WO2018035711A1 - 目标检测方法及系统 - Google Patents
目标检测方法及系统 Download PDFInfo
- Publication number
- WO2018035711A1 WO2018035711A1 PCT/CN2016/096358 CN2016096358W WO2018035711A1 WO 2018035711 A1 WO2018035711 A1 WO 2018035711A1 CN 2016096358 W CN2016096358 W CN 2016096358W WO 2018035711 A1 WO2018035711 A1 WO 2018035711A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- obstacle
- target
- type
- radar
- scan
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/255—Detecting or recognising potential candidate objects based on visual cues, e.g. shapes
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/93—Radar or analogous systems specially adapted for specific applications for anti-collision purposes
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
Definitions
- the present invention relates to the field of driverless technology, and in particular, to a target detection method, system and controller.
- the driverless car is a smart car that senses the road environment through the in-vehicle sensing system, automatically plans the driving route, and controls the vehicle to reach the predetermined target. It uses the on-board sensor to sense the surrounding environment of the vehicle, and controls the steering and speed of the vehicle based on the road, vehicle position and obstacle information obtained by the perception, so that the vehicle can travel safely and reliably on the road.
- a target detection method includes:
- the first scan data is obtained by the radar performing a first type scan on the first target area;
- the image data is obtained by imaging the second target area by the digital image device; and a coincidence region exists between the second target area and the first target area;
- a controller comprising a memory and a processor, the memory storing instructions that, when executed by the processor, cause the processor to perform the following steps:
- the first scan data is obtained by the radar performing a first type scan on the first target area;
- the image data is obtained by imaging the second target area by the digital image device; and a coincidence region exists between the second target area and the first target area;
- a target detection system includes: a radar for providing scan data; a digital image device for providing image data; a driving carrier control system for controlling an operating state of the driving carrier; and a controller respectively for the radar, Digital imaging device, driving carrier control system connection: the controller is configured to receive first scan data sent by the radar and image data sent by the digital image device; wherein the first scan data is used by the radar to target the first target After the first type of scanning is performed on the area, the image data is obtained by imaging the second target area by the digital imaging device; the controller is further configured to use the first scan data from the image data.
- the controller is further configured to Controlling, by the type of each obstacle object, that there is an obstacle target that needs to be circumvented, controlling the radar to the obstacle that needs to be circumvented Scanning a second type of material and target tracking, wherein the accuracy of the second type is greater than the accuracy of the scanning of the first scan type.
- the accuracy of the second type of scanning performed by the radar is greater than the accuracy of the first type of scanning, then the first type of scanning is equivalent to the coarse positioning, and the second type of scanning is equivalent to the fine positioning.
- the first scan data sent by the radar and the image data sent by the digital image device since the first scan data can provide preliminary positioning information of the obstacle object, the obstacles are searched for from the image data according to the first scan data.
- the image information corresponding to the object target can reduce unnecessary image search, speed up image processing speed, and thus can quickly identify the type of obstacle.
- the control radar performs a second type of scanning and tracking on the obstacle target that needs to be circumvented, that is, performing fine scanning and tracking, thereby utilizing the advantage that the radar can accurately locate.
- a second type of scanning and tracking on the obstacle target that needs to be circumvented that is, performing fine scanning and tracking, thereby utilizing the advantage that the radar can accurately locate.
- FIG. 1 is a hardware structural diagram of a target detection system according to an embodiment
- FIG. 2 is a flowchart of a target detection method provided by an embodiment
- FIG. 3 is a specific flowchart of step S500 in the target detecting method of the embodiment shown in FIG. 2;
- step S510 is a specific flowchart of step S510 in the target detecting method of the embodiment shown in FIG. 3;
- FIG. 5 is a flowchart of a target detection method provided by another embodiment
- Fig. 6 is a view showing the internal structure of a controller in the object detecting system of the embodiment shown in Fig. 1.
- the radar 100 is configured to provide scan data, which emits a certain measurement wave (where the measurement wave is an electromagnetic wave for a laser radar or a microwave/millimeter wave radar, and an ultrasonic wave for an ultrasonic radar) to scan a target area and receive echo information. Thereby obtaining the distance, speed, orientation and other information of the target object from the radar.
- the radar 100 can be a laser radar, an ultrasonic radar, a microwave/millimeter wave radar, or other types of radar.
- the radar 100 can provide accurate position information, it cannot provide visual information for accurate identification and detection of the target, and the power consumption is high, and the scanning speed has certain limitations.
- the digital imaging device 300 is configured to provide image data, which can provide rich visual information to identify and detect the target, but cannot ensure the accuracy of the position, and the tracking of the target for a long time is prone to the problem of losing.
- the controller 200 is a master control system for controlling the operation of the radar 100 and the digital imaging device 300 and performing related data processing procedures.
- the driving carrier control system 400 is used to control the operating state of the driving carrier, which is for example a smart pilot in a driverless car. In this embodiment, the controller 200 controls the operation of the driving carrier according to the detection and tracking of the obstacle object and by the driving carrier control system 400.
- the radar 100, the controller 200, and the digital imaging device 300 can be integrated into a single hardware to achieve data sharing, thereby shortening data transmission time and avoiding the occurrence of information transmission between independent hardware. Error and information delay, which lays the foundation for high-precision target positioning and tracking.
- the present embodiment combines the radar 100 with the digital imaging device 300, and firstly uses the radar 100 to scan to achieve coarse positioning of the obstacle target.
- the image data obtained by the digital imaging device 300 is used to identify and detect the obstacle target type to determine whether it is necessary to avoid.
- the radar 100 is used to accurately scan the obstacle target that needs to be avoided, thereby realizing real-time realization of the obstacle target. Position tracking.
- the target detection method provided by this embodiment is performed by the controller 200.
- the specific principle is as follows. Please refer to FIG. 2 .
- Step S200 receiving the first scan data transmitted by the radar 100.
- the first scan data is obtained by the radar 100 performing a first type scan on the first target area.
- the first type of scanning has a low precision and a wide scanning range, and the purpose is to initially locate the obstacle target for subsequent target recognition. From the first scan data, the approximate distribution of the obstacle targets within the scan range can be determined, so that the coarse positioning information of each obstacle target can be obtained.
- the coarse positioning information may include, for example, preliminary distance information and orientation information, wherein the orientation information may include, for example, a horizontal angle and a vertical pitch angle.
- the radar 100 may perform scanning at each time point separated by a set time period. At this time, the first scan data includes data scanned multiple times in different time periods. That includes multi-frame scan data, so that the accuracy of the scan data can be improved.
- Step S400 receiving image data transmitted by the digital image device 300.
- the image data is obtained by the digital image device 300 for imaging the second target area, and at the same time, there is a coincident area between the second target area and the first target area.
- the range of the first target area and the second target area may be the same or different, but at least there is a coincident area between the two, and the overlapping area may be correspondingly set according to different situations.
