WO2022156711A1 - 货台识别方法、装置、电子设备及计算机可读存储介质 - Google Patents

货台识别方法、装置、电子设备及计算机可读存储介质 Download PDF

Info

Publication number
WO2022156711A1
WO2022156711A1 PCT/CN2022/072754 CN2022072754W WO2022156711A1 WO 2022156711 A1 WO2022156711 A1 WO 2022156711A1 CN 2022072754 W CN2022072754 W CN 2022072754W WO 2022156711 A1 WO2022156711 A1 WO 2022156711A1
Authority
WO
WIPO (PCT)
Prior art keywords
point
cluster
measurement points
target
point cluster
Prior art date
Application number
PCT/CN2022/072754
Other languages
English (en)
French (fr)
Inventor
李大林
李毅
Original Assignee
长沙智能驾驶研究院有限公司
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 长沙智能驾驶研究院有限公司 filed Critical 长沙智能驾驶研究院有限公司
Publication of WO2022156711A1 publication Critical patent/WO2022156711A1/zh

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
    • 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
    • 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
    • G01S15/00—Systems using the reflection or reradiation of acoustic waves, e.g. sonar systems
    • G01S15/02—Systems using the reflection or reradiation of acoustic waves, e.g. sonar systems using reflection of acoustic waves
    • G01S15/06—Systems determining the position data of a target
    • 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
    • G01S15/00—Systems using the reflection or reradiation of acoustic waves, e.g. sonar systems
    • G01S15/86—Combinations of sonar systems with lidar systems; Combinations of sonar systems with systems not using wave reflection
    • 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
    • G01S17/02—Systems using the reflection of electromagnetic waves other than radio waves
    • G01S17/06—Systems determining position data of a target
    • 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
    • G01S17/86—Combinations of lidar systems with systems other than lidar, radar or sonar, e.g. with direction finders
    • 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
    • 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
    • G01S7/52—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S15/00
    • G01S7/539—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S15/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
    • G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/762—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using clustering, e.g. of similar faces in social networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
    • 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
    • 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 belongs to the technical field of automatic driving, and in particular, relates to a method, device, electronic device, and computer-readable storage medium for identifying a cargo bed.
  • Autonomous driving is usually a technology that relies on the cooperation of artificial intelligence, visual computing, radar, surveillance devices and global positioning systems to allow computers to operate motor vehicles automatically and safely without any human active operation.
  • the autonomous driving industry has developed rapidly, and autonomous driving technology has become increasingly mature.
  • Embodiments of the present application provide a cargo bed identification method, device, electronic device, and computer-readable storage medium, so as to solve the problems of relatively large errors and low accuracy in the related art when a vehicle is parked on a cargo bed.
  • an embodiment of the present application provides a method for identifying a cargo bed, including:
  • an initial measurement point sequence is obtained based on sensor collection, the initial measurement point sequence includes a plurality of measurement points, and each measurement point carries the corresponding first position information and the initial measurement point in the initial measurement point sequence.
  • the serial number information in the point sequence, the cargo bed area is an area located within the preset distance range of the cargo bed;
  • a first target point cluster matching the cargo bed is identified from the at least one point cluster.
  • an embodiment of the present application provides a pallet identification device, including:
  • the acquisition module is used to acquire an initial measurement point sequence based on sensor acquisition when the vehicle is located in the cargo bay area, the initial measurement point sequence includes a plurality of measurement points, and each measurement point carries the corresponding first position information and serial number information in the sequence of initial measurement points, the cargo bed area is an area located within a preset distance range of the cargo bed;
  • a determining module configured to determine a jump point from the plurality of measurement points according to the first position information and the sequence number information
  • a segmentation module configured to perform segmentation and clustering on the plurality of measurement points based on the jump points to obtain at least one point cluster
  • An identification module configured to identify a first target point cluster matching the cargo bed from the at least one point cluster.
  • an embodiment of the present application provides an electronic device, including: a processor and a memory storing computer program instructions;
  • an embodiment of the present application provides a computer-readable storage medium, where computer program instructions are stored on the computer-readable storage medium, and when the computer program instructions are executed by a processor, the above-mentioned method for identifying a cargo bed is implemented.
  • the method, device, electronic device, and computer-readable storage medium for cargo bed identification provided by the embodiments of the present application, when the vehicle is located in the cargo bed area, obtains the initial measurement point sequence collected by the sensor, and then selects the measurement points from the plurality of measurement points according to the measurement
  • the first position information of the point and the serial number information in the initial measurement point sequence determine the jump point, divide and cluster the plurality of measurement points into at least one point cluster based on the jump point, and finally identify the corresponding goods from the at least one point cluster.
  • the first target point cluster of the station matching when the vehicle is located in the cargo bed area, obtains the initial measurement point sequence collected by the sensor, and then selects the measurement points from the plurality of measurement points according to the measurement
  • the first position information of the point and the serial number information in the initial measurement point sequence determine the jump point, divide and cluster the plurality of measurement points into at least one point cluster based on the jump point, and finally identify the corresponding goods from the at least one point cluster.
  • the jump points determined from the multiple measurement points are used to divide and cluster the multiple measurement points into different point clusters, and then the first target point cluster matching the cargo platform is determined, so as to realize the Precise positioning, combined with the positioning system on the vehicle and the positioning of the cargo platform, reduces the error when the vehicle is parked on the cargo platform, and has high accuracy.
  • FIG. 1 is a schematic flowchart of a cargo bed identification method provided by an embodiment of the present application.
  • FIG. 2 is a schematic diagram of the principle of a cargo platform identification method in an embodiment of the present application
  • FIG. 3 is a schematic diagram of the docking relationship between a vehicle and a cargo bed in an embodiment of the present application
  • FIG. 4 is a schematic structural diagram of a cargo bed identification device provided by an embodiment of the present application.
  • FIG. 5 is a schematic structural diagram of an electronic device provided by an embodiment of the present application.
  • the embodiments of the present application provide a method, apparatus, electronic device, and computer-readable storage medium for identifying a cargo bed.
  • the cargo bed identification method provided by the embodiment of the present application is first introduced below.
  • FIG. 1 shows a schematic flowchart of a cargo bed identification method provided by an embodiment of the present application.
  • the cargo platform identification method includes:
  • the initial measurement point sequence when the vehicle is located in the cargo bay area, obtain an initial measurement point sequence based on sensor collection, the initial measurement point sequence includes a plurality of measurement points, and each measurement point carries the corresponding first position information and the initial measurement point sequence
  • the serial number information in , the cargo bed area is the area within the preset distance of the cargo bed;
  • S104 Identify a first target point cluster matching the cargo platform from at least one point cluster.
  • the selection of the sensor may be a multi-line laser radar, a single-line laser radar, or an ultrasonic radar, or the like.
  • the sensor can be installed on the vehicle, for example, in the upper, middle or lower part of the rear of the car, or it can be installed on the roof or the bottom of the vehicle.
  • the sensor determines the specific arrangement according to the location, such as downward, upward or horizontal.
  • the sensor can be installed horizontally at the rear of the vehicle, the installation position is lower than the height of the cargo bed, and located on the left and right centerlines of the vehicle, which can illuminate the facade of the cargo bed.
  • the vehicle is located in a cargo bed area
  • the cargo bed area may be an area within a preset distance range of the cargo bed.
  • the vehicle may determine whether the vehicle is located in the cargo bed area according to the indication of the high-precision map.
  • the cargo bed area may be an area within 10 m from the position where the vehicle is parked on the cargo bed, and the specific preset distance range may be determined according to the actual situation.
  • the vehicle is located in the cargo bay area, which generally may mean that the vehicle travels to the cargo bay area, or may mean that the vehicle itself is located in the cargo bay area.
  • the initial measurement point sequence is acquired based on the sensor acquisition.
  • the sensor starts to work, and a plurality of measurement points can be acquired according to the clockwise or counterclockwise ranging distance sequence, wherein clockwise or counterclockwise based on the sensor
  • clockwise or counterclockwise based on the sensor In the way of collecting the distance sequence of the hour-hand distance measurement, multiple measurement points form the initial measurement point sequence.
  • Each measurement point carries corresponding first position information and sequence number information in the initial measurement point sequence.
  • the senor is a single-line lidar
  • the single-line lidar is installed horizontally at the rear of the vehicle.
  • the single-line lidar emits laser beams according to the clockwise ranging distance sequence, and the laser beam is irradiated on the facade of the cargo bay or other obstacles to obtain measurement points, and each laser beam corresponds to a measurement point. Multiple measurement points form the initial measurement point sequence.
  • a jump point is determined from the plurality of measurement points according to the first position information and the serial number information.
  • each measurement point carries the corresponding first position information and the sequence number information in the initial sequence of measurement points.
  • the first position information corresponding to the two measurement points of adjacent sequence number information is quite different. , it can be considered that there is a transition point.
