具体实施方式

下面将参照附图更详细地描述本公开的实施例。虽然附图中显示了本公开的某些实施例，然而应当理解的是，本公开可以通过各种形式来实现，而且不应该被解释为限于这里阐述的实施例。相反，提供这些实施例是为了更加透彻和完整地理解本公开。应当理解的是，本公开的附图及实施例仅用于示例性作用，并非用于限制本公开的保护范围。

另外还需要说明的是，为了便于描述，附图中仅示出了与有关发明相关的部分。在不冲突的情况下，本公开中的实施例及实施例中的特征可以相互组合。

需要注意，本公开中提及的“第一”、“第二”等概念仅用于对不同的装置、模块或单元进行区分，并非用于限定这些装置、模块或单元所执行的功能的顺序或者相互依存关系。

需要注意，本公开中提及的“一个”、“多个”的修饰是示意性而非限制性的，本领域技术人员应当理解，除非在上下文另有明确指出，否则应该理解为“一个或多个”。

本公开实施方式中的多个装置之间所交互的消息或者信息的名称仅用于说明性的目的，而并不是用于对这些消息或信息的范围进行限制。

下面将参考附图并结合实施例来详细说明本公开。

图1是本公开的一些实施例的障碍物检测方法的一个应用场景的示意图。

在图1的应用场景中，首先，计算设备101可以获取环境点云数据集合102，其中，上述环境点云数据是通过安装在目标车辆上的激光雷达对周围环境扫描得到的，上述环境点云数据包括：横坐标值，纵坐标值，竖坐标值，雷达回波功率值，上述环境点云数据包括的横坐标值、纵坐标值和竖坐标值是在目标车辆坐标系下的坐标值，上述目标车辆坐标系是以上述目标车辆行进方向为横轴、以上述目标车辆的底盘中心为原点、以与上述目标车辆的后轴平行的线作为纵轴、以与地面垂直的线作为竖轴的坐标系。其次，计算设备101可以对上述环境点云数据集合102进行裁剪处理以生成裁剪后的环境点云数据集合103。然后，计算设备101可以对上述裁剪后的环境点云数据集合103进行降采样处理以生成降采样后的环境点云数据集合104。进而，计算设备101可以将上述降采样后的环境点云数据集合104输入至障碍物检测模型105以生成障碍物信息集合106。接着，计算设备101可以对上述障碍物信息集合106中的障碍物信息进行过滤处理以生成过滤障碍物信息集合107。最后，计算设备101可以通过车载通信模块108，将上述过滤障碍物信息集合107发送至控制规划终端109。

需要说明的是，上述计算设备101可以是硬件，也可以是软件。当计算设备为硬件时，可以实现成多个服务器或终端设备组成的分布式集群，也可以实现成单个服务器或单个终端设备。当计算设备体现为软件时，可以安装在上述所列举的硬件设备中。其可以实现成例如用来提供分布式服务的多个软件或软件模块，也可以实现成单个软件或软件模块。在此不做具体限定。

应该理解，图1中的计算设备的数目仅仅是示意性的。根据实现需要，可以具有任意数目的计算设备。

继续参考图2，示出了根据本公开的障碍物检测方法的一些实施例的流程200。该障碍物检测方法，包括以下步骤：

步骤201，获取环境点云数据集合。

在一些实施例，障碍物检测方法的执行主体(如图1上述的计算设备101)可以通过有线连接或无线连接的方式获取上述环境点云数据集合。其中，上述环境点云数据可以是通过安装在目标车辆上的激光雷达对周围环境扫描得到的，上述环境点云数据可以包括：横坐标值，纵坐标值，竖坐标值，雷达回波功率值，上述环境点云数据包括的横坐标值、纵坐标值和竖坐标值是在目标车辆坐标系下的坐标值，上述目标车辆坐标系可以是以上述目标车辆行进方向为横轴、以上述目标车辆的底盘中心为原点、以与上述目标车辆的后轴平行的线作为纵轴、以与地面垂直的线作为竖轴的坐标系。

步骤202，对环境点云数据集合进行裁剪处理以生成裁剪后的环境点云数据集合。

在一些实施例中，上述执行主体可以通过各种方式对环境点云数据集合进行裁剪处理以生成裁剪后的环境点云数据集合。

在一些实施例的可选的实现方式中，上述执行主体对环境点云数据集合进行裁剪处理以生成裁剪后的环境点云数据集合，可以包括以下步骤：

第一步，获取横向感知距离和上述激光雷达的感知半径。其中，上述横向感知距离可以是上述目标车辆左侧或右侧的最大感知距离。上述激光雷达的感知半径可以是上述激光雷达的最大感知距离。