- the target detection method provided by this embodiment is to detect the obstacle target in the overlapping area.
- step S200 and step S400 is not limited to the above-described one case, as long as the first scan data and the image data can be received.
- the order of the two steps may be interchanged, or both steps may be performed simultaneously.
- Step S500 searching for image information corresponding to each obstacle object from the image data according to the first scan data to identify the type of each obstacle object.
- the type identification and detection of the coarse positioning target is performed by the digital imaging device 300.
- the image data can be directly found in the corresponding region.
- the image information corresponding to each obstacle object is used to identify the type of the obstacle object based on the color, texture, and the like of the image. Therefore, the first embodiment of the present invention realizes the rough positioning of the obstacle object by the radar 100, thereby reducing the unnecessary image search process, thereby realizing the function of quickly identifying the obstacle target.
- step S600 when it is determined that there is an obstacle target that needs to be circumvented according to the type of each obstacle object, the control radar 100 performs a second type of scan and tracking on the obstacle object that needs to be circumvented. Among them, the accuracy of the second type of scanning is greater than the precision of the first type of scanning.
- the specific judgment manner of determining whether there is an obstacle target to be circumvented according to the type of each obstacle object is, for example, if the obstacle target is a pedestrian or a remitted vehicle, the evasion is required; if the obstacle target is When objects such as garbage plastic bags blown by the wind do not need to be circumvented.
- the second type of scanning corresponds to a fine scan, which can detect the motion trajectory, the moving direction, the moving speed, and the like of the obstacle object.
- the accuracy of the second type of scanning is greater than the precision of the first type of scanning.
- the scanning range of the second type of scanning is smaller and more concentrated, the angular resolution is higher, the scanning speed is slower, etc., compared with the first type of scanning.
- the relative speed measurement is also required in real time.
- the first scan data scanned by the radar 100 can perform preliminary positioning on the obstacle object, so that the target data corresponding to each obstacle object is searched for from the image data acquired by the digital image device 300 according to the first scan data.
- Image information can reduce the unnecessary image search process and speed up image processing, so that the type of obstacle can be quickly identified.
- the control radar 100 performs a second type of scan and tracking on the obstacle target that needs to be circumvented, that is, performs fine scan tracking, so that the radar 100 can be accurately positioned.
- the advantage is to achieve accurate positioning and tracking of obstacle targets. Therefore, the target detection method improves the accuracy and efficiency of detecting the obstacle target, thereby reducing the safety hazard of driving.
- the above-mentioned target detection method can not only speed up the recognition of the obstacle target type by the initial positioning information, but also utilize the radar 100 by recognizing the obstacle target. Accurate positioning and tracking of obstacle targets to be circumvented to improve the shortcomings of digital imaging equipment, such as the use of image sequences for tracking moving objects, and the ease of tracking after long-term tracking.
- the image data provided by the digital imaging device 300 can accurately determine the type of the obstacle target, thereby enabling rapid guidance of the radar. 100 is positioned to the target area, reducing power consumption. Therefore, the above-described target detection method can realize the recognition and tracking of the obstacle object faster, more accurate, more stable, and lower power consumption by appropriately utilizing the radar 100 and the digital image device 300.
- step S500 specifically includes the following content, please refer to FIG. 3.
- Step S510 prioritizing the obstacle objects according to the first scan data to obtain priority information.
- Step S510 specifically includes the following.
- Step S511 the distance between each obstacle object and the reference object is obtained by the first scan data.
- the reference object is, for example, a radar 100 or a driving carrier.
- Step S512 performing difference operation between the current frame data in the first scan data and the previous frame data.
- the first scan data includes multi-frame data, and the time of each frame data scan is different.
- the current frame data refers to the data obtained by the radar 100 in real time scanning
- the previous frame data refers to the data obtained by the radar 100 adjacent to the current time and scanned before the current time.
- step S515 it is determined whether there is an obstacle target in a moving state, and if so, step S515 is performed, otherwise step S514 is performed.
- step S512 it is determined whether there is an obstacle target in a moving state based on the result of the difference operation.
- the relationship between the corresponding current frame data and the previous frame data may also be different, so the result of the difference operation can be used to determine whether the obstacle object is in motion.
- the result of the difference operation can be used to determine whether the obstacle object is in motion.
- the motion direction and speed of the obstacle object can also be obtained according to the difference operation result.
- the current frame data may be divided with the previous frame data, as long as it can be determined whether the obstacle object is in motion. .
- step S514 the obstacle target in the moving state is prioritized higher than the obstacle target in the stationary state, and the obstacle target in the stationary state is prioritized according to the size of the distance.
- the obstacle target in motion Since the obstacle target in motion has variability and complexity, the obstacle target in motion is given a higher priority, and it is easier to improve driving safety.
- the priority may be further calculated according to parameters such as the moving direction and relative speed of each obstacle object.
- Step S515 prioritizing the obstacle objects according to the size of the distance.
- the nearest obstacle target is more threatening, so a higher priority is given.
- step S510 is not limited to the above one case, as long as each obstacle object can be prioritized according to the actual situation and according to the first scan data. For example, if it is known that the obstacle targets are all at rest, it is not necessary to provide step S513 and step S515.
- Step S520 sequentially searching for image information corresponding to each obstacle object from the image data according to the priority information to identify the type of each obstacle object.
- the priority information of the obstacle object since the priority information of the obstacle object has a certain regularity, according to the regularity of the priority information, it is easy to find the corresponding image information in the image data without performing full scanning on the image data, thereby further Speed up the recognition of obstacle target types.
- the obstacle object with greater threat can be judged in time, so that corresponding avoidance measures can be taken in time to further improve the safety of driving.
- step S500 is not limited to the above, as long as the image information corresponding to each obstacle object can be searched from the image data according to the first scan data to identify the type of each obstacle object.
- the specific implementation manner of the target detection method may also be the following.
- the method further includes:
- step S100 the control radar 100 performs a first type of scanning on the first target area.
- the method further includes:
- step S300 the digital imaging device 300 is controlled to scan and image the second target area.
- step S100 and step S400 are not limited to the above case.
- step S100 and step S300 may be simultaneously performed, that is, control radar 100 and digital imaging device 300 start collecting data at the same time, and then perform separately.
- Step S200 and step S400 may be performed by other devices on the premise that the target detection speed can be satisfied.
- step S600 specifically includes the following.
- step S610 it is determined whether there is an obstacle target that needs to be circumvented according to the type of each obstacle object. If yes, step S620 is performed; otherwise, execution from step S100 is continued.
- step S620 the control radar 100 performs a second type of scanning and tracking on the obstacle target that needs to be circumvented.
- the operational states of the radar 100 and the digital imaging device 300 are uniformly controlled by the controller 200, thereby facilitating timely detection of obstacle objects.
- step S620 the method further includes:
- Step S700 Output a corresponding evasive route or driving control suggestion according to the second scan data obtained by the radar 100 for performing the second type of scanning.
