关系技术结合，能够更加精确快速地自动化生成路口高精地图，提高制图效率。

与上述路口高精地图生成方法相对应，本发明实施例还提供一种路口高精地图生成装置，由于本发明实施例提供的路口高精地图生成装置与上述几种实施例提供的路口高精地图生成方法相对应，因此前述路口高精地图生成方法的实施方式也适用于本实施例提供的路口高精地图生成装置，在本发明实施例中不再赘述。

参考图6，其所示为本发明实施例提供的一种路口高精地图生成装置结构示意图，该装置具有实现上述方法实施例中路口高精地图生成方法的功能，所述功能可以由硬件实现，也可以由硬件执行相应的软件实现，该装置可以包括：

路口内参考线起始节点生成模块610，用于基于待处理路口的参数和路口外道路参考线拓扑关系，确定所述待处理路口的路口内参考线起始节点；

路口内参考线生成模块620，用于基于所述路口内参考线起始节点，确定所述待处理路口在不同道路转向上的路口内参考线；所述路口内参考线指示车道行驶方向；

路口内车道线生成模块630，用于根据所述路口内参考线确定对应车道的车道线起始节点，并基于所述车道线起始节点生成路口内车道线；

路口内车道中心线生成模块640，用于基于所述路口内车道线确定车道中心线起始节点，并基于所述车道中心线起始节点生成所述路口内车道线对应的路口内车道中心线；

高精地图生成模块650，用于基于所述路口内参考线、路口内车道线和所述路口内车道中心线，以及所述路口内参考线、所述路口内车道中心线与所述待处理路口相关的交通设施对象建立的逻辑关联模型，生成所述待处理路口的路口高精地图。

可选的，所述路口高精地图生成装置还包括：

第一参数确定模块，用于根据预先采集的所述待处理路口的图片数据，确定路口类型，将所述路口类型作为第一参数；

第二参数确定模块，用于基于与所述待处理路口相交道路的道路模型数据，生成驶入和驶出道路数量，将所述驶入和驶出道路数量作为第二参数；

第三参数确定模块，用于根据驶入路口道路与路标方向箭头的关联关系，确定驶入路口道路转向关系，将所述驶入路口道路转向关系作为第三参数；所述驶入路口道路转向关系包括直行、左转和右转；所述路标方向箭头指示车辆在所述驶入路口道路执行不同方向上的转向；

参数配置模块，用于基于所述第一参数、所述第二参数和所述第三参数，生成所述待处理路口的所述参数。

可选的，所述路口内参考线起始节点生成模块还包括：

四至范围确定模块，用于获取与所述待处理路口相交道路的路口外道路参考线和停止线，并基于所述停止线生成所述待处理路口的四至范围路段；所述四至范围路段包括与所述待处理路口相交的路段；

道路参考线划分模块，用于基于所述待处理路口的参数和所述四至范围路段内的路口外道路参考线拓扑关系将所述路口外道路参考线划分为满足预设条件的驶入路口道路参考线和驶出路口道路参考线；

道路参考线节点确定模块，用于确定所述驶入路口道路参考线和所述驶出路口道路参考线与所述待处理路口连接的节点集合，所述节点集合包括驶入节点集合和驶出节点集合；

道路参考线节点划分模块，用于根据所述驶入路口道路转向关系，确定所述节点集合中各驶入节点在不同道路转向关系中对应的各驶出节点；

路口内参考线起始节点确定模块，用于基于所述各驶入节点和所述各驶出节点确定所述路口内参考线起始节点。

可选的，所述道路参考线划分模块包括：

第一驶入驶出道路参考线确定单元，用于基于所述路口外道路参考线拓扑关系模型、所述路口类型及所述驶入和驶出道路数量，确定驶入路口道路参考线集合和驶出路口道路参考线集合；

第一计算单元，用于计算所述路口外道路参考线中各道路参考线与所述待处理路口的中心的平均距离；

第二驶入驶出道路参考线确定单元，用于将所述平均距离最小的所述道路参考线确定为所述满足预设条件的驶入路口道路参考线和驶出路口道路参考线。

可选的，所述道路参考线节点划分模块包括：

道路车道线确定单元，用于根据所述驶入路口道路转向关系确定与所述各驶入节点相交的道路车道线；

第二计算单元，用于计算所述各驶出节点与待匹配驶入节点相交的所述道路车道线的距离值和方向值；所述待匹配驶入节点为所述驶入节点集合中的任一驶入节点；

第一驶出节点划分单元，用于将所述距离值最大和所述方向值为正的对应驶出节点作为所述待匹配驶入节点在左转方向上的驶出节点；将所述距离值最大和所述方向值为负的对应驶出节点作为所述待匹配驶入节点在右转方向上的驶出节点；