第二步，基于横向感知距离和感知半径，确定非裁剪区域。

可选的，上述执行主体可以基于横向感知距离和感知半径，通过以下公式，确定非裁剪区域(如图3中所示的阴影部分)：

其中，TR表示上述横向感知距离。x表示上述环境点云数据集合中环境点云数据包括的横坐标。y表示上述环境点云数据集合中环境点云数据包括的纵坐标。R表示上述感知半径。CH表示上述目标车辆的车身长度。

第三步，从环境点云数据集合中选择落入非裁剪区域的环境点云数据作为裁剪后的环境点云数据，得到裁剪后的环境点云数据集合。

作为示例，可以从上述环境点云数据集合中选择包括的横坐标值和纵坐标值均落入非裁剪区域的环境点云数据作为裁剪后的环境点云数据，得到裁剪后的环境点云数据集合。

步骤203，对裁剪后的环境点云数据集合进行降采样处理以生成降采样后的环境点云数据集合。

在一些实施例中，上述执行主体对裁剪后的环境点云数据集合进行降采样处理以生成降采样后的环境点云数据集合，可以包括以下步骤：

第一步，获取高精度地图中的地面信息。

第二步，基于上述地面信息构建拟合平面。

第三步，去除上述环境点云数据集合中落入上述拟合平面的环境点云数据以生成降采样后的环境点云数据集合。

在一些实施例的一些可选的实现方式中，上述执行主体对裁剪后的环境点云数据集合进行降采样处理以生成降采样后的环境点云数据集合，可以包括以下步骤：

第一步，基于上述裁剪后的环境点云数据集合和预设的最大递归深度，构建八叉树。

第二步，将上述八叉树中包含的环境点云数据确定为降采样后的环境点云数据，得到降采样后的环境点云数据集合。

步骤204，将降采样后的环境点云数据集合输入至障碍物检测模型以生成障碍物信息集合。

在一些实施例中，上述执行主体可以将降采样后的环境点云数据集合输入至障碍物检测模型以生成障碍物信息集合。其中，上述障碍物检测模型可以包括：卷积层，池化层，全连接层。上述卷积层用于特征提取，上述池化层用于特征压缩。上述全连接层用于基于特征，进行分类。

在一些实施例的一些可选的实现方式中，上述执行主体将降采样后的环境点云数据集合输入至障碍物检测模型以生成障碍物信息集合，其中，上述障碍物检测模型可以包括：第一特征提取层，体素切分和特征拼接层，第二特征提取层，单元特征平铺层，第三特征提取层，障碍物属性回归层，可以包括以下步骤：

第一步，通过障碍物检测模型中的第一特征提取层401对环境点云数据集合进行稀疏卷积以生成第一特征。其中，上述环境点云数据集合是n×4的向量组。上述第一特征是n×m的向量组。

第二步，通过障碍物检测模型中的体素切分和特征拼接层402对第一特征进行体素划分以及特征拼接以生成第二特征。

第三步，通过障碍物检测模型中的第二特征提取层403对第二特征进行进一步的特征提取以生成第三特征。其中，上述第三特征是n×s的向量组。

第四步，通过障碍物检测模型中的单元特征平铺层404将第三特征平铺至对应的体素网格中以生成第四特征。

第五步，通过障碍物检测模型中的第三特征提取层405对第四特征进行二维卷积特征提取以生成第五特征。

第六步，基于第五特征，通过障碍物检测模型中的障碍物属性回归层406对障碍物属性进行回归处理以生成障碍物信息集合。

步骤205，对障碍物信息集合中的障碍物信息进行过滤处理以生成过滤障碍物信息集合。

在一些实施例中，上述执行主体可以从上述障碍物信息集合中过滤掉对应置信度数值不在预设范围内的障碍物信息以生成过滤障碍物信息集合。其中，上述预设范围可以是[0，0.2]。

步骤206，通过车载通信模块，将过滤障碍物信息集合发送至控制规划终端。

在一些实施例中，上述执行主体可以通过有线连接或无线连接的方式，通过车载通信模块，将过滤障碍物信息集合发送至控制规划终端。

本公开的上述各个实施例中具有如下有益效果：通过本公开的一些实施例的障碍物检测方法，提高了障碍物检测的结果的准确度，从而，为自动驾驶车辆对于障碍物的规避提供了更加精准的数据，降低了自动驾驶车辆在行驶过程中的风险程度。具体来说，发明人发现，造成障碍物检测的结果不够准确地原因在于：未对环境点云数据进行预处理，从而，导致聚类结构不够精准，即，生成的障碍物信息不够精准。基于此，本公开的一些实施例的障碍物检测方法通过对环境点云数据进行裁剪处理、降采样处理以及对生成的障碍物信息进行过滤处理。从而，使得最终生成的障碍物信息更加准确。除此之外，由于环境点云数据有着较为稀疏的特点，因此，本公开基于环境点云数据特点生成的障碍物检测模型，生成障碍物信息。从而，提高了生成的障碍物信息的准确度。