- the positioning information of the obstacle object relative to the speed and direction of the driving carrier may be measured according to the second scan data, and the corresponding planning route is made according to the type of the obstacle target. For example, if the obstacle target is a building or a pedestrian, when the relative speed is high and the distance is very close, it should be stopped immediately; if the obstacle target is to enter the vehicle or the vehicle driving ahead, reduce the speed and according to the safety distance Automatically adjust speed following.
- step S700 is not limited to the above case, and step S700 can also be performed, for example, by the driving carrier control system 400. At this time, the above target detection method does not need to provide step S700.
- FIGS. 2 to FIG. 5 are schematic flowcharts of a method according to an embodiment of the present invention. It should be understood that although the various steps in the flowcharts of FIGS. 2 through 5 are sequentially displayed in accordance with the indication of the arrows, these steps are not necessarily performed in the order indicated by the arrows. Except as explicitly stated herein, the execution of these steps is not strictly limited, and may be performed in other sequences. Moreover, at least some of the steps in FIGS.
- 2 to 5 may include a plurality of sub-steps or stages, which are not necessarily performed at the same time, but may be executed at different times, and the execution order thereof is also It is not necessarily performed sequentially, but may be performed alternately or alternately with at least a portion of other steps or sub-steps or stages of other steps.
- the controller 200 includes a memory 210 and a processor 220.
- the memory 210 stores instructions, which when executed by the processor 220, may cause the processor 220 to execute the processes including the embodiments of the target detection method, which are not described in detail herein.
- the memory 210 is a computer readable storage medium, and specifically may be a non-volatile storage medium such as a magnetic disk, an optical disk, a read-only memory (ROM), or a random storage memory ( Random –Access- Memory, RAM), etc.
- the above instructions may also be stored in one or more computer readable storage media, and the instructions may be executed by one or more processors including the processes of the various embodiments of the two-dimensional lidar ranging method described above.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Electromagnetism (AREA)
- Computer Networks & Wireless Communication (AREA)
- Radar Systems Or Details Thereof (AREA)
- Traffic Control Systems (AREA)
Abstract
一种目标检测方法、系统及控制器,包括:接收由雷达发送的第一扫描数据;所述第一扫描数据由所述雷达对第一目标区域进行第一类型扫描后得出;接收由数字影像设备发送的影像数据;所述影像数据由所述数字影像设备对第二目标区域进行成像得出;所述第二目标区域与所述第一目标区域之间存在重合区域;根据所述第一扫描数据从所述影像数据中寻找各障碍物目标对应的影像信息,以识别各所述障碍物目标的类型;根据各所述障碍物目标的类型判定存在需要规避的障碍物目标时,控制所述雷达对所述需要规避的障碍物目标进行第二类型扫描并跟踪;所述第二类型扫描的精度大于所述第一类型扫描的精度。
Description
【技术领域】
本发明涉及无人驾驶技术领域,特别是涉及一种目标检测方法、系统及控制器。
【背景技术】
随着科技的发展创新,无人驾驶技术日渐成熟。其中,无人驾驶汽车是通过车载传感系统感知道路环境,自动规划行车路线并控制车辆到达预定目标的智能汽车。它是利用车载传感器来感知车辆周围环境,并根据感知所获得的道路、车辆位置和障碍物信息,控制车辆的转向和速度,从而使车辆能够安全、可靠地在道路上行驶。
然而,由于道路交通的复杂、多变性,在行驶过程中,对于行人、车辆等障碍物目标需要实时精确得检测出来以进行合理避让,才可提高行车安全性。故如何提高检测障碍物目标的精确性是亟待解决的问题。
【发明内容】
基于此,有必要提供一种能够提高检测障碍物目标的精确性的目标检测方法、系统及控制器。
一种目标检测方法,包括:
接收由雷达发送的第一扫描数据;所述第一扫描数据由所述雷达对第一目标区域进行第一类型扫描后得出;