第三计算单元，用于计算所述各驶出节点与待匹配驶入节点形成的方向线与所述道路车道线的夹角值；

第二驶出节点划分单元，用于将所述夹角值最小的对应驶出节点作为所述待匹配驶入节点在直行方向上的驶出节点。

可选的，所述路口内参考线生成模块包括：

道路车道线获取单元，用于获取与驶入节点预设范围内相交的第一道路车道线以及与驶出节点预设范围内相交的第二道路车道线；

获取与驶入节点预设范围内相交的第一道路车道线以及与驶出节点预设范围内相交的第二道路车道线；

对于所述第一道路车道线，确定所述驶入节点与所述第一道路车道线连接的最近两节点连接线段的第一方向向量；

对于所述第二道路车道线，确定所述驶出节点与所述第二道路车道线连接的最近两节点连接线段的第二方向向量；

基于所述驶入节点、所述驶出节点、所述第一方向向量和所述第二方向向量，确定所述待处理路口在不同道路转向上的路口内参考线。

可选的，所述路口内车道中心线生成模块包括：

车道线长度确定单元，用于针对所述路口内车道线，获取其中相对应的左车道线和右车道线，确定左车道线长度和右车道线长度；

等距离比例法计算单元，用于针对所述左车道线上每个点，基于所述点与所述左车道线第一端点的距离、所述左车道线长度和右车道线长度确定所述右车道线上的各点中与所述点相对应的对应点；

车道中心线起始节点生成单元，用于确定所述左车道线上每个点和所述对应点连线的中点，得到多个所述车道中心线起始节点；

路口内车道中心线确定单元，用于基于所述车道中心线起始节点生成所述左车道线和右车道线对应的路口内车道中心线。

可选的，所述高精地图生成模块包括：

路口道路几何模型和拓扑关系生成单元，用于基于所述路口内参考线、路口内车道线和所述路口内车道中心线生成所述待处理路口的路口道路几何模型和拓扑关系；

逻辑关联关系生成单元，用于构建所述路口内参考线、所述路口内车道中心线与所述待处理路口内交通设施对象的逻辑关联关系；

高精地图生成单元，用于基于所述路口道路几何模型和拓扑关系以及所述逻辑关联关系生成所述待处理路口的路口高精地图。

本发明实施例还提供一种电子设备，包括处理器和存储器，所述存储器中存储有至少一条指令或者至少一段程序，所述至少一条指令或者所述至少一段程序由所述处理器加载并执行以实现如上述路口高精地图生成方法的步骤。

存储器可用于存储软件程序以及模块，处理器通过运行存储在存储器的软件程序以及模块，从而执行各种功能应用。存储器可主要包括存储程序区和存储数据区，其中，存储程序区可存储操作系统、功能所需的应用程序等；存储数据区可存储根据所述设备的使用所创建的数据等。此外，存储器可以包括高速随机存取存储器，还可以包括非易失性存储器，例如至少一个磁盘存储器件、闪存器件、或其他易失性固态存储器件。相应地，存储器还可以包括存储器控制器，以提供处理器对存储器的访问。处理器可以是中央处理单元，还可以是其他通用处理器、数字信号处理器、专用集成电路或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等，通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。

本发明实施例所提供的方法实施例可以在计算机终端、服务器或者类似的运算装置中执行。以运行在服务器上为例，图7是本发明实施例提供的运行一种路口高精地图生成方法的服务器的硬件结构示意图，如图7所示，该服务器700可因配置或性能不同而产生比较大的差异，可以包括一个或一个以上处理器（Central Processing Units，CPU）710（处理器710可以包括但不限于微处理器MCU或可编程逻辑器件FPGA等的处理装置）、用于存储数据的存储器730，一个或一个以上存储应用程序723或数据722的存储介质720（例如一个或一个以上海量存储设备）。其中，存储器730和存储介质720可以是短暂存储或持久存储。存储在存储介质720的程序可以包括一个或一个以上模块，每个模块可以包括对服务器中的一系列指令操作。更进一步地，处理器710可以设置为与存储介质720通信，在服务器700上执行存储介质720中的一系列指令操作。服务器700还可以包括一个或一个以上电源760，一个或一个以上有线或无线网络接口750，一个或一个以上输入输出接口740，和/或，一个或一个以上操作系统721，例如Windows ServerTM，Mac OS XTM，UnixTM, LinuxTM，FreeBSDTM等等。

输入输出接口740可以用于经由一个网络接收或者发送数据。上述的网络具体实例可包括服务器700的通信供应商提供的无线网络。在一个实例中，输入输出接口740包括一个网络适配器（Network Interface Controller，NIC），其可通过基站与其他网络设备相连从而可与互联网进行通讯。在一个实例中，输入输出接口740可以为射频（RadioFrequency，RF）模块，其用于通过无线方式与互联网进行通讯。

本领域普通技术人员可以理解，图7所示的结构仅为示意，其并不对上述电子装置的结构造成限定。例如，服务器700还可包括比图7中所示更多或者更少的组件，或者具有与图7所示不同的配置。

本发明实施例还提供了一种计算机可读存储介质，所述计算机可读存储介质中存储有至少一条指令或至少一段程序，所述至少一条指令或所述至少一段程序由处理器加载并执行以实现如上述路口高精地图生成方法的步骤。在本发明实施例中，所述计算机程序包括计算机程序代码，所述计算机程序代码可以为源代码形式、对象代码形式、可执行文件或某些中间形式等。所述计算机可读存储介质可以包括：能够携带所述计算机程序代码的任何实体或装置、记录介质、U盘、移动硬盘、磁碟、光盘、计算机存储器、只读存储器、随机存取存储器、电载波信号、电信信号以及软件分发介质等。