进一步参考图5，作为对上述各图所示方法的实现，本公开提供了一种障碍物检测装置的一些实施例，这些装置实施例与图2所示的那些方法实施例相对应，该装置具体可以应用于各种电子设备中。

如图5所示，一些实施例的障碍物检测装置500包括：获取单元单元501、裁剪处理单元502、降采样处理单元503、输入单元504、过滤处理单元505和发送单元506。其中，获取单元501，被配置成获取环境点云数据集合，其中，上述环境点云数据是通过安装在目标车辆上的激光雷达对周围环境扫描得到的，上述环境点云数据包括：横坐标值，纵坐标值，竖坐标值，雷达回波功率值，上述环境点云数据包括的横坐标值、纵坐标值和竖坐标值是在目标车辆坐标系下的坐标值，上述目标车辆坐标系是以上述目标车辆行进方向为横轴、以上述目标车辆的底盘中心为原点、以与上述目标车辆的后轴平行的线作为纵轴、以与地面垂直的线作为竖轴的坐标系。裁剪处理单元502，被配置成对上述环境点云数据集合进行裁剪处理以生成裁剪后的环境点云数据集合。降采样处理单元503，被配置成对上述裁剪后的环境点云数据集合进行降采样处理以生成降采样后的环境点云数据集合。输入单元504，被配置成将上述降采样后的环境点云数据集合输入至障碍物检测模型以生成障碍物信息集合。过滤处理单元505，被配置成对上述障碍物信息集合中的障碍物信息进行过滤处理以生成过滤障碍物信息集合。发送单元506，被配置成通过车载通信模块，将上述过滤障碍物信息集合发送至控制规划终端。

可以理解的是，该装置500中记载的诸单元与参考图2描述的方法中的各个步骤相对应。由此，上文针对方法描述的操作、特征以及产生的有益效果同样适用于装置500及其中包含的单元，在此不再赘述。

下面参考图6，其示出了适于用来实现本公开的一些实施例的电子设备(例如图1中的计算设备101)600的结构示意图。图6示出的电子设备仅仅是一个示例，不应对本公开的实施例的功能和使用范围带来任何限制。

如图6所示，电子设备600可以包括处理装置(例如中央处理器、图形处理器等)601，其可以根据存储在只读存储器(ROM)602中的程序或者从存储装置608加载到随机访问存储器(RAM)603中的程序而执行各种适当的动作和处理。在RAM 603中，还存储有电子设备600操作所需的各种程序和数据。处理装置601、ROM 602以及RAM603通过总线604彼此相连。输入/输出(I/O)接口605也连接至总线604。

通常，以下装置可以连接至I/O接口605：包括例如触摸屏、触摸板、键盘、鼠标、摄像头、麦克风、加速度计、陀螺仪等的输入装置606；包括例如液晶显示器(LCD)、扬声器、振动器等的输出装置607；包括例如磁带、硬盘等的存储装置608；以及通信装置609。通信装置609可以允许电子设备600与其他设备进行无线或有线通信以交换数据。虽然图6示出了具有各种装置的电子设备600，但是应理解的是，并不要求实施或具备所有示出的装置。可以替代地实施或具备更多或更少的装置。图6中示出的每个方框可以代表一个装置，也可以根据需要代表多个装置。

特别地，根据本公开的一些实施例，上文参考流程图描述的过程可以被实现为计算机软件程序。例如，本公开的一些实施例包括一种计算机程序产品，其包括承载在计算机可读介质上的计算机程序，该计算机程序包含用于执行流程图所示的方法的程序代码。在这样的一些实施例中，该计算机程序可以通过通信装置609从网络上被下载和安装，或者从存储装置608被安装，或者从ROM 602被安装。在该计算机程序被处理装置601执行时，执行本公开的一些实施例的方法中限定的上述功能。