接收由数字影像设备发送的影像数据;所述影像数据由所述数字影像设备对第二目标区域进行成像得出;所述第二目标区域与所述第一目标区域之间存在重合区域;
根据所述第一扫描数据从所述影像数据中寻找各障碍物目标对应的影像信息,以识别各所述障碍物目标的类型;
根据各所述障碍物目标的类型判定存在需要规避的障碍物目标时,控制所述雷达对所述需要规避的障碍物目标进行第二类型扫描并跟踪;所述第二类型扫描的精度大于所述第一类型扫描的精度。
一种控制器,包括存储器和处理器,所述存储器中储存有指令,所述指令被所述处理器执行时,可使得所述处理器执行以下步骤:
接收由雷达发送的第一扫描数据;所述第一扫描数据由所述雷达对第一目标区域进行第一类型扫描后得出;
接收由数字影像设备发送的影像数据;所述影像数据由所述数字影像设备对第二目标区域进行成像得出;所述第二目标区域与所述第一目标区域之间存在重合区域;
根据所述第一扫描数据从所述影像数据中寻找各障碍物目标对应的影像信息,以识别各所述障碍物目标的类型;
根据各所述障碍物目标的类型判定存在需要规避的障碍物目标时,控制所述雷达对所述需要规避的障碍物目标进行第二类型扫描并跟踪;所述第二类型扫描的精度大于所述第一类型扫描的精度。
一种目标检测系统,包括:雷达,用于提供扫描数据;数字影像设备,用于提供影像数据;驾驶载体控制系统,用于控制驾驶载体的运行状态;及控制器,分别与所述雷达、数字影像设备、驾驶载体控制系统连接:所述控制器用于接收由雷达发送的第一扫描数据及由数字影像设备发送的影像数据;其中,所述第一扫描数据由所述雷达对第一目标区域进行第一类型扫描后得出,所述影像数据由所述数字影像设备对第二目标区域进行成像得出;所述控制器还用于根据所述第一扫描数据从所述影像数据中寻找各障碍物目标对应的影像信息,以识别各所述障碍物目标的类型,其中,所述第二目标区域与所述第一目标区域之间存在重合区域;所述控制器还用于根据各所述障碍物目标的类型判定存在需要规避的障碍物目标时,控制所述雷达对所述需要规避的障碍物目标进行第二类型扫描并跟踪,其中,所述第二类型扫描的精度大于所述第一类型扫描的精度。
在该目标检测方法、系统及控制器中,由雷达执行的第二类型扫描的精度大于第一类型扫描的精度,那么第一类型扫描相当于粗定位,第二类型扫描相当于精细定位。同时,首先接收由雷达发送的第一扫描数据及由数字影像设备发送的影像数据,由于第一扫描数据可以提供障碍物目标的初步定位信息,故根据第一扫描数据从影像数据中寻找各障碍物目标对应的影像信息,能减少不必要的图像搜索,加快了图像处理速度,从而能够快速识别出障碍物的类型。之后,根据障碍物的类型判定存在需要规避的障碍物目标时,再控制雷达对需要规避的障碍物目标进行第二类型扫描并跟踪,即进行精细扫描跟踪,从而利用雷达能够精确定位的优势来实现后续运行过程中对障碍物目标的精确跟踪与检测。因此,该目标检测方法、系统及控制器提高了对障碍物目标进行检测的精度及效率,从而降低了行车安全隐患。
【附图说明】
为了更清楚地说明本发明实施例或现有技术中的技术方案,下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本发明的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他实施例的附图。
图1为一实施例提供的目标检测系统的硬件结构图;
图2为一实施例提供的目标检测方法的流程图;
图3为图2所示实施例的目标检测方法中步骤S500的其中一种具体流程图;
图4为图3所示实施例的目标检测方法中步骤S510的其中一种具体流程图;
图5为另一实施例提供的目标检测方法的流程图;
图6为图1所示实施例的目标检测系统中控制器的内部结构图。
【具体实施方式】
为了便于理解本发明,下面将参照相关附图对本发明进行更全面的描述。附图中给出了本发明的较佳实施例。但是,本发明可以以许多不同的形式来实现,并不限于本文所描述的实施例。相反地,提供这些实施例的目的是使对本发明的公开内容的理解更加透彻全面。
除非另有定义,本文所使用的所有的技术和科学术语与属于发明的技术领域的技术人员通常理解的含义相同。本文中在发明的说明书中所使用的术语只是为了描述具体的实施例的目的,不是旨在限制本发明。本文所使用的术语“和/或”包括一个或多个相关的所列项目的任意的和所有的组合。
一实施例提供了一种目标检测系统,其涉及到的硬件结构如图1所示
。雷达100,用于提供扫描数据,其发射某种测量波(其中,测量波对于激光雷达或者微波/毫米波雷达为电磁波,对于超声波雷达则为超声波)对目标区域进行扫描并接收回波信息,从而获得目标物体距离雷达的距离、速度、方位等信息。具体的,雷达100可以为激光雷达、超声波雷达、微波/毫米波雷达或其他类型的雷达。然而,雷达100虽然可提供精确的位置信息,却无法提供视觉信息以进行目标的精确识别与检测,并且功耗较高,扫描速度也有一定的局限性。数字影像设备300用于提供影像数据,其可提供丰富的视觉信息以对目标进行识别与检测,但无法保证位置的精确性,而且长时间对目标进行跟踪容易出现跟丢的问题。控制器200是主控系统,用于控制雷达100及数字影像设备300的运行,并进行相关的数据处理过程。驾驶载体控制系统400用于控制驾驶载体的运行状态,其例如为无人驾驶汽车中的智能驾驶仪。本实施例中,控制器200根据对障碍物目标的检测及跟踪并通过驾驶载体控制系统400来控制驾驶载体的运行过程。
进一步的,还可以将雷达100、控制器200及数字影像设备300集成到单一硬件内,以实现数据共享,从而缩短了数据传输时间,同时还可以避免发生各独立硬件之间信息传递而产生的误差及信息延迟现象,从而为高精度目标定位与跟踪打下硬件基础。
为了提高无人驾驶技术中对障碍物目标的检测精度及效率,降低行车安全隐患,本实施例将雷达100与数字影像设备300结合使用,首先利用雷达100扫描以实现障碍物目标的粗定位,再利用数字影像设备300得出的影像数据来识别并检测障碍物目标类型以判断是否需要避让,最后再次利用雷达100对需要避让的障碍物目标进行精确扫描,从而最终实现对障碍物目标的实时定位追踪。同时,本实施例提供的目标检测方法由控制器200来执行,具体原理如下,请参考图2。
步骤S200,接收由雷达100发送的第一扫描数据。其中,第一扫描数据由雷达100对第一目标区域进行第一类型扫描后得出。
该步骤中,第一类型扫描的精度较低且扫描范围较广,目的是对障碍物目标进行初步定位,以便于进行后续的目标识别。从第一扫描数据内可以确定扫描范围内障碍物目标的大致分布情况,从而可以得出各障碍物目标的粗定位信息。该粗定位信息例如可以包括初步的距离信息和方位信息,其中方位信息例如可以包括水平角度和纵向俯仰角度。另外,在时间允许的前提下,雷达100在进行第一类型扫描时,可以在相隔设定时间段的各时间点分别进行扫描,这时,第一扫描数据包括不同时间内多次扫描的数据,即包括多帧扫描数据,从而可以提高扫描数据的精确度。
步骤S400,接收由数字影像设备300发送的影像数据。其中,该影像数据由数字影像设备300对第二目标区域进行成像得出,同时,第二目标区域与第一目标区域之间存在重合区域。
其中,第一目标区域和第二目标区域的范围可以相同或者不同,但两者之间至少要有重合区域,且该重合区域可以根据不同的情况进行相应的设置。本实施例提供的目标检测方法就是针对该重合区域内的障碍物目标来进行检测的。
另外,步骤S200和步骤S400的顺序不限于上述一种情况,只要能够接收到第一扫描数据、影像数据即可。例如,根据雷达100、数字影像设备300实际的运行情况,也可将这两个步骤的顺序互换,或者同时执行这两个步骤。
步骤S500,根据上述第一扫描数据从上述影像数据中寻找各障碍物目标对应的影像信息,以识别各障碍物目标的类型。
在该步骤中,相当于通过数字影像设备300对粗定位目标进行类型识别与检测,具体为:根据上述由第一扫描数据得出的粗定位信息,可以直接从影像数据中找到位于相应区域内的各障碍物目标对应的影像信息,从而根据影像的颜色、纹理等信息来对障碍物目标的类型进行识别。因此,本实施例通过雷达100首先对障碍物目标实现粗定位能够减少不必要的图像搜索过程,从而实现快速对障碍物目标进行识别的功能。
