WO2018218680A1 - 一种障碍物检测方法及设备 - Google Patents

一种障碍物检测方法及设备 Download PDF

Info

Publication number
WO2018218680A1
WO2018218680A1 PCT/CN2017/087041 CN2017087041W WO2018218680A1 WO 2018218680 A1 WO2018218680 A1 WO 2018218680A1 CN 2017087041 W CN2017087041 W CN 2017087041W WO 2018218680 A1 WO2018218680 A1 WO 2018218680A1
Authority
WO
WIPO (PCT)
Prior art keywords
obstacle
information
target area
obstacle information
target
Prior art date
Application number
PCT/CN2017/087041
Other languages
English (en)
French (fr)
Inventor
曹彤彤
邵云峰
Original Assignee
华为技术有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to CN201780091513.XA priority Critical patent/CN110691990B/zh
Priority to EP17911618.1A priority patent/EP3623844A4/en
Priority to PCT/CN2017/087041 priority patent/WO2018218680A1/zh
Publication of WO2018218680A1 publication Critical patent/WO2018218680A1/zh
Priority to US16/700,679 priority patent/US20200110173A1/en

Links

Images

Classifications

    • 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
    • G01S13/931—Radar or analogous systems specially adapted for specific applications for anti-collision purposes of land vehicles
    • 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
    • G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
    • G01S17/88—Lidar systems specially adapted for specific applications
    • G01S17/93—Lidar systems specially adapted for specific applications for anti-collision purposes
    • G01S17/931—Lidar systems specially adapted for specific applications for anti-collision purposes of land vehicles
    • 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
    • G01S13/66—Radar-tracking systems; Analogous systems
    • G01S13/72—Radar-tracking systems; Analogous systems for two-dimensional tracking, e.g. combination of angle and range tracking, track-while-scan radar
    • G01S13/723—Radar-tracking systems; Analogous systems for two-dimensional tracking, e.g. combination of angle and range tracking, track-while-scan radar by using numerical data
    • G01S13/726—Multiple target tracking
    • 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
    • G01S13/86—Combinations of radar systems with non-radar systems, e.g. sonar, direction finder
    • G01S13/865—Combination of radar systems with lidar systems
    • 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
    • G01S13/86—Combinations of radar systems with non-radar systems, e.g. sonar, direction finder
    • G01S13/867—Combination of radar systems with cameras
    • 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
    • G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
    • G01S7/48—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S17/00
    • G01S7/4802—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S17/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR 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 application relates to the field of detection technologies, and in particular, to an obstacle detection method and device.
  • Obstacle detection and identification is an important factor to ensure the safety of vehicles in the state of automatic driving. Only by correctly identifying the obstacles such as vehicles, pedestrians, buildings and other vehicles around the vehicle can the vehicle's accessible area and the current road condition be judged, and then the vehicle can be driven. Propose follow-up planning and decision making. Therefore, the obstacle detection system in the vehicle plays an important role in ensuring the driving safety of the vehicle under the automatic driving state.
  • common sensors include vision sensors, millimeter wave radar sensors, lidar sensors, and the like.
  • Vision sensors such as monocular cameras or binocular cameras, can collect video information around the vehicle, and obtain information such as the position and velocity of obstacles by identifying and extracting obstacles in the video, which are characterized by azimuth information. Accurate, but the accuracy of distance, speed and other information is poor.
  • the millimeter wave radar sensor can obtain information such as the position and speed of the obstacle, which is characterized by accurate position and speed, but does not contain the contour information of the obstacle, that is, the information such as the size and shape of the obstacle cannot be obtained, and there is an obstacle error. The situation of the inspection.
  • the lidar sensor can obtain point cloud data of the surrounding environment of the vehicle, and can obtain information such as the position and contour of the obstacle according to the point cloud data clustering, and the feature is that the position and the contour are accurate, but the speed information is not included.
  • existing obstacle detection systems often use multi-sensor fusion to detect obstacles to compensate for the deficiencies of individual sensors.
  • multiple sensors are used to separately collect obstacle information in the surrounding environment of the vehicle, and then the respective acquired obstacle information is converted into the same coordinate system for data fusion, such as the vehicle body coordinate system, and the multi-dimensional obstacle can be extracted through data fusion.
  • the attribute features include both speed information acquired by the millimeter wave radar sensor and contour information acquired by the lidar sensor.
  • this method has poor recognition accuracy for distant obstacles.
  • the recognition ability of distant obstacles can be improved by improving the scanning angle resolution of the lidar sensor, a large amount of point cloud data is collected from the lidar sensor. Extracting the contour information of the obstacle consumes huge processor resources, which in turn causes the obstacle information to update at a slower speed and a longer delay.
  • the embodiment of the present application provides an obstacle detection method and device for improving the recognition accuracy of an obstacle while avoiding a large amount of data processing.
  • an obstacle detecting method includes: the obstacle information processing device receives description information of n obstacles collected by the obstacle detecting data collector for the target vehicle, where n is a positive integer.
  • the obstacle information processing device updates the obstacle information list according to the description information of the n obstacles, obtains the updated obstacle information list, and then obtains the target area information and the non-target area information according to the updated obstacle information list, and obtains the target
  • the regional information and non-target area information are sent to the lidar data collector to make the lidar data collector in each target area
  • the number of sample points acquired within the volume is greater than the number of sample points collected per unit volume of each non-target area.
  • the target area information is used to indicate at least one target area
  • the non-target area information is used to indicate at least one non-target area, wherein the target area is an area including an obstacle, and the non-target area is an area not including an obstacle, or
  • the target area is an area containing an obstacle that satisfies a preset screening condition
  • the non-target area is an area that does not include an obstacle that satisfies a preset screening condition.
  • the obstacle information processing device receives the set of sampling points collected by the laser radar data collector, and the sampling point set includes the number of sampling points collected by the lidar data collector in the at least one target area and the number of sampling points collected in the at least one non-target area.
  • the obstacle information processing device obtains the obstacle detection result based on the set of sampling points and the updated list of obstacle information. Therefore, the obstacle detection data collector first obtains the description information of the obstacle collected by the obstacle detection data collector, and then updates the obstacle information list, and the obstacle information processing device obtains the target area according to the updated obstacle information list. Information and non-target area information, so that the laser radar data collector can obtain different precision sampling point sets for different regions, avoiding large data processing amount, and ensuring obstacle recognition accuracy, and the obstacle information processing device according to the obtained sampling Point collection and updated list of obstacle information to obtain the final obstacle detection result. Therefore, the method provided by the embodiment of the present application can improve the accuracy of the obstacle recognition while avoiding a large amount of data processing, and effectively avoid excessive consumption of computing resources caused by excessive data processing, thereby causing processing delay and Downtime and other issues.
  • the obstacle detection data collector includes at least one visual sensor and/or at least one millimeter wave radar sensor. It should be understood that the obstacle detection data collector herein can also be other types of sensors, and the design provided by the embodiments of the present application ensures the flexibility of the system structure and the diversity of sampling results.
  • the target area information further includes a first signal transmission parameter
  • the non-target area information further includes a second signal transmission parameter, wherein the first signal transmission parameter is greater than the second signal transmission parameter
  • the target The area information further includes a first signal receiving sampling parameter
  • the non-target area information further includes a second signal receiving sampling parameter, wherein the first signal receiving sampling parameter is greater than the second signal receiving sampling parameter. Therefore, by using the method provided by the embodiment of the present application, more sampling points are obtained in a unit volume of the target area, which can effectively ensure the accuracy of the contour information of the obstacle, and obtain fewer sampling points in the unit volume of the non-target area. Can effectively reduce the amount of data processing.
  • the target area information includes a first signal transmission parameter
  • the non-target area information includes a second signal transmission parameter
  • the target area information further includes a first signal receiving sampling parameter
  • the target area information further includes a second signal receiving sampling parameter, that is, providing two sets of optional parameters, or providing more sets of optional parameters for the lidar data collector to select one of the groups or selecting two of the groups (for example, Transmit and receive each selected set of parameters) to perform data acquisition.
  • the lidar controller in the lidar data collector can be applied to the data acquisition of the target area and the non-target area according to the pre-configured two signal transmission parameters and/or the two signal reception sampling parameters.
  • the obstacle information list includes m obstacle information, m is a positive integer; the obstacle information processing device updates the obstacle information list according to the description information of the n obstacles, and obtains the updated obstacle information.
  • the list may be, but is not limited to, the following method: the obstacle information processing device acquires n first distances according to the description information of the n obstacles, and the n first distances refer to n obstacles corresponding to the description information of the n obstacles The distance between the object and the target vehicle.
  • the obstacle information processing device acquires m second distances according to the m obstacle information included in the obstacle information list, and the m second distances refer to the m obstacles corresponding to the m obstacle information and the target vehicle respectively Distance between from.
  • the obstacle information processing device updates the obstacle information list for the ith first distance, i is an arbitrary integer of 1 to n, and includes the following possible cases: (1) when the ith first distance and each When the difference of the second distances is greater than the preset threshold, the obstacle information processing device increases the description information of the obstacle corresponding to the ith first distance to the obstacle information list; (2) when the ith first distance When the t difference values calculated by the t second distances are both less than or equal to the preset threshold, the obstacle information processing apparatus determines that the obstacle corresponding to the i-th first distance is the same as the obstacle corresponding to the second distance of the target, according to the The description information of the obstacle corresponding to the first distances updates the obstacle information of the obstacle corresponding to the second distance of the target, wherein the target second distance refers to the second distance corresponding to the smallest difference among the t differences, t ⁇ m, t is a positive integer; (3) when the difference between the i-th first distance and the j-th second distance is less than or equal
  • the above update rule is based on the distance between the obstacle and the target vehicle as an update basis, and may also be used or combined with other parameters as an update basis, such as the direction angle of the obstacle.
  • the obstacle detection data collector is a sensor combination, it is necessary to update the obstacle information list by using the obstacle description information collected by the multiple sensors, and the specific process will not be described again.
  • the obstacle information information list and the obstacle information list may be combined with each obstacle information in the obstacle information list. Performing prediction to ensure time synchronization between the various information, and then matching the predicted obstacle information list with the description information of the obstacle collected by the obstacle detection data collector, thereby ensuring the updated obstacle information list. accuracy.
  • the updated list of obstacle information includes at least one obstacle information, each obstacle information including an orientation angle of the obstacle.
  • the obstacle information processing apparatus obtains the target area information and the non-target area information based on the updated obstacle information list, and may adopt a method in which the obstacle information processing apparatus determines a target according to the direction angle of the obstacle included in each obstacle information.
  • the direction angle section of the area is used as the target area information, wherein the target area is an area including an obstacle.
  • the obstacle information processing device determines the direction angle interval of the at least one non-target area according to the target area information and the maximum direction angle acquisition range of the lidar data collector, as the non-target area information, wherein the non-target area is not including the obstacle region. Therefore, it is ensured that the contour information of each obstacle to obtain an obstacle has high precision.
  • the updated list of obstacle information includes at least one obstacle information, each obstacle information including an orientation angle of the obstacle.
  • the obstacle information processing device obtains the target area information and the non-target area information according to the updated obstacle information list, and may adopt a method in which the obstacle information processing apparatus selects at least one obstacle according to the at least one obstacle information and the preset screening condition. At least one obstacle information that satisfies the preset screening condition is filtered out in the information.
  • the obstacle information processing device determines a direction angle interval of the target area according to the direction angle of the obstacle included in each obstacle information that satisfies the preset screening condition, as the target area information, wherein the target area is included to satisfy the preset screening condition The area of the obstacle.
  • the obstacle information processing device determines the direction angle range of the at least one non-target area according to the target area information and the maximum direction angle acquisition range of the lidar data collector, as the non-target area information, wherein the non-target area does not include the preset The area of the obstacle that screens the condition. Therefore, by screening out obstacle information that satisfies the preset condition, the number of target areas is reduced, and the amount of data processing is further reduced.
  • the obstacle information processing device can also refer to the motion state of the target vehicle, that is, combined with the target vehicle.
  • the motion state and the updated obstacle information list predicting each obstacle information in the updated obstacle information list to ensure time synchronization between the respective information, and then according to the predicted obstacle information list,
  • the target area information and the non-target area information are obtained, so that the accuracy of the determined target area can be better ensured.
  • the updated obstacle information list includes at least one obstacle information, each obstacle information includes a direction angle of the obstacle; the obstacle information processing device according to the sample point set and the updated obstacle information
  • the obstacle information processing device determines the direction angle interval of the kth region based on the direction angle of the obstacle included in the kth obstacle information, wherein the kth obstacle The information is any one of at least one obstacle information.
  • the obstacle information processing device determines a kth target sample point set from the sample point set according to the direction angle interval of the kth area, and the kth target sample point set includes a direction angle of the kth area in the sample point set. Sample points collected within the interval.
  • the obstacle information processing device obtains contour information of at least one obstacle based on the k-th target sample point set.
  • the obstacle information processing device obtains at least one determination distance according to the contour information of the at least one obstacle, wherein each determination distance refers to a distance between the obstacle corresponding to the contour information of the obstacle and the target vehicle, and the kth The difference between the obstacle corresponding to the obstacle information and the target vehicle.
  • the first detection result is used as an obstacle detection result, wherein the first detection result includes at least one contour information of the obstacle corresponding to the determination distance greater than the preset distance threshold in the determination distance; and/or, the second detection is performed
  • the result is the obstacle detection result with the kth obstacle information, wherein the second detection result refers to the contour information of the obstacle corresponding to the determination distance of the at least one determination distance that is less than or equal to the preset distance threshold. Therefore, the accuracy and effectiveness of the finally obtained obstacle detection result can be ensured.
  • the method further includes: when the obstacle information processing device does not determine the k-th target sampling point set, determining that the k-th obstacle information is not included in the obstacle detection result; or when the obstacle information processing When the device does not determine the kth target sampling point set, if the target area information includes the direction angle interval of the kth target area, the obstacle detection result does not include the kth obstacle information; if the target area information is not The direction angle section including the kth target area reduces the existence probability of the obstacle included in the kth obstacle information by a preset value, and obtains the updated kth obstacle information, and the updated kth obstacle The information is used as an obstacle detection result, and each obstacle information includes the probability of existence of the obstacle. Therefore, the comprehensiveness of the obstacle detection result can be ensured.
  • the obstacle information processing device determines the non-target sampling point set according to the set of target sample points corresponding to the set of sampling points and the at least one obstacle information respectively.
  • the obstacle information processing apparatus determines contour information of the at least one obstacle based on the non-target sampling point set as the third detection result.
  • the obstacle information processing apparatus uses the third detection result as an obstacle detection result. Therefore, the entire sampling point set can be effectively utilized, and the comprehensiveness of the obstacle detection result can be ensured.
  • the obstacle information processing apparatus first obtains contour information of at least one obstacle as a contour information set based on the set of sampling points. At this time, the obstacle information processing apparatus determines the direction angle section of the kth area based on the direction angle of the obstacle included in the kth obstacle information. Then, the obstacle information processing apparatus determines the k-th target contour information set from the contour information set according to the direction angle section of the k-th area, and then the obstacle information processing apparatus includes according to the k-th target contour information set At least one judgment distance is obtained from the contour information of at least one obstacle, and the subsequent processing is the same as the above method, and will not be described again.
  • the obstacle information processing device may remove at least one obstacle information from the contour information set.
  • the contour information outside the corresponding target contour information set is also used as the obstacle detection result.
  • the obstacle information processing apparatus may also not obtain the outline of any obstacle according to the set of sampling points.
  • the information indicates that no obstacles were found within the acquisition range of the lidar data collector.
  • the obstacle information processing apparatus may refer to the motion state of the target vehicle, that is, in combination with the motion state of the target vehicle and the updated obstacle information list, and perform each obstacle information in the updated obstacle information list.
  • the prediction is to ensure the time synchronization between the various information, and then the obstacle detection result is obtained according to the sampling point set and the predicted obstacle information list, so that the accuracy of the obstacle detection result can be better ensured.
  • the obstacle information processing apparatus replaces the updated obstacle information list with the obstacle detection result. Therefore, the timely update of the obstacle information list can be ensured, and the accuracy of the obstacle detection result can be ensured.
  • an obstacle detecting method comprising: a laser radar data collector receiving target area information and non-target area information transmitted by an obstacle information processing device, wherein the target area information is used to indicate at least one target The area, the non-target area information is used to indicate at least one non-target area, the target area is an area containing an obstacle, the non-target area is an area not including an obstacle, or the target area is an obstacle including an obstacle that meets a preset screening condition.
  • the area, the non-target area is an area that does not include an obstacle satisfying the preset screening condition; the lidar data collector performs data collection according to the target area information and the non-target area information, and obtains a set of sampling points, and the sampling point set includes laser radar data collection.
  • the number of sampling points collected in at least one target area and the number of sampling points collected in at least one non-target area, and the number of sampling points collected in a unit volume per target area is larger than the unit volume in each non-target area.
  • Number of sampling points; the lidar data collector will The set of sampling points is sent to the obstacle information processing device. Therefore, with the method provided by the embodiment of the present application, the lidar data collector acquires a set of sampling points with high precision in the target area, and acquires a set of sampling points with low precision in the non-target area, thereby ensuring the accuracy of obstacle recognition. It also avoids a large amount of data processing, and effectively avoids excessive consumption of computing resources caused by excessive data processing, which in turn leads to problems such as processing delay and downtime.
  • the target area information further includes a first signal transmission parameter
  • the non-target area information further includes a second signal transmission parameter, wherein the first signal transmission parameter is greater than the second signal transmission parameter
  • the target The area information further includes a first signal receiving sampling parameter
  • the non-target area information further includes a second signal receiving sampling parameter, wherein the first signal receiving sampling parameter is greater than the second signal receiving sampling parameter. Therefore, it can be ensured that the lidar data collector acquires a set of sampling points with high precision in the target area, and acquires a set of sampling points with low precision in the non-target area.
  • an obstacle information processing apparatus includes a transceiver, a processor, a memory, a transceiver for transmitting and receiving information, a memory for storing a program, an instruction, or a code, and the processor is configured to execute a program, an instruction, or Code to perform the method of any of the above first aspects or any of the possible implementations of the first aspect.
  • a lidar data collector includes a signal transmitter, a lidar sensor, a lidar controller, and a signal receiver, wherein the signal transmitter is connected to the lidar sensor, the signal receiver and the lidar controller Connected, the lidar sensor is connected to the lidar controller.
  • a lidar controller is used to control the lidar sensor to perform the method of any of the possible implementations of the second aspect or the second aspect above.
  • an obstacle detecting system comprising: the system comprising the obstacle information processing device of the third aspect, the laser radar data collector and the obstacle detecting data collector of the fourth aspect.
  • the present application provides a computer readable storage medium having stored therein instructions that, when executed on a computer, cause the computer to perform any of the above aspects or any of the possible aspects of the first aspect Methods.