  • the first position information can be used to reflect the first distance between the vehicle and an obstacle such as a cargo bed, and among the multiple measurement points, there are three consecutive points with serial number information: the first point, the second point and the third point , when the gap between the first distances corresponding to the first point and the second point is relatively small, and the gap between the first distances corresponding to the second point and the third point is relatively large, then the Three points are determined as jump points.
  • the distinction between smaller and larger here can be based on a set threshold.
  • the jump point can be determined by comparing the first position information of the measurement points of the adjacent serial number information.
  • At least one point cluster is obtained by dividing and clustering a plurality of measurement points based on the jump points.
  • segmentation clustering is essentially to separate measurement points belonging to different obstacles, so as to obtain point clusters corresponding to the same obstacle.
  • the jump point is determined again, it is considered that the jump point and the point with the serial number information before the jump point belong to different point clusters.
  • the points before the sequence number information of the third point are divided and clustered into a point cluster.
  • At least one point cluster is obtained based on the segmentation and clustering, until the segmentation and clustering of all the measurement points is completed.
  • a first target point cluster matching the cargo bed is identified from the at least one point cluster.
  • each point cluster in the at least one point cluster can be matched with a cargo bed.
  • the size of the cargo bed is fixed, and usually has specific characteristics, such as continuous measurement points, smoothness, etc. , the length exceeds a certain value, etc., the preset conditions can be determined based on these characteristics. Then, it is determined that the point cluster that meets the preset conditions is the first target point cluster, wherein the first target point cluster corresponds to the cargo platform, and the other point clusters correspond to other obstacles.
  • the above cargo platform identification method obtains the initial measurement point sequence collected by the sensor, and then determines the jump point from the plurality of measurement points according to the first position information of the measurement point and the serial number information in the initial measurement point sequence, and then determines the jump point based on the jump point.
  • the plurality of measurement points are divided and clustered into at least one point cluster, and finally a first target point cluster matching the cargo bed is identified from the at least one point cluster. Since the jump points determined from multiple measurement points are used, the multiple measurement points are divided and clustered into different point clusters, and then the first target point cluster matching the cargo platform is determined, so as to realize the precise positioning of the cargo platform. Combined with the positioning system on the vehicle and the positioning of the cargo platform, the error when the vehicle is parked on the cargo platform is reduced, and the accuracy is high.
  • the distances between the multiple cargo platforms are similar. Therefore, in the initial measurement point sequence collected by the sensor, the multiple point clusters after the segmentation and clustering all meet the preset conditions. At this time, the corresponding There are multiple cargo platforms. At this time, the cargo platform closest to the vehicle is determined as the cargo platform that needs to be finally identified, that is, among the multiple point clusters that meet the preset conditions, the one closest to the vehicle is determined as the first target point cluster.
  • the cargo platform identification method further includes:
  • a vehicle travel path is generated, the second position information indicates the position of the cargo bed, and the vehicle travel path is used to instruct the vehicle to travel to the cargo bed.
  • the second position information indicates the position of the cargo bed, which can be obtained according to the first position information corresponding to each of the multiple measurement points included in the first target point cluster.
  • the second position information can be the first target point.
  • the horizontal center of gravity in the cluster can be the first target point.
  • the vehicle travel path is generated.
  • the second position information can be taken as the horizontal center of gravity in the first target point cluster as an example to calculate the distance from the horizontal center of gravity to the sensor, and the horizontal center of gravity. From the center of gravity to the angle of the sensor, get the distance and angle between the cargo bed and the sensor. Combined with the position information of the sensor, the distance and angle between the first target point cluster and the sensor are transformed into the distance and angle between the vehicle and the cargo bed, and the vehicle travel path is generated.
  • the vehicle travel path is used to instruct the vehicle to travel to the cargo bed.
  • the vehicle travels to the cargo bed according to the distance and angle between the vehicle and the cargo bed until the vehicle stops at the cargo bed.
  • the RANSAC algorithm is used to perform further straight line extraction calculation according to the first target point cluster, and the optimal straight line equation is calculated.
  • the distance between the sensor center and the fitted straight line of the cargo platform can be calculated as the distance from the sensor to the cargo platform, and then according to the external parameters from the sensor to the vehicle, it can be converted to the distance value between the vehicle and the cargo platform.
  • the vehicle travel path is adjusted according to the angle between the body direction and the normal direction of the cargo bed and the distance between the vehicle and the cargo bed, and the vehicle is parked at the cargo bed based on the vehicle travel path.
  • the vehicle can obtain the distance value between the vehicle and the cargo platform and the angle between the body direction and the normal direction of the cargo platform in real time, and can adjust it in time according to the data obtained in real time.
  • the sensor can be installed horizontally, and the installation is calibrated with the direction of the vehicle body, so that the direction of the coordinate axis of the sensor coordinate system is the direction of the vehicle body, and the normal direction of the cargo platform can be obtained by calculating the vertical line of the first target point cluster line.
  • the vehicle meets the parking requirements at this time, that is, the vehicle It has been accurately parked at the cargo bay.
  • the senor is a single-line lidar.
  • the senor is a single-line lidar, which has the advantages of high accuracy, no blind spots, and a lower cost of single-line lidar. Therefore, in this example, a single-line lidar is used to collect the initial measurement point sequence, which further reduces the error, increases the accuracy, and effectively reduces the cost.
  • the single-line lidar collects the initial measurement point sequence, which can be triggered when a certain instruction is obtained.
  • the lidar sends the acquisition command, and the single-line lidar responds to the acquisition command and performs the acquisition according to the clockwise or counterclockwise ranging distance sequence. It can also be that when it is recognized that the vehicle is located in the cargo area, the acquisition instruction is automatically triggered, and the single-line lidar is instructed to collect.
  • a first target point cluster matching a preset attribute of the cargo bed is identified from at least one point cluster, and the preset attribute includes at least one of length, horizontal center of gravity, and maximum residual error.
  • the preset attributes include at least one of length, horizontal center of gravity, and maximum residual.
  • the distance D(p 0 , pn ) between the first point p 0 and the end point pn in the point cluster C k is calculated, and the distance D( p 0 , pn ) between the first and last points of each point cluster C k ) is the length.
  • the center of gravity of the measurement points included in the point cluster C k is the horizontal center of gravity.
  • the maximum residual is the maximum distance from the midpoint of the point cluster C k to the line segment composed of the first and last points.
  • the first target point cluster that matches the preset attribute of the pallet is identified from at least one point cluster.
  • the point cluster satisfies a preset length attribute, that is, D(p 0 , pn ) is greater than the preset length d
  • the point cluster is identified as the first target point cluster.
  • the size of the cargo bed is fixed, and the preset length d can be set according to the actual width of the cargo bed. d is set to 90% of the actual cargo bed width, ie, 1.6m.
  • D(p 0 , pn ) is greater than 1.6m
  • the point cluster is the first target point cluster.
  • the point cluster satisfies the preset horizontal center of gravity attribute, that is, the distance from the horizontal center of gravity to the vehicle body direction line passing through the center point of the sensor is less than the threshold dh , the point cluster is identified as the first target point cluster.
  • the threshold d h is generally small, such as 10cm, that is, when the distance from the horizontal center of gravity to the body direction line passing through the sensor center point is less than 10cm, the point cluster is the first target. point cluster.
  • the point cluster satisfies the preset maximum residual attribute, that is, the maximum distance rd k from the midpoint of the point cluster C k to the line segment composed of the first and last points is less than the preset residual value rd thres , the point cluster is identified as the first A target point cluster.
  • the preset residual value rd thres may be 5 cm, and when rd k is less than 5 cm, the point cluster is the first target point cluster.
  • the point cluster when the point cluster satisfies any two different attributes described above, the point cluster is identified as the first target point cluster. For example, when the point cluster satisfies that D(p 0 , pn ) is greater than 1.6m and rd k is less than 5cm, the point cluster is the first target point cluster.
  • the point cluster when the point cluster satisfies all the above attributes at the same time, the point cluster is identified as the first target point cluster. For example, when the point cluster satisfies that D(p 0 , p n ) is greater than 1.6m at the same time, the distance from the horizontal center of gravity to the body direction straight line passing through the sensor center point is less than 10cm and rd k is less than 5cm, the point cluster is the first target point cluster .
  • the position and attributes of the cargo bay can be used as the conditions for the vehicle to perform the docking operation on the cargo bay, and further steps can be made accordingly. Do the calculation of the docking relationship.
  • the point cluster is not the first target point cluster, it indicates other obstacles.
  • obstacle information such as position, size, angle, etc.
  • Control processes such as obstacle avoidance.
  • the cargo bed identification method further includes:
  • a sequence of target measurement points is determined from the sequence of initial measurement points, the sequence of target measurement points includes a plurality of target measurement points, and the target measurement points are measurement points within the preset sensing range;
  • the jump point is determined from the plurality of measurement points according to the first position information and the serial number information, including:
  • the jump point is determined from the plurality of target measurement points according to the first position information and the serial number information.