步骤S600,根据各障碍物目标的类型判定存在需要规避的障碍物目标时,控制雷达100对需要规避的障碍物目标进行第二类型扫描并跟踪。其中,第二类型扫描的精度大于第一类型扫描的精度。
该步骤中,根据各障碍物目标的类型判定是否存在需要规避的障碍物目标的具体判断方式例如为:若障碍物目标为行人或者汇入的车辆时,则需要规避;若障碍物目标是被风吹起的垃圾塑料袋等物体时,则无需规避。
另外,第二类型扫描相当于精细扫描,其可以检测障碍物目标的运动轨迹、运动方向、运动速度等。第二类型扫描的精度大于第一类型扫描的精度,具体可以表现在:与第一类型扫描相比,第二类型扫描的扫描范围更小更集中、角分辨率更高、扫描速度更慢等,从而实现对需规避的障碍物目标的精确定位。另外,若障碍物目标为运动物体,还需实时进行相对速度的测算。
因此,在上述目标检测方法中,雷达100扫描的第一扫描数据可以对障碍物目标进行初步的定位,故根据第一扫描数据从数字影像设备300获取的影像数据中寻找各障碍物目标对应的影像信息,能够减少不必要的图像搜索过程,加快了图像处理速度,从而能够快速识别出障碍物的类型。之后,根据障碍物的类型判定存在需要规避的障碍物目标时,再控制雷达100对需要规避的障碍物目标进行第二类型扫描并跟踪,即进行精细扫描跟踪,从而可以利用雷达100具备精确定位的优势来实现后续对障碍物目标的精确定位与跟踪。因此,该目标检测方法提高了对障碍物目标进行检测的精度及效率,从而降低了行车安全隐患。
另外,上述目标检测方法与单纯使用数字成像设备进行检测的方法相比,不仅可以通过初定位信息加快了对障碍物目标类型进行识别的速度,还能通过识别出障碍物目标后再利用雷达100对需规避的障碍物目标进行精确定位与跟踪的方式,来改善数字成像设备例如利用图像序列对运动物体跟踪具有的稳定性较差且长时间跟踪后容易跟丢的缺陷。同时,上述目标检测方法与单纯使用雷达进行检测的方法相比,由于雷达的功耗通常较高,因此通过数字影像设备300提供的影像数据可以精确判断障碍物目标的类型,从而能够快速指引雷达100定位到目标区域,减少了功耗。因此,上述目标检测方法通过对雷达100与数字影像设备300进行恰当的利用,从而能够实现对障碍物目标进行更快、更准确、更稳定、更低功耗的识别和跟踪。
具体的,步骤S500具体包括以下内容,请参考图3。
步骤S510,根据上述第一扫描数据对各障碍物目标进行优先级排序以得出优先级信息。
其中,可以根据设定的准则来对障碍物目标进行优先级排序。本实施例提供了一种具体优先级排序准则,请参考图4。步骤S510具体包括以下内容。
步骤S511,通过第一扫描数据得出各障碍物目标与参考物体之间的距离。
其中,参考物体例如为雷达100或驾驶载体。
步骤S512,将第一扫描数据中的当前帧数据与前一帧数据进行差分运算。其中,第一扫描数据包括多帧数据,且各帧数据扫描的时间不同。
在该步骤中,当前帧数据是指雷达100实时扫描得出的数据,前一帧数据是指雷达100在与当前时刻相邻且先于当前时刻进行扫描而得出的数据。
步骤S513,判断是否存在处于运动状态的障碍物目标,若是,执行步骤S515,否则执行步骤S514。
在步骤S512得出差分运算的结果后,即可根据该差分运算的结果来判断是否存在处于运动状态的障碍物目标。其中,障碍物目标分别处于运动状态和静止状态时,其对应的当前帧数据与前一帧数据之间的关系也会存在差别,因此通过差分运算的结果即可判断障碍物目标是否处于运动状态。例如,在已补偿驾驶载体自身位移和速度的基础上,若障碍物目标的当前帧数据和前一帧数据相同,则表明该障碍物目标为静止状态;否则,表明该障碍物目标处于运动状态。另外,若判断该障碍物目标处于运动状态,那么根据差分运算结果还可以得出该障碍物目标的运动方向和速度。
可以理解的是,不限于通过步骤S512的方法来判断障碍物目标是否处于运动状态,例如也可以将当前帧数据与前一帧数据进行除法运算,只要能够判断障碍物目标是否处于运动状态即可。
步骤S514,使处于运动状态的障碍物目标的优先级高于处于静止状态的障碍物目标,同时将处于静止状态的障碍物目标根据距离的大小进行优先级排序。
由于处于运动状态的障碍物目标具有多变性和复杂性,因此将处于运动状态的障碍物目标给予较高的优先级,更易于提高行车安全性。
另外,若处于运动状态的障碍物目标存在多个,可以进一步按照各障碍物目标的运动方向、相对速度等参数来计算优先级。
步骤S515,根据距离的大小对各障碍物目标进行优先级排序。
若障碍物目标都处于静止状态,那么距离最近的障碍物目标威胁较大,因此给予较高的优先级。
可以理解的是,步骤S510的具体实现方式不限于上述一种情况,只要能够根据实际情况并根据上述第一扫描数据对各障碍物目标进行优先级排序即可。例如,若在已知障碍物目标均处于静止状态时,则无需设置步骤S513和步骤S515。
步骤S520,根据上述优先级信息依次从影像数据中寻找各障碍物目标对应的影像信息,以识别各障碍物目标的类型。
其中,由于障碍物目标的优先级信息存在一定的规律性,因此根据优先级信息具有的规律性就会很容易在影像数据中找到相应的影像信息,而无需对影像数据进行全部扫描,从而进一步加快了障碍物目标类型的识别速度。
另外,根据优先级信息还可以及时判断出威胁较大的障碍物目标,从而能够及时采取相应的避让措施,进一步提高行车的安全性。
可以理解的是,步骤S500的具体执行方式不限于上述情况,只要能够根据第一扫描数据从影像数据中寻找各障碍物目标对应的影像信息,以识别各障碍物目标的类型即可。
进一步的,如图5所示,目标检测方法的具体实现方式还可以为以下情况。
步骤S200之前还包括:
步骤S100,控制雷达100对第一目标区域进行第一类型扫描。
步骤S400之前还包括:
步骤S300,控制数字影像设备300对第二目标区域进行扫描成像。
需要说明的是,步骤S100至步骤S400之间的顺序并不限于上述情况,例如也可以使步骤S100与步骤S300同时执行,即控制雷达100与数字成像设备300同时开始采集数据,之后再分别执行步骤S200和步骤S400。另外,根据实际情况,在能够满足目标检测速度的前提下,步骤S100或步骤S400也可由其他设备来执行。
另外,步骤S600具体包括以下内容。
步骤S610,根据各障碍物目标的类型判断是否存在需要规避的障碍物目标,若是,执行步骤S620,否则,继续从步骤S100开始执行。
步骤S620,控制雷达100对需要规避的障碍物目标进行第二类型扫描并跟踪。
因此,在整个目标检测的过程中,雷达100和数字影像设备300的运行状态都由控制器200来统一控制,从而便于及时对障碍物目标进行检测。
进一步的,步骤S620之后还包括:
步骤S700,根据上述雷达100进行第二类型扫描得出的第二扫描数据输出相应的规避路线或驾驶控制建议。
其中,可以根据第二扫描数据测算障碍物目标相对驾驶载体的速度、方向等定位信息,并结合障碍物目标的类型做出相应的规划路线。例如:如果障碍物目标是建筑物或者行人,当其相对速度较高并且距离很近时,应立即刹停;如果障碍物目标是汇入车辆或者前方行驶的车辆,则降低速度并根据安全距离自动调整速度跟随。
可以理解的是,步骤S700不限于上述情况,例如也可由驾驶载体控制系统400来执行步骤S700。这时,上述目标检测方法则无需设置步骤S700。
图2至图5为本发明实施例的方法的流程示意图。应该理解的是,虽然图2至图5的流程图中的各个步骤按照箭头的指示依次显示,但是这些步骤并不是必然按照箭头指示的顺序依次执行。除非本文中有明确的说明,这些步骤的执行并没有严格的顺序限制,其可以以其他的顺序执行。而且,图2至图5中的至少一部分步骤可以包括多个子步骤或者多个阶段,这些子步骤或者阶段并不必然是在同一时刻执行完成,而是可以在不同的时刻执行,其执行顺序也不必然是依次进行,而是可以与其他步骤或者其他步骤的子步骤或者阶段的至少一部分轮流或者交替地执行。
进一步的,如图6所示,控制器200包括存储器210和处理器220。其中,存储器210中储存有指令,该指令被处理器220执行时,可使得处理器220执行包括上述目标检测方法各实施例的流程,这里就不再一一详述。其中,存储器210是一种计算机可读取存储介质,具体可为磁碟、光盘、只读存储记忆体(Read-Only-Memory,ROM)等非易失性存储介质,或随机存储记忆体(Random