  • the present application also provides a computer program product comprising instructions which, when run on a computer, cause the computer to perform the method of any of the first aspect or the first aspect of the first aspect described above.
  • FIG. 1 is a schematic diagram of an obstacle detection system in an embodiment of the present application.
  • FIG. 2 is a schematic diagram of an obstacle information processing apparatus according to an embodiment of the present application.
  • FIG. 3 is a schematic diagram of an obstacle detection data collector in an embodiment of the present application.
  • FIG. 4 is a schematic diagram of a laser radar data collector in an embodiment of the present application.
  • FIG. 5 is a flowchart of an overview of an obstacle detection method according to an embodiment of the present application.
  • FIG. 6 is a schematic diagram of a target area and a non-target area in the embodiment of the present application.
  • FIG. 7 is a second schematic diagram of a target area and a non-target area in the embodiment of the present application.
  • FIG. 8 is a specific flowchart of implementing obstacle detection for different configurations of an obstacle detection system according to an embodiment of the present application.
  • FIG. 9 is a second flowchart of the obstacle detection detection for different configurations of the obstacle detection system according to the embodiment of the present application.
  • FIG. 10 is a third flowchart of the obstacle detection detection for different configurations of the obstacle detection system according to the embodiment of the present application.
  • FIG. 11 is a fourth flowchart of a specific flowchart for implementing obstacle detection in different configurations of an obstacle detection system according to an embodiment of the present application.
  • the embodiment of the present application provides an obstacle detection system.
  • the obstacle detection system 100 includes an obstacle information processing device 110, an obstacle detection data collector 120, and a laser radar data collector 130.
  • the number of the obstacle detection data collectors 120 and the number of the laser radar data collectors 130 may be plural, and FIG. 1 is only a schematic diagram.
  • the obstacle detection data collector can be a single sensor, for example, a vision sensor or a millimeter wave radar sensor, and the obstacle detection data collector can also be a sensor combination, including multiple vision sensors and/or multiple millimeter waves. Radar sensor. It should be understood that the aforementioned visual sensors and millimeter wave radar sensors are by way of example only, and may be other types of sensors.
  • the obstacle information processing device 110 can communicate with the obstacle detection data collector 120 and the laser radar data collector 130.
  • the obstacle information processing apparatus 200 includes a transceiver 210, a processor 220, and a memory 230.
  • each obstacle detection data collector 300 may include a signal transmitter 310, a sensor 320, and the like. It should be noted that the obstacle detection data collector in the embodiment of the present application does not involve improvement.
  • the laser radar data collector 400 includes a signal transmitter 410, a lidar sensor 420, a lidar controller 430, and a signal receiver 440, wherein the signal transmitter 410 is connected to the lidar sensor 420, and the signal is received.
  • the 440 is coupled to a lidar controller 430 that is coupled to a lidar controller 430.
  • the lidar sensor herein can be a solid state lidar sensor with controllable characteristics of laser signal emission.
  • the lidar data collector involved in the embodiment of the present application has a certain difference from the lidar sensor mentioned in the prior art, and related improvements of the lidar data collector will be described in detail later.
  • target vehicles involved in the present application may be various vehicles, bicycles, motorcycles, electric Car and so on.
  • the present application provides an obstacle detection method for improving obstacle recognition accuracy while avoiding large data processing amount. As shown in FIG. 5, the method includes:
  • Step 500 The obstacle detection data collector collects description information of n obstacles for the target vehicle, where n is a positive integer.
  • Step 510 The obstacle detection data collector transmits the description information of the n obstacles to the obstacle information processing device.
  • the obstacle detection data collector is assembled on the target vehicle, and the data collection may be periodically performed to obtain the description information of the obstacle, and then the acquired obstacle description information is sent to the obstacle information processing device.
  • the description information of the obstacle here is different according to the obstacle detection data collector.
  • the obstacle detection data collector is a millimeter wave radar sensor
  • the millimeter wave radar sensor can collect the distance, azimuth and speed information of the obstacle.
  • the description information of the obstacle for example, the millimeter wave radar sensor collects the description information of the three obstacles, which are respectively recorded as A, B, and C, and are sent to the obstacle information processing device, and specifically include:
  • ⁇ represents the azimuth of the obstacle
  • d represents the distance between the obstacle and the target vehicle
  • s represents the speed of the obstacle
  • Step 520 The obstacle information processing device receives the description information of the n obstacles sent by the obstacle detection data collector, updates the obstacle information list according to the description information of the n obstacles, and obtains the updated obstacle information list.
  • Step 530 The obstacle information processing device obtains the target area information and the non-target area information according to the updated obstacle information list, and transmits the target area information and the non-target area information to the lidar data collector.
  • the target area information is used to indicate at least one target area
  • the non-target area information is used to indicate at least one non-target area, wherein the target area is an area including an obstacle, and the non-target area is an area not including an obstacle, or
  • the target area is an area containing an obstacle that satisfies a preset screening condition, and the non-target area is an area that does not include an obstacle that satisfies a preset screening condition.
  • Step 540 The laser radar data collector receives the target area information and the non-target area information sent by the obstacle information processing device, performs data collection according to the target area information and the non-target area information, and obtains a sampling point set.
  • the sample point set includes the number of sample points collected by the lidar data collector in at least one target area and the number of sample points collected in at least one non-target area, and the number of sample points collected in a unit volume of each target area is greater than The number of sample points collected per unit volume of each non-target area.
  • Step 550 The lidar data collector sends the set of sampling points to the obstacle information processing device.
  • Step 560 The obstacle information processing device receives the set of sampling points sent by the laser radar data collector, and obtains an obstacle detection result according to the set of sampling points and the updated list of obstacle information.
  • the obstacle information processing device after the obstacle information processing device obtains the obstacle detection result, the obstacle information processing device replaces the updated obstacle information list with the obstacle detection result as the obstacle adopted in the next execution step 520. List of information.
  • the obstacle information processing device transmits the obstacle detection result to other modules in the target vehicle, for example, a planning decision module in the vehicle travel controller for reference.
  • the target area information further includes a first signal transmission parameter, and the non-target area information A second signal transmission parameter is also included, wherein the first signal transmission parameter is greater than the second signal transmission parameter.
  • This parameter configuration is mainly used in the scene where the laser signal emission of the lidar collector is easier to control.
  • a large signal transmission parameter is used to perform laser signal transmission, and more sampling points are obtained in a unit volume, which can effectively ensure The accuracy of the contour information of the obstacle, the laser signal is emitted by using a small signal transmission parameter in the non-target area, and the number of sampling points is obtained in a unit volume, thereby reducing the amount of data processing.
  • the target area information further includes a first signal receiving sampling parameter
  • the non-target area information further includes a second signal receiving sampling parameter, wherein the first signal receiving sampling parameter is greater than the second signal receiving sampling parameter
  • This parameter configuration is mainly used in the scene where the laser emission signal of the lidar collector is difficult to control.
  • a larger signal is used to receive sampling parameters, that is, a high sampling rate, and more sampling points are obtained in a unit volume, which can effectively ensure the accuracy of the contour information of the obstacle, and adopt a smaller signal receiving sampling in the non-target area.
  • the parameter, the low sampling rate achieves fewer sample points per unit volume, thus reducing the amount of data processing.
  • the lidar data collector adopts different parameters for the target area and the non-target area, so that the accuracy of the contour information of the obstacle obtained for the target area is higher than the accuracy of the contour information of the obstacle obtained for the non-target area, avoiding A large amount of data processing can further improve the resource utilization rate of the obstacle information processing device while ensuring the accuracy of obstacle recognition.
  • the target area information includes a first signal transmission parameter
  • the non-target area information includes a second signal transmission parameter
  • the target area information further includes a first signal receiving sampling parameter
  • the target area information further includes a second signal receiving sampling parameter, that is, providing two sets of optional parameters, or providing more sets of optional parameters for the lidar data collector to select one of the groups or selecting two of the groups (for example, Transmit and receive each selected set of parameters) to perform data acquisition.
  • the lidar controller in the lidar data collector can be applied to the data acquisition of the target area and the non-target area according to the pre-configured two signal transmission parameters and/or the two signal reception sampling parameters.
  • step 520 The possible implementations of step 520, step 530 and step 560 mentioned above are described in detail below.
  • the obstacle information processing device updates the obstacle information list according to the description information of the n obstacles, and obtains the updated obstacle information list, which may include the following two cases:
  • the first case the list of obstacle information is empty.
  • the obstacle information processing device directly adds the description information of the n obstacles to the obstacle information list, and obtains the updated obstacle information list.
  • the second case the list of obstacle information is not empty.
  • the obstacle information processing apparatus may update the obstacle information list by using, but not limited to, the following method.
  • the obstacle information processing device acquires n first distances according to the description information of the n obstacles, where the n first distances refer to the n obstacles corresponding to the description information of the n obstacles and the target vehicle respectively The distance between them.
  • the distance between the obstacle and the target vehicle is not included in the description information of the obstacle, it can also be calculated by other parameters in the description information of the obstacle.
  • the obstacle information list includes m obstacle information, and m is a positive integer.
  • the obstacle information processing device also needs to acquire m second distances according to m obstacle information included in the obstacle information list, and m second distances refer to m obstacles The distance between the m obstacles corresponding to the object information and the target vehicle.
  • the obstacle information does not include the distance between the obstacle and the target vehicle, it can also be calculated by other parameters in the obstacle information.
  • the following takes the i-th first distance as an example, and i is an arbitrary integer of 1 to n, indicating that the obstacle information processing device may update the obstacle information list in several cases:
  • the obstacle information processing device increases the description information of the obstacle corresponding to the ith first distance to the obstacle information. List.
  • the obstacle corresponding to the i-th first distance is a newly detected obstacle.
  • the obstacle information processing device determines the obstacle corresponding to the i-th first distance and the j-th second distance.
  • the obstacle information of the obstacle corresponding to the jth second distance is updated according to the description information of the obstacle corresponding to the i-th first distance, and the jth second distance is m second distances. Any one of them.
  • the obstacle information processing apparatus updates the obstacle information corresponding to the obstacle in the obstacle information list by determining the same obstacle as the obstacle corresponding to the i-th first distance.
  • the obstacle information processing device determines the obstacle and the target corresponding to the i-th first distance
  • the obstacles corresponding to the two distances are the same, and the obstacle information corresponding to the obstacle corresponding to the second distance of the target is updated according to the description information of the obstacle corresponding to the i-th first distance, wherein the target second distance refers to the smallest of the t differences
  • the second distance corresponding to the difference, t ⁇ m, t is a positive integer.
  • the obstacle corresponding to the second distance having the smallest difference from the i-th first distance is determined as the same obstacle as the obstacle corresponding to the i-th first distance, and the obstacle corresponding to the obstacle information list is updated. Obstacle information.
  • each of the first distances is judged, the obstacle information list is updated, and the obstacle information processing apparatus obtains the updated obstacle information list.
  • the following describes how the obstacle information processing device updates the obstacle information list to obtain an updated obstacle information list in combination with a specific example.
  • the millimeter-wave radar sensor collects the description information of three obstacles, which are respectively recorded as A, B, and C, including:
  • ⁇ represents the azimuth of the obstacle
  • d represents the distance between the obstacle and the target vehicle
  • s represents the speed of the obstacle
  • the obstacle information processing apparatus can obtain three first distances based on the description information of the above three obstacles, which are d1, d2, and d3, respectively.
  • the first obstacle information list includes three obstacle information, which are respectively recorded as X, Y, and Z, and specifically include:
  • the obstacle information processing device can obtain three second distances according to the above three obstacle information, respectively Dx, dy, dz.
  • the following describes how to update the first obstacle information list by using d1 as an example.
  • A( ⁇ 1, d1, s1) is added to the obstacle information list.
  • the difference between d1 and dx, the difference between d1 and dy, the difference between d1 and dz are greater than a preset threshold, and the difference between d2 and dx is less than or equal to a preset threshold, d3 and The difference of dz is less than or equal to a preset threshold, and the difference between d3 and dy is less than or equal to a preset threshold, and the difference between d3 and dz is smaller than the difference between d3 and dy. Therefore, the obstacle information processing device updates the obstacle information list.
  • the updated list of obstacle information including:
  • A( ⁇ 1, d1, s1) the difference between d1 and dx, dy, dz is greater than a preset threshold, and A( ⁇ 1, d1, s1) is the newly added obstacle information
  • B( ⁇ 2, d2, s2) the difference between d2 and dx is less than or equal to a preset threshold, that is, B( ⁇ 2, d2, s2) is used to update X( ⁇ x, dx, sx);
  • update rule is based on the distance between the obstacle and the target vehicle as an update basis, and may also be used or combined with other parameters as an update basis, such as the direction angle of the obstacle.
  • the obstacle detection data collector is a sensor combination, it is necessary to update the obstacle information list by using the obstacle description information collected by the multiple sensors, and the specific process will not be described again.
  • the obstacle information information list and the obstacle information list may be combined with each obstacle information in the obstacle information list. Performing prediction to ensure time synchronization between the various information, and then matching the predicted obstacle information list with the description information of the obstacle collected by the obstacle detection data collector, thereby ensuring the updated obstacle information list. accuracy.
  • the obstacle information processing device obtains the target area information and the non-target area information according to the updated obstacle information list, and the updated obstacle information list includes at least one obstacle information, and each obstacle information includes an obstacle Direction angle.
  • obtaining the target area information and the non-target area information may be, but not limited to, the following two methods:
  • the obstacle information processing apparatus determines a direction angle section of a target area as the target area information based on the direction angle of the obstacle included in each obstacle information, wherein the target area is an area including the obstacle.
  • the obstacle information processing device determines the direction angle interval of the at least one non-target area as the non-target area information according to the target area information and the maximum direction angle acquisition range of the lidar data collector, wherein the non-target area is not included The area of the obstacle.
  • the updated obstacle information list includes two obstacle information
  • the obstacle information processing device expands a preset angle to each side according to the direction angle parameter included in each obstacle information, and then serves as a target area.
  • Direction angle range Therefore, as shown in FIG. 6, the two obstacles respectively correspond to the two target area direction angle sections, and the two target area direction angle sections are used as the target area information, and are further collected according to the maximum direction angle of the lidar data collector.
  • the range uses the other areas except the target area as the non-target areas, and obtains the direction angle sections of the three non-target areas as the non-target area information.
  • the second method the obstacle information processing device selects at least one obstacle information that meets the preset screening condition from the at least one obstacle information according to the at least one obstacle information and the preset screening condition, and then satisfies the preset according to each
  • the obstacle information of the screening condition includes a direction angle of the obstacle, and a direction angle section of the target area is determined as the target area information, wherein the target area is an area including an obstacle satisfying the preset screening condition.
  • the obstacle information processing device determines the direction angle interval of the at least one non-target area as the non-target area information according to the target area information and the maximum direction angle acquisition range of the lidar data collector, wherein the non-target area is not included The area of the obstacle that meets the preset screening criteria.
  • the obstacle information processing device may determine, based on the at least one obstacle information, whether the obstacle is a dynamic obstacle (ie, a preset screening condition) based on the speed of the obstacle, and when determining that the obstacle is a dynamic obstacle, The obstacle is used as an obstacle satisfying the preset screening condition, and a direction angle section of the target area is obtained according to the direction angle of the obstacle included in the obstacle information of the obstacle.
  • a dynamic obstacle ie, a preset screening condition
  • the obstacle information processing apparatus may determine, based on the distance between the obstacle and the target vehicle, whether the obstacle is a nearby obstacle (ie, a preset screening condition) according to the at least one obstacle information, for example, setting one a threshold value, when the distance between the obstacle and the target vehicle is less than the threshold, determining that the obstacle is an obstacle located near the obstacle, and using the obstacle as an obstacle satisfying the preset screening condition, according to the obstacle of the obstacle.
  • the direction angle of the obstacle included in the object information obtains a range of direction angles of the target area.
  • the obstacle information processing device may determine, based on the at least one obstacle information, whether the obstacle is an obstacle in the direction of the target vehicle driving direction (ie, a preset screening condition) based on the azimuth angle of the obstacle, and when the obstacle is determined as the target When an obstacle is in the vehicle traveling direction, the obstacle is used as an obstacle satisfying the preset screening condition, and a direction angle section of the target area is obtained according to the direction angle of the obstacle included in the obstacle information of the obstacle.
  • a preset screening condition ie, a preset screening condition
  • the obstacle information processing apparatus may use an obstacle located close to and dynamic as an obstacle satisfying the preset screening condition, and obtain a direction of the target area according to the direction angle of the obstacle included in the obstacle information of the obstacle. Corner interval.
  • the obstacle information processing device first selects an obstacle that satisfies the preset screening condition according to the at least one obstacle information, and then includes the obstacle information corresponding to the obstacle that meets the preset screening condition, according to the preset screening condition.
  • the direction angle of the obstacle obtaining the direction angle interval of the target area. Therefore, the obstacle information processing device can reduce the number of target regions finally obtained by first screening out the obstacles satisfying the preset screening conditions, thereby reducing the number of sampling points that the lidar data collector needs to collect, and reducing the obstacle information processing device.
  • the second obstacle information list includes four obstacle information, and the obstacle information processing device can use the obstacle located in the vicinity and in dynamic as an obstacle satisfying the preset screening condition, as shown in FIG.
  • the obstacles for the screening conditions include two, and the direction angles of the two target regions are determined according to the direction angles of the obstacles respectively included in the obstacle information of the two obstacles, and further according to the maximum direction angle of the lidar data collector.
  • the acquisition range obtains a range of direction angles for three non-target areas. It should be noted that obstacles are also included in the non-target area at this time.
  • the obstacle information processing apparatus may refer to the motion state of the target vehicle, that is, in combination with the motion state of the target vehicle and the updated obstacle information list, and perform each obstacle information in the updated obstacle information list.
  • the prediction is to ensure time synchronization between the various information, and then the target area information and the non-target area information are obtained according to the predicted obstacle information list, so that the accuracy of the determined target area can be better ensured.
  • the obstacle information processing device obtains according to the sample point set and the updated obstacle information list.
  • the obstacle detection result can be used but is not limited to the following method.
  • the kth obstacle information is any one of at least one obstacle information included in the updated obstacle information column, indicating the obstacle How the information processing device obtains the obstacle detection result.
  • the obstacle information processing apparatus determines the direction angle section of the kth area based on the direction angle of the obstacle included in the kth obstacle information. Then, the obstacle information processing device determines a kth target sample point set from the sample point set according to the direction angle interval of the kth region, wherein the kth target sample point set includes the kth in the sample point set The sampling points collected in the direction angle range of the area.
  • the k-th obstacle information includes an obstacle having a direction angle of ⁇ k, the k-th region having a direction angle interval of ( ⁇ k-5° to ⁇ k+5°), and the kth target sample point set includes a sample point set. Sample points acquired in the region ( ⁇ k-5° to ⁇ k+5°).
  • the obstacle information processing device obtains contour information of the at least one obstacle according to the k-th target sample point set.
  • the obstacle information processing device processes the k-th target sample point set, for example, data clustering, to obtain contour information of at least one obstacle, for example, due to a region in ( ⁇ k-5° to ⁇ k+5°)
  • a plurality of obstacles may be included therein, and the distance between each obstacle and the target vehicle is different, and thus it is possible to obtain contour information of a plurality of obstacles.
  • the obstacle information processing device obtains at least one determination distance according to the contour information of the at least one obstacle, wherein each determination distance refers to a distance between the obstacle corresponding to the contour information of the obstacle and the target vehicle, The difference between the distance between the obstacle corresponding to the kth obstacle information and the target vehicle.