  • a preset perception range associated with the location information of the sensor is determined. For example, some invalid data are discarded according to the sensor installation position and preset sensing range requirements.
  • the field of view of some single-line lidars is 270°.
  • the preset sensing range is 180°, and the other 90° data will actually be affected. The vehicle itself is obscured.
  • the target measurement point sequence is determined from the initial measurement point sequence based on the preset sensing range.
  • the field of view of the single-line lidar can be 270°
  • the initial measurement point sequence collected by the single-line lidar includes all the data measured by the single-line lidar within the sensing range of 270°, naturally including the data measured within the sensing range of 90°. data on the vehicle itself.
  • the data of the vehicle itself is eliminated from the initial measurement point sequence, and the target measurement point sequence within the preset perception range is determined, and the target measurement point sequence includes a plurality of target measurement points within the preset perception range.
  • the jump point is determined from multiple target measurement points, which eliminates the interference of the data of the vehicle itself, effectively saves computing resources, and improves the efficiency of data processing.
  • the jump point is determined from the plurality of measurement points according to the first position information and the serial number information, including:
  • the third point is determined as the jumping point.
  • the first point, the second point and the third point whose serial number information is consecutive and sequentially arranged are selected from the multiple measurement points, wherein the serial number information of the first point is p i-2 , and the serial number of the second point is The information is p i-1 , and the serial number information of the third point is p i .
  • the raw output data of the single-line lidar can be a clockwise or counter-clockwise ranging distance sequence
  • the sensor coordinate system can be established with the position of the single-line lidar as the origin, and calculated according to the distance and the corresponding angle value.
  • the data of single-line lidar is a fixed amount, ranging information in a certain order, and the angular resolution is also fixed, and the angle corresponding to each data can be calculated to calculate the coordinates.
  • the distance of the laser scanning line, and its scanning angle resolution is generally a fixed and small value. Therefore, it is possible to calculate three consecutive points. The distance difference between two adjacent points is converted into a comparison of the difference in scan line length between two consecutive points.
  • the distance difference between the second point p i-1 and the sensor and the distance between the third point p i and the sensor L(p i-1 , p i ) is calculated, and then compared with the first threshold T(p i-1 , p i ) Compare the magnitudes to determine whether the third point p i is a transition point. If L(p i-1 , p i )>T(p i-1 , p i ), then the third point p i can be a jump point to generate a new point cluster, otherwise the third point p i and the second point p i Point p i-1 belongs to the same point cluster.
  • the calculation formula of the first threshold T(pi -1 , pi ) is:
  • l i-1 is the scan line length of the second point p i-1
  • l i is the scan line length of the third point p i
  • ⁇ 0 is the angular resolution of the sensor
  • ⁇ ⁇ (l i-1 )+ ⁇ (l i ) also represents noise with radial error terms.
  • the third point p i can be a jump point, and a new point cluster is generated; otherwise, the third point p i and the second point p i-1 belong to the same point cluster.
  • the calculation formula of the included angle ⁇ is:
  • k 1 is the slope of the connection line P i P i-1
  • k 2 is the slope of the connection line P i-1 P i-2
  • the value range of ⁇ t can be determined according to the actual situation. Generally speaking, ⁇ t can be about 90°, for example, the value range of ⁇ t is 80-100°.
  • the difference between the distance between the second point and the sensor and the distance between the third point and the sensor is greater than a preset first threshold, and the difference between the first point and the third
  • the third point is determined as the jumping point. That is, when L(pi -1 , pi )> T (pi -1 , pi ) and the included angle ⁇ > ⁇ t, the third point pi is a jump point, and a new point cluster is generated; otherwise, the third point The three points p i and the second point p i-1 belong to the same point cluster.
  • the jump point can be judged by the distance and the angle.
  • the distance In the process of dividing the cluster, not only the distance is considered, but also the angle factor is added, which can divide the cluster and identify the shipping station more accurately and effectively.
  • identifying the first target point cluster matching the cargo bed from at least one point cluster including:
  • At least one preliminary selection point cluster is determined from at least one point cluster, and the number of measurement points in the preliminary selection point cluster satisfies the preset number condition;
  • a first target point cluster that matches the pallet is identified from the at least one preliminary point cluster.
  • At least one preliminary selection point cluster is determined from at least one point cluster, and the number of measurement points in the preliminary selection point cluster satisfies the preset number condition .
  • the potential objects corresponding to the point clusters may be very small, so the number of measurement points contained in some point clusters may be very large.
  • point clusters whose number of measurement points does not meet the preset number condition can be eliminated.
  • point clusters that cannot meet the minimum number of points, such as at least three measurement points will be eliminated.
  • the remaining point clusters are extracted, and the first target point cluster matching the cargo platform is identified, which can effectively improve the efficiency of cargo platform identification.
  • the cargo platform identification method further includes:
  • a second target point cluster matching the rubber block is identified from the at least one point cluster.
  • the data detected by the sensor can be used to identify the rubber block, and then the distance from the center of the sensor to the rubber block can be estimated.
  • the positional relationship between the vehicle and the center of the cargo platform if the difference between the distance from the center of the sensor to the rubber blocks on both sides is less than the second threshold, if the difference between the distance from the center of the sensor to the rubber blocks on both sides is less than 5cm, it is considered that the vehicle is parked on the cargo platform accurately , otherwise, it is considered that the vehicle is parked on the cargo platform incorrectly.
  • identifying a second target point cluster matching the rubber block from at least one point cluster may specifically be determined by determining a point cluster that satisfies a preset condition with the first target point cluster as the second target point cluster.
  • the second target point cluster and the first target point cluster can satisfy the following three conditions:
  • the distance from the cargo platform is very close: the minimum distance from the measurement point in the first target point cluster is less than the preset third threshold;
  • the angle with the cargo platform it is basically perpendicular to the first target point cluster, the angle is around 90 degrees, and the setting range is 80-100 degrees;
  • the distance from the center of the cargo platform: the horizontal center of gravity of the second target point cluster and the horizontal center of gravity of the first target point cluster are less than the fourth threshold.
  • the position of the rubber blocks on both sides of the cargo platform is determined to ensure that the vehicle is in the center of the cargo platform when it is parked, which further ensures the accuracy of the vehicle parking on the cargo platform.
  • FIG. 2 is a schematic diagram of the principle of the cargo platform identification method in the embodiment of the present application, wherein the sensor is a single-line laser radar.
  • the single-line laser radar data is the initial measurement point sequence collected by the single-line laser radar, the initial measurement point sequence is preprocessed, and the data of the vehicle itself is eliminated according to the installation position and preset sensing range of the single-line laser radar. , retain the measurement data within the preset sensing range, that is, retain the target measurement point sequence. Then, a jump point is determined according to the distance and angular relationship between the multiple target measurement points in the target measurement point sequence, and the multiple target measurement points are divided and clustered based on the jump point to obtain at least one point cluster.
  • the point clusters are all two-dimensional point clusters, and then feature extraction is performed on these two-dimensional point clusters.
  • Feature extraction mainly extracts the length, horizontal center of gravity and maximum residual error of the point cluster.
  • FIG. 3 is a schematic diagram of the parking relationship between the vehicle and the cargo platform in the embodiment of the present application.
  • the cargo is obtained according to multiple measurement points of the first target point cluster.
  • the platform is fitted with a straight line, and the normal line of the fitted straight line of the cargo platform can be obtained through the horizontal center of gravity of the first target point cluster.
  • the distance between the vehicle and the cargo platform can be calculated by the distance between the sensor and the fitted straight line of the cargo platform, and the angle between the straight line in the direction of the body passing through the center point of the sensor and the normal line of the fitted straight line of the cargo platform can reflect the distance between the vehicle and the cargo platform. angle.
  • the calculation result of the parking relationship can be used as a criterion for judging whether the parking position of the vehicle is in place, and can also be directly used for the real-time feedback results of the vehicle reversing and docking with the cargo platform, and the vehicle can be controlled accordingly to further ensure the accuracy of the vehicle parking on the cargo platform.
  • FIG. 4 shows a schematic structural diagram of a cargo bed identification device provided by an embodiment of the present application. For convenience of description, only parts related to the embodiment of the present application are shown.
  • the cargo bed identification device includes:
  • the acquisition module 401 is configured to acquire an initial measurement point sequence based on sensor collection when the vehicle is located in the cargo bay area, the initial measurement point sequence includes a plurality of measurement points, and each measurement point carries the corresponding first position information and The serial number information in the initial measurement point sequence, the cargo bed area is the area within the preset distance range of the cargo bed;
  • a determination module 402 configured to determine a jump point from a plurality of measurement points according to the first position information and the serial number information;
  • a segmentation module 403, configured to perform segmentation and clustering on a plurality of measurement points based on the jump points to obtain at least one point cluster;
  • the identification module 404 is configured to identify the first target point cluster matching the cargo pallet from the at least one point cluster.