需要说明的是，本公开的一些实施例上述的计算机可读介质可以是计算机可读信号介质或者计算机可读存储介质或者是上述两者的任意组合。计算机可读存储介质例如可以是——但不限于——电、磁、光、电磁、红外线、或半导体的系统、装置或器件，或者任意以上的组合。计算机可读存储介质的更具体的例子可以包括但不限于：具有一个或多个导线的电连接、便携式计算机磁盘、硬盘、随机访问存储器(RAM)、只读存储器(ROM)、可擦式可编程只读存储器(EPROM或闪存)、光纤、便携式紧凑磁盘只读存储器(CD-ROM)、光存储器件、磁存储器件、或者上述的任意合适的组合。在本公开的一些实施例中，计算机可读存储介质可以是任何包含或存储程序的有形介质，该程序可以被指令执行系统、装置或者器件使用或者与其结合使用。而在本公开的一些实施例中，计算机可读信号介质可以包括在基带中或者作为载波一部分传播的数据信号，其中承载了计算机可读的程序代码。这种传播的数据信号可以采用多种形式，包括但不限于电磁信号、光信号或上述的任意合适的组合。计算机可读信号介质还可以是计算机可读存储介质以外的任何计算机可读介质，该计算机可读信号介质可以发送、传播或者传输用于由指令执行系统、装置或者器件使用或者与其结合使用的程序。计算机可读介质上包含的程序代码可以用任何适当的介质传输，包括但不限于：电线、光缆、RF(射频)等等，或者上述的任意合适的组合。

在一些实施方式中，客户端、服务器可以利用诸如HTTP(HyperText TransferProtocol，超文本传输协议)之类的任何当前已知或未来研发的网络协议进行通信，并且可以与任意形式或介质的数字数据通信(例如，通信网络)互连。通信网络的示例包括局域网(“LAN”)，广域网(“WAN”)，网际网(例如，互联网)以及端对端网络(例如，ad hoc端对端网络)，以及任何当前已知或未来研发的网络。

上述计算机可读介质可以是上述装置中所包含的；也可以是单独存在，而未装配入该电子设备中。上述计算机可读介质承载有一个或者多个程序，当上述一个或者多个程序被该电子设备执行时，使得该电子设备：获取环境点云数据集合，其中，上述环境点云数据是通过安装在目标车辆上的激光雷达对周围环境扫描得到的，上述环境点云数据包括：横坐标值，纵坐标值，竖坐标值，雷达回波功率值，上述环境点云数据包括的横坐标值、纵坐标值和竖坐标值是在目标车辆坐标系下的坐标值，上述目标车辆坐标系是以上述目标车辆行进方向为横轴、以上述目标车辆的底盘中心为原点、以与上述目标车辆的后轴平行的线作为纵轴、以与地面垂直的线作为竖轴的坐标系。对上述环境点云数据集合进行裁剪处理以生成裁剪后的环境点云数据集合。对上述裁剪后的环境点云数据集合进行降采样处理以生成降采样后的环境点云数据集合。将上述降采样后的环境点云数据集合输入至障碍物检测模型以生成障碍物信息集合。对上述障碍物信息集合中的障碍物信息进行过滤处理以生成过滤障碍物信息集合。通过车载通信模块，将上述过滤障碍物信息集合发送至控制规划终端。

可以以一种或多种程序设计语言或其组合来编写用于执行本公开的一些实施例的操作的计算机程序代码，上述程序设计语言包括面向对象的程序设计语言—诸如Java、Smalltalk、C++，还包括常规的过程式程序设计语言—诸如“C”语言或类似的程序设计语言。程序代码可以完全地在用户计算机上执行、部分地在用户计算机上执行、作为一个独立的软件包执行、部分在用户计算机上部分在远程计算机上执行、或者完全在远程计算机或服务器上执行。在涉及远程计算机的情形中，远程计算机可以通过任意种类的网络——包括局域网(LAN)或广域网(WAN)——连接到用户计算机，或者，可以连接到外部计算机(例如利用因特网服务提供商来通过因特网连接)。

附图中的流程图和框图，图示了按照本公开各种实施例的系统、方法和计算机程序产品的可能实现的体系架构、功能和操作。在这点上，流程图或框图中的每个方框可以代表一个模块、程序段、或代码的一部分，该模块、程序段、或代码的一部分包含一个或多个用于实现规定的逻辑功能的可执行指令。也应当注意，在有些作为替换的实现中，方框中所标注的功能也可以以不同于附图中所标注的顺序发生。例如，两个接连地表示的方框实际上可以基本并行地执行，它们有时也可以按相反的顺序执行，这依所涉及的功能而定。也要注意的是，框图和/或流程图中的每个方框、以及框图和/或流程图中的方框的组合，可以用执行规定的功能或操作的专用的基于硬件的系统来实现，或者可以用专用硬件与计算机指令的组合来实现。