  • the obstacle information processing apparatus obtains the outline information of the two obstacles, respectively denoted as A1 and A2, and further, obtains the distance S1 between the obstacle corresponding to A1 and the target vehicle according to A1 and A2, and corresponds to A2.
  • the distance S2 between the obstacle and the target vehicle calculates the distance difference between the obstacle corresponding to the kth obstacle information and the target vehicle, and the obstacle corresponding to the kth obstacle information. The difference distance between the target vehicle and the target vehicle is obtained.
  • comparing each judgment distance with the preset distance threshold may include the following situations:
  • the obstacle information processing device regards the first detection result as an obstacle detection result, wherein the first detection result includes contour information of the obstacle corresponding to the determination distance of the at least one determination distance that is greater than the preset distance threshold.
  • the obstacle corresponding to the determination distance may be inferred to be a newly detected obstacle, and the contour information of the obstacle is used as an obstacle detection result.
  • the contour information of the obstacle is P1
  • the obstacle is denoted by X1
  • the obstacle detection result is recorded as X1 (P1).
  • the obstacle information processing device uses the second detection result and the kth obstacle information as the obstacle detection result, wherein the second detection result refers to the determination distance corresponding to the determination distance of the at least one determination distance that is less than or equal to the preset distance threshold Outline information of obstacles.
  • the contour information of the obstacle corresponding to the judgment distance and the kth obstacle may be The object information is collectively used as an obstacle detection result.
  • the contour information of the obstacle is P1
  • the kth obstacle information is written as K( ⁇ k, dk, sk)
  • the obstacle detection result is written as K( ⁇ k, dk, sk, P1).
  • the obstacle information processing device does not determine the target sampling point set
  • the following three processing methods may also be included:
  • the obstacle information processing apparatus determines that the kth obstacle information is not included in the obstacle detection result.
  • the obstacle information processing device does not determine the kth target sampling point set, it indicates that the obstacle corresponding to the kth obstacle information is not collected by the lidar data collector, and the obstacle corresponding to the kth obstacle information The object may have moved away from the target vehicle and, therefore, no longer serves as an obstacle detection result.
  • the obstacle information processing apparatus determines the obstacle detection The kth obstacle information is not included in the results.
  • the corresponding angular interval of the kth region is the target region, and a large number of sampling points have been acquired by the lidar data collector.
  • the k-th target sampling point set is not determined at this time, so it can be basically determined that the obstacle corresponding to the k-th obstacle information has been far away from the target vehicle.
  • the kth obstacle information is included
  • the probability of existence of the obstacle is lowered by a preset value, the updated kth obstacle information is obtained, and the updated kth obstacle information is used as an obstacle detection result, wherein each obstacle information includes the existence of the obstacle Probability.
  • the obstacle corresponding to the kth obstacle information is not an obstacle satisfying the preset screening condition. Since it exists in the non-target area, the number of sampling points obtained by the lidar data collector is small, which may not be sufficient. The contour information of the obstacle is obtained. Therefore, the obstacle information corresponding to the obstacle is not deleted, but the probability of the obstacle included in the obstacle information is lowered by a preset value, thereby ensuring the comprehensiveness of the obstacle detection result.
  • the obstacle information processing device may further generate the at least one obstacle information according to the sample point set and the at least one obstacle information.
  • the corresponding set of target sample points respectively determines a set of non-target sample points, that is, a set of remaining sample points.
  • the obstacle information processing apparatus determines contour information of at least one obstacle, for example, data clustering, as a third detection result, according to the non-target sampling point set.
  • the obstacle information processing apparatus uses the third detection result as an obstacle detection result. Therefore, the comprehensiveness of the obstacle detection results is ensured.
  • the lidar collector has a large amount of data acquisition and high precision, and the millimeter wave radar sensor has an obstacle detection error. Therefore, even if the non-target area does not contain obstacles, the obstacle can be detected by the lidar collector, and the target vehicle is always in the running state. When the lidar collector collects data, it is also in the non-target area at this time. New obstacles may appear.
  • the obstacle information processing apparatus first obtains contour information of at least one obstacle as a contour information set based on the set of sampling points. At this time, the obstacle information processing apparatus determines the direction angle section of the kth area based on the direction angle of the obstacle included in the kth obstacle information. Then, the obstacle information processing apparatus determines the k-th target contour information set from the contour information set according to the direction angle section of the k-th area, and then the obstacle information processing apparatus includes according to the k-th target contour information set At least one judgment distance is obtained from the contour information of at least one obstacle, and the subsequent processing is the same as the above method, and will not be described again.
  • the obstacle information processing device may remove at least one obstacle information from the contour information set.
  • the contour information outside the corresponding target contour information set is also used as the obstacle detection result.
  • the obstacle information processing apparatus may also not obtain contour information of any obstacle according to the set of sampling points, indicating that no obstacle is found within the collection range of the lidar data collector.
  • the obstacle information processing apparatus may refer to the motion state of the target vehicle, that is, in combination with the motion state of the target vehicle and the updated obstacle information list, and perform each obstacle information in the updated obstacle information list.
  • the prediction is to ensure the time synchronization between the various information, and then the obstacle detection result is obtained according to the sampling point set and the predicted obstacle information list, so that the accuracy of the obstacle detection result can be better ensured.
  • a millimeter wave radar sensor is included in Figure 8.
  • Step 1 The description information of the three obstacles collected by the millimeter wave radar sensor is sent to the obstacle information processing device, which are respectively recorded as A, B, and C, and specifically include:
  • ⁇ represents the azimuth of the obstacle
  • d represents the distance between the obstacle and the target vehicle
  • s represents the speed of the obstacle
  • Step 2 When the obstacle information list is empty, the obstacle information processing device directly adds the description information of the above three obstacles to the obstacle information list, and obtains the updated obstacle information list.
  • Step 3 The obstacle information processing device obtains the target area information and the non-target area information according to the updated obstacle information list and transmits the information to the lidar data collector.
  • the target area information also includes a horizontal resolution scan of 0.5° for the target area
  • the non-target area information also includes a horizontal resolution scan of 2° for the non-target area.
  • Step 4 The laser radar data collector performs data collection according to the target area information and the non-target area information, and feeds back the sampling point set to the obstacle information processing device.
  • Step 5 The obstacle information processing device obtains an obstacle detection result according to the set of sampling points and the updated list of obstacle information.
  • the obstacle information processing apparatus obtains X( ⁇ 1, d1, p1) according to the set of sampling points, where p1 represents contour information of the obstacle.
  • X( ⁇ 1, d1, p1) is successfully matched with A( ⁇ 1, d1, s1) in the second obstacle information list, so the obstacle A remains and the information is updated as A( ⁇ 1, d1, s1, p1),
  • the obstacle B fails to match the contour information of the obstacle successfully, and B is an obstacle that satisfies the preset screening condition, the obstacle information of the obstacle B does not serve as an obstacle detection result.
  • the obstacle C fails to match the contour information of the obstacle successfully, it is temporarily retained because it is an obstacle that does not satisfy the preset screening condition, but reduces the existence probability of the obstacle to C'( ⁇ 3, d3, s3). Said as the result of obstacle detection.
  • the obstacle detection result finally obtained in the embodiment shown in FIG. 8 is used as the next obstacle information list updated according to the description information of the obstacle, that is, the obstacle information.
  • the feedback of the last obstacle detection result is added to the list, which can further improve the reliability of the obstacle information list.
  • FIG. 10 includes two millimeter wave radar sensors, the others being identical to the embodiment shown in FIG.
  • Step1 The millimeter wave radar sensor 1 collects the description information of the three obstacles and sends them to the obstacle information processing device, which are respectively recorded as A, B, and C.
  • Step1' The description information of the obstacle acquired by the millimeter wave radar sensor 2 is sent to the obstacle information processing device, and is recorded as D, D ( ⁇ 4, d4, s4).
  • Step 2 When the obstacle information list is empty, the obstacle information processing device directly adds the description information of the above four obstacles to the obstacle information list to obtain the second obstacle information list.
  • Step 3 The obstacle information processing device obtains the target area information and the non-target area information according to the second obstacle information list and transmits the information to the lidar data collector.
  • the obstacle information processing apparatus selects the obstacle A and the obstacle B as obstacles satisfying the preset screening condition, and obtains the target area information and the non-target area information and transmits the data to the lidar data. Collector.
  • Step 4 The laser radar data collector performs data collection according to the target area information and the non-target area information, and feeds back the sampling point set to the obstacle information processing device.
  • Step 5 The obstacle information processing device obtains an obstacle detection result according to the set of sampling points and the updated list of obstacle information.
  • the obstacle information processing apparatus obtains X( ⁇ 1, d1, p1) based on the set of sampling points, and therefore, the obstacle A remains and updates its information to A ( ⁇ 1, d1). , s1, p1), the obstacle information of the obstacle B cannot be used as the obstacle detection result, the obstacle information of the obstacle C and the obstacle existence probability of the obstacle D are lowered by a preset value, respectively, by C '( ⁇ 3, d3, s3), D'( ⁇ 4, d4, s4).
  • FIG. 11 includes two lidar data collectors, the others being identical to the embodiment shown in FIG.
  • Step1 The millimeter wave radar sensor 1 collects the description information of the three obstacles and sends them to the obstacle information processing device, which are respectively recorded as A, B, and C.
  • Step 2 When the obstacle information list is empty, the obstacle information processing device directly adds the description information of the above three obstacles to the obstacle information list, and obtains the updated obstacle information list.
  • Step 3 The obstacle information processing device obtains the target area information and the non-target area information according to the updated obstacle information list and transmits the information to the lidar data collector.
  • the obstacle information processing apparatus selects the obstacle A and the obstacle B as obstacles satisfying the preset screening condition, and obtains the target area information and the non-target area information and transmits the data to the lidar data.
  • Step 4 The laser radar data collector 1 performs data acquisition according to the target area information and the non-target area information, and feeds back the sampling point set to the obstacle information processing apparatus.
  • Step 4' The laser radar data collector 2 performs data acquisition based on the target area information and the non-target area information, and feeds back the sampling point set to the obstacle information processing apparatus.
  • Step 5 The obstacle information processing device obtains an obstacle detection result according to the set of sampling points and the updated list of obstacle information.
  • the obstacle information processing apparatus obtains X( ⁇ 1, d1, p1) according to the set of sampling points fed back by the laser radar data collector 1, and is fed back according to the laser radar data collector 2.
  • the obstacle existence probability in the obstacle information of the obstacle C is lowered by a preset value expressed by C' ( ⁇ 3, d3, s3).
  • the present application further provides an obstacle information processing device, which can be used to perform the method embodiment corresponding to the obstacle information processing device in FIG. 5, and thus the obstacle information processing device provided by the embodiment of the present application
  • an obstacle information processing device which can be used to perform the method embodiment corresponding to the obstacle information processing device in FIG. 5, and thus the obstacle information processing device provided by the embodiment of the present application
  • the implementation of the method refer to the implementation manner of the method, and the repeated description is not repeated.
  • an embodiment of the present application provides an obstacle information processing apparatus 200, including a transceiver 210, a processor 220, and a memory 230.
  • the memory 230 is used to store programs, instructions or codes, and the processor 220 is configured to execute a memory. a program, instruction or code in 230;
  • the transceiver 210 is configured to receive description information of the n obstacles collected by the obstacle detection data collector for the target vehicle, where n is a positive integer;
  • the processor 220 is configured to update the obstacle information list according to the description information of the n obstacles, obtain the updated obstacle information list, and obtain the target area information and the non-target area information according to the updated obstacle information list;
  • the transceiver 210 is configured to send the target area information and the non-target area information to the lidar data collector, so that the number of sampling points collected by the lidar data collector in a unit volume of each target area is greater than that in each non-target area.
  • a set of sampling points, the set of sampling points comprising a number of sampling points collected by the lidar data collector in at least one target area and sampling points collected in at least one non-target area;
  • the processor 220 is configured to obtain an obstacle detection result according to the sample point set and the updated obstacle information list.
  • the obstacle detection data collector includes at least one visual sensor and/or at least one millimeter wave radar sensor.
  • the target area information further includes a first signal transmission parameter
  • the non-target area information further includes a second signal transmission parameter, wherein the first signal transmission parameter is greater than the second signal transmission parameter
  • the target area information further includes a first signal receiving sampling parameter
  • the non-target area information further includes a second signal receiving sampling parameter, wherein the first signal receiving sampling parameter is greater than the second signal receiving sampling parameter
  • the obstacle information list includes m obstacle information, and m is a positive integer
  • the processor 220 is configured to:
  • n first distances are distances between the n obstacles corresponding to the description information of the n obstacles and the target vehicle respectively;
  • m second distances are distances between the m obstacles corresponding to the m obstacle information and the target vehicle, m Is a positive integer;
  • i is an arbitrary integer of 1 to n
  • the obstacle information list is updated, including:
  • the description information of the obstacle corresponding to the ith first distance is added to the obstacle information list;
  • t-th difference calculated by the i-th first distance and the t second distances are both less than or equal to a preset threshold, determining that the obstacle corresponding to the i-th first distance is the same as the obstacle corresponding to the second distance of the target, Updating the obstacle information of the obstacle corresponding to the second distance of the target according to the description information of the obstacle corresponding to the ith first distance, where the target second distance is the second distance corresponding to the smallest difference among the t differences T ⁇ m, t is a positive integer;
  • the difference between the i-th first distance and the j-th second distance is less than or equal to the preset threshold, determining that the obstacle corresponding to the i-th first distance is the same as the obstacle corresponding to the j-th second distance, according to
  • the description information of the obstacle corresponding to the i-th first distance updates the obstacle information of the obstacle corresponding to the jth second distance, and the j-th second distance refers to any one of the m second distances.
  • the updated obstacle information list includes at least one obstacle information, and each obstacle information includes an orientation angle of the obstacle;
  • the processor 220 is configured to:
  • a direction angle interval of a target area as target area information according to a direction angle of the obstacle included in each obstacle information, wherein the target area is an area including an obstacle;
  • the updated obstacle information list includes at least one obstacle information, and each obstacle information includes an orientation angle of the obstacle;
  • the processor 220 is configured to:
  • a direction angle interval of a target area as the target area information according to the direction angle of the obstacle included in each of the obstacle information satisfying the preset screening condition, wherein the target area is an area including an obstacle satisfying the preset screening condition ;
  • the updated obstacle information list includes at least one obstacle information, and each obstacle information includes an orientation angle of the obstacle;
  • the processor 220 is configured to:
  • the kth target sample point set is determined from the sample point set, and the kth target sample point set includes the sample collected in the direction angle interval of the kth region in the sample point set. point;
  • each judgment distance refers to a distance between the obstacle corresponding to the contour information of the obstacle and the target vehicle, and a distance between the obstacle corresponding to the kth obstacle information and the target vehicle;
  • the first detection result includes at least one contour information of the obstacle corresponding to the determination distance greater than the preset distance threshold in the determination distance; and/or, the second detection result and the second detection result
  • the k obstacle information is used as the obstacle detection result, wherein the second detection result refers to the contour information of the obstacle corresponding to the determination distance of the at least one determination distance that is less than or equal to the preset distance threshold.
  • the processor 220 is also used to:
  • the obstacle detection result does not include the kth obstacle information; if the target area information is not The direction angle section including the kth target area reduces the existence probability of the obstacle included in the kth obstacle information by a preset value, and obtains the updated kth obstacle information, and the updated kth obstacle
  • the information is used as an obstacle detection result, and each obstacle information includes the probability of existence of the obstacle.
  • the processor 220 is also used to:
  • the third detection result is used as an obstacle detection result.
  • the processor 220 is also used to:
  • the updated obstacle information list is replaced with the obstacle detection result.
  • the present application further provides a laser radar data collector, which can be used to perform the method embodiment corresponding to the laser radar data collector in FIG. 5 above. Therefore, the laser radar data collector provided by the embodiment of the present application is provided.
  • the implementation of the method refer to the implementation manner of the method, and the repeated description is not repeated.
  • the present application provides a laser radar data collector 400, including: a signal transmitter 410, a lidar sensor 420, a lidar controller 430, a signal receiver 440, wherein the signal transmitter 410 and the laser radar
  • the sensor 420 is connected, the signal receiver 440 is connected to the lidar controller 430, and the lidar sensor 420 is connected to the lidar controller 430;
  • the signal receiver 440 is configured to receive target area information and non-target area information sent by the obstacle information processing device, where the target area information is used to indicate at least one target area, and the non-target area information is used to indicate at least one non-target area,
  • the target area is an area containing an obstacle, the non-target area is an area not including an obstacle, or the target area is an area containing an obstacle satisfying a preset screening condition, and the non-target area is an obstacle not including a predetermined screening condition Area of matter;
  • a laser radar controller 430 configured to control the lidar sensor 420 to perform data acquisition according to the target area information and the non-target area information;
  • the lidar sensor 420 is configured to obtain a set of sampling points, where the set of sampling points includes the number of sampling points collected in the at least one target area and the number of sampling points collected in the at least one non-target area, and is collected in a unit volume of each target area.
  • the number of sampling points is greater than the number of sampling points collected in a unit volume per non-target area;
  • the signal transmitter 410 is configured to send the set of sampling points to the obstacle information processing device.
  • the target area information further includes a first signal transmission parameter
  • the non-target area information further includes a second signal transmission parameter, wherein the first signal transmission parameter is greater than the second signal transmission parameter
  • the target area information further includes a first signal receiving sampling parameter
  • the non-target area information further includes a second signal receiving sampling parameter, wherein the first signal receiving sampling parameter is greater than the second signal receiving sampling parameter
  • the obstacle detection data collector first obtains the description information of the obstacle collected by the obstacle detection data collector, and then updates the obstacle information list, and obtains the target area information according to the updated obstacle information list. And non-target area information, so that the lidar data collector can obtain different sets of sampling points for different regions for different regions, thereby avoiding a large amount of data processing.
  • the obstacle information processing device obtains the final obstacle detection result based on the obtained sample point set and the updated obstacle information list.
  • the accuracy of the contour information of the obstacle acquired by the lidar data collector in the target area is high, and the accuracy of the contour information of the obstacle acquired in the non-target area is low, thereby ensuring The obstacle recognition accuracy, while avoiding a large amount of data processing, can effectively avoid problems such as processing delay and downtime caused by excessive consumption of computing resources.
  • embodiments of the present application can be provided as a method, system, or computer program product. Therefore, the embodiments of the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Moreover, embodiments of the present application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) including computer usable program code.
  • computer-usable storage media including but not limited to disk storage, CD-ROM, optical storage, etc.
  • Embodiments of the present application are described with reference to flowchart illustrations and/or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present application. It will be understood that each flow and/or block of the flowchart illustrations and/or FIG.
  • These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing device to produce a machine for the execution of instructions for execution by a processor of a computer or other programmable data processing device.
  • the computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture comprising the instruction device.
  • the apparatus implements the functions specified in one or more blocks of a flow or a flow and/or block diagram of the flowchart.
  • These computer program instructions can also be loaded onto a computer or other programmable data processing device such that a series of operational steps are performed on a computer or other programmable device to produce computer-implemented processing for execution on a computer or other programmable device.
  • the instructions provide steps for implementing the functions specified in one or more of the flow or in a block or blocks of a flow diagram.