本发明实施例还提供了一种计算机可读存储介质，所述计算机可读存储介质中存储有至少一条指令或至少一段程序，所述至少一条指令或所述至少一段程序由处理器加载并执行以实现如上述路口高精地图生成方法的步骤。在本发明实施例中，所述计算机程序包括计算机程序代码，所述计算机程序代码可以为源代码形式、对象代码形式、可执行文件或某些中间形式等。所述计算机可读存储介质可以包括：能够携带所述计算机程序代码的任何实体或装置、记录介质、U盘、移动硬盘、磁碟、光盘、计算机存储器、只读存储器、随机存取存储器、电载波信号、电信信号以及软件分发介质等。

本申请实施例还提供一种计算机存储介质，所述计算机存储介质中存储有至少一条指令或至少一段程序，所述至少一条指令或至少一段程序由处理器加载并执行以实现上述的方法。在本发明实施例中，所述计算机程序包括计算机程序代码，所述计算机程序代码可以为源代码形式、对象代码形式、可执行文件或某些中间形式等。所述计算机可读存储介质可以包括但不限于：能够携带所述计算机程序代码的任何实体或装置、记录介质、U盘、移动硬盘、磁碟、光盘、计算机存储器、只读存储器、随机存取存储器、电载波信号、电信信号以及软件分发介质等。

本发明实施例还提供一种计算机程序产品或计算机程序，该计算机程序产品或计算机程序包括计算机指令，该计算机指令存储在计算机可读存储介质中。计算机设备的处理器从计算机可读存储介质读取该计算机指令，处理器执行该计算机指令，使得该计算机设备执行上述方法的步骤。

以上所述仅为本发明的较佳实施例，并不用以限制本发明，凡在本发明的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本发明的保护范围之内。

Specific embodiments

The following will be combined with the accompanying drawings in an embodiment of the present invention, the technical solution in an embodiment of the present invention is clearly and completely described. Obviously, the embodiments described are only a portion of the embodiments of the present invention, and not all embodiments. Based on embodiments in the present invention, all other embodiments obtained by those of ordinary skill in the art without doing creative work, are within the scope of protection of the present invention.

As used herein, "an embodiment" or "embodiment" refers to a particular feature, structure, or characteristic that may be included in at least one embodiment of the present invention. In the description of the present invention, it is to be understood that the term "up", "down", "top", "bottom" and the like indicates the orientation or position relationship based on the drawings, only to facilitate the description of the present invention and simplify the description, and not to indicate or imply that the means or elements referred to must have a specific orientation, structured and operated in a particular orientation, and therefore cannot be understood as a limitation of the present invention. Further, the terms "first", "second" are for descriptive purposes only and cannot be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, the characteristics limited to "first" and "second" may include one or more such features expressly or implicitly. Moreover, the terms "first", "second", etc. are used to distinguish similar objects, rather than to describe a particular order or sequencing. It should be understood that the data used in this manner are interchangeable in appropriate cases, so that the embodiments of the present invention described herein can be implemented in order other than those illustrated or described herein.

It should be noted that, referring to FIG. 1, the high-precision map road data model based on the present invention is a data model formulated in combination with the navigation electronic map and the high-precision map production specification to meet the needs of automatic driving, including road reference lines, lane boundary lines, lane centerlines, ground and aboveground road ancillary facilities, parking areas and speed limit signs.

The topological relationship referred to hereinafter referred to in the present invention refers primarily to the connecting relationship between the reference line, the lane line and the lane center line; The correlation relationship mainly refers to the relationship between the traffic auxiliary facilities and the road or lane, such as the logical relationship between the stop line and the lane, the logical relationship between the road boundary and the reference line, etc.; The intersection to be treated is a regular intersection, which refers to an intersection that can be parameterized.

Referring to FIG. 2, which shows a schematic diagram of a high-precision map generation method at an intersection provided by an embodiment of the present invention. The present specification provides a method of operation as described in an embodiment or flowchart, but based on conventional or unconventional labor may include more or fewer operating steps. The sequence of steps enumerated in an embodiment is only one of the many steps in the order of execution, does not represent a unique order of execution. In practice, when the system apparatus or product is executed, it may be executed sequentially or in parallel according to an embodiment or the method shown in the accompanying drawings (e.g., a parallel processor or a multithreaded environment). Embodiments of the present invention provides a method for generating a high-precision map of the intersection comprising:

S201, based on the parameters of the intersection to be processed and the topological relationship of the road reference line outside the intersection, determine the starting node of the reference line in the intersection to be processed;