  • the above-mentioned cargo bed identification device further includes:
  • a location acquisition module used to determine a preset sensing range associated with the location information of the sensor
  • a screening module configured to determine a sequence of target measurement points from the sequence of initial measurement points based on a preset sensing range, where the sequence of target measurement points includes a plurality of target measurement points, and the target measurement points are measurement points within the preset sensing range;
  • the above determination module 402 can be specifically used for:
  • the jump point is determined from the plurality of target measurement points according to the first position information and the serial number information.
  • the above determination module 402 includes:
  • the selection unit is used to select the first point, the second point and the third point whose serial number information is continuous and sequentially arranged from the plurality of measurement points;
  • the calculation unit is configured to calculate, based on the first position information and the position information of the sensor, the difference between the distance between the second point and the sensor and the distance between the third point and the sensor, and the connection between the first point and the second point. the angle between the line and the line connecting the second and third points;
  • a determination unit configured to determine the third point as a jump point when the difference value is greater than the preset first threshold value, and/or the included angle is greater than the preset angle value.
  • the above-mentioned identification module 404 can be specifically used for:
  • a first target point cluster matching a preset attribute of the cargo bed is identified from at least one point cluster, and the preset attribute includes at least one of length, horizontal center of gravity, and maximum residual error.
  • the above-mentioned identification module 404 includes:
  • a primary selection unit configured to determine at least one primary selection point cluster from at least one point cluster, and the number of measurement points in the primary selection point cluster satisfies a preset number condition
  • the identification unit is used for identifying the first target point cluster matching the cargo platform from the at least one preliminary selection point cluster.
  • the above-mentioned cargo platform identification device further includes:
  • Rubber block identification module used to identify the second target point cluster matching the rubber block from at least one point cluster.
  • the above-mentioned cargo bed identification device further includes:
  • the parking module is configured to generate a vehicle travel path according to the second position information and the position information of the sensor, the second position information indicates the position of the cargo bed, and the vehicle travel path is used to instruct the vehicle to travel to the cargo bed.
  • Module completion means dividing the internal structure of the device into different functional units or modules to complete all or part of the functions described above.
  • Each functional unit and module in the embodiment may be integrated in one processing unit, or each unit may exist physically alone, or two or more units may be integrated in one unit, and the above-mentioned integrated units may adopt hardware. It can also be realized in the form of software functional units.
  • the specific names of the functional units and modules are only for the convenience of distinguishing from each other, and are not used to limit the protection scope of the present application. For the specific working processes of the units and modules in the above-mentioned system, reference may be made to the corresponding processes in the foregoing method embodiments, which will not be repeated here.
  • FIG. 5 shows a schematic diagram of a hardware structure of an electronic device provided by an embodiment of the present application.
  • the electronic device may include a processor 501 and a memory 502 storing computer program instructions.
  • the computer program may be divided into one or more modules/units, and the one or more modules/units are stored in the memory 502 and executed by the processor 501 to complete the present application.
  • One or more modules/units may be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program in the electronic device.
  • the above-mentioned processor 501 may include a central processing unit (CPU), or a specific integrated circuit (Application Specific Integrated Circuit, ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.
  • CPU central processing unit
  • ASIC Application Specific Integrated Circuit
  • Memory 502 may include mass storage for data or instructions.
  • memory 502 may include a Hard Disk Drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a Universal Serial Bus (USB) drive or two or more A combination of more than one of the above.