Landscapes

  • Engineering & Computer Science (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Electromagnetism (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Traffic Control Systems (AREA)
  • Optical Radar Systems And Details Thereof (AREA)
  • Radar Systems Or Details Thereof (AREA)

Abstract

一种障碍物检测方法及系统（100），用以提升障碍物识别精度的同时避免较大的数据处理量，方法包括：障碍物信息处理设备（110）接收障碍物检测数据采集器（120）针对目标交通工具采集的n个障碍物的描述信息，根据n个障碍物的描述信息更新障碍物信息列表，获得更新后的障碍物信息列表；障碍物信息处理设备（110）根据更新后的障碍物信息列表获得目标区域信息和非目标区域信息，并将目标区域信息和非目标区域信息发送给激光雷达数据采集器（130）；障碍物信息处理设备（110）接收激光雷达数据采集器（130）采集的采样点集合；最后，障碍物信息处理设备（110）根据采样点集合和更新后的障碍物信息列表，获得障碍物检测结果。因此，避免较大的数据处理量，同时保证了障碍物识别精度。

Description

一种障碍物检测方法及设备 技术领域
本申请涉及检测技术领域，尤其涉及一种障碍物检测方法及设备。
背景技术
障碍物检测和识别是保障处于自动驾驶状态的车辆安全的重要因素，只有正确识别车辆周围的车辆、行人、建筑物等障碍物才能判断出车辆的可通行区域以及当前路况状态，进而为车辆行驶提出后续规划与决策。因此，车辆中的障碍物检测系统对于保障车辆在自动驾驶状态下的行车安全具有重要作用。
在障碍物检测系统中，常见的传感器包括视觉传感器、毫米波雷达传感器、激光雷达传感器等。
视觉传感器，例如单目摄像头或者双目摄像头等，可以采集车辆周围的视频信息，通过对视频中的障碍物进行识别和特征提取从而获取障碍物的位置、速度等信息，其特点是方位角信息准确，但距离、速度等信息精度较差。
毫米波雷达传感器能够获得障碍物的位置、速度等信息，其特点是位置和速度较准确，但是不包含障碍物的轮廓信息，即无法获得障碍物的大小，形状等信息，而且存在障碍物误检的情况。
激光雷达传感器能够获得车辆周围环境的点云数据，根据点云数据聚类可以获得障碍物的位置、轮廓等信息，其特点是位置、轮廓准确，但是不包含速度信息。
因此，为了克服单传感器检测障碍物的缺陷，现有障碍物检测系统往往采用多传感器融合检测障碍物的方法，以弥补单个传感器的不足。
例如，利用多个传感器分别采集车辆周围环境中的障碍物信息，然后将各自采集得到的障碍物信息转换到同一坐标系中进行数据融合，如车身坐标系，通过数据融合可以提取障碍物的多维属性特征，例如，既包含毫米波雷达传感器采集的速度信息，又包括激光雷达传感器采集的轮廓信息。但是，该方法对远处障碍物的识别精度差，尽管可以通过提升激光雷达传感器的扫描角分辨率来提高对远处障碍物的识别能力，但是从激光雷达传感器采集得到大量的点云数据中提取障碍物的轮廓信息要消耗巨大的处理器资源，进而造成障碍物信息的更新速度较慢，延迟较长。
发明内容
本申请实施例提供一种障碍物检测方法及设备，用以提升障碍物识别精度的同时避免较大的数据处理量。
第一方面，一种障碍物检测方法，包括:障碍物信息处理设备接收障碍物检测数据采集器针对目标交通工具采集的n个障碍物的描述信息，n为正整数。障碍物信息处理设备根据n个障碍物的描述信息更新障碍物信息列表，获得更新后的障碍物信息列表，然后根据更新后的障碍物信息列表获得目标区域信息和非目标区域信息，并将目标区域信息和非目标区域信息发送给激光雷达数据采集器，以使激光雷达数据采集器在每个目标区域的单位 体积内采集的采样点数大于在每个非目标区域的单位体积内采集的采样点数。其中，目标区域信息用于指示至少一个目标区域，非目标区域信息用于指示至少一个非目标区域，其中，目标区域为包含障碍物的区域，非目标区域为不包含障碍物的区域，或者，目标区域为包含满足预设筛选条件的障碍物的区域，非目标区域为不包含满足预设筛选条件的障碍物的区域。障碍物信息处理设备接收激光雷达数据采集器采集的采样点集合，采样点集合包括激光雷达数据采集器在至少一个目标区域内采集的采样点数和在至少一个非目标区域内采集的采样点数。最后，障碍物信息处理设备根据采样点集合和更新后的障碍物信息列表，获得障碍物检测结果。因此，障碍物检测数据采集器首先获得由障碍物检测数据采集器采集的障碍物的描述信息，然后对障碍物信息列表进行更新，障碍物信息处理设备根据更新后的障碍物信息列表获得目标区域信息和非目标区域信息，以使激光雷达数据采集器针对不同区域获取不同精度的采样点集合，避免较大的数据处理量，同时保证了障碍物识别精度，障碍物信息处理设备根据获得的采样点集合和更新后的障碍物信息列表，获得最终的障碍物检测结果。因此，采用本申请实施例提供的方法在提升障碍物识别精度的同时避免较大的数据处理量，并有效避免因数据处理量过大导致的计算资源过度消耗，进而带来的处理延时和宕机等问题。
在一种可能的设计中，障碍物检测数据采集器包括至少一个视觉传感器和/或至少一个毫米波雷达传感器。应理解的是，这里的障碍物检测数据采集器还可以为其他类型的传感器，采用本申请实施例提供的设计保证了系统结构的灵活性和采样结果的多样性。
在一种可能的设计中，目标区域信息还包括第一信号发射参数，非目标区域信息还包括第二信号发射参数，其中，第一信号发射参数大于第二信号发射参数；和/或，目标区域信息还包括第一信号接收采样参数，非目标区域信息还包括第二信号接收采样参数，其中，第一信号接收采样参数大于第二信号接收采样参数。因此，采用本申请实施例提供的方法在目标区域的单位体积内获得较多的采样点数，可以有效保证障碍物的轮廓信息的精度，在非目标区域的单位体积内获得较少的采样点数，可以有效减少数据处理量。
应理解的是，在一种可能的实现方式中，目标区域信息包括第一信号发射参数，非目标区域信息包括第二信号发射参数，同时，目标区域信息还包括第一信号接收采样参数，非目标区域信息还包括第二信号接收采样参数，即提供两组可选的参数，或者提供更多组可选的参数，以供激光雷达数据采集器选择其中一组或选择其中两组(例如针对发射和接收各选择一组参数)执行数据采集。
此外，当目标区域信息不包括第一信号发射参数，以及非目标区域信息不包括第二信号发射参数时，且当目标区域信息不包括第一信号接收采样参数，以及非目标区域信息不包括第二信号接收采样参数时，激光雷达数据采集器中的激光雷达控制器可以根据预配置的两个信号发射参数和/或两个信号接收采样参数分别应用于目标区域和非目标区域的数据采集。
在一种可能的设计中，障碍物信息列表包括m个障碍物信息，m为正整数；障碍物信息处理设备根据n个障碍物的描述信息更新障碍物信息列表，得到更新后的障碍物信息列表，可以采用但不限于以下方法：障碍物信息处理设备根据n个障碍物的描述信息，获取n个第一距离，n个第一距离是指n个障碍物的描述信息对应的n个障碍物分别与目标交通工具之间的距离。障碍物信息处理设备根据障碍物信息列表中包括的m个障碍物信息，获取m个第二距离，m个第二距离是指m个障碍物信息对应的m个障碍物分别与目标交通工具之间的距 离。障碍物信息处理设备针对第i个第一距离，i为取遍1～n的任意整数，更新障碍物信息列表，包括以下几种可能的情况：(1)当第i个第一距离与每个第二距离的差值均大于预设阈值时，障碍物信息处理设备将第i个第一距离对应的障碍物的描述信息增加至障碍物信息列表；(2)当第i个第一距离与t个第二距离计算得到的t个差值均小于等于预设阈值时，障碍物信息处理设备确定第i个第一距离对应的障碍物与目标第二距离对应的障碍物相同，根据第i个第一距离对应的障碍物的描述信息更新目标第二距离对应的障碍物的障碍物信息，其中，目标第二距离是指t个差值中最小差值对应的第二距离，t≤m，t为正整数；(3)当第i个第一距离仅与第j个第二距离的差值小于等于预设阈值时，障碍物信息处理设备确定第i个第一距离对应的障碍物与第j个第二距离对应的障碍物相同，根据第i个第一距离对应的障碍物的描述信息更新第j个第二距离对应的障碍物的障碍物信息，第j个第二距离是指m个第二距离中的任意一个。因此，保证了更新后的障碍物信息列表的准确性。
应理解的是，上述更新规则是以障碍物与目标交通工具之间的距离作为更新依据，也可采用或结合其他参数作为更新依据，例如障碍物的方向角等。此外，当障碍物检测数据采集器为传感器组合时，需要利用多个传感器分别采集到的障碍物描述信息对障碍物信息列表进行更新，具体过程不再赘述。
若障碍物检测数据采集器针对目标交通工具采集障碍物的描述信息的时间间隔较大，还可以结合目标交通工具的运动状态以及障碍物信息列表，对障碍物信息列表中的每个障碍物信息进行预测，以保证各个信息之间的时间同步，然后根据预测后的障碍物信息列表与障碍物检测数据采集器采集到的障碍物的描述信息进行匹配，从而保证更新后的障碍物信息列表的准确性。
在一种可能的设计中，更新后的障碍物信息列表包括至少一个障碍物信息，每个障碍物信息包括障碍物的方向角。障碍物信息处理设备根据更新后的障碍物信息列表获得目标区域信息和非目标区域信息，可以采用如下方法：障碍物信息处理设备根据每个障碍物信息包括的障碍物的方向角，确定一个目标区域的方向角区间，作为目标区域信息，其中，目标区域为包含障碍物的区域。障碍物信息处理设备根据目标区域信息和激光雷达数据采集器的最大方向角采集范围，确定至少一个非目标区域的方向角区间，作为非目标区域信息，其中，非目标区域为不包含障碍物的区域。因此，保证每个障碍物获得障碍物的轮廓信息都具有较高的精度。
在一种可能的设计中，更新后的障碍物信息列表包括至少一个障碍物信息，每个障碍物信息包括障碍物的方向角。障碍物信息处理设备根据更新后的障碍物信息列表获得目标区域信息和非目标区域信息，可以采用如下方法：障碍物信息处理设备根据至少一个障碍物信息和预设筛选条件，从至少一个障碍物信息中筛选出至少一个满足预设筛选条件的障碍物信息。障碍物信息处理设备根据每个满足预设筛选条件的障碍物信息包括的障碍物的方向角，确定一个目标区域的方向角区间，作为目标区域信息，其中，目标区域为包含满足预设筛选条件的障碍物的区域。障碍物信息处理设备根据目标区域信息和激光雷达数据采集器的最大方向角采集范围，确定至少一个非目标区域的方向角区间，作为非目标区域信息，其中，非目标区域为不包含满足预设筛选条件的障碍物的区域。因此，通过筛选出满足预设条件的障碍物信息，以减少目标区域的数目，进一步降低数据处理量。
此外，障碍物信息处理设备还可参考目标交通工具的运动状态，即结合目标交通工具 的运动状态以及更新后的障碍物信息列表，对更新后的障碍物信息列表中的每个障碍物信息进行预测，以保证各个信息之间的时间同步，然后根据预测后的障碍物信息列表，获得目标区域信息和非目标区域信息，从而可以更好地保证确定的目标区域的准确性。
在一种可能的设计中，更新后的障碍物信息列表包括至少一个障碍物信息，每个障碍物信息包括障碍物的方向角；障碍物信息处理设备根据采样点集合和更新后的障碍物信息列表，获得障碍物检测结果，可以采用如下方法：障碍物信息处理设备根据第k个障碍物信息包括的障碍物的方向角，确定第k个区域的方向角区间，其中，第k个障碍物信息为至少一个障碍物信息中的任意一个。障碍物信息处理设备根据第k个区域的方向角区间，从采样点集合中确定出第k个目标采样点集合，第k个目标采样点集合包括采样点集合中在第k个区域的方向角区间内采集的采样点。障碍物信息处理设备根据第k个目标采样点集合获得至少一个障碍物的轮廓信息。障碍物信息处理设备根据至少一个障碍物的轮廓信息，获得至少一个判断距离，其中，每个判断距离是指该障碍物的轮廓信息对应的障碍物与目标交通工具之间的距离，与第k个障碍物信息对应的障碍物与目标交通工具之间的距离的差值。进一步地，将第一检测结果作为障碍物检测结果，其中，第一检测结果包括至少一个判断距离中大于预设距离阈值的判断距离对应的障碍物的轮廓信息；和/或，将第二检测结果与第k个障碍物信息作为障碍物检测结果，其中，第二检测结果是指至少一个判断距离中小于等于预设距离阈值的判断距离对应的障碍物的轮廓信息。因此，可以保证最终获得的障碍物检测结果的准确性和有效性。
在一种可能的设计中，还包括：当障碍物信息处理设备未确定出第k个目标采样点集合时，确定障碍物检测结果中不包括第k个障碍物信息；或者当障碍物信息处理设备未确定出第k个目标采样点集合时，若目标区域信息中包括第k个目标区域的方向角区间，则障碍物检测结果中不包括第k个障碍物信息；若目标区域信息中不包括第k个目标区域的方向角区间，将第k个障碍物信息中包括的障碍物存在概率降低预设数值，获得更新后的第k个障碍物信息，将更新后的第k个障碍物信息作为障碍物检测结果，每个障碍物信息中包括障碍物的存在概率。因此，可以保证障碍物检测结果的全面性。
在一种可能的设计中，障碍物信息处理设备根据采样点集合和至少一个障碍物信息分别对应的目标采样点集合，确定非目标采样点集合。障碍物信息处理设备根据非目标采样点集合确定至少一个障碍物的轮廓信息，作为第三检测结果。障碍物信息处理设备将第三检测结果作为障碍物检测结果。因此，能够使整个采样点集合得到有效利用，可以保证障碍物检测结果的全面性。
此外，基于与上述方法的相同思路，障碍物信息处理设备还根据采样点集合首先获得至少一个障碍物的轮廓信息，作为轮廓信息集合。此时，障碍物信息处理设备根据第k个障碍物信息包括的障碍物的方向角，确定第k个区域的方向角区间。然后，障碍物信息处理设备根据第k个区域的方向角区间，从轮廓信息集合中确定出第k个目标轮廓信息集合，接着，障碍物信息处理设备根据第k个目标轮廓信息集合中包括的至少一个障碍物的轮廓信息，获得至少一个判断距离，后续处理过程与上述方法相同，不再赘述。同理，在障碍物信息处理设备根据轮廓信息集合和至少一个障碍物信息分别对应的目标轮廓信息集合获得障碍物检测结果后，障碍物信息处理设备可以将轮廓信息集合中除去至少一个障碍物信息分别对应的目标轮廓信息集合外的轮廓信息也作为障碍物检测结果。
应理解的是，障碍物信息处理设备还可能根据采样点集合没有获得任何障碍物的轮廓 信息，此时表明在激光雷达数据采集器的采集范围内没有发现障碍物。
此外，障碍物信息处理设备还可参考目标交通工具的运动状态，即结合目标交通工具的运动状态以及更新后的障碍物信息列表，对更新后的障碍物信息列表中的每个障碍物信息进行预测，以保证各个信息之间的时间同步，然后根据采样点集合和预测后的障碍物信息列表，获得障碍物检测结果，从而可以更好地保证障碍物检测结果的准确性。
在一种可能的设计中，障碍物信息处理设备采用障碍物检测结果替换更新后的障碍物信息列表。因此，可以保证障碍物信息列表的及时更新，保证障碍物检测结果的准确性。
第二方面，一种障碍物检测方法，其特征在于，包括:激光雷达数据采集器接收障碍物信息处理设备发送的目标区域信息和非目标区域信息，其中，目标区域信息用于指示至少一个目标区域，非目标区域信息用于指示至少一个非目标区域，目标区域为包含障碍物的区域，非目标区域为不包含障碍物的区域，或者，目标区域为包含满足预设筛选条件的障碍物的区域，非目标区域为不包含满足预设筛选条件的障碍物的区域；激光雷达数据采集器根据目标区域信息和非目标区域信息执行数据采集，获得采样点集合，采样点集合包括激光雷达数据采集器在至少一个目标区域内采集的采样点数和在至少一个非目标区域内采集的采样点数，且在每个目标区域的单位体积内采集的采样点数大于在每个非目标区域的单位体积内采集的采样点数；激光雷达数据采集器将采样点集合发送至障碍物信息处理设备。因此，采用本申请实施例提供的方法，激光雷达数据采集器在目标区域获取精度较高的采样点集合，在非目标区域获取精度较低的采样点集合，这样既保证了障碍物识别精度，也避免较大的数据处理量，并有效避免因数据处理量过大导致的计算资源过度消耗，进而带来的处理延时和宕机等问题。
在一种可能的设计中，目标区域信息还包括第一信号发射参数，非目标区域信息还包括第二信号发射参数，其中，第一信号发射参数大于第二信号发射参数；和/或，目标区域信息还包括第一信号接收采样参数，非目标区域信息还包括第二信号接收采样参数，其中，第一信号接收采样参数大于第二信号接收采样参数。因此，可以保证激光雷达数据采集器在目标区域获取精度较高的采样点集合，在非目标区域获取精度较低的采样点集合。
第三方面，一种障碍物信息处理设备，包括收发器、处理器、存储器，收发器，用于收发信息，存储器用于存储程序、指令或代码，处理器用于执行存储器中的程序、指令或代码，以完成上述第一方面或第一方面的任意可能的实现方式中的方法。
第四方面，一种激光雷达数据采集器，包括信号发送器，激光雷达传感器，激光雷达控制器，和信号接收器，其中，信号发送器与激光雷达传感器连接，信号接收器与激光雷达控制器连接，激光雷达传感器与激光雷达控制器连接。激光雷达控制器用于控制激光雷达传感器，以完成上述第二方面或第二方面的任意可能的实现方式中的方法。
第五方面，一种障碍物检测系统，包括：该系统包括上述第三方面的障碍物信息处理设备，上述第四方面的激光雷达数据采集器和障碍物检测数据采集器。
第六方面，本申请提供了一种计算机可读存储介质，计算机可读存储介质中存储有指令，当其在计算机上运行时，使得计算机执行上述第一方面或第一方面的任意可能的设计的方法。
第七方面，本申请还提供了一种包含指令的计算机程序产品，当其在计算机上运行时，使得计算机执行上述第一方面或第一方面的任意可能的设计的方法。
附图说明
图1为本申请实施例中障碍物检测系统的示意图；
图2为本申请实施例中障碍物信息处理设备的示意图；
图3为本申请实施例中障碍物检测数据采集器的示意图；
图4为本申请实施例中激光雷达数据采集器的示意图；
图5为本申请实施例中障碍物检测方法的概述流程图之一；
图6为本申请实施例中目标区域和非目标区域的示意图之一；
图7为本申请实施例中目标区域和非目标区域的示意图之二；
图8为本申请实施例中针对障碍物检测系统的不同配置实现障碍物检测的具体流程图之一；
图9为本申请实施例中针对障碍物检测系统的不同配置实现障碍物检测的具体流程图之二；
图10为本申请实施例中针对障碍物检测系统的不同配置实现障碍物检测的具体流程图之三；
图11为本申请实施例中针对障碍物检测系统的不同配置实现障碍物检测的具体流程图之四。
具体实施方式
下面结合附图，对本申请的实施例进行描述。
本申请实施例提供一种障碍物检测系统，如图1所示，障碍物检测系统100包括：障碍物信息处理设备110，障碍物检测数据采集器120和激光雷达数据采集器130。其中，障碍物检测数据采集器120的数量和激光雷达数据采集器130的数量均可以为多个，图1仅为示意图。例如，障碍物检测数据采集器可以为单个的传感器，例如，一个视觉传感器或一个毫米波雷达传感器，障碍物检测数据采集器还可以为传感器组合，包括多个视觉传感器和/或多个毫米波雷达传感器。应理解的是，上述提到的视觉传感器和毫米波雷达传感器仅作为举例，还可以为其他类型的传感器。
其中，障碍物信息处理设备110能够与障碍物检测数据采集器120和激光雷达数据采集器130进行通信。
具体的，如图2所示，障碍物信息处理设备200，包括收发器210、处理器220、存储器230。
如图3所示，每个障碍物检测数据采集器300可以包括信号发送器310、传感器320等。须知，本申请实施例中的障碍物检测数据采集器不涉及改进。