–Access-
Memory,RAM)等。另外,上述指令也可存储于一个或多个计算机可读取存储介质中,且该指令可以被一个或多个处理器执行包括上述二维激光雷达测距方法各实施例的流程。
以上所述实施例的各技术特征可以进行任意的组合,为使描述简洁,未对上述实施例中的各个技术特征所有可能的组合都进行描述,然而,只要这些技术特征的组合不存在矛盾,都应当认为是本说明书记载的范围。
以上所述实施例仅表达了本发明的几种实施方式,其描述较为具体和详细,但并不能因此而理解为对发明专利范围的限制。应当指出的是,对于本领域的普通技术人员来说,在不脱离本发明构思的前提下,还可以做出若干变形和改进,这些都属于本发明的保护范围。因此,本发明专利的保护范围应以所附权利要求为准。
Claims (19)
- 一种目标检测方法,包括:接收由雷达发送的第一扫描数据;所述第一扫描数据由所述雷达对第一目标区域进行第一类型扫描后得出;接收由数字影像设备发送的影像数据;所述影像数据由所述数字影像设备对第二目标区域进行成像得出;所述第二目标区域与所述第一目标区域之间存在重合区域;根据所述第一扫描数据从所述影像数据中寻找各障碍物目标对应的影像信息,以识别各所述障碍物目标的类型;及根据各所述障碍物目标的类型判定存在需要规避的障碍物目标时,控制所述雷达对所述需要规避的障碍物目标进行第二类型扫描并跟踪;所述第二类型扫描的精度大于所述第一类型扫描的精度。
- 根据权利要求1所述的方法,其特征在于,根据所述第一扫描数据从所述影像数据中寻找各障碍物目标对应的影像信息,以识别各所述障碍物目标的类型的步骤包括:根据所述第一扫描数据对各所述障碍物目标进行优先级排序以得出优先级信息;根据所述优先级信息依次从所述影像数据中寻找各所述障碍物目标对应的影像信息,以识别各所述障碍物目标的类型。
- 根据权利要求2所述的方法,其特征在于,根据所述第一扫描数据对各所述障碍物目标进行优先级排序以得出优先级信息的步骤包括:通过所述第一扫描数据得出各所述障碍物目标与参考物体之间的距离;根据所述距离的大小对各所述障碍物目标进行优先级排序。
- 根据权利要求3所述的方法,其特征在于,根据所述距离的大小对各所述障碍物目标进行优先级排序的步骤之前还包括:判断是否存在处于运动状态的障碍物目标,若是,使所述处于运动状态的障碍物目标的优先级高于处于静止状态的障碍物目标,同时将处于静止状态的障碍物目标根据所述距离的大小进行优先级排序;否则,执行根据所述距离的大小对各所述障碍物目标进行优先级排序的步骤。
- 根据权利要求4所述的方法,其特征在于,判断是否存在处于运动状态的障碍物目标的步骤前还包括:将所述第一扫描数据中的当前帧数据与前一帧数据进行差分运算;其中,所述第一扫描数据包括在不同时间扫描得出的多帧数据;同时,判断是否存在处于运动状态的障碍物目标的步骤为:根据所述差分运算的结果判断是否存在处于运动状态的障碍物目标。
- 根据权利要求1所述的方法,其特征在于,在接收由雷达发送的第一扫描数据的步骤前还包括:控制所述雷达对所述第一目标区域进行第一类型扫描。
- 根据权利要求6所述的方法,其特征在于,根据各所述障碍物目标的类型判定存在需要规避的障碍物目标时,控制所述雷达对所述需要规避的障碍物目标进行第二类型扫描并跟踪的步骤包括:根据各所述障碍物目标的类型判断是否存在需要规避的障碍物目标,若是,控制所述雷达对所述需要规避的障碍物目标进行第二类型扫描并跟踪;否则,继续从控制所述雷达对所述第一目标区域进行第一类型扫描的步骤开始执行。
- 根据权利要求1所述的方法,其特征在于,接收由数字影像设备发送的影像数据的步骤之前还包括:控制所述数字影像设备对所述第二目标区域进行扫描成像。
- 根据权利要求1所述的方法,其特征在于,根据各所述障碍物目标的类型判定存在需要规避的障碍物目标时,控制所述雷达对所述需要规避的障碍物目标进行第二类型扫描并跟踪的步骤之后还包括:根据所述第二类型扫描得出的第二扫描数据输出相应的规避路线或驾驶控制建议。
- 一种控制器,包括存储器和处理器,所述存储器中储存有指令,所述指令被所述处理器执行时,可使得所述处理器执行以下步骤:接收由雷达发送的第一扫描数据;所述第一扫描数据由所述雷达对第一目标区域进行第一类型扫描后得出;接收由数字影像设备发送的影像数据;所述影像数据由所述数字影像设备对第二目标区域进行成像得出;所述第二目标区域与所述第一目标区域之间存在重合区域;根据所述第一扫描数据从所述影像数据中寻找各障碍物目标对应的影像信息,以识别各所述障碍物目标的类型;根据各所述障碍物目标的类型判定存在需要规避的障碍物目标时,控制所述雷达对所述需要规避的障碍物目标进行第二类型扫描并跟踪;所述第二类型扫描的精度大于所述第一类型扫描的精度。
- 根据权利要求10所述的控制器,其特征在于,根据所述第一扫描数据从所述影像数据中寻找各障碍物目标对应的影像信息,以识别各所述障碍物目标的类型的步骤包括:根据所述第一扫描数据对各所述障碍物目标进行优先级排序以得出优先级信息;根据所述优先级信息依次从所述影像数据中寻找各所述障碍物目标对应的影像信息,以识别各所述障碍物目标的类型。
- 根据权利要求11所述的控制器,其特征在于,根据所述第一扫描数据对各所述障碍物目标进行优先级排序以得出优先级信息的步骤包括:通过所述第一扫描数据得出各所述障碍物目标与参考物体之间的距离;根据所述距离的大小对各所述障碍物目标进行优先级排序。
- 根据权利要求12所述的控制器,其特征在于,根据所述距离的大小对各所述障碍物目标进行优先级排序的步骤之前还包括:判断是否存在处于运动状态的障碍物目标,若是,使所述处于运动状态的障碍物目标的优先级高于处于静止状态的障碍物目标,同时将处于静止状态的障碍物目标根据所述距离的大小进行优先级排序;否则,执行根据所述距离的大小对各所述障碍物目标进行优先级排序的步骤。
- 根据权利要求13所述的控制器,其特征在于,判断是否存在处于运动状态的障碍物目标的步骤前还包括:将所述第一扫描数据中的当前帧数据与前一帧数据进行差分运算;其中,所述第一扫描数据包括在不同时间扫描得出的多帧数据;同时,判断是否存在处于运动状态的障碍物目标的步骤为:根据所述差分运算的结果判断是否存在处于运动状态的障碍物目标。
- 根据权利要求10所述的控制器,其特征在于,在接收由雷达发送的第一扫描数据的步骤前还包括:控制所述雷达对所述第一目标区域进行第一类型扫描。
- 根据权利要求15所述的控制器,其特征在于,根据各所述障碍物目标的类型判定存在需要规避的障碍物目标时,控制所述雷达对所述需要规避的障碍物目标进行第二类型扫描并跟踪的步骤包括:根据各所述障碍物目标的类型判断是否存在需要规避的障碍物目标,若是,控制所述雷达对所述需要规避的障碍物目标进行第二类型扫描并跟踪;否则,继续从控制所述雷达对所述第一目标区域进行第一类型扫描的步骤开始执行。
- 根据权利要求10所述的控制器,其特征在于,接收由数字影像设备发送的影像数据的步骤之前还包括:控制所述数字影像设备对所述第二目标区域进行扫描成像。
- 根据权利要求10所述的控制器,其特征在于,根据各所述障碍物目标的类型判定存在需要规避的障碍物目标时,控制所述雷达对所述需要规避的障碍物目标进行第二类型扫描并跟踪的步骤之后还包括:根据所述第二类型扫描得出的第二扫描数据输出相应的规避路线或驾驶控制建议。
- 一种目标检测系统,包括:雷达,用于提供扫描数据;数字影像设备,用于提供影像数据;驾驶载体控制系统,用于控制驾驶载体的运行状态;及控制器,分别与所述雷达、数字影像设备、驾驶载体控制系统连接:所述控制器用于接收由雷达发送的第一扫描数据及由数字影像设备发送的影像数据;其中,所述第一扫描数据由所述雷达对第一目标区域进行第一类型扫描后得出,所述影像数据由所述数字影像设备对第二目标区域进行成像得出;所述控制器还用于根据所述第一扫描数据从所述影像数据中寻找各障碍物目标对应的影像信息,以识别各所述障碍物目标的类型,其中,所述第二目标区域与所述第一目标区域之间存在重合区域;所述控制器还用于根据各所述障碍物目标的类型判定存在需要规避的障碍物目标时,控制所述雷达对所述需要规避的障碍物目标进行第二类型扫描并跟踪,其中,所述第二类型扫描的精度大于所述第一类型扫描的精度。