Specifically, first based on the road section prior information and manual visual interpretation of the parameterized configuration rules of the pending intersection, the parameters of the to be treated intersection comprising the type of intersection, the number of roads entering and exiting the road and the turning relationship of the road at the intersection, in one possible embodiment, according to the pre-collected picture data of the intersection to be processed, determine the intersection type, the type of the intersection as the first parameter P₁; Based on the road model data of the intersection road with the intersection of the intersection to be processed, the number of inbound and outbound roads is generated, and the number of roads driven in and out is taken as the second parameter P2; According to the correlation relationship between the road at the intersection and the direction arrow of the road sign, the steering relationship of the road at the intersection is determined, and the turning relationship of the road at the intersection is taken as the third parameter P₃; The road turning relationship at the intersection includes a straight, left and right turn; The signpost direction arrow indicates that the vehicle performs a turn in different directions on the road at the intersection; Based on the first parameter, the second parameter and the third parameter, the parameters of the to be treated intersection are generated, with {P₁, P₂, P₃} as the parameter marking the to be treated intersection, for the subsequent automatic generation of the intersection HIGH-precision map

Referring to FIG. 3, in one possible embodiment, step S201 further comprises:

S301, obtain the intersection of the road outside the intersection and stop line with the intersection of the intersection to be processed, and generate a four-to-range road section of the road to be treated based on the stop line; The four to the range of road sections includes a section of road intersecting with the intersection to be treated;

S302, based on the parameters of the intersection to be treated and the intersection outside the road reference line topology relationship within the four to the range of road sections, the road reference line outside the intersection is divided into driving intersection road reference line and driving out of the intersection road reference line that meets the preset conditions;

Specifically, according to the geometric intersection relationship of the road space, query all the road reference lines and stop lines outside the intersection of the intersection to be processed, and automatically generate the four to the range of the road section of the intersection to be processed based on the stop line, and the road data model such as the road data model of the road reference line outside the intersection mentioned subsequently refers to the road data model of the four to the range of the road section of the road to be processed; Based on the intersection type, the number of inbound and outbound roads and the intersection road reference line topology relationship outside the intersection road reference line grouping, in one possible embodiment, step S302 further comprises:

(1) Based on the topological relationship model of the road reference line outside the intersection, the type of the intersection and the number of roads entering and leaving, the collection of road reference lines and the collection of road reference lines of the road exit of the intersection are determined;

(2) Calculate the average distance between each road reference line outside the intersection and the center of the intersection to be treated;

(3) The said road reference line with the smallest average distance is determined as the driving intersection road reference line and the driving intersection road reference line that meets the preset conditions.

Specifically, with reference to Figure 5, the road reference line of the driving intersection is the arrow pointing to the arrow line of the intersection to be processed, and the set of road reference lines of the driving intersection is {L₁, L₂, L₃, L₄}; The exit road reference line is an arrow away from the arrow line of the intersection to be processed, and the set of road reference lines for the exit road is {K₁, K₂, K₃, K₄}. where L_i or K_i contain one or more guides, i= 1, 2, 3, 4. If L_i or K_i contains more than one reference line, calculate the average distance from the center of the intersection to be processed according to Formula (1), retain the reference line corresponding to the minimum average distance, and realize the set L_i or K_i Contains only one driving intersection road guide that meets the preset conditions (i.e., the arrow points to the solid arrow line of the intersection to be processed), the exit road guide (i.e., the arrow away from the arrow solid line of the to be processed intersection) and a plurality of dotted reference lines corresponding to the connection, wherein the preset condition is the road reference line with the lowest average distance from the center of the to be processed intersection.

（1）

where [X, Y] represents the planar coordinates of the center of the intersection to be processed; [X_i,Yi] is the planar coordinate of the nodes on the road reference line contained in L_i or K_i, and n is the number of nodes on the corresponding road reference line, which is the average distance.

S303, determining the set of nodes connected to the road reference line and the road reference line of the exit road and the intersection to be processed, the node set comprises a set of driving in nodes and a set of driving out nodes;

Specifically, according to the inbound road reference line and the exit road reference line obtained in step S302, find the node connecting the reference line with the intersection to be processed, including the dotted reference line node, as shown in Figure 5. In the road guide outside the intersection that meets the preset conditions, if its node is located at the starting point of the reference line, the node is added to the outbound node V_i concentration; If its node is located at the end of the reference line, the node is added to the driving node U_i set, according to the location of the reference line in the road in or out of the intersection, with reference to S302, the i value can be set to 1, 2, 3, 4. There is no lane line ground identification in the intersection to be processed, it is difficult to automatically extract the lane model based on the image and point cloud data source, and the entry node and exit node are determined by basing the road reference line outside the intersection and the parameters of the intersection to be processed, which is used to determine the node of the reference line in the intersection and further determine the reference line in the intersection.

S304, according to the road steering relationship of the driving intersection, determine the corresponding driving nodes of each driving node in the set of nodes in different road steering relationships;

In one possible embodiment, step S304 further comprises:

(1) according to the said road turning relationship at the intersection to determine the road lane line intersecting with the respective entry node;

(2) Calculate the distance value and direction value of the road lane line where each driving node intersects the node to be matched; The to be matched drive-in node is any of the incoming nodes in the set of driven-in nodes;

(3) The corresponding driving out node of the maximum distance value and the positive direction value as the driving node to be matched in the left turning direction of the driving node; The corresponding driving out node of the maximum distance value and the negative value of the direction is the driving out node of the node to be matched in the right turning direction;

(4) Calculate the angle value between the direction line formed by the each exit node and the node to be matched and the road lane line;

(5) The corresponding driving out node with the smallest angle value is the driving out node in the straight direction of the node to be matched.