描述于本公开的一些实施例中的单元可以通过软件的方式实现，也可以通过硬件的方式来实现。所描述的单元也可以设置在处理器中，例如，可以描述为：一种处理器包括获取单元、裁剪处理单元、降采样处理单元、输入单元、过滤处理单元、发送单元。其中，这些单元的名称在某种情况下并不构成对该单元本身的限定，例如，获取单元还可以被描述为“获取环境点云数据集合的单元”。

本文中以上描述的功能可以至少部分地由一个或多个硬件逻辑部件来执行。例如，非限制性地，可以使用的示范类型的硬件逻辑部件包括：现场可编程门阵列(FPGA)、专用集成电路(ASIC)、专用标准产品(ASSP)、片上系统(SOC)、复杂可编程逻辑设备(CPLD)等等。

以上描述仅为本公开的一些较佳实施例以及对所运用技术原理的说明。本领域技术人员应当理解，本公开的实施例中所涉及的发明范围，并不限于上述技术特征的特定组合而成的技术方案，同时也应涵盖在不脱离上述发明构思的情况下，由上述技术特征或其等同特征进行任意组合而形成的其它技术方案。例如上述特征与本公开的实施例中公开的(但不限于)具有类似功能的技术特征进行互相替换而形成的技术方案。

Mode of execution

Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While certain embodiments of the present disclosure are shown in the drawings, it is to be understood that the present disclosure may be embodied in various forms and should not be construed as being limited to the embodiments set forth herein. Rather, these embodiments are provided for the thorough and complete understanding of the present disclosure. It should be understood that the drawings and examples of the present disclosure are for illustrative purposes only and are not intended to limit the scope of the present disclosure.

It should also be noted that, for ease of description, only parts related to the relevant invention are shown in the drawings. In the case of no conflict, the embodiments in the present disclosure and the features in the embodiments may be combined with each other.

It is to be noted that '1st' is referred to in the present disclosure. The concept of '2nd' is only used to distinguish different devices, modules, or units, and is not intended to limit the order of functions performed by these devices, modules or units or dependency relationship to each other.

It is to be noted that the modifications of 'one', 'a plurality' mentioned in this disclosure are schematic and non-limiting, and it will be understood by those skilled in the art that 'one or more' should be understood unless the context clearly dictates otherwise.

The names of messages or information that are interacted among the multiple devices in the embodiment of the present disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

The present disclosure will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

1 Is a schematic diagram of an application scenario of an obstacle detection method according to some embodiments of the present disclosure.

In the application scenario of FIG. 1, first, the computing device 101 may acquire the environmental point cloud data set 102, wherein the environmental point cloud data includes a coordinate value, a vertical coordinate value, a vertical coordinate value, and a radar echo power value. Second, the invention will be described. The computing device 101 may perform a cropping process on the set 102 of environmental points cloud data to generate a set 103 of cropped environmental points. The computing device 101 may then perform downsampling processing on the clipped environmental point cloud data set 103 to generate a reduced sampled environmental point cloud data set 104. Further, the computing device 101 may input the downsampled environmental point cloud data set 104 to the obstacle detection model 105 to generate an obstacle information set 106. Next, the computing device 101 may filter the obstacle information in the obstacle information set 106 described above to generate a set 107 of filtered obstacle information. Finally. The computing device 101 may send the filtered obstacle information set 108 to the control planning terminal 107 through the onboard communication module 109.

It should be noted that the computing device 101 may be hardware or software. When the computing device is hardware, a distributed cluster consisting of multiple servers or terminal devices may be implemented, or a single server or a single terminal device may also be implemented. When the computing device is embodied as software, it may be installed in the hardware devices listed above. It may be implemented, for example, as a plurality of software or software modules for providing distributed services, or may be implemented as a single software or software module. This is not specifically defined herein.

It should be understood that the number of computing devices in FIG. 1 is merely illustrative. There may be any number of computing devices depending on the implementation requirements.

With continued reference to FIG. 2, a flow 200 of some embodiments of an obstacle detection method according to the present disclosure is shown. The obstacle detection method includes the following steps.

Step 201: obtaining an environmental point cloud data set.

In some embodiments, the execution subject of the obstacle detection method (the computing device 1 described above in FIG. 101) may acquire the set of environmental point cloud data by a wired connection or a wireless connection. The target vehicle coordinate system may include a coordinate value, a vertical coordinate value, a vertical coordinate value, a radar echo power value, and a coordinate value included in a target vehicle coordinate system.