如图4所示，激光雷达数据采集器400包括信号发送器410，激光雷达传感器420，激光雷达控制器430，和信号接收器440，其中，信号发送器410与激光雷达传感器420连接，信号接收器440与激光雷达控制器430连接，激光雷达传感器420与激光雷达控制器430连接。例如，这里的激光雷达传感器可以为具有激光信号发射可控制特性的固态激光雷达传感器。此外，本申请实施例中涉及的激光雷达数据采集器与现有技术中提到的激光雷达传感器存在一定差异，激光雷达数据采集器的相关改进将在后文详细描述。
应理解的是，本申请中涉及的目标交通工具可以为各种车辆，自行车，摩托车，电动 车等。
结合如图1～图4所示的障碍物检测系统和障碍物检测系统中的各个设备，本申请提供一种障碍物检测方法，用以提升障碍物识别精度的同时避免较大的数据处理量，如图5所示，该方法包括：
步骤500：障碍物检测数据采集器针对目标交通工具采集到n个障碍物的描述信息，n为正整数。
步骤510：障碍物检测数据采集器将n个障碍物的描述信息发送至障碍物信息处理设备。
具体的，障碍物检测数据采集器装配于目标交通工具上，可以周期性地执行数据采集，获取障碍物的描述信息，然后将获取到的障碍物描述信息发送至障碍物信息处理设备。这里的障碍物的描述信息根据障碍物检测数据采集器的不同而不同，当障碍物检测数据采集器为一个毫米波雷达传感器时，毫米波雷达传感器可以采集障碍物的距离，方位角和速度信息等作为障碍物的描述信息，例如，毫米波雷达传感器采集了3个障碍物的描述信息，分别记为A、B、C，发送至障碍物信息处理设备，具体包括：
A(θ1,d1,s1)
B(θ2,d2,s2)
C(θ3,d3,s3)
其中θ表示障碍物的方位角，d表示障碍物与目标交通工具之间的距离，s表示障碍物的速度。
步骤520：障碍物信息处理设备接收障碍物检测数据采集器发送的n个障碍物的描述信息，根据n个障碍物的描述信息更新障碍物信息列表，获得更新后的障碍物信息列表。
步骤530：障碍物信息处理设备根据更新后的障碍物信息列表获得目标区域信息和非目标区域信息，将目标区域信息和非目标区域信息发送至激光雷达数据采集器。
其中，目标区域信息用于指示至少一个目标区域，非目标区域信息用于指示至少一个非目标区域，其中，目标区域为包含障碍物的区域，非目标区域为不包含障碍物的区域，或者，目标区域为包含满足预设筛选条件的障碍物的区域，非目标区域为不包含满足预设筛选条件的障碍物的区域。
步骤540：激光雷达数据采集器接收障碍物信息处理设备发送的目标区域信息和非目标区域信息，根据目标区域信息和非目标区域信息执行数据采集，获得采样点集合。
其中，采样点集合包括激光雷达数据采集器在至少一个目标区域内采集的采样点数和在至少一个非目标区域内采集的采样点数，且在每个目标区域的单位体积内采集的采样点数大于在每个非目标区域的单位体积内采集的采样点数。
步骤550：激光雷达数据采集器将采样点集合发送至障碍物信息处理设备。
步骤560：障碍物信息处理设备接收激光雷达数据采集器发送的采样点集合，根据采样点集合和更新后的障碍物信息列表，获得障碍物检测结果。
在一种可能的实现方式中，在障碍物信息处理设备获得障碍物检测结果之后，障碍物信息处理设备采用障碍物检测结果替换更新后的障碍物信息列表，作为下一次执行步骤520采用的障碍物信息列表。
进一步地，障碍物信息处理设备将障碍物检测结果发送至目标交通工具中的其他模块，例如，车辆行驶控制器中的规划决策模块，以供参考。
在一种可能的实现方式中，目标区域信息还包括第一信号发射参数，非目标区域信息 还包括第二信号发射参数，其中，第一信号发射参数大于第二信号发射参数。
这种参数配置主要用于激光雷达采集器的激光信号发射较易控制的场景，在目标区域采用较大的信号发射参数执行激光信号发射，在单位体积内获得较多的采样点数，可以有效保证障碍物的轮廓信息的精度，在非目标区域采用较小的信号发射参数执行激光信号发射，在单位体积内获得较少的采样点数，因此可以减少数据处理量。
在一种可能的实现方式中，目标区域信息还包括第一信号接收采样参数，非目标区域信息还包括第二信号接收采样参数，其中，第一信号接收采样参数大于第二信号接收采样参数。
这种参数配置主要用于激光雷达采集器的激光发射信号较难控制的场景。在目标区域采用较大的信号接收采样参数，即高采样率，在单位体积内获得较多的采样点数，可以有效保证障碍物的轮廓信息的精度，在非目标区域采用较小的信号接收采样参数，即低采样率，在单位体积内获得较少的采样点数，因此可以减少数据处理量。
因此，激光雷达数据采集器针对目标区域和非目标区域采用不同的参数，以实现针对目标区域获得的障碍物的轮廓信息的精度高于针对非目标区域获得的障碍物的轮廓信息的精度，避免了较大的数据处理量，进而可以实现在保证障碍物识别精度的同时提升障碍物信息处理设备的资源利用率。
应理解的是，在一种可能的实现方式中，目标区域信息包括第一信号发射参数，非目标区域信息包括第二信号发射参数，同时，目标区域信息还包括第一信号接收采样参数，非目标区域信息还包括第二信号接收采样参数，即提供两组可选的参数，或者提供更多组可选的参数，以供激光雷达数据采集器选择其中一组或选择其中两组(例如针对发射和接收各选择一组参数)执行数据采集。
此外，当目标区域信息不包括第一信号发射参数，以及非目标区域信息不包括第二信号发射参数时，且当目标区域信息不包括第一信号接收采样参数，以及非目标区域信息不包括第二信号接收采样参数时，激光雷达数据采集器中的激光雷达控制器可以根据预配置的两个信号发射参数和/或两个信号接收采样参数分别应用于目标区域和非目标区域的数据采集。
下面针对上述提到的步骤520、步骤530和步骤560的可能实现方式分别进行详细描述。
针对步骤520，障碍物信息处理设备根据n个障碍物的描述信息更新障碍物信息列表，获得更新后的障碍物信息列表，可以包括以下两种情况：
第一种情况：障碍物信息列表为空。
此时，障碍物信息处理设备直接将n个障碍物的描述信息添加到障碍物信息列表，获得更新后的障碍物信息列表。
第二种情况：障碍物信息列表为非空。
此时，障碍物信息处理设备可以采用但不限于以下方法，更新障碍物信息列表。
具体的，障碍物信息处理设备根据n个障碍物的描述信息，获取n个第一距离，n个第一距离是指n个障碍物的描述信息对应的n个障碍物分别与目标交通工具之间的距离。
应理解的是，当障碍物的描述信息中不包括障碍物与目标交通工具之间的距离时，还可以通过障碍物的描述信息中的其他参数计算获得。
假设障碍物信息列表包括m个障碍物信息，m为正整数。障碍物信息处理设备还需要根据障碍物信息列表中包括的m个障碍物信息，获取m个第二距离，m个第二距离是指m个障碍 物信息对应的m个障碍物分别与目标交通工具之间的距离。
应理解的是，这里获取n个第一距离和获取m个第二距离没有必然的先后顺序。当障碍物信息不包括障碍物与目标交通工具之间的距离时，还可以通过障碍物信息中的其他参数计算获得。
进一步地，下面以第i个第一距离为例，i为取遍1～n的任意整数，说明障碍物信息处理设备更新障碍物信息列表可能存在的几种情况：
(1)当第i个第一距离与每个第二距离的差值均大于预设阈值时，障碍物信息处理设备将第i个第一距离对应的障碍物的描述信息增加至障碍物信息列表。
由上可以推断，第i个第一距离对应的障碍物为新检测到的障碍物。
(2)当第i个第一距离仅与第j个第二距离的差值小于等于预设阈值时，障碍物信息处理设备确定第i个第一距离对应的障碍物与第j个第二距离对应的障碍物相同，根据第i个第一距离对应的障碍物的描述信息更新第j个第二距离对应的障碍物的障碍物信息，第j个第二距离为m个第二距离中的任意一个。
因此，障碍物信息处理设备通过确定与第i个第一距离对应的障碍物相同的障碍物，更新障碍物信息列表中该障碍物对应的障碍物信息。
(3)当第i个第一距离与t个第二距离计算得到的t个差值均小于等于预设阈值时，障碍物信息处理设备确定第i个第一距离对应的障碍物与目标第二距离对应的障碍物相同，根据第i个第一距离对应的障碍物的描述信息更新目标第二距离对应的障碍物的障碍物信息，其中，目标第二距离是指t个差值中最小差值对应的第二距离，t≤m，t为正整数。
此时，由于与第i个第一距离的差值小于等于预设阈值的第二距离存在多个，所以需要首先确定出与第i个第一距离对应的障碍物相同的障碍物。此处将与第i个第一距离的差值最小的第二距离对应的障碍物确定为与第i个第一距离对应的障碍物相同的障碍物，更新障碍物信息列表中该障碍物对应的障碍物信息。
因此，通过上述过程，对每个第一距离进行判断，更新障碍物信息列表，障碍物信息处理设备获得更新后的障碍物信息列表。
下面结合具体实例说明障碍物信息处理设备如何更新障碍物信息列表获得更新后的障碍物信息列表。
假设毫米波雷达传感器采集了3个障碍物的描述信息，分别记为A、B、C，具体包括：
A(θ1,d1,s1)
B(θ2,d2,s2)
C(θ3,d3,s3)
其中，θ表示障碍物的方位角，d表示障碍物与目标交通工具之间的距离，s表示障碍物的速度。
因此，障碍物信息处理设备可以根据上述3个障碍物的描述信息，获得3个第一距离，分别为d1、d2、d3。
假设第一障碍物信息列表中包括3个障碍物信息，分别记为X、Y、Z，具体包括：
X(θx,dx,sx)
Y(θy,dy,sy)
Z(θz,dz,sz)
因此，障碍物信息处理设备可以根据上述3个障碍物信息，获得3个第二距离，分别为 dx，dy，dz。
作为一个可选的实施例，下面以d1为例说明如何更新第一障碍物信息列表。
(1)当d1与dx的差值，d1与dy的差值，d1与dz的差值均大于预设阈值时，将A(θ1,d1,s1)加入到障碍物信息列表。
(2)当d1仅与dx的差值小于等于预设阈值时，用A(θ1,d1,s1)更新X(θx,dx,sx)。
(3)当d1与dx的差值，d1与dy的差值均小于等于预设阈值，且d1与dx的差值小于d1与dy的差值时，用A(θ1,d1,s1)更新X(θx,dx,sx)。
作为一个可选的实施例，假设d1与dx的差值，d1与dy的差值，d1与dz的差值均大于预设阈值，d2仅与dx的差值小于等于预设阈值，d3与dz的差值小于等于预设阈值，d3与dy的差值小于等于预设阈值，且d3与dz的差值小于d3与dy的差值，因此，障碍物信息处理设备更新障碍物信息列表，更新后的障碍物信息列表，具体包括：
A(θ1,d1,s1)，对应d1与dx，dy，dz的差值均大于预设阈值，A(θ1,d1,s1)为新增的障碍物信息；
B(θ2,d2,s2)，对应d2与仅dx的差值小于等于预设阈值，即用B(θ2,d2,s2)更新X(θx,dx,sx)；
C(θ3,d3,s3)，对应d3与dz，dy的差值小于等于预设阈值，且d3与dz的差值小于d3与dy的差值，即用C(θ3,d3,s3)更新Z(θz,dz,sz)；
Y(θy,dy,sy)，保留。
应理解的是，上述更新规则是以障碍物与目标交通工具之间的距离作为更新依据，也可采用或结合其他参数作为更新依据，例如障碍物的方向角等。
此外，当障碍物检测数据采集器为传感器组合时，需要利用多个传感器分别采集到的障碍物描述信息对障碍物信息列表进行更新，具体过程不再赘述。
若障碍物检测数据采集器针对目标交通工具采集障碍物的描述信息的时间间隔较大，还可以结合目标交通工具的运动状态以及障碍物信息列表，对障碍物信息列表中的每个障碍物信息进行预测，以保证各个信息之间的时间同步，然后根据预测后的障碍物信息列表与障碍物检测数据采集器采集到的障碍物的描述信息进行匹配，从而保证更新后的障碍物信息列表的准确性。
针对步骤530，障碍物信息处理设备根据更新后的障碍物信息列表获得目标区域信息和非目标区域信息，更新后的障碍物信息列表包括至少一个障碍物信息，每个障碍物信息包括障碍物的方向角。具体的，获得目标区域信息和非目标区域信息可以采用但不限于以下两种方式：
第一种方式：障碍物信息处理设备根据每个障碍物信息包括的障碍物的方向角，确定一个目标区域的方向角区间，作为目标区域信息，其中，目标区域为包含障碍物的区域。
进一步地，障碍物信息处理设备根据目标区域信息和激光雷达数据采集器的最大方向角采集范围，确定至少一个非目标区域的方向角区间，作为非目标区域信息，其中，非目标区域为不包含障碍物的区域。
例如，更新后的障碍物信息列表中包括两个障碍物信息，则障碍物信息处理设备根据每个障碍物信息中包括的方向角参数向两侧各拓展预设角度后，作为一个目标区域的方向角区间。因此，如图6所示，两个障碍物分别对应两个目标区域方向角区间，将这两个目标区域方向角区间作为目标区域信息，进一步根据激光雷达数据采集器的最大方向角采集 范围将除目标区域外的其他区域作为非目标区域，获得三个非目标区域的方向角区间，作为非目标区域信息。
第二种方式：障碍物信息处理设备根据至少一个障碍物信息和预设筛选条件，从至少一个障碍物信息中筛选出至少一个满足预设筛选条件的障碍物信息，然后根据每个满足预设筛选条件的障碍物信息包括的障碍物的方向角，确定一个目标区域的方向角区间，作为目标区域信息，其中，目标区域为包含满足预设筛选条件的障碍物的区域。
进一步地，障碍物信息处理设备根据目标区域信息和激光雷达数据采集器的最大方向角采集范围，确定至少一个非目标区域的方向角区间，作为非目标区域信息，其中，非目标区域为不包含满足预设筛选条件的障碍物的区域。
例如，障碍物信息处理设备根据至少一个障碍物信息，可以基于障碍物的速度判断障碍物是否为处于动态的障碍物(即预设筛选条件)，当确定障碍物为处于动态的障碍物时，将该障碍物作为满足预设筛选条件的障碍物，根据该障碍物的障碍物信息包括的障碍物的方向角获得一个目标区域的方向角区间。
或者，障碍物信息处理设备根据至少一个障碍物信息，可以基于障碍物与目标交通工具之间的距离判断障碍物是否为位于近处的障碍物(即预设筛选条件)，例如，设定一个阈值，当障碍物与目标交通工具之间的距离小于该阈值时，确定障碍物为位于近处的障碍物时，将该障碍物作为满足预设筛选条件的障碍物，根据该障碍物的障碍物信息包括的障碍物的方向角获得一个目标区域的方向角区间。
又或者，障碍物信息处理设备根据至少一个障碍物信息，可以基于障碍物的方位角判断障碍物是否为目标交通工具行车方向上的障碍物(即预设筛选条件)，当确定障碍物为目标交通工具行车方向上的障碍物时，将该障碍物作为满足预设筛选条件的障碍物，根据该障碍物的障碍物信息包括的障碍物的方向角获得一个目标区域的方向角区间。
又或者，障碍物信息处理设备可以将位于近处且处于动态的障碍物作为满足预设筛选条件的障碍物，根据该障碍物的障碍物信息包括的障碍物的方向角获得一个目标区域的方向角区间。
通过上述方法，障碍物信息处理设备首先采用预设筛选条件，根据至少一个障碍物信息筛选出满足预设筛选条件的障碍物，然后根据满足预设筛选条件的障碍物对应的障碍物信息包括的障碍物的方向角，获得目标区域的方向角区间。因此，障碍物信息处理设备通过首先筛选出满足预设筛选条件的障碍物，可以减少最终得到的目标区域的数量，进而能够减少激光雷达数据采集器需要采集的采样点数，降低障碍物信息处理设备的数据处理量。
例如，第二障碍物信息列表中包括四个障碍物信息，障碍物信息处理设备可以将位于近处且处于动态的障碍物作为满足预设筛选条件的障碍物，如图7所示，满足预设筛选条件的障碍物包括两个，根据这两个障碍物的障碍物信息分别包括的障碍物的方向角确定两个目标区域的方向角区间，并进一步根据激光雷达数据采集器的最大方向角采集范围获得三个非目标区域的方向角区间。须知，此时非目标区域中也包括障碍物。
此外，障碍物信息处理设备还可参考目标交通工具的运动状态，即结合目标交通工具的运动状态以及更新后的障碍物信息列表，对更新后的障碍物信息列表中的每个障碍物信息进行预测，以保证各个信息之间的时间同步，然后根据预测后的障碍物信息列表，获得目标区域信息和非目标区域信息，从而可以更好地保证确定的目标区域的准确性。
针对步骤560，障碍物信息处理设备根据采样点集合和更新后的障碍物信息列表，获 得障碍物检测结果，可以采用但不限于以下方法。
以更新后的障碍物信息列表中包括的第k个障碍物信息为例，第k个障碍物信息为更新后的障碍物信息列中包括的至少一个障碍物信息中的任意一个，说明障碍物信息处理设备如何获得障碍物检测结果。
障碍物信息处理设备根据第k个障碍物信息包括的障碍物的方向角，确定第k个区域的方向角区间。然后，障碍物信息处理设备根据第k个区域的方向角区间，从采样点集合中确定出第k个目标采样点集合，其中，第k个目标采样点集合包括采样点集合中在第k个区域的方向角区间内采集的采样点。
例如，第k个障碍物信息包括的障碍物的方向角为θk，第k个区域的方向角区间为(θk-5°～θk+5°)，第k个目标采样点集合包括采样点集合中在(θk-5°～θk+5°)区域内采集的采样点。
进一步地，障碍物信息处理设备根据第k个目标采样点集合获得至少一个障碍物的轮廓信息。
具体的，障碍物信息处理设备对第k个目标采样点集合进行处理，例如，数据聚类，获得至少一个障碍物的轮廓信息，例如，由于在(θk-5°～θk+5°)区域内可能包括多个障碍物，每个障碍物与目标交通工具之间的距离不同，因此，可能获得多个障碍物的轮廓信息。
接着，障碍物信息处理设备根据上述至少一个障碍物的轮廓信息，获得至少一个判断距离，其中，每个判断距离是指该障碍物的轮廓信息对应的障碍物与目标交通工具之间的距离，与第k个障碍物信息对应的障碍物与目标交通工具之间的距离的差值。
例如，假设障碍物信息处理设备获得2个障碍物的轮廓信息，分别记为A1和A2，进一步地，根据A1和A2获得A1对应的障碍物与目标交通工具之间的距离S1，以及A2对应的障碍物与目标交通工具之间的距离S2，计算S1与第k个障碍物信息对应的障碍物与目标交通工具之间的距离差值，以及S2与第k个障碍物信息对应的障碍物与目标交通工具之间的距离差值，获得两个判断距离。
此时，将每个判断距离与预设距离阈值进行比较，可能包括以下几种情况：
情况1：障碍物信息处理设备将第一检测结果作为障碍物检测结果，其中，第一检测结果包括至少一个判断距离中大于预设距离阈值的判断距离对应的障碍物的轮廓信息。
具体的，当判断距离大于预设距离阈值时，可以推断该判断距离对应的障碍物为新检测到的障碍物，将该障碍物的轮廓信息作为障碍物检测结果。例如，该障碍物的轮廓信息为P1，该障碍物记为X1，则障碍物检测结果记为X1(P1)。
情况2：障碍物信息处理设备将第二检测结果与第k个障碍物信息作为障碍物检测结果，其中，第二检测结果是指至少一个判断距离中小于等于预设距离阈值的判断距离对应的障碍物的轮廓信息。
当判断距离小于等于预设距离阈值时，该判断距离对应的障碍物与第k个障碍物信息对应的障碍物相同，因此，可以将该判断距离对应的障碍物的轮廓信息与第k个障碍物信息共同作为障碍物检测结果。例如，该障碍物的轮廓信息为P1，第k个障碍物信息记为K(θk,dk,sk)，则障碍物检测结果记为K(θk,dk,sk,P1)。
此外，当障碍物信息处理设备未确定出目标采样点集合时，还可以包括以下三种处理方式：
(1)障碍物信息处理设备确定障碍物检测结果中不包括第k个障碍物信息。
由于障碍物信息处理设备未确定出第k个目标采样点集合，则表明第k个障碍物信息对应的障碍物未被激光雷达数据采集器采集到相关数据，第k个障碍物信息对应的障碍物可能已经远离目标交通工具，因此，不再作为障碍物检测结果。
(2)若目标区域信息中包括第k个区域的方向角区间，即第k个障碍物信息对应的障碍物为满足预设筛选条件的障碍物时，则障碍物信息处理设备确定障碍物检测结果中不包括第k个障碍物信息。