Specifically, according to Formula (2), calculate the distance value Of the road lane line where each driving node V_i intersects with the incoming node to be matched, the Dist and direction values Dir, if the driving node matches the driving node in the left turn direction, then look for the point where The Maximum Dist and Dir are positive; If the matching inbound node is driven out of the right direction, look for the point where The Maximum Of Dist and Dir are negative; If the inbound node is matched in the straight direction of the driving node, the formula (2) is used to calculate the angle between the direction line formed by the matching inbound node and the driving node V_i and the road lane line Ang, and the driving node corresponding to the minimum value of Ang is the resulting.

（2）

where { A, B, C } represents the linear equation expression coefficient of the road lane line; [X_0,Y0] is the planar coordinate of the exit node V_i; [X_1,Y1] and [X_2,Y2] are the nodes of node Node1 and Node2 coordinates for road lane lines 2 is located at the next node of Node1, and Node1 and Node2 are the two nodes closest to the exit node V_i on the road lane line, and Dir greater than 0 indicates the exit node V_i Located on the left side of the road lane line, less than 0 is on the right side, equal to 0 on the road lane line.

S305, based on the respective driving node and the respective driving out node to determine the starting node of the reference line within the intersection.

S202, based on the starting node of the reference line in the intersection, determine the reference line within the intersection of the intersection to be processed on different road turns; The reference line within the intersection indicates the direction of travel of the lane;

Referring to FIG. 4, in one possible embodiment, step S202 further comprises:

S401, to obtain the first road lane line that intersects within the preset range of the incoming node and the second road lane line that intersects within the preset range of the exit node;

S402, for the first road lane line, determine the first direction vector of the entry node connected to the first road lane line of the nearest two node connection segments;

S403, for the second road lane line, determine the second direction vector of the nearest two node connection segments of the outgoing node and the second road lane line connection;

S404, based on the driving node, the driving out node, the first direction vector and the second direction vector, to determine the intersection to be processed on different road turns within the reference line.

Specifically, the preset range in an embodiment of the present invention refers to each driving node and driving out node as the center, respectively constructing a buffer zone with a radius of 0.5 meters, based on the vector space intersection method, to obtain a buffer zone intersecting the first road lane line and the second road lane line, calculating the first road lane line and the shortest line connected to the driving node first direction vector Vl₁ (x, y) And the second vector Vr₂ (x,y) of the shortest line connected to the exit node of the second road lane line will drive into the node coordinates, the exit node coordinates, and the first direction vector Vl₁(x,y), the second vector Vr₂ (x,y) Brought into the curve equation (3), solve the parameters of the reference line curve equation. where [X(p),Y(p)] is the parameter cubic curve function for the solution; [aU, bU,cU, dU] and [aV, bV, cV, dV] are polynomial parameters; Parameter p defines the point on the curve, the p value corresponding to the starting point is 0, the p value corresponding to the end point is 1, the aforementioned driving node is the starting point of the curve, and the driving node is the end point of the curve; The neutralization of formula (3) is a derivative of p, substituting the tangential direction values Vl₁(x,y) and Vr₂(x,y) of the curve's start and end points to solve for 8 polynomial parameters. After the curve function expression is found, the node is interpolated at 1m intervals, and based on the interpolated node, the point cloud closest to the node is searched, and the point cloud Z value is assigned to the node, and the reference line in the intersection is generated. By solving the curve equation method to determine the reference line in the intersection, in the intersection where there are multiple traffic directions intersecting the traffic complex intersection, the in-intersection reference line topology relationship model can be independently based on the driving node and the driving node.

（3）

S203, according to the intersection reference line to determine the lane line starting node corresponding to the lane, and based on the lane line starting node to generate the lane line in the intersection;

Specifically, according to the driving direction of the lane corresponding to the road and the driving direction corresponding to the road, determine the number of lanes corresponding to the left turn, straight and right turn road reference line in the intersection, based on the principle that the number of lanes on the left side of the reference line is always 1, determine the driving node and the driving node connected by the reference line, further determine the driving node and the driving node and tangent direction vector, the same method step S404, based on formula (3) to solve the lane line curve equation, interpolate the node at 1m interval, Then, based on the interpolated node, search for the point cloud closest to the node, assign the point cloud Z value to the node, and generate the lane line in the intersection.

S204, based on the lane line in the intersection to determine the lane centerline starting node, and based on the lane centerline starting node to generate the intersection lane centerline corresponding to the intersection lane centerline;

In one possible embodiment, step S204 further comprises:

(1) For the lane line in the intersection, obtain the corresponding left lane line and right lane line, and determine the length of the left lane line and the length of the right lane line;

(2) For each point on the left lane line, based on the distance between the point and the first endpoint of the left lane line, the length of the left lane line and the length of the right lane line, the corresponding point of each point on the right lane line corresponding to the point is determined;

(3) Determine the midpoint of each point on the left lane line and the corresponding point of the connection line, and obtain a plurality of the lane centerline starting nodes;

(4) Based on the starting node of the lane centerline, the lane centerline in the intersection corresponding to the left lane line and the right lane line is generated.