  • Memory 502 may include removable or non-removable (or fixed) media, where appropriate.
  • Storage 502 may be internal or external to the integrated gateway disaster recovery device, where appropriate.
  • memory 502 is non-volatile solid state memory.
  • Memory may include read only memory (ROM), random access memory (RAM), magnetic disk computer readable storage media devices, optical computer readable storage media devices, flash memory devices, electrical, optical or other physical/tangible memory storage devices.
  • ROM read only memory
  • RAM random access memory
  • the processor 501 reads and executes the computer program instructions stored in the memory 502 to implement any one of the pallet identification methods in the foregoing embodiments.
  • the electronic device may also include a communication interface 503 and a bus 504 .
  • the processor 501, the memory 502, and the communication interface 503 are connected through the bus 504 and complete the mutual communication.
  • the communication interface 503 is mainly used to implement communication between modules, apparatuses, units and/or devices in the embodiments of the present application.
  • the bus 504 includes hardware, software, or both, coupling the components of the online data flow metering device to each other.
  • the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, Enhanced Industry Standard Architecture (EISA) bus, Front Side Bus (FSB), HyperTransport (HT) Interconnect, Industry Standard Architecture (ISA) Bus, Infiniband Interconnect, Low Pin Count (LPC) Bus, Memory Bus, Microchannel Architecture (MCA) Bus, Peripheral Component Interconnect (PCI) Bus, PCI-Express (PCI-X) Bus, Serial Advanced Technology Attachment (SATA) bus, Video Electronics Standards Association Local (VLB) bus or other suitable bus or a combination of two or more of the above.
  • Bus 504 may include one or more buses, where appropriate. Although embodiments of this application describe and illustrate a particular bus, this application contemplates any suitable bus or interconnect.
  • the embodiments of the present application may provide a computer-readable storage medium for implementation.
  • Computer program instructions are stored on the computer-readable storage medium; when the computer program instructions are executed by the processor, any one of the cargo pallet identification methods in the foregoing embodiments is implemented.
  • the functional blocks shown in the above-described structural block diagrams may be implemented as hardware, software, firmware, or a combination thereof.
  • it When implemented in hardware, it may be, for example, an electronic circuit, an application specific integrated circuit (ASIC), suitable firmware, a plug-in, a function card, or the like.
  • ASIC application specific integrated circuit
  • elements of the present application are programs or code segments used to perform the required tasks.
  • the program or code segments may be stored in a machine-readable medium or transmitted over a transmission medium or communication link by a data signal carried in a carrier wave.
  • a "machine-readable medium” may include any medium that can store or transmit information.
  • machine-readable media examples include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, and the like.
  • the code segments may be downloaded via a computer grid such as the Internet, an intranet, or the like.
  • processors may be, but are not limited to, general purpose processors, special purpose processors, application specific processors, or field programmable logic circuits. It will also be understood that each block of the block diagrams and/or flowchart illustrations, and combinations of blocks in the block diagrams and/or flowchart illustrations, can also be implemented by special purpose hardware for performing the specified functions or actions, or by special purpose hardware and/or A combination of computer instructions is implemented.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Remote Sensing (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Multimedia (AREA)
  • Electromagnetism (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Computing Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Acoustics & Sound (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • General Engineering & Computer Science (AREA)
  • Evolutionary Biology (AREA)
  • Traffic Control Systems (AREA)
  • Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
  • Warehouses Or Storage Devices (AREA)

Abstract

一种货台识别方法、装置、电子设备及计算机可读存储介质。其中，货台识别方法包括：在车辆位于货台区域的情况下，获取基于传感器采集得到初始测量点序列，初始测量点序列包括多个测量点，且每一测量点携带对应的第一位置信息和在初始测量点序列中的序号信息，货台区域为位于货台预设距离范围内的区域（S101）；根据第一位置信息和序号信息从多个测量点中确定出跳变点（S102）；基于跳变点对多个测量点进行分割聚类得到至少一个点簇（S103）；从至少一个点簇中识别与货台匹配的第一目标点簇（S104）。

Description

货台识别方法、装置、电子设备及计算机可读存储介质
相关申请的交叉引用
本申请要求享有于2021年01月22日提交的中国专利申请202110089750.0的优先权，该申请的全部内容通过引用并入本文中。
技术领域
本申请属于自动驾驶技术领域，尤其涉及一种货台识别方法、装置、电子设备及计算机可读存储介质。