由于第k个障碍物信息对应的障碍物为满足预设筛选条件的障碍物，其对应的第k个区域的方向角区间为目标区域，已由激光雷达数据采集器采集获得了大量的采样点，但是此时未确定出第k个目标采样点集合，因此基本可以断定第k个障碍物信息对应的障碍物已经远离目标交通工具。
(3)若目标区域信息中不包括第k个区域的方向角区间，即第k个障碍物信息对应的障碍物不是满足预设筛选条件的障碍物时，将第k个障碍物信息中包括的障碍物存在概率降低预设数值，获得更新后的第k个障碍物信息，将更新后的第k个障碍物信息作为障碍物检测结果，其中，每个障碍物信息中包括障碍物的存在概率。
此时，第k个障碍物信息对应的障碍物不为满足预设筛选条件的障碍物，由于它存在于非目标区域中，此时激光雷达数据采集器获得的采样点数较少，可能不足以获得该障碍物的轮廓信息，因此，不删除该障碍物对应的障碍物信息，而是将障碍物信息中包括的障碍物存在概率降低预设数值，进而保证障碍物检测结果的全面性。
进一步地，在障碍物信息处理设备根据采样点集合和至少一个障碍物信息分别对应的目标采样点集合获得障碍物检测结果后，障碍物信息处理设备还可根据采样点集合和至少一个障碍物信息分别对应的目标采样点集合确定非目标采样点集合，即剩余采样点集合。
障碍物信息处理设备根据非目标采样点集合确定至少一个障碍物的轮廓信息，例如，数据聚类，作为第三检测结果。障碍物信息处理设备将第三检测结果作为障碍物检测结果。因此，保证了障碍物检测结果的全面性。
应理解的是，激光雷达采集器的数据采集量大，精度高，毫米波雷达传感器存在障碍物误检的情况。因此，即使非目标区域内不包含障碍物，通过激光雷达采集器还是可能检测到障碍物，而且目标交通工具一直处于行驶状态，当激光雷达采集器采集数据时，此时在非目标区域内还可能出现新的障碍物。
此外，基于与上述方法的相同思路，障碍物信息处理设备还根据采样点集合首先获得至少一个障碍物的轮廓信息，作为轮廓信息集合。此时，障碍物信息处理设备根据第k个障碍物信息包括的障碍物的方向角，确定第k个区域的方向角区间。然后，障碍物信息处理设备根据第k个区域的方向角区间，从轮廓信息集合中确定出第k个目标轮廓信息集合，接着，障碍物信息处理设备根据第k个目标轮廓信息集合中包括的至少一个障碍物的轮廓信息，获得至少一个判断距离，后续处理过程与上述方法相同，不再赘述。同理，在障碍物信息处理设备根据轮廓信息集合和至少一个障碍物信息分别对应的目标轮廓信息集合获得障碍物检测结果后，障碍物信息处理设备可以将轮廓信息集合中除去至少一个障碍物信息分别对应的目标轮廓信息集合外的轮廓信息也作为障碍物检测结果。
应理解的是，障碍物信息处理设备还可能根据采样点集合没有获得任何障碍物的轮廓信息，此时表明在激光雷达数据采集器的采集范围内没有发现障碍物。
此外，障碍物信息处理设备还可参考目标交通工具的运动状态，即结合目标交通工具的运动状态以及更新后的障碍物信息列表，对更新后的障碍物信息列表中的每个障碍物信息进行预测，以保证各个信息之间的时间同步，然后根据采样点集合和预测后的障碍物信息列表，获得障碍物检测结果，从而可以更好地保证障碍物检测结果的准确性。
作为一个可选的实施例，参阅图8所示，图8中包括一个毫米波雷达传感器。
步骤(step)1：毫米波雷达传感器采集了3个障碍物的描述信息发送至障碍物信息处理设备，分别记为A、B、C，具体包括：
A(θ1,d1,s1)
B(θ2,d2,s2)
C(θ3,d3,s3)
其中，θ表示障碍物的方位角，d表示障碍物与目标交通工具之间的距离，s表示障碍物的速度。
step2：当障碍物信息列表为空时，障碍物信息处理设备直接将上述3个障碍物的描述信息添加到障碍物信息列表，获得更新后的障碍物信息列表。
Step3：障碍物信息处理设备根据更新后的障碍物信息列表，获得目标区域信息和非目标区域信息发送至激光雷达数据采集器。
例如，障碍物信息处理设备选取速度大于0的障碍物作为满足预设筛选条件的障碍物，并根据障碍物信息中每个满足预设筛选条件的障碍物的方位角，将该方向角顺时针和逆时针各拓展一角度(例如5°)作为目标区域，其他区域作为非目标区域。假设s3＝0，s1>0，s2>0，则选取障碍物A和障碍物B作为满足预设筛选条件的障碍物，并获得两个目标区域的方向角区间，分别对应(θ1-5°～θ1+5°)和(θ2-5°～θ2+5°)。目标区域信息还包括对目标区域采用0.5°的水平分辨率扫描，非目标区域信息还包括对非目标区域采用2°的水平分辨率扫描。
Step4：激光雷达数据采集器根据目标区域信息和非目标区域信息执行数据采集，并将采样点集合反馈至障碍物信息处理设备。
Step5：障碍物信息处理设备根据采样点集合和更新后的障碍物信息列表获得障碍物检测结果。
具体的，障碍物信息处理设备根据采样点集合，获得X(θ1,d1,p1)，其中p1代表障碍物的轮廓信息。X(θ1,d1,p1)与第二障碍物信息列表中的A(θ1,d1,s1)匹配成功，因此障碍物A保留，并更新其信息为A(θ1,d1,s1,p1)，作为障碍物检测结果。由于障碍物B未能与障碍物的轮廓信息匹配成功，且B为满足预设筛选条件的障碍物，则障碍物B的障碍物信息不作为障碍物检测结果。障碍物C虽然未能与障碍物的轮廓信息匹配成功，但是由于其为不满足预设筛选条件的障碍物，因此暂时保留，但是降低障碍物存在概率，以C’(θ3,d3,s3)表示，作为障碍物检测结果。
最终，障碍物信息获得的障碍物检测结果为：
A(θ1,d1,s1,p1)
C’(θ3,d3,s3)
作为一个可选的实施例，参阅图9所示，将图8所示实施例中最终获得的障碍物检测结果作为下一次根据障碍物的描述信息进行更新的障碍物信息列表，即障碍物信息列表中加入了上一次障碍物检测结果的反馈，可以进一步提高障碍物信息列表的可靠性。
作为一个可选的实施例，参阅图10所示，图10中包括两个毫米波雷达传感器，其他与图8所示实施例一致。
step1：毫米波雷达传感器1采集了3个障碍物的描述信息发送至障碍物信息处理设备，分别记为A、B、C。
step1’：毫米波雷达传感器2采集了1个障碍物的描述信息发送至障碍物信息处理设备，记为D，D(θ4,d4,s4)。
step2：当障碍物信息列表为空时，障碍物信息处理设备直接将上述4个障碍物的描述信息添加到障碍物信息列表，获得第二障碍物信息列表。
Step3：障碍物信息处理设备根据第二障碍物信息列表，获得目标区域信息和非目标区域信息发送至激光雷达数据采集器。
假设此处与图8所示的实施例一致，障碍物信息处理设备选取障碍物A和障碍物B作为满足预设筛选条件的障碍物，获得目标区域信息和非目标区域信息发送至激光雷达数据采集器。
Step4：激光雷达数据采集器根据目标区域信息和非目标区域信息执行数据采集，并将采样点集合反馈至障碍物信息处理设备。
Step5：障碍物信息处理设备根据采样点集合和更新后的障碍物信息列表获得障碍物检测结果。
假设此处与图8所示的实施例一致，障碍物信息处理设备根据采样点集合，获得X(θ1,d1,p1)，因此，障碍物A保留，并更新其信息为A(θ1,d1,s1,p1)，障碍物B的障碍物信息不能作为障碍物检测结果，障碍物C的障碍物信息和障碍物D的障碍物信息中的障碍物存在概率均降低预设数值，分别以C’(θ3,d3,s3)，D’(θ4,d4,s4)表示。
最终，障碍物信息获得的障碍物检测结果为：
A(θ1,d1,s1,p1)
C’(θ3,d3,s3)
D’(θ4,d4,s4)
作为一个可选的实施例，参阅图11所示，图11中包括两个激光雷达数据采集器，其他与图8所示实施例一致。
step1：毫米波雷达传感器1采集了3个障碍物的描述信息发送至障碍物信息处理设备，分别记为A、B、C。
step2：当障碍物信息列表为空时，障碍物信息处理设备直接将上述3个障碍物的描述信息添加到障碍物信息列表，获得更新后的障碍物信息列表。
Step3：障碍物信息处理设备根据更新后的障碍物信息列表，获得目标区域信息和非目标区域信息发送至激光雷达数据采集器。
假设此处与图8所示的实施例一致，障碍物信息处理设备选取障碍物A和障碍物B作为满足预设筛选条件的障碍物，获得目标区域信息和非目标区域信息发送至激光雷达数据采集器1和激光雷达数据采集器2。
Step4：激光雷达数据采集器1根据目标区域信息和非目标区域信息执行数据采集，并将采样点集合反馈至障碍物信息处理设备。
Step4’：激光雷达数据采集器2根据目标区域信息和非目标区域信息执行数据采集，并将采样点集合反馈至障碍物信息处理设备。
Step5：障碍物信息处理设备根据采样点集合和更新后的障碍物信息列表获得障碍物检测结果。
假设此处与图8所示的实施例一致，障碍物信息处理设备根据激光雷达数据采集器1反馈的采样点集合，获得X(θ1,d1,p1)，根据激光雷达数据采集器2反馈的采样点集合，获得Y(θ2,d2,p2)。因此，障碍物A保留，并更新其信息为A(θ1,d1,s1,p1)，障碍物B保留，并更新其信息为B(θ2,d2,s2,p2)。障碍物C的障碍物信息中的障碍物存在概率降低预设数值，以C’(θ3,d3,s3)表示。
最终，障碍物信息获得的障碍物检测结果为：
A(θ1,d1,s1,p1)
B(θ2,d2,s2,p2)
C’(θ3,d3,s3)
基于同一构思，本申请还提供了一种障碍物信息处理设备，该设备可以用于执行上述图5中障碍物信息处理设备对应的方法实施例，因此本申请实施例提供的障碍物信息处理设备的实施方式可以参见该方法的实施方式，重复之处不再赘述。
参阅图2所示，本申请实施例提供一种障碍物信息处理设备200，包括收发器210、处理器220、存储器230；存储器230用于存储程序、指令或代码；处理器220用于执行存储器230中的程序、指令或代码；
收发器210，用于接收障碍物检测数据采集器针对目标交通工具采集的n个障碍物的描述信息，n为正整数；
处理器220，用于根据n个障碍物的描述信息更新障碍物信息列表，获得更新后的障碍物信息列表；根据更新后的障碍物信息列表获得目标区域信息和非目标区域信息；
收发器210，用于将目标区域信息和非目标区域信息发送给激光雷达数据采集器，以使激光雷达数据采集器在每个目标区域的单位体积内采集的采样点数大于在每个非目标区域的单位体积内采集的采样点数，其中，目标区域信息用于指示至少一个目标区域，非目标区域信息用于指示至少一个非目标区域，其中，目标区域为包含障碍物的区域，非目标区域为不包含障碍物的区域，或者，目标区域为包含满足预设筛选条件的障碍物的区域，非目标区域为不包含满足预设筛选条件的障碍物的区域；以及接收激光雷达数据采集器采集的采样点集合，采样点集合包括激光雷达数据采集器在至少一个目标区域内采集的采样点数和在至少一个非目标区域内采集的采样点数；
处理器220，用于根据采样点集合和更新后的障碍物信息列表，获得障碍物检测结果。
在一种可能的设计中，障碍物检测数据采集器包括至少一个视觉传感器和/或至少一个毫米波雷达传感器。
在一种可能的设计中，目标区域信息还包括第一信号发射参数，非目标区域信息还包括第二信号发射参数，其中，第一信号发射参数大于第二信号发射参数；
和/或，目标区域信息还包括第一信号接收采样参数，非目标区域信息还包括第二信号接收采样参数，其中，第一信号接收采样参数大于第二信号接收采样参数。
在一种可能的设计中，障碍物信息列表包括m个障碍物信息，m为正整数；
处理器220用于：
根据n个障碍物的描述信息，获取n个第一距离，n个第一距离是指n个障碍物的描述信息对应的n个障碍物分别与目标交通工具之间的距离；
根据障碍物信息列表中包括的m个障碍物信息，获取m个第二距离，m个第二距离是指m个障碍物信息对应的m个障碍物分别与目标交通工具之间的距离，m为正整数；
针对第i个第一距离，i为取遍1～n的任意整数，更新障碍物信息列表，包括:
当第i个第一距离与每个第二距离的差值均大于预设阈值时，将第i个第一距离对应的障碍物的描述信息增加至障碍物信息列表；
当第i个第一距离与t个第二距离计算得到的t个差值均小于等于预设阈值时，确定第i个第一距离对应的障碍物与目标第二距离对应的障碍物相同，根据第i个第一距离对应的障碍物的描述信息更新目标第二距离对应的障碍物的障碍物信息，其中，目标第二距离是指t个差值中最小差值对应的第二距离，t≤m，t为正整数；
当第i个第一距离仅与第j个第二距离的差值小于等于预设阈值时，确定第i个第一距离对应的障碍物与第j个第二距离对应的障碍物相同，根据第i个第一距离对应的障碍物的描述信息更新第j个第二距离对应的障碍物的障碍物信息，第j个第二距离是指m个第二距离中的任意一个。
在一种可能的设计中，更新后的障碍物信息列表包括至少一个障碍物信息，每个障碍物信息包括障碍物的方向角；
处理器220用于：
根据每个障碍物信息包括的障碍物的方向角，确定一个目标区域的方向角区间，作为目标区域信息，其中，目标区域为包含障碍物的区域；
根据目标区域信息和激光雷达数据采集器的最大方向角采集范围，确定至少一个非目标区域的方向角区间，作为非目标区域信息，其中，非目标区域为不包含障碍物的区域。
在一种可能的设计中，更新后的障碍物信息列表包括至少一个障碍物信息，每个障碍物信息包括障碍物的方向角；
处理器220用于：
根据至少一个障碍物信息和预设筛选条件，从至少一个障碍物信息中筛选出至少一个满足预设筛选条件的障碍物信息；
根据每个满足预设筛选条件的障碍物信息包括的障碍物的方向角，确定一个目标区域的方向角区间，作为目标区域信息，其中，目标区域为包含满足预设筛选条件的障碍物的区域；
根据目标区域信息和激光雷达数据采集器的最大方向角采集范围，确定至少一个非目标区域的方向角区间，作为非目标区域信息，其中，非目标区域为不包含满足预设筛选条件的障碍物的区域。
在一种可能的设计中，更新后的障碍物信息列表包括至少一个障碍物信息，每个障碍物信息包括障碍物的方向角；
处理器220用于：
根据第k个障碍物信息包括的障碍物的方向角，确定第k个区域的方向角区间，其中，第k个障碍物信息为至少一个障碍物信息中的任意一个；
根据第k个区域的方向角区间，从采样点集合中确定出第k个目标采样点集合，第k个目标采样点集合包括采样点集合中在第k个区域的方向角区间内采集的采样点；
根据第k个目标采样点集合获得至少一个障碍物的轮廓信息；
根据至少一个障碍物的轮廓信息，获得至少一个判断距离，其中，每个判断距离是指 该障碍物的轮廓信息对应的障碍物与目标交通工具之间的距离，与第k个障碍物信息对应的障碍物与目标交通工具之间的距离的差值；
将第一检测结果作为障碍物检测结果，其中，第一检测结果包括至少一个判断距离中大于预设距离阈值的判断距离对应的障碍物的轮廓信息；和/或，将第二检测结果与第k个障碍物信息作为障碍物检测结果，其中，第二检测结果是指至少一个判断距离中小于等于预设距离阈值的判断距离对应的障碍物的轮廓信息。
在一种可能的设计中，处理器220还用于：
当未确定出第k个目标采样点集合时，确定障碍物检测结果中不包括第k个障碍物信息；或者
当未确定出第k个目标采样点集合时，若目标区域信息中包括第k个目标区域的方向角区间，则障碍物检测结果中不包括第k个障碍物信息；若目标区域信息中不包括第k个目标区域的方向角区间，将第k个障碍物信息中包括的障碍物存在概率降低预设数值，获得更新后的第k个障碍物信息，将更新后的第k个障碍物信息作为障碍物检测结果，每个障碍物信息中包括障碍物的存在概率。
在一种可能的设计中，处理器220还用于：
根据采样点集合和至少一个障碍物信息分别对应的目标采样点集合，确定非目标采样点集合；
根据非目标采样点集合确定至少一个障碍物的轮廓信息，作为第三检测结果；
将第三检测结果作为障碍物检测结果。
在一种可能的设计中，处理器220还用于：
在根据采样点集合和更新后的障碍物信息列表，获得障碍物检测结果之后，采用障碍物检测结果替换更新后的障碍物信息列表。
基于同一构思，本申请还提供了一种激光雷达数据采集器，该设备可以用于执行上述图5中激光雷达数据采集器对应的方法实施例，因此本申请实施例提供的激光雷达数据采集器的实施方式可以参见该方法的实施方式，重复之处不再赘述。
参阅图4所示，本申请提供一种激光雷达数据采集器400，包括:信号发送器410，激光雷达传感器420，激光雷达控制器430，信号接收器440，其中，信号发送器410与激光雷达传感器420连接，信号接收器440与激光雷达控制器430连接，激光雷达传感器420与激光雷达控制器430连接；
信号接收器440，用于接收障碍物信息处理设备发送的目标区域信息和非目标区域信息，其中，目标区域信息用于指示至少一个目标区域，非目标区域信息用于指示至少一个非目标区域，目标区域为包含障碍物的区域，非目标区域为不包含障碍物的区域，或者，目标区域为包含满足预设筛选条件的障碍物的区域，非目标区域为不包含满足预设筛选条件的障碍物的区域；
激光雷达控制器430，用于根据目标区域信息和非目标区域信息控制激光雷达传感器420执行数据采集；
激光雷达传感器420，用于获得采样点集合，采样点集合包括在至少一个目标区域内采集的采样点数和在至少一个非目标区域内采集的采样点数，且在每个目标区域的单位体积内采集的采样点数大于在每个非目标区域的单位体积内采集的采样点数；
信号发送器410，用于将采样点集合发送至障碍物信息处理设备。
在一种可能的设计中，目标区域信息还包括第一信号发射参数，非目标区域信息还包括第二信号发射参数，其中，第一信号发射参数大于第二信号发射参数；
和/或，目标区域信息还包括第一信号接收采样参数，非目标区域信息还包括第二信号接收采样参数，其中，第一信号接收采样参数大于第二信号接收采样参数。
综上所述，障碍物检测数据采集器首先获得由障碍物检测数据采集器采集的障碍物的描述信息，然后对障碍物信息列表进行更新，并根据更新后的障碍物信息列表获得目标区域信息和非目标区域信息，以使激光雷达数据采集器针对不同区域获取不同精度的采样点集合，避免了较大的数据处理量。障碍物信息处理设备根据获得的采样点集合和更新后的障碍物信息列表，获得最终的障碍物检测结果。因此，采用本申请实施例提供的方法，激光雷达数据采集器在目标区域获取的障碍物的轮廓信息的精度较高，在非目标区域获取的障碍物的轮廓信息的精度较低，这样既保证了障碍物识别精度，同时避免较大的数据处理量，可以有效避免因计算资源过度消耗导致的处理延时和宕机等问题。
本领域内的技术人员应明白，本申请实施例可提供为方法、系统、或计算机程序产品。因此，本申请实施例可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且，本申请实施例可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。
本申请实施例是参照根据本申请实施例的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器，使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理设备以特定方式工作的计算机可读存储器中，使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品，该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理设备上，使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理，从而在计算机或其他可编程设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
显然，本领域的技术人员可以对本申请实施例进行各种改动和变型而不脱离本申请的精神和范围。这样，倘若本申请实施例的这些修改和变型属于本申请权利要求及其等同技术的范围之内，则本申请也意图包含这些改动和变型在内。