In step 202, the set of environmental point cloud data is tailored to generate the cropped environmental point cloud data set.

In some embodiments, the execution subject may perform a cropping process on the set of environmental point cloud data in various ways to generate the cropped environmental point cloud data set.

In an alternative implementation of some embodiments, the execution subject performs a trimming process on the set of environmental point cloud data to generate the cropped environmental point cloud data set, and may include the following steps.

The first Steps to acquire a lateral sensing distance and a sensing radius of the laser radar. The lateral sensing distance may be a maximum sensing distance on the left side or the right side of the target vehicle. The sensing radius of the laser radar may be a maximum sensing distance of the laser radar.

The second The non-cropping area is determined based on the lateral sensing distance and the sensing radius.

Alternatively, the above-described execution body may determine the non-cropping area (shaded part as shown in 3) by the following formula based on the lateral sensing distance and the sensing radius.

Herein TR represents the lateral sensing distance described above. X represents abscissas included in the cloud data of the environmental point in the set of environmental point cloud data. Y represents ordinates included in the cloud data of the environmental point in the set of environmental point cloud data. R represents the above perceived radius. CH denotes a vehicle body length of the target vehicle.

The third The environmental point cloud data falling into the non-cutting area is selected as the cropped environmental point cloud data from the environment point cloud data set, and the cropped environment point cloud data set is obtained.

As an example, the cropped environment point cloud data set may be obtained from the environment point cloud data included in the cloud data set of the environment point and the environmental point cloud data which fall into the non-cutting area as the cropped environmental point cloud data.

Step 203: downsampling the cropped ambient point cloud data set to generate a set of downsampled environmental point cloud data.

In some embodiments, the execution subject performs down-sampling processing on the cropped ambient point cloud data set to generate a reduced-sampled environmental point cloud data set, which may include the following steps.

The first Steps to obtain ground information in a high-precision map.

The second Next, a fitting plane is constructed based on the above ground information.

The third The environmental point cloud data falling into the fitting plane in the set of environmental point cloud data is removed to generate a reduced sampled environmental point cloud data set.

In some alternative implementations of some embodiments, the execution subject performs down-sampling processing on the cropped ambient point cloud data set to generate a reduced-sampled environmental point cloud data set, which may include the following steps.

The first An octree is constructed based on the clipped environment point cloud data set and a preset maximum recursion depth.

The second The environmental point cloud data contained in the octree is determined as the downsampled environmental point cloud data, and the sampled environmental point cloud data set is obtained.

In step 204, the downsampled environmental point cloud data set is input to an obstacle detection model to generate an obstacle information set.

In some embodiments, the execution subject may input a set of downsampled environmental point cloud data to an obstacle detection model to generate an obstacle information set. The obstacle detection model may include a roll-up layer, a pooling layer, and a full-connection layer. The volume layer is used for feature extraction, and the pooling layer is used for feature compression. The above all-connecting layer is used for classifying based on the features.

In some alternative implementations of some embodiments, the execution subject inputs a reduced set of environmental point cloud data to an obstacle detection model to generate an obstacle information set, wherein the obstacle detection model may include first feature extraction layers, voxel segmentation and feature splicing layers, second feature extraction layers, unit feature smoothing layers, third feature extraction layers, and barrier property regression layers.

1st: Sparse convolution of the set of environmental point cloud data first is performed by 401 feature extraction layer (s) in the obstacle detection model to generate first features. Wherein, the set of environmental point cloud data is a vector group n×4. The first feature is a vector group n×m.

2nd: Voxel segmentation and feature stitching layer 402 in the obstacle detection model performs voxel division and feature stitching on first features to generate second features.

The third Next, second features are extracted by 403 feature extraction layer second in the obstacle detection model to generate third features. The third feature is a vector group n×s.

The fourth 404 Features are tiled into the corresponding voxel grid by the unit feature tiling layer third in the obstacle detection model to generate fourth features.

5th: A feature extraction layer third of 405 features in the obstacle detection model performs two-dimensional convolution feature extraction on fourth features to generate fifth features.

6th: Based on fifth features, an obstacle attribute regression layer 406 in the obstacle detection model performs regression processing on the obstacle attributes to generate an obstacle information set.

Step 205: filtering the obstacle information in the obstacle information set to generate a set of filtered obstacle information.

In some embodiments, the execution subject may filter out obstacle information corresponding to a confidence value that is not within a preset range from the set of obstacle information to generate a set of filtered obstacle information. Here, the preset range may be [0, 0.2].