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/CN2016/096358 WO2018035711A1 (zh) | 2016-08-23 | 2016-08-23 | 目标检测方法及系统 |
| US16/327,685 US10891499B2 (en) | 2016-08-23 | 2016-08-23 | Target detection method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/CN2016/096358 WO2018035711A1 (zh) | 2016-08-23 | 2016-08-23 | 目标检测方法及系统 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2018035711A1 true WO2018035711A1 (zh) | 2018-03-01 |
Family
ID=61246007
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2016/096358 Ceased WO2018035711A1 (zh) | 2016-08-23 | 2016-08-23 | 目标检测方法及系统 |
Country Status (2)
| Country | Link |
|---|---|
| US (1) | US10891499B2 (zh) |
| WO (1) | WO2018035711A1 (zh) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112541416A (zh) * | 2020-12-02 | 2021-03-23 | 深兰科技(上海)有限公司 | 跨雷达障碍物跟踪方法、装置、电子设备和存储介质 |
| CN113635864A (zh) * | 2021-07-27 | 2021-11-12 | 际络科技(上海)有限公司 | 车载激光雷达的清洗方法、装置、电子设备及存储介质 |
| CN115201804A (zh) * | 2021-04-12 | 2022-10-18 | 武汉智行者科技有限公司 | 目标速度估计方法、装置以及存储介质 |
Families Citing this family (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10962980B2 (en) * | 2018-08-30 | 2021-03-30 | Ford Global Technologies, Llc | System and methods for reverse braking during automated hitch alignment |
| CN112837529B (zh) * | 2019-11-25 | 2022-12-02 | 斑马智行网络(香港)有限公司 | 数据处理方法及系统、采集装置、处理器以及存储介质 |
| JP7581837B2 (ja) * | 2019-12-20 | 2024-11-13 | 株式会社デンソー | レーダ装置 |
| KR20220027359A (ko) * | 2020-08-26 | 2022-03-08 | 현대자동차주식회사 | 자율주행을 위한 높이 맵 격자간격 조정방법 |
| CN112385398A (zh) * | 2020-10-22 | 2021-02-23 | 成都大学 | 一种用于自动采摘农机的高精度定位方法 |
| CN114667459A (zh) * | 2020-11-10 | 2022-06-24 | 深圳市大疆创新科技有限公司 | 雷达及雷达的数据处理方法、可移动平台、存储介质 |
| CN113378628B (zh) * | 2021-04-27 | 2023-04-14 | 阿里云计算有限公司 | 一种道路障碍物区域检测方法 |
| CN113311837A (zh) * | 2021-05-26 | 2021-08-27 | 上海十方生态园林股份有限公司 | 一种基于水上水下环境识别的无人船避障系统 |
| CN113567972B (zh) * | 2021-08-03 | 2024-07-02 | 广州海事科技有限公司 | 基于雷达的海上监控方法、系统、设备及存储介质 |
| CN114037972B (zh) * | 2021-10-08 | 2024-08-13 | 岚图汽车科技有限公司 | 目标检测方法、装置、设备及可读存储介质 |
| CN115016481A (zh) * | 2022-06-16 | 2022-09-06 | 中路高科交通检测检验认证有限公司 | 一种用于隧道衬砌内部检测雷达的避障系统和避障方法 |
| CN115834563A (zh) * | 2022-11-24 | 2023-03-21 | 广州市申迪计算机系统有限公司 | 一种可远程向web端发送雷达数据的方法和系统 |
| CN119288303A (zh) * | 2024-10-25 | 2025-01-10 | 广州小鹏自动驾驶科技有限公司 | 车门避障控制方法、装置、车辆和存储介质 |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2005080119A1 (en) * | 2004-01-28 | 2005-09-01 | Toyota Jidosha Kabushiki Kaisha | Running support system for vehicle |
| CN101626493A (zh) * | 2009-08-06 | 2010-01-13 | 北京北大千方科技有限公司 | 一种结合激光扫描和视频的行人前进方向判断方法 |
| CN102508246A (zh) * | 2011-10-13 | 2012-06-20 | 吉林大学 | 车辆前方障碍物检测跟踪方法 |
| CN104978568A (zh) * | 2015-06-18 | 2015-10-14 | 奇瑞汽车股份有限公司 | 前方车辆检测方法 |
| CN105223956A (zh) * | 2015-11-09 | 2016-01-06 | 中山大学 | 一种全向移动机器人的动态避障方法 |
| CN105480227A (zh) * | 2015-12-29 | 2016-04-13 | 大连楼兰科技股份有限公司 | 主动驾驶技术中基于红外雷达与视频图像信息融合的方法 |
| CN106249239A (zh) * | 2016-08-23 | 2016-12-21 | 深圳市速腾聚创科技有限公司 | 目标检测方法及装置 |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5341344A (en) * | 1992-10-14 | 1994-08-23 | Optical Detective Systems, Inc. | Obstacle-detection system |
| JP3861781B2 (ja) * | 2002-09-17 | 2006-12-20 | 日産自動車株式会社 | 前方車両追跡システムおよび前方車両追跡方法 |
| US7474389B2 (en) * | 2006-12-19 | 2009-01-06 | Dean Greenberg | Cargo dimensional and weight analyzing system |