背景技术
自动驾驶通常是一种依靠人工智能、视觉计算、雷达、监控装置和全球定位系统协同合作，让电脑可以在没有任何人类主动的操作下，自动安全地操作机动车辆的技术。近年来，自动驾驶行业飞速发展，自动驾驶技术日趋成熟。
在相关技术中，自动驾驶汽车需要停靠货台时，主要是通过车辆上的定位系统进行定位，但是在实际应用中，仅依赖车辆上的定位系统进行停靠，仍然会存在较大误差，导致停靠的精准度低。
发明内容
本申请实施例提供一种货台识别方法、装置、电子设备及计算机可读存储介质，以解决相关技术中车辆停靠货台时存在较大误差，精准度低的问题。
第一方面，本申请实施例提供一种货台识别方法，包括：
在车辆位于货台区域的情况下，获取基于传感器采集得到初始测量点序列，所述初始测量点序列包括多个测量点，且每一测量点携带对应的第一位置信息和在所述初始测量点序列中的序号信息，所述货台区域为位于货台预设距离范围内的区域；
根据所述第一位置信息和所述序号信息从所述多个测量点中确定出跳变点；
基于所述跳变点对所述多个测量点进行分割聚类得到至少一个点簇；
从所述至少一个点簇中识别与所述货台匹配的第一目标点簇。
第二方面，本申请实施例提供了一种货台识别装置，包括：
获取模块，用于在车辆位于货台区域的情况下，获取基于传感器采集得到初始测量点序列，所述初始测量点序列包括多个测量点，且每一测量点携带对应的第一位置信息和在所述初始测量点序列中的序号信息，所述货台区域为位于货台预设距离范围内的区域；
确定模块，用于根据所述第一位置信息和所述序号信息从所述多个测量点中确定出跳变点；
分割模块，用于基于所述跳变点对所述多个测量点进行分割聚类得到至少一个点簇；
识别模块，用于从所述至少一个点簇中识别与所述货台匹配的第一目标点簇。
第三方面，本申请实施例提供了一种电子设备，包括：处理器以及存储有计算机程序指令的存储器；
所述处理器执行所述计算机程序指令时实现上述的货台识别方法。
第四方面，本申请实施例提供了一种计算机可读存储介质，所述计算机可读存储介质上存储有计算机程序指令，所述计算机程序指令被处理器执行时实现上述的货台识别方法。
本申请实施例提供的货台识别方法、装置、电子设备及计算机可读存储介质在车辆位于货台区域的情况下，通过获取传感器采集的初始测量点序列，再从多个测量点中根据测量点的第一位置信息和在初始测量点序列中的序号信息确定出跳变点，基于跳变点将多个测量点分割聚类成至少一个点簇，最后从至少一个点簇中识别与货台匹配的第一目标点簇。本申请实施例利用从多个测量点中确定的跳变点，将多个测量点分割聚类成不同的点簇，再确定出与货台匹配的第一目标点簇，从而实现货台的精准定位，结合车辆上的定位系统和对货台的定位进行停靠，使得减少车辆停靠货台 时的误差，精准度高。
附图说明
为了更清楚地说明本申请实施例的技术方案，下面将对本申请实施例中所需要使用的附图作简单的介绍，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1是本申请实施例提供的货台识别方法的流程示意图；
图2是本申请实施例中货台识别方法的原理示意图；
图3是本申请实施例中车辆与货台的停靠关系示意图；
图4是本申请实施例提供的货台识别装置的结构示意图；
图5是本申请实施例提供的电子设备的结构示意图。
具体实施方式
下面将详细描述本申请的各个方面的特征和示例性实施例，为了使本申请的目的、技术方案及优点更加清楚明白，以下结合附图及具体实施例，对本申请进行进一步详细描述。应理解，此处所描述的具体实施例仅意在解释本申请，而不是限定本申请。对于本领域技术人员来说，本申请可以在不需要这些具体细节中的一些细节的情况下实施。下面对实施例的描述仅仅是为了通过示出本申请的示例来提供对本申请更好的理解。
需要说明的是，在本文中，诸如第一和第二等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来，而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。而且，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者设备不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者设备所固有的要素。在没有更多限制的情况下，由语句“包括……”限定的要素，并不排除在包括所述要素的过程、方法、物品或者设备中还存在另外的相同要素。
为了解决相关技术中的技术问题，本申请实施例提供了一种货台识别 方法、装置、电子设备及计算机可读存储介质。下面首先对本申请实施例所提供的货台识别方法进行介绍。
图1示出了本申请一个实施例提供的货台识别方法的流程示意图。
如图1所示，该货台识别方法，包括：
S101，在车辆位于货台区域的情况下，获取基于传感器采集得到初始测量点序列，初始测量点序列包括多个测量点，且每一测量点携带对应的第一位置信息和在初始测量点序列中的序号信息，货台区域为位于货台预设距离范围内的区域；
S102，根据第一位置信息和序号信息从多个测量点中确定出跳变点；
S103，基于跳变点对多个测量点进行分割聚类得到至少一个点簇；
S104，从至少一个点簇中识别与货台匹配的第一目标点簇。
本申请实施例中，对于传感器的选择，可以是多线激光雷达，或者单线激光雷达，或者超声波雷达等。传感器可以安装在车辆上，例如可以安装在车尾的上部、中部或下部，或者，可以安装在车顶或车底，传感器根据位置确定具体的设置方式，例如向下、向上或水平设置，此处亦不做具体限定。当然，一般来说，为了保证测量的有效性，传感器可以水平安装在车尾，安装位置低于货台高度，且位于车辆左右中心线上，能够照射到货台立面。
在S101中，车辆位于货台区域，货台区域可以是位于货台预设距离范围内的区域。示例地，可以是车辆根据高精度地图的指示，确定车辆是否位于货台区域。例如，货台区域可以为距离车辆停靠在货台位置处的10m以内的区域，具体预设距离范围可以根据实际情况决定。其中，车辆位于货台区域，一般可以指车辆行驶至货台区域中，也可以是指车辆本身就位于该货台区域内。
当然，在一些可行的实施方式中，也可以使通过用户手动确认车辆位于货台区域，例如，当车辆的驾驶员判断车辆到达货台附近时，手动触发货台识别的执行过程。
当车辆位于货台区域时，获取基于传感器采集得到初始测量点序列，示例地，传感器开始工作，可以按照顺时针或逆时针测距距离序列采集到 多个测量点，其中基于传感器顺时针或逆时针测距距离序列采集的方式，多个测量点组成初始测量点序列。每一测量点携带对应的第一位置信息和在初始测量点序列中的序号信息。
例如，传感器为单线激光雷达，且单线激光雷达水平安装在车尾。在车辆位于货台区域的情况下，单线激光雷达按照顺时针测距距离序列发射激光束，激光束照射在货台或其他障碍物的立面上得到测量点，每一个激光束对应得到一个测量点，多个测量点组成初始测量点序列。
在S102中，根据第一位置信息和序号信息从多个测量点中确定出跳变点。示例地，每一测量点携带对应的第一位置信息和在初始测量点序列中的序号信息，一般来说，当相邻序号信息的两个测量点所对应的第一位置信息差别较大时，可以认为存在跳变点。
例如，第一位置信息可以用于反映车辆到例如货台等障碍物之间的第一距离，多个测量点中存在第一点、第二点和第三点这三个序号信息连续的点，当第一点和第二点所分别对应的第一距离之间的差距较小，而第二点和第三点所分别对应的第一距离之间的差距较大，此时可以将第三点确定为跳变点。这里较小与较大的区分，可以是基于设置的阈值进行的。换而言之，可以根据相邻序号信息的测量点的第一位置信息进行对比，来确定出跳变点。
在S103中，基于跳变点对多个测量点进行分割聚类得到至少一个点簇。示例地，分割聚类本质上就是为了将属于不同障碍物的测量点分开来，从而获得对应相同障碍物的点簇。再确定出跳变点的情况下，认为跳变点与序号信息在跳变点之前的点属于不同的点簇。
例如，根据第一位置信息和序号信息确认出第三点为跳变点后，其中在第三点序号信息之前的点分割聚类为一个点簇。对包括第三点序号信号在内的之后的点重复S102，确定出下一个跳变点，再将下一个跳变点序号信息之前的点分割聚类成另一个点簇。基于此分割聚类得到至少一个点簇，直至多个测量点全部分割聚类完成。
在S104中，从至少一个点簇中识别与货台匹配的第一目标点簇。示例地，可以将至少一个点簇中的每一点簇与货台进行匹配，对于货台来说， 一般情况下，货台的大小是固定的，通常具有特定的特征，例如测量点连续、平滑，长度超过特定值等，可以基于这些特征来确定预设条件。然后确定出符合预设条件的点簇为第一目标点簇，其中第一目标点簇对应的就是货台，其他点簇对应的则是其他障碍物。
上述货台识别方法通过获取传感器采集的初始测量点序列，再从多个测量点中根据测量点的第一位置信息和在初始测量点序列中的序号信息确定出跳变点，基于跳变点将多个测量点分割聚类成至少一个点簇，最后从至少一个点簇中识别与货台匹配的第一目标点簇。由于利用从多个测量点中确定的跳变点，将多个测量点分割聚类成不同的点簇，再确定出与货台匹配的第一目标点簇，从而实现货台的精准定位，结合车辆上的定位系统和对货台的定位进行停靠，使得减少车辆停靠货台时的误差，精准度高。
在实施例的一种可能实施方式中，多个货台距离相近，因此在传感器采集到的初始测量点序列中，分割聚类后的多个点簇皆符合预设条件，此时，对应的货台有多个，此时，将距离车辆最近的货台确定为最终需要识别的货台，即在符合预设条件的多个点簇中，确定距离车辆最近的为第一目标点簇。
可选的，在一个示例中，从至少一个点簇中识别与货台匹配的第一目标点簇之后，该货台识别方法还包括：
根据第二位置信息和传感器的位置信息，生成车辆行驶路径，第二位置信息指示货台的位置，车辆行驶路径用于指示车辆行驶至货台。
在本示例中，第二位置信息指示货台的位置，可以根据第一目标点簇所包含的多个测量点各自对应的第一位置信息得到，例如，第二位置信息可以是第一目标点簇中的水平重心。
根据第二位置信息和传感器的位置信息，生成车辆行驶路径，示例地，可以以第二位置信息为第一目标点簇中的水平重心为例，计算该水平重心到传感器的距离，以及该水平重心到传感器的角度，得到货台与传感器的距离和角度。结合传感器的位置信息，将第一目标点簇与传感器的距离和角度变换到车辆与货台之间的距离和角度，生成车辆行驶路径。
车辆行驶路径用于指示车辆行驶至货台，示例地，车辆根据车辆与货 台之间的距离和角度，按照车辆行驶路径行驶至货台，直至将车辆停靠在货台处。
在本示例的一种可能实施方式中，如图3所示，识别出第一目标点簇后，根据第一目标点簇，利用RANSAC算法进行进一步的直线提取计算，计算出最优直线方程，可计算传感器中心到货台拟合直线的距离，作为传感器到货台的距离，再根据传感器到车辆的外参，可以变换到车辆与货台之间的距离值。再根据车身方向与货台法线方向所呈的角度和车辆与货台之间的距离值调整车辆行驶路径，基于此车辆行驶路径将车辆停靠在货台处。
通常来说，车辆在停靠过程中可以实时获取到车辆与货台之间的距离值和车身方向与货台法线方向所呈的角度，并可根据实时获取的数据及时进行调整。其中，传感器可采用水平安装，安装时与车身方向进行标定，使得传感器坐标系的坐标轴方向即为车身方向，而货台法线方向可通过计算第一目标点簇直线的垂线得到。
一般情况下，对于车辆与货台之间的距离差小于设定距离阈值，同时车身方向与货台法线方向所呈的角度小于设定角度阈值时，认为此时车辆符合停靠要求，即车辆已经准确的停靠在货台处。
可选的，在一个示例中，传感器为单线激光雷达。
本示例中传感器为单线激光雷达，单线激光雷达具有精度高，无盲区的优势，且单线激光雷达成本更低。因此，本示例中采用单线激光雷达采集初始测量点序列，进一步减小了误差、精度更高且有效降低了成本。