Claims (25)

  1. 一种障碍物检测方法，其特征在于，包括:
    障碍物信息处理设备接收障碍物检测数据采集器针对目标交通工具采集的n个障碍物的描述信息，n为正整数；
    所述障碍物信息处理设备根据所述n个障碍物的描述信息更新障碍物信息列表，获得更新后的障碍物信息列表；
    所述障碍物信息处理设备根据所述更新后的障碍物信息列表获得目标区域信息和非目标区域信息，并将所述目标区域信息和所述非目标区域信息发送给激光雷达数据采集器，以使所述激光雷达数据采集器在每个目标区域的单位体积内采集的采样点数大于在每个非目标区域的单位体积内采集的采样点数，其中，所述目标区域信息用于指示至少一个目标区域，所述非目标区域信息用于指示至少一个非目标区域，其中，所述目标区域为包含障碍物的区域，所述非目标区域为不包含障碍物的区域，或者，所述目标区域为包含满足预设筛选条件的障碍物的区域，所述非目标区域为不包含满足所述预设筛选条件的障碍物的区域；
    所述障碍物信息处理设备接收所述激光雷达数据采集器采集的采样点集合，所述采样点集合包括所述激光雷达数据采集器在所述至少一个目标区域内采集的采样点数和在所述至少一个非目标区域内采集的采样点数；
    所述障碍物信息处理设备根据所述采样点集合和所述更新后的障碍物信息列表，获得障碍物检测结果。
  2. 如权利要求1所述的方法，其特征在于，所述障碍物检测数据采集器包括至少一个视觉传感器和/或至少一个毫米波雷达传感器。
  3. 如权利要求1或2所述的方法，其特征在于，所述目标区域信息还包括第一信号发射参数，所述非目标区域信息还包括第二信号发射参数，其中，所述第一信号发射参数大于所述第二信号发射参数；
    和/或，所述目标区域信息还包括第一信号接收采样参数，所述非目标区域信息还包括第二信号接收采样参数，其中，所述第一信号接收采样参数大于所述第二信号接收采样参数。
  4. 如权利要求1-3任一项所述的方法，其特征在于，所述障碍物信息列表包括m个障碍物信息，m为正整数；
    所述障碍物信息处理设备根据所述n个障碍物的描述信息更新障碍物信息列表，得到更新后的障碍物信息列表，包括：
    所述障碍物信息处理设备根据所述n个障碍物的描述信息，获取n个第一距离，所述n个第一距离是指所述n个障碍物的描述信息对应的n个障碍物分别与所述目标交通工具之间的距离；
    所述障碍物信息处理设备根据所述障碍物信息列表中包括的m个障碍物信息，获取m个第二距离，所述m个第二距离是指所述m个障碍物信息对应的m个障碍物分别与所述目标交通工具之间的距离；
    所述障碍物信息处理设备针对第i个第一距离，i为取遍1～n的任意整数，更新所述障碍物信息列表，包括:
    当第i个第一距离与每个第二距离的差值均大于预设阈值时，所述障碍物信息处理设备将所述第i个第一距离对应的障碍物的描述信息增加至所述障碍物信息列表；
    当所述第i个第一距离与t个第二距离计算得到的t个差值均小于等于所述预设阈值时，所述障碍物信息处理设备确定所述第i个第一距离对应的障碍物与目标第二距离对应的障碍物相同，根据所述第i个第一距离对应的障碍物的描述信息更新所述目标第二距离对应的障碍物的障碍物信息，其中，所述目标第二距离是指所述t个差值中最小差值对应的第二距离，t≤m，t为正整数；
    当所述第i个第一距离仅与第j个第二距离的差值小于等于所述预设阈值时，所述障碍物信息处理设备确定所述第i个第一距离对应的障碍物与所述第j个第二距离对应的障碍物相同，根据所述第i个第一距离对应的障碍物的描述信息更新所述第j个第二距离对应的障碍物的障碍物信息，所述第j个第二距离是指所述m个第二距离中的任意一个。
  5. 如权利要求1-4任一项所述的方法，其特征在于，所述更新后的障碍物信息列表包括至少一个障碍物信息，每个障碍物信息包括障碍物的方向角；
    所述障碍物信息处理设备根据所述更新后的障碍物信息列表获得目标区域信息和非目标区域信息，包括：
    所述障碍物信息处理设备根据每个障碍物信息包括的障碍物的方向角，确定一个目标区域的方向角区间，作为目标区域信息，其中，所述目标区域为包含障碍物的区域；
    所述障碍物信息处理设备根据所述目标区域信息和所述激光雷达数据采集器的最大方向角采集范围，确定至少一个非目标区域的方向角区间，作为非目标区域信息，其中，所述非目标区域为不包含障碍物的区域。
  6. 如权利要求1-4任一项所述的方法，其特征在于，所述更新后的障碍物信息列表包括至少一个障碍物信息，每个障碍物信息包括障碍物的方向角；
    所述障碍物信息处理设备根据所述更新后的障碍物信息列表获得目标区域信息和非目标区域信息，包括：
    所述障碍物信息处理设备根据所述至少一个障碍物信息和所述预设筛选条件，从所述至少一个障碍物信息中筛选出至少一个满足所述预设筛选条件的障碍物信息；
    所述障碍物信息处理设备根据每个满足所述预设筛选条件的障碍物信息包括的障碍物的方向角，确定一个目标区域的方向角区间，作为目标区域信息，其中，所述目标区域为包含满足所述预设筛选条件的障碍物的区域；
    所述障碍物信息处理设备根据所述目标区域信息和所述激光雷达数据采集器的最大方向角采集范围，确定至少一个非目标区域的方向角区间，作为非目标区域信息，其中，所述非目标区域为不包含满足所述预设筛选条件的障碍物的区域。
  7. 如权利要求1-6任一项所述的方法，其特征在于，所述更新后的障碍物信息列表包括至少一个障碍物信息，每个障碍物信息包括障碍物的方向角；
    所述障碍物信息处理设备根据所述采样点集合和所述更新后的障碍物信息列表，获得障碍物检测结果，包括：
    所述障碍物信息处理设备根据第k个障碍物信息包括的障碍物的方向角，确定第k个区域的方向角区间，其中，所述第k个障碍物信息为所述至少一个障碍物信息中的任意一个；
    所述障碍物信息处理设备根据所述第k个区域的方向角区间，从所述采样点集合中确定出第k个目标采样点集合，所述第k个目标采样点集合包括所述采样点集合中在所述第k 个区域的方向角区间内采集的采样点；
    所述障碍物信息处理设备根据所述第k个目标采样点集合获得至少一个障碍物的轮廓信息；
    所述障碍物信息处理设备根据所述至少一个障碍物的轮廓信息，获得至少一个判断距离，其中，每个判断距离是指该障碍物的轮廓信息对应的障碍物与所述目标交通工具之间的距离，与所述第k个障碍物信息对应的障碍物与所述目标交通工具之间的距离的差值；
    将第一检测结果作为障碍物检测结果，其中，所述第一检测结果包括所述至少一个判断距离中大于预设距离阈值的判断距离对应的障碍物的轮廓信息；和/或，将第二检测结果与所述第k个障碍物信息作为障碍物检测结果，其中，所述第二检测结果是指所述至少一个判断距离中小于等于所述预设距离阈值的判断距离对应的障碍物的轮廓信息。
  8. 如权利要求7所述的方法，其特征在于，还包括：
    当所述障碍物信息处理设备未确定出所述第k个目标采样点集合时，确定障碍物检测结果中不包括所述第k个障碍物信息；或者
    当所述障碍物信息处理设备未确定出所述第k个目标采样点集合时，若所述目标区域信息中包括所述第k个目标区域的方向角区间，则障碍物检测结果中不包括所述第k个障碍物信息；若所述目标区域信息中不包括所述第k个目标区域的方向角区间，将所述第k个障碍物信息中包括的障碍物存在概率降低预设数值，获得更新后的第k个障碍物信息，将所述更新后的第k个障碍物信息作为障碍物检测结果，每个障碍物信息中包括障碍物的存在概率。
  9. 如权利要求7或8所述的方法，其特征在于，还包括：
    所述障碍物信息处理设备根据所述采样点集合和所述至少一个障碍物信息分别对应的目标采样点集合，确定非目标采样点集合；
    所述障碍物信息处理设备根据所述非目标采样点集合确定至少一个障碍物的轮廓信息，作为第三检测结果；
    所述障碍物信息处理设备将所述第三检测结果作为障碍物检测结果。
  10. 如权利要求1-9任一项所述的方法，其特征在于，在所述障碍物信息处理设备根据所述采样点集合和所述更新后的障碍物信息列表，获得障碍物检测结果之后，还包括：
    所述障碍物信息处理设备采用所述障碍物检测结果替换所述更新后的障碍物信息列表。
  11. 一种障碍物检测方法，其特征在于，包括:
    激光雷达数据采集器接收障碍物信息处理设备发送的目标区域信息和非目标区域信息，其中，所述目标区域信息用于指示至少一个目标区域，所述非目标区域信息用于指示至少一个非目标区域，所述目标区域为包含障碍物的区域，所述非目标区域为不包含障碍物的区域，或者，所述目标区域为包含满足预设筛选条件的障碍物的区域，所述非目标区域为不包含满足所述预设筛选条件的障碍物的区域；
    所述激光雷达数据采集器根据所述目标区域信息和所述非目标区域信息执行数据采集，获得采样点集合，所述采样点集合包括所述激光雷达数据采集器在所述至少一个目标区域内采集的采样点数和在所述至少一个非目标区域内采集的采样点数，且在每个目标区域的单位体积内采集的采样点数大于在每个非目标区域的单位体积内采集的采样点数；
    所述激光雷达数据采集器将所述采样点集合发送至所述障碍物信息处理设备。
  12. 如权利要求11所述的方法，其特征在于，所述目标区域信息还包括第一信号发射参数，所述非目标区域信息还包括第二信号发射参数，其中，所述第一信号发射参数大于所述第二信号发射参数；
    和/或，所述目标区域信息还包括第一信号接收采样参数，所述非目标区域信息还包括第二信号接收采样参数，其中，所述第一信号接收采样参数大于所述第二信号接收采样参数。
  13. 一种障碍物信息处理设备，其特征在于，包括收发器、处理器、存储器；所述存储器用于存储程序、指令或代码；所述处理器用于执行所述存储器中的程序、指令或代码；
    所述收发器，用于接收障碍物检测数据采集器针对目标交通工具采集的n个障碍物的描述信息，n为正整数；
    所述处理器，用于根据所述n个障碍物的描述信息更新障碍物信息列表，获得更新后的障碍物信息列表；根据所述更新后的障碍物信息列表获得目标区域信息和非目标区域信息；
    所述收发器，用于将所述目标区域信息和所述非目标区域信息发送给激光雷达数据采集器，以使所述激光雷达数据采集器在每个目标区域的单位体积内采集的采样点数大于在每个非目标区域的单位体积内采集的采样点数，其中，所述目标区域信息用于指示至少一个目标区域，所述非目标区域信息用于指示至少一个非目标区域，其中，所述目标区域为包含障碍物的区域，所述非目标区域为不包含障碍物的区域，或者，所述目标区域为包含满足预设筛选条件的障碍物的区域，所述非目标区域为不包含满足所述预设筛选条件的障碍物的区域；以及接收所述激光雷达数据采集器采集的采样点集合，所述采样点集合包括所述激光雷达数据采集器在所述至少一个目标区域内采集的采样点数和在所述至少一个非目标区域内采集的采样点数；
    所述处理器，用于根据所述采样点集合和所述更新后的障碍物信息列表，获得障碍物检测结果。
  14. 如权利要求13所述的障碍物信息处理设备，其特征在于，所述障碍物检测数据采集器包括至少一个视觉传感器和/或至少一个毫米波雷达传感器。
  15. 如权利要求13或14所述的障碍物信息处理设备，其特征在于，所述目标区域信息还包括第一信号发射参数，所述非目标区域信息还包括第二信号发射参数，其中，所述第一信号发射参数大于所述第二信号发射参数；
    和/或，所述目标区域信息还包括第一信号接收采样参数，所述非目标区域信息还包括第二信号接收采样参数，其中，所述第一信号接收采样参数大于所述第二信号接收采样参数。
  16. 如权利要求13-15任一项所述的障碍物信息处理设备，其特征在于，所述障碍物信息列表包括m个障碍物信息，m为正整数；
    所述处理器用于：
    根据所述n个障碍物的描述信息，获取n个第一距离，所述n个第一距离是指所述n个障碍物的描述信息对应的n个障碍物分别与所述目标交通工具之间的距离；
    根据所述障碍物信息列表中包括的m个障碍物信息，获取m个第二距离，所述m个第二距离是指所述m个障碍物信息对应的m个障碍物分别与所述目标交通工具之间的距离，m为正整数；
    针对第i个第一距离，i为取遍1～n的任意整数，更新所述障碍物信息列表，包括:
    当第i个第一距离与每个第二距离的差值均大于预设阈值时，将所述第i个第一距离对应的障碍物的描述信息增加至所述障碍物信息列表；
    当所述第i个第一距离与t个第二距离计算得到的t个差值均小于等于所述预设阈值时，确定所述第i个第一距离对应的障碍物与目标第二距离对应的障碍物相同，根据所述第i个第一距离对应的障碍物的描述信息更新所述目标第二距离对应的障碍物的障碍物信息，其中，所述目标第二距离是指所述t个差值中最小差值对应的第二距离，t≤m，t为正整数；
    当所述第i个第一距离仅与第j个第二距离的差值小于等于所述预设阈值时，确定所述第i个第一距离对应的障碍物与所述第j个第二距离对应的障碍物相同，根据所述第i个第一距离对应的障碍物的描述信息更新所述第j个第二距离对应的障碍物的障碍物信息，所述第j个第二距离是指所述m个第二距离中的任意一个。
  17. 如权利要求13-16任一项所述的障碍物信息处理设备，其特征在于，所述更新后的障碍物信息列表包括至少一个障碍物信息，每个障碍物信息包括障碍物的方向角；
    所述处理器用于：
    根据每个障碍物信息包括的障碍物的方向角，确定一个目标区域的方向角区间，作为目标区域信息，其中，所述目标区域为包含障碍物的区域；
    根据所述目标区域信息和所述激光雷达数据采集器的最大方向角采集范围，确定至少一个非目标区域的方向角区间，作为非目标区域信息，其中，所述非目标区域为不包含障碍物的区域。
  18. 如权利要求13-16任一项所述的障碍物信息处理设备，其特征在于，所述更新后的障碍物信息列表包括至少一个障碍物信息，每个障碍物信息包括障碍物的方向角；
    所述处理器用于：
    根据所述至少一个障碍物信息和所述预设筛选条件，从所述至少一个障碍物信息中筛选出至少一个满足所述预设筛选条件的障碍物信息；
    根据每个满足所述预设筛选条件的障碍物信息包括的障碍物的方向角，确定一个目标区域的方向角区间，作为目标区域信息，其中，所述目标区域为包含满足所述预设筛选条件的障碍物的区域；
    根据所述目标区域信息和所述激光雷达数据采集器的最大方向角采集范围，确定至少一个非目标区域的方向角区间，作为非目标区域信息，其中，所述非目标区域为不包含满足所述预设筛选条件的障碍物的区域。
  19. 如权利要求13-18任一项所述的障碍物信息处理设备，其特征在于，所述更新后的障碍物信息列表包括至少一个障碍物信息，每个障碍物信息包括障碍物的方向角；
    所述处理器用于：
    根据第k个障碍物信息包括的障碍物的方向角，确定第k个区域的方向角区间，其中，所述第k个障碍物信息为所述至少一个障碍物信息中的任意一个；
    根据所述第k个区域的方向角区间，从所述采样点集合中确定出第k个目标采样点集合，所述第k个目标采样点集合包括所述采样点集合中在所述第k个区域的方向角区间内采集的采样点；
    根据所述第k个目标采样点集合获得至少一个障碍物的轮廓信息；
    根据所述至少一个障碍物的轮廓信息，获得至少一个判断距离，其中，每个判断距离 是指该障碍物的轮廓信息对应的障碍物与所述目标交通工具之间的距离，与所述第k个障碍物信息对应的障碍物与所述目标交通工具之间的距离的差值；
    将第一检测结果作为障碍物检测结果，其中，所述第一检测结果包括所述至少一个判断距离中大于预设距离阈值的判断距离对应的障碍物的轮廓信息；和/或，将第二检测结果与所述第k个障碍物信息作为障碍物检测结果，其中，所述第二检测结果是指所述至少一个判断距离中小于等于所述预设距离阈值的判断距离对应的障碍物的轮廓信息。
  20. 如权利要求19所述的障碍物信息处理设备，其特征在于，所述处理器还用于：
    当未确定出所述第k个目标采样点集合时，确定障碍物检测结果中不包括所述第k个障碍物信息；或者
    当未确定出所述第k个目标采样点集合时，若所述目标区域信息中包括所述第k个目标区域的方向角区间，则障碍物检测结果中不包括所述第k个障碍物信息；若所述目标区域信息中不包括所述第k个目标区域的方向角区间，将所述第k个障碍物信息中包括的障碍物存在概率降低预设数值，获得更新后的第k个障碍物信息，将所述更新后的第k个障碍物信息作为障碍物检测结果，每个障碍物信息中包括障碍物的存在概率。
  21. 如权利要求19或20所述的障碍物信息处理设备，其特征在于，所述处理器还用于：
    根据所述采样点集合和所述至少一个障碍物信息分别对应的目标采样点集合，确定非目标采样点集合；
    根据所述非目标采样点集合确定至少一个障碍物的轮廓信息，作为第三检测结果；
    将所述第三检测结果作为障碍物检测结果。
  22. 如权利要求13-21任一项所述的障碍物信息处理设备，其特征在于，所述处理器还用于：
    在根据所述采样点集合和所述更新后的障碍物信息列表，获得障碍物检测结果之后，采用所述障碍物检测结果替换所述更新后的障碍物信息列表。
  23. 一种激光雷达数据采集器，其特征在于，包括:信号发送器，激光雷达传感器，激光雷达控制器，信号接收器，其中，所述信号发送器与所述激光雷达传感器连接，所述信号接收器与所述激光雷达控制器连接，所述激光雷达传感器与所述激光雷达控制器连接；
    所述信号接收器，用于接收障碍物信息处理设备发送的目标区域信息和非目标区域信息，其中，所述目标区域信息用于指示至少一个目标区域，所述非目标区域信息用于指示至少一个非目标区域，所述目标区域为包含障碍物的区域，所述非目标区域为不包含障碍物的区域，或者，所述目标区域为包含满足预设筛选条件的障碍物的区域，所述非目标区域为不包含满足所述预设筛选条件的障碍物的区域；
    所述激光雷达控制器，用于根据所述目标区域信息和所述非目标区域信息控制所述激光雷达传感器执行数据采集；
    所述激光雷达传感器，用于获得采样点集合，所述采样点集合包括在所述至少一个目标区域内采集的采样点数和在所述至少一个非目标区域内采集的采样点数，且在每个目标区域的单位体积内采集的采样点数大于在每个非目标区域的单位体积内采集的采样点数；
    所述信号发送器，用于将所述采样点集合发送至所述障碍物信息处理设备。
  24. 如权利要求23所述的激光雷达数据采集器，其特征在于，所述目标区域信息还包括第一信号发射参数，所述非目标区域信息还包括第二信号发射参数，其中，所述第一信号发射参数大于所述第二信号发射参数；
    和/或，所述目标区域信息还包括第一信号接收采样参数，所述非目标区域信息还包括第二信号接收采样参数，其中，所述第一信号接收采样参数大于所述第二信号接收采样参数。
  25. 一种障碍物检测系统，其特征在于，包括如权利要求13-22所述的障碍物信息处理设备，障碍物检测数据采集器和如权利要求23-24所述的激光雷达数据采集器。
PCT/CN2017/087041 2017-06-02 2017-06-02 一种障碍物检测方法及设备 WO2018218680A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
CN201780091513.XA CN110691990B (zh) 2017-06-02 2017-06-02 一种障碍物检测方法及设备
EP17911618.1A EP3623844A4 (en) 2017-06-02 2017-06-02 METHOD AND DEVICE FOR DETECTING OBSTACLES
PCT/CN2017/087041 WO2018218680A1 (zh) 2017-06-02 2017-06-02 一种障碍物检测方法及设备
US16/700,679 US20200110173A1 (en) 2017-06-02 2019-12-02 Obstacle detection method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2017/087041 WO2018218680A1 (zh) 2017-06-02 2017-06-02 一种障碍物检测方法及设备