In step 206, the filtered obstacle information set is transmitted to the control planning terminal through the vehicle-mounted communication module.

In some embodiments, the execution main body may transmit the filtered obstacle information collection to the control planning terminal through the vehicle-mounted communication module in a wired connection or a wireless connection manner.

The above-described various embodiments of the present disclosure have the following advantageous effects. To the obstacle detection method, the accuracy of the obstacle detection result is improved, more accurate data is provided for the avoidance of the obstacle of the automatic driving vehicle, and the risk degree of the automatic driving vehicle in the driving process is reduced. In particular, the inventor found that the result of the obstacle detection is not exactly the reason that the cloud data of the environment point is not pre-processed, so that the clustering structure is not accurate enough, that is, the generated obstacle information is not accurate enough. Based on this, the obstacle detection method according to some embodiments of the present disclosure performs clipping processing on the cloud data of the environment point. A downsampling process and a filtering process of the generated obstacle information are performed. , Finally generated obstacle information is more accurate. Besides, since the cloud data of the environmental point has a relatively sparse characteristic, the obstacle detection model generated based on the characteristic of the cloud data of the environmental point is used for generating obstacle information. , The accuracy of the generated obstacle information is improved.

Further to FIG. 5, as an implementation of the method shown in each of the figures, the present disclosure provides some embodiments of an obstacle detection device that corresponds to those of the method embodiment shown 2, which may specifically be applied to various electronic devices.

As shown 5, the obstacle detection apparatus 500 of some embodiments includes an acquisition unit unit 501, a cropping processing unit 502, a downsampling processing unit 503, an input unit 504, a filter processing unit 505, and a transmission unit 506. The target vehicle coordinate system 501 includes a coordinate value, a vertical coordinate value, a vertical coordinate value, and a radar echo power value, and the target vehicle coordinate system includes a coordinate value, a vertical coordinate value, a vertical coordinate value, and a radar echo power value. The processing unit 502 is cropped. The set of environmental point cloud data is tailored to generate the cropped environmental point cloud data set. The downsampling processing unit 503 is configured to perform down-sampling processing on the clipped environmental point cloud data set to generate a downsampled environmental point cloud data set. The input unit 504 is configured to input the downsampled environmental point cloud data set to an obstacle detection model to generate an obstacle information set. The filter processing unit 505 is configured to filter the obstacle information in the above-described obstacle information set to generate a set of filtered obstacle information. The transmission unit 506. The filtering obstacle information set is sent to a control planning terminal through a vehicle-mounted communication module.

It will be appreciated that the units described in the apparatus 500 correspond to the respective steps in the method described with reference to FIG. 2. Thus, the operations, features and the beneficial effects described above for the method apply equally to the device 500 and the elements contained therein, which will not be described again herein.

Now to FIG. 6, a schematic diagram of a structure of an electronic device (e.g. computing device 1 in 101) suitable for implementing some embodiments of 600 the present disclosure is shown. The electronic device illustrated in FIG. 6 is merely an example, and is not intended to limit the functions and the scope of use of the embodiments of the present disclosure.

As shown 6, the electronic device 600 may include a processing device (e.g. a central processing unit, a graphics processor, etc.) 601 that may perform various appropriate actions and processes according to programs stored in read-only memory (ROM) 602 or programs loaded from storage device 608 (RAM) 603. In RAM 6603, various programs and data necessary for the operation of the electronic device 600 are also stored. The processing apparatus 601, ROM 666666602, and RAM603 are connected to each other via a bus 604. An input/output (I/O) interface 605 is also connected to the bus 604.

, The following devices may be connected to I/O interface 605: an input device 606 including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, and the like. Output devices such as liquid crystal displays (LCD), speakers, vibrators 607, and the like are included. A storage device 608 such as a magnetic tape, a hard disk, or the like is included. And communication device 609. The communication device 609 may allow the electronic device 600 to communicate wirelessly or wirelessly with other devices to exchange data. While FIG. 6 illustrates an electronic device 600 having various devices, it is to be understood that not all illustrated devices are required to be implemented or provided. Or fewer devices may alternatively be implemented or provided. Each block shown in FIG. 6 may represent a device or may represent a plurality of devices as desired.

In particular, according to some embodiments of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, some embodiments of the present disclosure include a computer program product comprising a computer program carried on a computer readable medium containing program code for executing the method shown in the flowchart. In such embodiments, the computer program may be downloaded and installed from the network through the communication device 609, or installed from the storage device 608, or installed from ROM 6602. When the computer program is executed by the processing apparatus 601, the above-described functions defined in the method of some embodiments of the present disclosure are performed.