Specifically, the two corresponding left lane lines and right lane lines in the lane line in the intersection are selected, all the points N_t on the left lane line of the line are traversed, the distance D of each N_t from the first endpoint of the left lane line is calculated, and the total length of the left lane line is determined by Length1 and the total length of the right lane line, Length2, based on equation (4), in D * Length2/Length1 invert the corresponding point M t corresponding to each N_t in each point on the right lane line , which calculates the center points of shape points N_andM_t as nodes of the lane centerline within the intersection.

（4）

where D represents the distance of N_t from the first endpoint of the left lane line; Length1 and Length2 represent the total length of the left lane line and the total length of the right lane line; [X_1,Y1] is the coordinate of point N_t, and [X_2,Y2] is the point M _{coordinates of t}; [X,Y] is the node coordinate of the lane centerline in the generated intersection;dCurMe represents the plane distance value of O_t from the first endpoint of the right lane line on the right lane line, the first and second endpoints are the corresponding endpoints of the left lane line and the right lane line, O_t is the distance M_t on the right lane line The nearest point, the plane distance between the point O_t and the first endpoint of the right lane line should be less than the plane distance of M_t from the first endpoint of the right lane line; [X_S, Y_S] is the coordinate of the point O_t, and dCurSegLength is rightlaneLine with M_The plane distance of the two nodes O_t and O_{t+1 before} and after t. The equal proportional distance method is used to determine the corresponding points on the left and right lane lines, and the starting node of the lane centerline is determined based on the corresponding points, and the curvature and slope of the lane centerline in the intersection are further generated, which better serves the driving of autonomous vehicles.

S205, based on the intersection within the reference line, the intersection lane line and the intersection within the lane centerline, and the intersection within the reference line, the intersection lane centerline and the traffic facility object to be treated intersection related to the logical association model established, to generate the intersection of the intersection to be treated high-precision map.

In one possible embodiment, step S205 further comprises:

(1) Based on the reference line in the intersection, the lane line within the intersection and the lane centerline within the intersection, the intersection road geometry model and topological relationship of the intersection to be processed are generated;

(2) Construct the logical relationship between the reference line in the intersection, the lane centerline in the intersection and the traffic facility object in the intersection to be processed;

(3) Based on the intersection road geometry model and topological relationship and the logical association relationship, the intersection high-precision map of the intersection to be processed is generated.

Specifically, according to the spatial position intersection relationship between the stop line and the lane centerline in the intersection, the logical association relationship between the lane centerline and the stop line in the intersection is constructed; Taking the node ending the lane centerline within the intersection as the reference, searching for the nearest traffic light combination object at a certain distance in front of the road driving direction, a certain distance set in the embodiment of the present invention is set to 30m, based on the found traffic light location and the vehicle driving direction of the lane centerline within the intersection, the logical association relationship between the lane centerline and each traffic light element within the intersection is completed, and finally based on the intersection road geometry model and topology relationship and the logical association relationship to generate the intersection HD map of the intersection to be processed. The combination of geometric model, topological relationship and association relationship technology can more accurately and quickly automate the generation of intersection high-precision maps, improving the efficiency of drawing.

Corresponding to the above-described intersection high-precision map generation method, the embodiment of the present invention also provides an intersection high-precision map generation device, since the intersection high-precision map generation device provided by the embodiment of the present invention corresponds to the intersection high-precision map generation method provided by the above-mentioned embodiments, so the embodiment of the foregoing intersection high-precision map generation method is also applicable to the intersection high-precision map generation device provided in the present embodiment, which will not be repeated in the embodiment of the present invention.

Referring to FIGURE 6, which shows a schematic structural diagram of an intersection high-precision map generation device provided by an embodiment of the present invention, the apparatus having the function of implementing the intersection high-precision map generation method in the above-described method embodiment, the function may be implemented by hardware, or may be implemented by hardware corresponding software, the apparatus may include:

The starting node of the reference line within the intersection generates module 610, which is used to determine the starting node of the reference line within the intersection of the intersection based on the parameters of the intersection to be processed and the topological relationship of the road reference line outside the intersection;

Junction in-junction reference line generation module 620, for based on the intersection within the reference line starting node, to determine the intersection to be processed on different road turns within the intersection reference line; The reference line within the intersection indicates the direction of travel of the lane;

Lane line generation module 630 in the intersection, for determining the lane line start node of the corresponding lane according to the intersection reference line, and generating the lane line within the intersection based on the lane line starting node;

Lane centerline generation module 640 within the intersection, for determining the lane centerline starting node based on the lane line within the intersection, and generating the lane centerline corresponding to the laneline within the intersection corresponding to the lane centerline within the intersection based on the lane centerline starting node;

HD map generation module 650 for basing the intersection within the reference line, the intersection lane line and the intersection within the lane centerline, and the intersection within the reference line, the intersection within the lane centerline and the traffic facility object to be processed intersection to establish a logical association model, to generate the intersection HD map of the intersection to be processed.