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US16/700,679 Continuation US20200110173A1 (en) 2017-06-02 2019-12-02 Obstacle detection method and device

Publications (1)

Publication Number Publication Date
WO2018218680A1 true WO2018218680A1 (zh) 2018-12-06

Family

ID=64454212

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/087041 WO2018218680A1 (zh) 2017-06-02 2017-06-02 一种障碍物检测方法及设备

Country Status (4)

Country Link
US (1) US20200110173A1 (zh)
EP (1) EP3623844A4 (zh)
CN (1) CN110691990B (zh)
WO (1) WO2018218680A1 (zh)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110889350A (zh) * 2019-11-18 2020-03-17 四川西南交大铁路发展股份有限公司 一种基于三维成像的线路障碍物监测报警系统及方法
CN112306061A (zh) * 2020-10-28 2021-02-02 深圳优地科技有限公司 一种机器人控制方法以及机器人
CN112764035A (zh) * 2020-12-28 2021-05-07 南京市德赛西威汽车电子有限公司 一种基于bsd雷达左右通信的虚假目标检测优化方法
CN115047876A (zh) * 2022-06-07 2022-09-13 达闼机器人股份有限公司 可移动设备避障方法、装置、设备和存储介质

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP7327257B2 (ja) 2020-04-13 2023-08-16 トヨタ自動車株式会社 車載センサシステム
CN111474560B (zh) * 2020-04-16 2023-11-24 苏州大学 一种障碍物定位方法、装置及设备
US20230156568A1 (en) * 2020-05-08 2023-05-18 Lg Electronics Inc. Method and apparatus for updating data transfer disturbance in multi-agent multi-device ml operations
CN112132929B (zh) * 2020-09-01 2024-01-26 北京布科思科技有限公司 一种基于深度视觉和单线激光雷达的栅格地图标记方法
CN113075668B (zh) * 2021-03-25 2024-03-08 广州小鹏自动驾驶科技有限公司 一种动态障碍物对象识别方法和装置
CN113552575B (zh) * 2021-07-16 2024-05-07 铁将军汽车电子股份有限公司 泊车障碍物侦测方法及装置
CN113469159B (zh) * 2021-09-06 2021-12-07 禾多科技（北京）有限公司 障碍物信息生成方法、装置、电子设备和计算机可读介质
CN116204791B (zh) * 2023-04-25 2023-08-11 山东港口渤海湾港集团有限公司 车辆行为预测场景数据集的构建和管理方法及其系统

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2012063236A (ja) * 2010-09-16 2012-03-29 Ricoh Co Ltd レーザレーダ装置
CN104111460A (zh) * 2013-04-22 2014-10-22 苏州宝时得电动工具有限公司 自动行走设备及其障碍检测方法
CN104793215A (zh) * 2014-01-17 2015-07-22 欧姆龙汽车电子株式会社 激光雷达装置、物体检测方法
CN105893931A (zh) * 2015-02-16 2016-08-24 松下知识产权经营株式会社 物体检测装置和物体检测方法
CN105911553A (zh) * 2016-04-15 2016-08-31 北京信息科技大学 一种道路可行域确定方法及系统
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
CN106291736A (zh) * 2016-08-16 2017-01-04 张家港长安大学汽车工程研究院 无人驾驶汽车车道动态障碍物检测方法
CN106767852A (zh) * 2016-12-30 2017-05-31 东软集团股份有限公司 一种生成探测目标信息的方法、装置和设备

Family Cites Families (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US3427611A (en) * 1962-08-15 1969-02-11 Litton Industries Inc Laser system
JP2523633B2 (ja) * 1987-05-13 1996-08-14 株式会社日立製作所 レ−ザレ−ダの走査方法
US6877691B2 (en) * 2002-03-12 2005-04-12 Bae Systems Information And Electronic Systems Integration Inc. High altitude stripping for threat discrimination
US7046187B2 (en) * 2004-08-06 2006-05-16 Time Domain Corporation System and method for active protection of a resource
CN102012908B (zh) * 2010-11-12 2012-11-21 浙江大学 障碍物环境中可视移动近邻的查询方法
JP5558440B2 (ja) * 2011-09-08 2014-07-23 三菱電機株式会社 物体検出装置
DK2866047T3 (da) * 2013-10-23 2021-03-29 Ladar Ltd Detekteringssystem til detektering af en genstand på en vandoverflade
US9594378B2 (en) * 2015-07-31 2017-03-14 Delphi Technologies, Inc. Variable object detection field-of-focus for automated vehicle control
US10120385B2 (en) * 2016-03-30 2018-11-06 Intel Corporation Comfort ride vehicle control system

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2012063236A (ja) * 2010-09-16 2012-03-29 Ricoh Co Ltd レーザレーダ装置
CN104111460A (zh) * 2013-04-22 2014-10-22 苏州宝时得电动工具有限公司 自动行走设备及其障碍检测方法
CN104793215A (zh) * 2014-01-17 2015-07-22 欧姆龙汽车电子株式会社 激光雷达装置、物体检测方法
CN105893931A (zh) * 2015-02-16 2016-08-24 松下知识产权经营株式会社 物体检测装置和物体检测方法
CN105911553A (zh) * 2016-04-15 2016-08-31 北京信息科技大学 一种道路可行域确定方法及系统
CN106291736A (zh) * 2016-08-16 2017-01-04 张家港长安大学汽车工程研究院 无人驾驶汽车车道动态障碍物检测方法
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
CN106767852A (zh) * 2016-12-30 2017-05-31 东软集团股份有限公司 一种生成探测目标信息的方法、装置和设备

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3623844A4 *

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110889350A (zh) * 2019-11-18 2020-03-17 四川西南交大铁路发展股份有限公司 一种基于三维成像的线路障碍物监测报警系统及方法
CN110889350B (zh) * 2019-11-18 2023-05-23 四川西南交大铁路发展股份有限公司 一种基于三维成像的线路障碍物监测报警系统及方法
CN112306061A (zh) * 2020-10-28 2021-02-02 深圳优地科技有限公司 一种机器人控制方法以及机器人
CN112764035A (zh) * 2020-12-28 2021-05-07 南京市德赛西威汽车电子有限公司 一种基于bsd雷达左右通信的虚假目标检测优化方法
CN112764035B (zh) * 2020-12-28 2024-01-30 南京市德赛西威汽车电子有限公司 一种基于bsd雷达左右通信的虚假目标检测优化方法
CN115047876A (zh) * 2022-06-07 2022-09-13 达闼机器人股份有限公司 可移动设备避障方法、装置、设备和存储介质

Also Published As

Publication number Publication date
US20200110173A1 (en) 2020-04-09
EP3623844A4 (en) 2020-05-20
CN110691990B (zh) 2021-10-26
EP3623844A1 (en) 2020-03-18
CN110691990A (zh) 2020-01-14

Similar Documents

Publication Publication Date Title
WO2018218680A1 (zh) 一种障碍物检测方法及设备
CN110268413B (zh) 低电平传感器融合
US10585409B2 (en) Vehicle localization with map-matched sensor measurements
CN111052132B (zh) 利用多个传感器的基于运动的车道检测的验证模块系统和方法
US10279809B2 (en) Travelled-route selecting apparatus and method
WO2020133217A1 (zh) 连续障碍物检测方法、设备、系统及存储介质
JP5713106B2 (ja) 車両特定システム及び車両特定装置
US20140297063A1 (en) Vehicle specifying apparatus
EP4148385A1 (en) Vehicle navigation positioning method and apparatus, and base station, system and readable storage medium
TWI604980B (zh) 載具控制系統及載具控制方法
JP2017227529A (ja) レーダ装置および追従対象決定方法
CN112286049A (zh) 运动轨迹预测方法和装置
WO2020028138A1 (en) Lidar system design to mitigate lidar crosstalk
CN110596708A (zh) 一种车辆间距检测方法、装置、车辆和存储介质
WO2021057324A1 (zh) 数据处理方法、装置、芯片系统及介质
CN113008237A (zh) 一种路径规划方法及装置、飞行器
CN113611112B (zh) 一种目标关联方法、装置、设备及存储介质
US20210063165A1 (en) Adaptive map-matching-based vehicle localization
JP2022537557A (ja) 運転可能な領域情報を決定するための方法及び装置
EP4160269A1 (en) Systems and methods for onboard analysis of sensor data for sensor fusion
CN104020468A (zh) 一种基于二次雷达的近距离目标更新周期的计算方法
CN111480165A (zh) 在考虑对象的特征结构的情况下创建用于车辆的基于特征的定位地图的方法
CN113313654B (zh) 激光点云滤波去噪方法、系统、设备及存储介质
Li et al. FARFusion: A Practical Roadside Radar-Camera Fusion System for Far-Range Perception
CN108983247B (zh) 物体目标识别系统、物体目标识别方法及存储介质

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: 17911618

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2017911618

Country of ref document: EP

Effective date: 20191209