It should be noted that the computer readable medium of some embodiments of the present disclosure may be a computer readable signal medium or a computer readable storage medium or any combination of the foregoing. The computer readable storage medium may, for example, be, but is not limited to, electricity. A magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of computer readable storage media may include, but are not limited to, electrical connections, portable computer disks, hard disks, random access memory (RAM), read only memory (ROM), erasable and programmable read only memory (EPROM or flash memory), optical fibers, portable compact disk read only memory (CD-ROM), optical storage devices, magnetic memory devices, or any suitable combination of the foregoing. In some embodiments of the present disclosure, a computer readable storage medium may be any tangible medium that includes or stores a program that may be used by or in connection with an instruction execution system, apparatus, or device. While in some embodiments of the present disclosure, a computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take a variety of forms, including, but not limited to, an electromagnetic signal, an optical signal, or any suitable combination of the foregoing. The computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium, and the computer readable signal medium may transmit, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any suitable medium, including but not limited to wire, optical cable, RF (radio frequency), etc, or any suitable combination of the foregoing.

In some embodiments, the client, the server may communicate using any currently known or future developed network protocol, such as HTTP (Hyper Text Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g. a communication network). Examples of communication networks include local area networks ('LAN'), wide area networks ('WAN'), Internet (e.g. Internet) and end-to-end networks (e.g. ad hoc end-to-end networks), and any currently known or future developed network.

The above-described computer-readable medium may be contained in the above-described apparatus. It may also be separately present without being fitted into the electronic device. When the one or more programs are executed by the electronic device, the environment point cloud data is obtained by scanning a surrounding environment through a laser radar mounted on a target vehicle. In the target vehicle coordinate system, the target vehicle coordinate value is a coordinate value in a target vehicle coordinate system, and the target vehicle coordinate system uses a line parallel to a rear axle of the target vehicle as a vertical axis and a line perpendicular to the ground as a vertical axis. The set of environmental point cloud data is tailored to generate the cropped environmental point cloud data set. The cropped ambient point cloud data set is subjected to downsampling processing to generate a reduced sampled environmental point cloud data set. The downsampled environmental point cloud data set is input to an obstacle detection model to generate an obstacle information set. The obstacle information in the above-described obstacle information set is filtered to generate a set of filtered obstacle information. The filtering obstacle information set is sent to a control planning terminal through a vehicle-mounted communication module.

Computer program code for performing operations of some embodiments of the present disclosure may be written in one or more program design languages, including program design languages facing an object, such as Java, Smallallen, C + and also including conventional over-program design languages such as' C ' language or similar program design language. The program code may be executed entirely on the user computer, partly on the user computer, as an independent software package, partly on the user's computer, partly on the remote computer, or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user computer via any kind of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g. connected over the Internet using an Internet service provider).

The flowchart and block diagrams in the Figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products in accordance with various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, the module, program segment, or portion of code containing one or more executable instructions for implementing the specified logical function. It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or they may sometimes be executed in the reverse order, depending upon the functionality involved. It should also be noted that each block in the block diagrams and/or flow diagrams, and combinations of blocks in the block diagrams and/or flow diagrams, may be implemented with dedicated hardware-based systems that perform the specified functions or operations, or may be implemented in a combination of dedicated hardware and computer instructions.

The elements described in some embodiments of the present disclosure may be implemented in software, or may be implemented in hardware. The described units may also be provided in a processor, for example, may be described as: a processor including an acquisition unit, a cropping processing unit, a downsampling processing unit, an input unit, a filtering processing unit, and a transmitting unit. Here, the names of the units do not constitute restrictions on the unit itself in some cases, for example, the acquisition unit may also be described as a unit for acquiring an environmental point cloud data set.

The functionality described herein may be performed at least in part by one or more hardware logic components. For example, non-limiting, exemplary types of hardware logic that may be used include field-programmable gate arrays (FPGA), application-specific integrated circuits (ASIC), application-specific standard products (ASSP), system-on-chip (SOC), complex programmable logic devices (CPLD), and the like.

The above description is merely a few of the preferred embodiments of the present disclosure and descriptions of the applied technical principles. It should be understood by those skilled in the art that the scope of the invention referred to in the embodiments of the present disclosure is not limited to the specific combinations of the above-described technical features, and it is also intended to cover other technical solutions formed by any combination of the above technical features or equivalent features without departing from the inventive concept. For example, the features described above and the technical features disclosed in the embodiments of the present disclosure (but not limited to) having similar functions are replaced with each other.