Optionally, the intersection HIGH-precision map generation apparatus further comprises:

The first parameter determination module for determining the intersection type according to the pre-collected picture data of the intersection to be processed, and the intersection type as the first parameter;

The second parameter determination module for generating the number of roads in and out of the road based on the road model data of the intersection of the intersection with the intersection of the intersection to be processed, the number of roads driven in and out as the second parameter;

The third parameter determination module, used to determine the turning relationship of the road at the intersection according to the correlation relationship between the road entering the intersection and the direction arrow of the road sign, and the turning relationship of the road at the driving intersection is taken as the third parameter; The road turning relationship at the intersection includes a straight, left and right turn; The signpost direction arrow indicates that the vehicle performs a turn in different directions on the road at the intersection;

Parameter configuration module for generating the parameters of the to be processed junction based on the first parameter, the second parameter and the third parameter.

Optionally, the reference line start node generation module within the intersection further comprises:

Four to range determination module for obtaining the intersection of the intersection road intersecting the road to be processed and the stop line, and based on the stop line to generate the four to range of the road section of the road to be treated; The four to the range of road sections includes a section of road intersecting with the intersection to be treated;

Road reference line division module for dividing the road reference line outside the intersection into driving intersection road reference lines and exit road reference lines that meet the preset conditions based on the parameters of the intersection to be treated and the intersection outside the road reference line topology relationship within the four to the range of road segments;

Road reference line node determination module for determining the set of nodes connected to the road reference line and the road reference line of the road exit road and the intersection to be processed, the node set comprises a collection of driving in nodes and a collection of driving out nodes;

Road reference line node division module, for the purpose of the road steering relationship according to the driving intersection road steering relationship, to determine the node in the set of the driving node in different road steering relationship corresponding to each driving node;

The reference line within the intersection start node determination module for determining the starting node of the reference line in the intersection based on the respective entry node and the exit node.

Optionally, the road guide division module comprises:

The first inbound and outbound road reference line determination unit, for determining the collection of inlet road reference lines and the collection of driving intersection road reference lines based on the topological relationship model of the road reference line outside the intersection, the type of the intersection and the number of roads entering and exiting the road;

The first calculation unit for calculating the average distance of each road reference line outside the intersection to the center of the intersection to be treated;

The second inbound and outbound road reference line determination unit for determining the road reference line with the smallest average distance is determined to be the driving intersection road reference line and the driving intersection road reference line that meets the preset conditions.

Optionally, the road guide node division module comprises:

Road lane line determination unit, for determining the road lane line intersecting with each entry node according to the road steering relationship of the driving intersection;

The second calculation unit, for calculating the distance value and direction value of the road lane line where each driving node intersects the driving node to be matched; The to be matched drive-in node is any of the incoming nodes in the set of driven-in nodes;

The first driving out node is divided into units for the corresponding driving out node of the maximum distance value and the positive direction value as the driving node to be matched in the left turning direction of the driving node; The corresponding driving out node of the maximum distance value and the negative value of the direction is the driving out node of the node to be matched in the right turning direction;

The third calculation unit, for calculating the angle value between the direction line formed by each driving node and the driving node to be matched with the road lane line;

The second drive-out node is divided into units for the corresponding driving-out node with the smallest angle value as the driving-out node to be matched in the straight direction of the driving-out node.

Optionally, the reference line generation module within the intersection comprises:

Road lane line acquisition unit, which is used to obtain the first road lane line that intersects within the preset range of the incoming node and the second road lane line that intersects within the preset range of the exit node;

Obtain the first road lane line that intersects within the preset range of the incoming node and the second road lane line that intersects within the preset range of the exit node;

For the first road lane line, determine the first direction vector of the nearest two node connection segments of the driving node and the first road lane line connection;

For the second road lane line, determine the second direction vector of the nearest two node connection segments of the driving node and the second road lane line connection;

Based on the driving node, the driving out node, the first direction vector and the second direction vector, the intersection reference line to be processed on different road turns is determined.

Optionally, the lane centerline generation module within the intersection comprises:

Lane line length determination unit, used to obtain the corresponding left lane line and right lane line for the lane line in the intersection, determine the length of the left lane line and the length of the right lane line;

Equidistance proportional method calculation unit for each point on the left lane line, based on the distance of the point to the first endpoint of the left lane line, the length of the left lane line and the length of the right lane line to determine the corresponding point of each point on the right lane line corresponding to the point;

Lane centerline start node generates a unit for determining the midpoint of each point on the left lane line and the corresponding point of the connection line, and obtaining a plurality of said lane centerline starting nodes;

The lane centerline determination unit within the intersection is used to generate the lane centerline within the intersection corresponding to the left lane line and the right lane line based on the starting node of the lane centerline.

Optionally, the HD map generation module includes:

Intersection road geometry model and topological relationship generation unit, for generating the intersection road geometry model and topological relationship of the intersection to be processed based on the reference line within the intersection, the lane line within the intersection and the lane centerline within the intersection;

Logical association relationship generation unit for constructing the reference line within the intersection, the lane centerline within the intersection and the traffic facility object in the intersection to be processed;

HD map generation unit for generating the intersection HD map of the intersection to be processed based on the intersection road geometry model and topological relationship and the logical association relationship.