在本示例中，单线激光雷达采集初始测量点序列，可以是在获取到某一指令的情况下触发的，例如车辆驾驶员或管理人员在车辆位于货台区域的情况下进行相关操作，向单线激光雷达发送采集指令，单线激光雷达响应采集指令按照顺时针或者逆时针测距距离序列进行采集。还可以是在识别到车辆位于货台区域时，自动触发采集指令，指示单线激光雷达进行采集。
可选的，在一个示例中，S104，从至少一个点簇中识别与货台匹配的第一目标点簇，包括：
从至少一个点簇中识别与货台的预设属性匹配的第一目标点簇，预设属性包括长度、水平重心和最大残差中的至少一项。
在本示例中，预设属性包括长度、水平重心和最大残差中的至少一项。示例地，计算点簇C k中首点p 0到末点p n之间的距离D(p 0，p n)，每一点簇C k首末点之间的距离D(p 0，p n)即为长度。点簇C k中所包含的测量点的重心即为水平重心。最大残差为点簇C k的中点到首末点组成的线段的最大距离。
本示例中，从至少一个点簇中识别与货台预设属性匹配的第一目标点簇。示例地，当点簇满足预设的长度属性，即D(p 0，p n)大于预设长度d时，将该点簇识别为第一目标点簇。例如，一般情况下，货台的大小是固定的，预设长度d可根据实际货台的宽度设置，如货台的宽度为1.8m，考虑到测量不可避免的微误差，可以将预设长度d设置为实际货台宽度的90％，即1.6m，当D(p 0，p n)大于1.6m时，该点簇为第一目标点簇。
或当点簇满足预设的水平重心属性，即水平重心到过传感器中心点的车身方向直线的距离小于阈值d h时，将该点簇识别为第一目标点簇。例如，在实际应用中，为提高车辆停靠准确度，阈值d h一般很小，如10cm，即当水平重心到过传感器中心点的车身方向直线的距离小于10cm时，该点簇为第一目标点簇。
或当点簇满足预设的最大残差属性，即点簇C k的中点到首末点组成的线段的最大距离rd k小于预设残差值rd thres时，将该点簇识别为第一目标点簇。例如，在实际应用中，预设残差值rd thres可以为5cm，当rd k小于5cm时，该点簇为第一目标点簇。
或者，当点簇满足上述任意两个不同属性的情况下，将该点簇识别为第一目标点簇。例如，当点簇满足D(p 0，p n)大于1.6m，且rd k小于5cm时，该点簇为第一目标点簇。
又或者，当点簇同时满足上述所有属性的情况下，将该点簇识别为第一目标点簇。例如，当点簇同时满足D(p 0，p n)大于1.6m，水平重心到过传感器中心点的车身方向直线的距离小于10cm且rd k小于5cm时，该点簇为第一目标点簇。
对于分类聚类结果，若识别到该点簇为第一目标点簇，说明存在货台 对象，则可将货台的位置、属性等作为车辆执行停靠货台操作的条件，并可以据此进一步进行停靠关系计算。
若该点簇不为第一目标点簇，说明为其它障碍物，在一些可行的实施方式中，可以直接输出障碍物信息，如位置、大小、角度等，这些障碍物信息可以用于车辆的避障等控制过程。
可选的，在一个示例中，获取基于传感器采集得到初始测量点序列之后，该货台识别方法还包括：
确定与传感器的位置信息关联的预设感知范围；
基于预设感知范围，从初始测量点序列中确定出目标测量点序列，目标测量点序列包括多个目标测量点，目标测量点为在预设感知范围内的测量点；
相应的，根据第一位置信息和序号信息从多个测量点中确定出跳变点，包括：
根据第一位置信息和序号信息从多个目标测量点中确定出跳变点。
在本示例中，确定与传感器的位置信息关联的预设感知范围。示例地，根据传感器安装位置和预设感知范围需求，舍弃掉部分无效数据。以单线激光雷达安装在车尾为例，部分单线激光雷达视场角为270°，当单线激光雷达安装在车尾时，此时预设感知范围为180°，另外90°的数据实际会受到车辆本身遮挡。
在本示例中，基于预设感知范围，从初始测量点序列中确定出目标测量点序列。示例地，单线激光雷达的视场角可为270°，其采集的初始测量点序列包括单线激光雷达在270°的感知范围内测量的所有数据，自然包括其在90°的感知范围内测量的车辆本身的数据。此时，从初始测量点序列中剔除车辆本身的数据，确定出预设感知范围内的目标测量点序列，目标测量点序列包括多个在预设感知范围内的目标测量点。
在本示例中，跳变点从多个目标测量点中确定，消除了车辆本身数据的干扰，有效节省了运算资源，提升了数据处理的效率。
可选的，在一个示例中，根据第一位置信息和序号信息从多个测量点中确定出跳变点，包括：
从多个测量点中选出序号信息连续且依次排列的第一点、第二点和第三点；
基于第一位置信息和传感器的位置信息，计算第二点到传感器之间的距离与第三点到传感器之间的距离的差值，以及第一点和第二点的连线与第二点和第三点的连线之间的夹角；
在差值大于预设第一阈值，和/或夹角大于预设角度值的情况下，确定第三点为跳变点。
在本示例中，从多个测量点中选出序号信息连续且依次排列的第一点、第二点和第三点，其中第一点的序号信息为p i-2，第二点的序号信息为p i-1，第三点的序号信息为p i。
以单线激光雷达为例，单线激光雷达的原始输出数据可以为顺时针或者逆时针测距距离序列，可以以单线激光雷达的位置为原点建立传感器坐标系，根据距离和对应的角度值，计算出该点相对单线激光雷达的坐标。其中，单线激光雷达的数据是固定数量，按一定顺序的测距信息，角度分辨率也是固定的，可以推算出每一个数据对应的角度，从而计算出坐标。
考虑到对于同一单线激光雷达来说，其数据中仅含有激光扫描线距离这个单一变量，且其扫描角度分辨率一般是一个固定不变的较小值，因此，可以将计算连续三个点中相邻两点间的距离差转化为比较连续两点之间扫描线长度差异。
即计算第二点p i-1到传感器之间与第三点p i到传感器之间的距离差值L(p i-1，p i)，再与第一阈值T(p i-1，p i)比较大小来判断第三点p i是否为跳变点。若L(p i-1，p i)＞T(p i-1，p i)，则第三点p i可为跳变点，生成新的点簇，否则第三点p i与第二点p i-1属于同一点簇。其中，第一阈值T(p i-1，p i)的计算公式为：
T(p i-1，p i)＝min(l i-1，l i)·2sin(Δθ/2)+ε
其中：l i-1是第二点p i-1的扫描线长度，l i是第三点p i的扫描线长度；Δ0是传感器的角度分辨率；ε＝ε(l i-1)+ε(l i)也表示具有径向误差项的噪声。
或者，计算第一点p i-2和第二点p i-1的连线P i-1P i-2与第二点p i-1和第三 点p i的连线P iP i-1之间的夹角α，再根据夹角α判断第三点p i是否为跳变点。若夹角α＞δ t，则第三点p i可为跳变点，生成新的点簇；否则第三点p i与第二点p i-1属于同一点簇。其中，夹角α的计算公式为：
Figure PCTCN2022072754-appb-000001
其中，k 1为连线P iP i-1的斜率，k 2为连线P i-1P i-2的斜率，δ t的取值范围可根据实际情况而定，一般而言，δ t可以为90°左右，例如δ t的取值范围为80-100°。
或者，在本示例的一种可能实现方式中，同时满足第二点到传感器之间的距离与第三点到传感器之间的距离的差值大于预设第一阈值，以及第一点和第二点的连线与第二点和第三点的连线之间的夹角大于预设角度值的情况下，确定第三点为跳变点。即当L(p i-1，p i)＞T(p i-1，p i)且夹角α＞δ t时，第三点p i为跳变点，生成新的点簇；否则第三点p i与第二点p i-1属于同一点簇。
本示例可以通过距离和角度两个方面来判断跳变点，分割聚类的过程中不仅考虑了距离，还加入了角度因子，能够更加准确有效地分割聚类并识别出货台。
可选的，在一个示例中，从至少一个点簇中识别与货台匹配的第一目标点簇，包括：
从至少一个点簇中确定出至少一个初选点簇，初选点簇中的测量点的数量满足预设数量条件；
从至少一个初选点簇中识别与货台匹配的第一目标点簇。
在本示例中，在识别与货台匹配的第一目标点簇之前，先从至少一个点簇中确定出至少一个初选点簇，初选点簇中的测量点的数量满足预设数量条件。示例地，在根据跳变点对多个测量点进行分割聚类得到至少一个点簇的过程中，基于点簇对应的潜在对象可能很小，因此某些点簇所包含的测量点数量可能很少，在此情况下，可以剔除掉测量点数量不满足预设数量条件的点簇。例如，为了满足特征提取的需要，一般对于无法达到最小点数要求，如至少三个测量点的点簇都将被剔除。
剔除掉无法满足预设数量条件的点簇后，在对剩余的点簇进行特征提 取，识别出与货台匹配的第一目标点簇，可以有效提升货台识别的效率。
可选的，在一个示例中，在货台配设有橡胶块的情况下，从至少一个点簇中识别与货台匹配的第一目标点簇之后，该货台识别方法还包括：
从至少一个点簇中识别与橡胶块匹配的第二目标点簇。
在一些场景中货台两侧还有对称的其它标志物，如货台两端配设有橡胶块，可利用传感器检测到的数据识别出橡胶块，再计算传感器中心到橡胶块的距离来估计车辆与货台中心的位置关系，若传感器中心到两侧橡胶块的距离的差值小于第二阈值，如传感器中心到两侧橡胶块的距离的差值小于5cm，则认为车辆停靠货台准确，否则则认为车辆停靠货台有误。
在本示例中，从至少一个点簇中识别与橡胶块匹配的第二目标点簇，具体可以通过确定满足与第一目标点簇之间预设条件的点簇为第二目标点簇。例如，第二目标点簇与第一目标点簇可满足以下三个条件：
1、与货台距离很近：与第一目标点簇中的测量点的最小距离小于预设第三阈值；
2、与货台的角度：与第一目标点簇基本垂直，角度在90度附近，设置范围如80-100°；
3、与货台中心的距离：第二目标点簇的水平重心与第一目标点簇的水平重心小于第四阈值。
本示例中通过确定货台两侧橡胶块的位置保证车辆停靠时位于货台中心，进一步保证车辆停靠货台的准确性。
如图2所示，图2是本申请实施例中货台识别方法的原理示意图，其中传感器为单线激光雷达。
在本申请实施例中，单线激光雷达数据为单线激光雷达采集到的初始测量点序列，对初始测量点序列进行预处理，根据单线激光雷达的安装位置和预设感知范围剔除掉车辆本身的数据，保留预设感知范围内的测量数据，即保留目标测量点序列。再根据目标测量点序列中多个目标测量点之间的距离和角度关系确定跳变点，基于跳变点对多个目标测量点进行分割聚类，得到至少一个点簇。其中点簇皆为二维点簇，再对这些二维点簇进行特征提取。
特征提取主要提取点簇的长度、水平重心和最大残差，当点簇提取的特征满足货台预设属性时，判断该点簇为货台，其他不满足预设属性的点簇为其他障碍物。
如图3所示，图3是本申请实施例中车辆与货台的停靠关系示意图，识别出与货台匹配的第一目标点簇后，根据第一目标点簇的多个测量点得到货台拟合直线，并且可以通过第一目标点簇的水平重心得到货台拟合直线法线。通过传感器到货台拟合直线的距离可以计算出车辆离货台的距离，而通过过传感器中心点的车身方向的直线与货台拟合直线法线的角度可以反映车辆与货台之间的角度。停靠关系的计算结果可作为车辆停靠位置是否到位的评判标准，也可直接用于车辆倒车对接货台的实时反馈结果，并依此控制车辆，进一步保证车辆停靠货台的准确度。
图4示出了本申请一个实施例提供的货台识别装置的结构示意图，为了便于说明，仅示出了与本申请实施例相关的部分。
参照图4，货台识别装置包括：
获取模块401，用于在车辆位于货台区域的情况下，获取基于传感器采集得到初始测量点序列，初始测量点序列包括多个测量点，且每一测量点携带对应的第一位置信息和在初始测量点序列中的序号信息，货台区域为位于货台预设距离范围内的区域；
确定模块402，用于根据第一位置信息和序号信息从多个测量点中确定出跳变点；
分割模块403，用于基于跳变点对多个测量点进行分割聚类得到至少一个点簇；
识别模块404，用于从至少一个点簇中识别与货台匹配的第一目标点簇。
可选的，上述货台识别装置还包括：
位置获取模块，用于确定与传感器的位置信息关联的预设感知范围；
筛选模块，用于基于预设感知范围，从初始测量点序列中确定出目标测量点序列，目标测量点序列包括多个目标测量点，目标测量点为在预设感知范围内的测量点；