| JP5566076B2 (ja) | 2009-10-14 | 2014-08-06 | 古野電気株式会社 | 航行支援装置 |
| CN103176185B (zh) | 2011-12-26 | 2015-01-21 | 上海汽车集团股份有限公司 | 用于检测道路障碍物的方法及系统 |
| CN105216792A (zh) | 2014-06-12 | 2016-01-06 | 株式会社日立制作所 | 对周围环境中的障碍物目标进行识别跟踪的方法和设备 |
| CN105718888B (zh) | 2016-01-22 | 2019-09-13 | 北京中科慧眼科技有限公司 | 障碍物预警方法和障碍物预警装置 |
-
2016
- 2016-08-23 WO PCT/CN2016/096358 patent/WO2018035711A1/zh not_active Ceased
- 2016-08-23 US US16/327,685 patent/US10891499B2/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2005080119A1 (en) * | 2004-01-28 | 2005-09-01 | Toyota Jidosha Kabushiki Kaisha | Running support system for vehicle |
| CN101626493A (zh) * | 2009-08-06 | 2010-01-13 | 北京北大千方科技有限公司 | 一种结合激光扫描和视频的行人前进方向判断方法 |
| CN102508246A (zh) * | 2011-10-13 | 2012-06-20 | 吉林大学 | 车辆前方障碍物检测跟踪方法 |
| CN104978568A (zh) * | 2015-06-18 | 2015-10-14 | 奇瑞汽车股份有限公司 | 前方车辆检测方法 |
| CN105223956A (zh) * | 2015-11-09 | 2016-01-06 | 中山大学 | 一种全向移动机器人的动态避障方法 |
| CN105480227A (zh) * | 2015-12-29 | 2016-04-13 | 大连楼兰科技股份有限公司 | 主动驾驶技术中基于红外雷达与视频图像信息融合的方法 |
| CN106249239A (zh) * | 2016-08-23 | 2016-12-21 | 深圳市速腾聚创科技有限公司 | 目标检测方法及装置 |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112541416A (zh) * | 2020-12-02 | 2021-03-23 | 深兰科技(上海)有限公司 | 跨雷达障碍物跟踪方法、装置、电子设备和存储介质 |
| CN112541416B (zh) * | 2020-12-02 | 2023-07-14 | 深兰科技(上海)有限公司 | 跨雷达障碍物跟踪方法、装置、电子设备和存储介质 |
| CN115201804A (zh) * | 2021-04-12 | 2022-10-18 | 武汉智行者科技有限公司 | 目标速度估计方法、装置以及存储介质 |
| CN113635864A (zh) * | 2021-07-27 | 2021-11-12 | 际络科技(上海)有限公司 | 车载激光雷达的清洗方法、装置、电子设备及存储介质 |
| CN113635864B (zh) * | 2021-07-27 | 2023-08-15 | 际络科技(上海)有限公司 | 车载激光雷达的清洗方法、装置、电子设备及存储介质 |
Also Published As
| Publication number | Publication date |
|---|---|
| US10891499B2 (en) | 2021-01-12 |
| US20190228239A1 (en) | 2019-07-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2018035711A1 (zh) | 目标检测方法及系统 | |
| CN109606354B (zh) | 一种基于分层规划的自动泊车方法及辅助系统 | |
| CN108058713B (zh) | 信息显示装置、信息显示方法及信息显示程序的记录介质 | |
| CN106297342B (zh) | 一种提前、实时提示交通信号灯信息的提醒装置及方法 | |
| CN106249239B (zh) | 目标检测方法及装置 | |
| Lei et al. | Automated lane change behavior prediction and environmental perception based on SLAM technology | |
| US20180267548A1 (en) | Route determination device, vehicle control device, route determination method, and storage medium | |
| US20170008531A1 (en) | Driving assistant for vehicles | |
| CN108688660B (zh) | 运行范围确定装置 | |
| WO2021017504A1 (zh) | 运动轨迹预测方法和装置 | |
| US10029699B2 (en) | Control apparatus for hybrid vehicle | |
| CN110060467A (zh) | 预测装置、预测方法及存储介质 | |
| CN110211420A (zh) | 判断当前场景是否为停车场的方法、系统及计算机可读存储介质 | |
| CN110087964A (zh) | 车辆控制系统、车辆控制方法及车辆控制程序 | |
| CN111959499A (zh) | 一种车辆控制方法及装置 | |
| CN108058707B (zh) | 信息显示装置、信息显示方法及信息显示程序的记录介质 | |
| US20200005632A1 (en) | Traffic light adaptive learning and mapping method and system for improving vehicle energy efficiency and driving comfort | |
| CN116508083A (zh) | 停车辅助方法及停车辅助装置 | |
| US20240240963A1 (en) | Parking assistance device and parking assistance method | |
| WO2023028764A1 (zh) | 一种自动泊车的方法、装置以及车辆 | |
| US20240418855A1 (en) | Vehicle control device | |
| JP2023139740A (ja) | 道路認識装置および運転支援装置 | |
| JP7141478B2 (ja) | 地図生成装置 | |
| CN115237113B (zh) | 机器人导航的方法、机器人、机器人系统及存储介质 | |
| US20220258737A1 (en) | Map generation apparatus and vehicle control apparatus |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 16913735 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 16913735 Country of ref document: EP Kind code of ref document: A1 |