Embodiments of the present invention further provides an electronic device, comprising a processor and a memory, the memory stores at least one instruction or at least one program, the at least one instruction or at least one program is loaded by the processor and executed to achieve the steps of the high-precision map generation method such as the above intersection.

Memory can be used to store software programs as well as modules, and the processor performs a variety of functional applications by running software programs and modules stored in memory. The memory may mainly include a storage program area and a storage data area, wherein the storage program area can store the operating system, functions required applications, etc.; The storage data area may store data created according to the use of the device and the like. Further, the memory may include a high-speed random access memory, may also include a nonvolatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory may further include a memory controller to provide processor access to the memory. The processor may be a central processing unit, may also be another general-purpose processor, digital signal processor, as-applied integrated circuit or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc., the general purpose processor may be a microprocessor or the processor may also be any conventional processor and the like.

Embodiments of the present invention provides a method embodiment may be performed in a computer terminal, server, or similar computing device. Taking running on the server as an example, FIG. 7 is a schematic hardware structure of the server provided by an embodiment of the present invention running an intersection high-precision map generation method, as shown in FIG. 7, the server 700 may produce a relatively large difference due to different configurations or performance, may include one or more processors (Central Processing Units, CPU) 710 (processor 710 may include, but is not limited to, microprocessor MCU or programmable logic device FPGA, etc.), a memory 730 for storing data, one or more storage applications 723 or data 722 storage medium 720 (e.g., one or one storage device in shanghai). Wherein, the memory 730 and the storage medium 720 may be transient storage or persistent storage. The program stored in the storage medium 720 may include one or more modules, each module may include a series of instructions to the server operation. Further, the processor 710 may be configured to communicate with the storage medium 720, performing a series of instruction operations in the storage medium 720 on the server 700. Server 700 may further include one or more power supplies 760, one or more wired or wireless network interfaces 750, one or more input and output interfaces 740, and / or, one or more operating systems 721, such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM and the like.

Input and output interface 740 may be used to receive or transmit data via a network. Specific examples of the above-described networks may include a wireless network provided by the communication provider of the server 700. In one example, the input and output interface 740 includes a network adapter (Network Interface Controller, NIC), which may be connected to other network devices through the base station to communicate with the Internet. In one example, the input and output interface 740 may be a radio frequency (RadioFrequency, RF) module for communicating with the Internet by wireless means.

Those of ordinary skill in the art will appreciate that the structure shown in FIG. 7 is only illustrative, which does not qualify the structure of the above-described electronic device. For example, the server 700 may further comprise more or fewer components than shown in FIG. 7, or having a different configuration than shown in FIG. 7.

Embodiments of the present invention further provides a computer-readable storage medium, the computer-readable storage medium stores at least one instruction or at least one program, the at least one instruction or at least one program is loaded by the processor and executed to achieve the steps of the high-precision map generation method such as the above intersection. In an embodiment of the present invention, the computer program comprises a computer program code, the computer program code may be in source code form, object code form, executable file, or some intermediate form and the like. The computer-readable storage medium may include: any entity or apparatus capable of carrying the computer program code, recording medium, U disk, removable hard disk, magnetic disk, optical disk, computer memory, read-only memory, random access memory, electric carrier signal, telecommunications signal and software distribution medium and the like.

Embodiments of the present invention further provides a computer-readable storage medium, the computer-readable storage medium stores at least one instruction or at least one program, the at least one instruction or at least one program is loaded by the processor and executed to achieve the steps of the high-precision map generation method such as the above intersection. In an embodiment of the present invention, the computer program comprises a computer program code, the computer program code may be in source code form, object code form, executable file, or some intermediate form and the like. The computer-readable storage medium may include: any entity or apparatus capable of carrying the computer program code, recording medium, U disk, removable hard disk, magnetic disk, optical disk, computer memory, read-only memory, random access memory, electric carrier signal, telecommunications signal and software distribution medium and the like.

Embodiments of the present application further provide a computer storage medium, the computer storage medium stores at least one instruction or at least one program, the at least one instruction or at least one program is loaded by the processor and executed to achieve the method described above. In an embodiment of the present invention, the computer program comprises a computer program code, the computer program code may be in source code form, object code form, executable file, or some intermediate form and the like. The computer-readable storage medium may include, but is not limited to: any entity or device capable of carrying the computer program code, recording medium, U disk, removable hard disk, magnetic disk, optical disk, computer memory, read-only memory, random access memory, electric carrier signal, telecommunications signal and software distribution medium and the like.

Embodiments of the present invention further provides a computer program product or computer program, the computer program product or computer program comprising a computer instruction, the computer instruction stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, the processor executes the computer instructions, such that the computer device performs the steps of the above method.

The above is only a preferred embodiment of the present invention and is not intended to limit the present invention, where within the spirit and principles of the present invention, any modifications, equivalent substitutions, improvements, etc., should be included within the scope of the present invention.