相应地，上述确定模块402，可具体用于：
根据第一位置信息和序号信息从多个目标测量点中确定出跳变点。
可选的，上述确定模块402包括：
选择单元，用于从多个测量点中选出序号信息连续且依次排列的第一点、第二点和第三点；
计算单元，用于基于第一位置信息和传感器的位置信息，计算第二点到传感器之间的距离与第三点到传感器之间的距离的差值，以及第一点和第二点的连线与第二点和第三点的连线之间的夹角；
确定单元，用于在差值大于预设第一阈值，和/或夹角大于预设角度值的情况下，确定第三点为跳变点。
可选的，上述识别模块404，可具体用于：
从至少一个点簇中识别与货台的预设属性匹配的第一目标点簇，预设属性包括长度、水平重心和最大残差中的至少一项。
可选的，上述识别模块404，包括：
初选单元，用于从至少一个点簇中确定出至少一个初选点簇，初选点簇中的测量点的数量满足预设数量条件；
识别单元，用于从至少一个初选点簇中识别与货台匹配的第一目标点簇。
可选的，在货台配设有橡胶块的情况下，上述货台识别装置还包括：
橡胶块识别模块：用于从至少一个点簇中识别与橡胶块匹配的第二目标点簇。
可选的，上述货台识别装置还包括：
停靠模块，用于根据第二位置信息和传感器的位置信息，生成车辆行驶路径，第二位置信息指示货台的位置，车辆行驶路径用于指示车辆行驶至货台。
需要说明的是，上述装置/单元之间的信息交互、执行过程等内容，与本申请方法实施例基于同一构思，是与上述货台识别方法对应的装置，上述方法实施例中所有实现方式均适用于该装置的实施例中，其具体功能及带来的技术效果，具体可参见方法实施例部分，此处不再赘述。
所属领域的技术人员可以清楚地了解到，为了描述的方便和简洁，仅以上述各功能单元、模块的划分进行举例说明，实际应用中，可以根据需要而将上述功能分配由不同的功能单元、模块完成，即将装置的内部结构划分成不同的功能单元或模块，以完成以上描述的全部或者部分功能。实施例中的各功能单元、模块可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中，上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。另外，各功能单元、模块的具体名称也只是为了便于相互区分，并不用于限制本申请的保护范围。上述系统中单元、模块的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
图5示出了本申请实施例提供的电子设备的硬件结构示意图。
在电子设备可以包括处理器501以及存储有计算机程序指令的存储器502。
处理器501执行计算机程序时实现上述任意各个方法实施例中的步骤。
示例性的，计算机程序可以被分割成一个或多个模块/单元，一个或者多个模块/单元被存储在存储器502中，并由处理器501执行，以完成本申请。一个或多个模块/单元可以是能够完成特定功能的一系列计算机程序指令段，该指令段用于描述计算机程序在电子设备中的执行过程。
具体地，上述处理器501可以包括中央处理器(CPU)，或者特定集成电路(Application Specific Integrated Circuit，ASIC)，或者可以被配置成实施本申请实施例的一个或多个集成电路。
存储器502可以包括用于数据或指令的大容量存储器。举例来说而非限制，存储器502可包括硬盘驱动器(Hard Disk Drive，HDD)、软盘驱动器、闪存、光盘、磁光盘、磁带或通用串行总线(Universal Serial Bus，USB)驱动器或者两个或更多个以上这些的组合。在合适的情况下，存储器502可包括可移除或不可移除(或固定)的介质。在合适的情况下，存储器502可在综合网关容灾设备的内部或外部。在特定实施例中，存储器502是非易失性固态存储器。
存储器可包括只读存储器(ROM)，随机存取存储器(RAM)，磁盘计算机可读存储介质设备，光计算机可读存储介质设备，闪存设备，电气、光学或其他物理/有形的存储器存储设备。因此，通常，存储器包括一个或多个编码有包括计算机可执行指令的软件的有形(非暂态)计算机可读计算机可读存储介质(例如，存储器设备)，并且当该软件被执行(例如，由一个或多个处理器)时，其可操作来执行参考根据本公开的一方面的方法所描述的操作。
处理器501通过读取并执行存储器502中存储的计算机程序指令，以实现上述实施例中的任意一种货台识别方法。
在一个示例中，电子设备还可包括通信接口503和总线504。其中，处理器501、存储器502、通信接口503通过总线504连接并完成相互间的通信。
通信接口503，主要用于实现本申请实施例中各模块、装置、单元和/或设备之间的通信。
总线504包括硬件、软件或两者，将在线数据流量计费设备的部件彼此耦接在一起。举例来说而非限制，总线可包括加速图形端口(AGP)或其他图形总线、增强工业标准架构(EISA)总线、前端总线(FSB)、超传输(HT)互连、工业标准架构(ISA)总线、无限带宽互连、低引脚数(LPC)总线、存储器总线、微信道架构(MCA)总线、外围组件互连(PCI)总线、PCI-Express(PCI-X)总线、串行高级技术附件(SATA)总线、视频电子标准协会局部(VLB)总线或其他合适的总线或者两个或更多个以上这些的组合。在合适的情况下，总线504可包括一个或多个总线。尽管本申请实施例描述和示出了特定的总线，但本申请考虑任何合适的总线或互连。
另外，结合上述实施例中的货台识别方法，本申请实施例可提供一种计算机可读存储介质来实现。该计算机可读存储介质上存储有计算机程序指令；该计算机程序指令被处理器执行时实现上述实施例中的任意一种货台识别方法。
需要明确的是，本申请并不局限于上文所描述并在图中示出的特定配 置和处理。为了简明起见，这里省略了对已知方法的详细描述。在上述实施例中，描述和示出了若干具体的步骤作为示例。但是，本申请的方法过程并不限于所描述和示出的具体步骤，本领域的技术人员可以在领会本申请的精神后，作出各种改变、修改和添加，或者改变步骤之间的顺序。
以上所述的结构框图中所示的功能块可以实现为硬件、软件、固件或者它们的组合。当以硬件方式实现时，其可以例如是电子电路、专用集成电路(ASIC)、适当的固件、插件、功能卡等等。当以软件方式实现时，本申请的元素是被用于执行所需任务的程序或者代码段。程序或者代码段可以存储在机器可读介质中，或者通过载波中携带的数据信号在传输介质或者通信链路上传送。“机器可读介质”可以包括能够存储或传输信息的任何介质。机器可读介质的例子包括电子电路、半导体存储器设备、ROM、闪存、可擦除ROM(EROM)、软盘、CD-ROM、光盘、硬盘、光纤介质、射频(RF)链路，等等。代码段可以经由诸如因特网、内联网等的计算机网格被下载。
还需要说明的是，本申请中提及的示例性实施例，基于一系列的步骤或者装置描述一些方法或系统。但是，本申请不局限于上述步骤的顺序，也就是说，可以按照实施例中提及的顺序执行步骤，也可以不同于实施例中的顺序，或者若干步骤同时执行。
上面参考根据本公开的实施例的方法、装置(系统)和计算机程序产品的流程图和/或框图描述了本公开的各方面。应当理解，流程图和/或框图中的每个方框以及流程图和/或框图中各方框的组合可以由计算机程序指令实现。这些计算机程序指令可被提供给通用计算机、专用计算机、或其它可编程数据处理装置的处理器，以产生一种机器，使得经由计算机或其它可编程数据处理装置的处理器执行的这些指令使能对流程图和/或框图的一个或多个方框中指定的功能/动作的实现。这种处理器可以是但不限于是通用处理器、专用处理器、特殊应用处理器或者现场可编程逻辑电路。还可理解，框图和/或流程图中的每个方框以及框图和/或流程图中的方框的组合，也可以由执行指定的功能或动作的专用硬件来实现，或可由专用硬件和计算机指令的组合来实现。
以上所述，仅为本申请的具体实施方式，所属领域的技术人员可以清楚地了解到，为了描述的方便和简洁，上述描述的系统、模块和单元的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。应理解，本申请的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本申请揭露的技术范围内，可轻易想到各种等效的修改或替换，这些修改或替换都应涵盖在本申请的保护范围之内。

Claims (18)

  1. 一种货台识别方法，包括：
    在车辆位于货台区域的情况下，获取基于传感器采集得到初始测量点序列，所述初始测量点序列包括多个测量点，且每一测量点携带对应的第一位置信息和在所述初始测量点序列中的序号信息，所述货台区域为位于货台预设距离范围内的区域；
    根据所述第一位置信息和所述序号信息从所述多个测量点中确定出跳变点；
    基于所述跳变点对所述多个测量点进行分割聚类得到至少一个点簇；
    从所述至少一个点簇中识别与所述货台匹配的第一目标点簇。
  2. 根据权利要求1所述的方法，其中，
    所述获取基于传感器采集得到初始测量点序列之后，所述方法还包括：
    确定与所述传感器的位置信息关联的预设感知范围；
    基于所述预设感知范围，从所述初始测量点序列中确定出目标测量点序列，所述目标测量点序列包括多个目标测量点，所述目标测量点为在所述预设感知范围内的测量点；
    所述根据所述第一位置信息和所述序号信息从所述多个测量点中确定出跳变点，包括：
    根据所述第一位置信息和所述序号信息从所述多个目标测量点中确定出跳变点。
  3. 根据权利要求1或2所述的方法，其中，
    所述根据所述第一位置信息和所述序号信息从所述多个测量点中确定出跳变点，包括：
    从所述多个测量点中选出所述序号信息连续且依次排列的第一点、第二点和第三点；
    基于所述第一位置信息和所述传感器的位置信息，计算所述第二点到所述传感器之间的距离与所述第三点到所述传感器之间的距离的差值，以及所述第一点和所述第二点的连线与所述第二点和所述第三点的连线之间的夹角；
    在所述差值大于预设第一阈值，和/或所述夹角大于预设角度值的情况下，确定所述第三点为跳变点。
  4. 根据权利要求1所述的方法，其中，
    所述从所述至少一个点簇中识别与所述货台匹配的第一目标点簇，包括：
    从所述至少一个点簇中识别与所述货台的预设属性匹配的第一目标点簇，所述预设属性包括长度、水平重心和最大残差中的至少一项。
  5. 根据权利要求1所述的方法，其中，
    所述从所述至少一个点簇中识别与所述货台匹配的第一目标点簇，包括：
    从所述至少一个点簇中确定出至少一个初选点簇，所述初选点簇中的测量点的数量满足预设数量条件；
    从所述至少一个初选点簇中识别与所述货台匹配的第一目标点簇。
  6. 根据权利要求1所述的方法，其中，
    在所述货台配设有橡胶块的情况下，所述从所述至少一个点簇中识别与所述货台匹配的第一目标点簇之后，所述方法还包括：
    从所述至少一个点簇中识别与所述橡胶块匹配的第二目标点簇。
  7. 根据权利要求1所述的方法，其中，
    所述从所述至少一个点簇中识别与所述货台匹配的第一目标点簇之后，所述方法还包括：
    根据第二位置信息和所述传感器的位置信息，生成车辆行驶路径，所述第二位置信息指示所述货台的位置，所述车辆行驶路径用于指示车辆行驶至所述货台。
  8. 根据权利要求1所述的方法，其中，
    所述传感器为单线激光雷达。
  9. 一种货台识别装置，包括：
    获取模块，用于在车辆位于货台区域的情况下，获取基于传感器采集得到初始测量点序列，所述初始测量点序列包括多个测量点，且每一测量点携带对应的第一位置信息和在所述初始测量点序列中的序号信息，所述 货台区域为位于货台预设距离范围内的区域；
    确定模块，用于根据所述第一位置信息和所述序号信息从所述多个测量点中确定出跳变点；
    分割模块，用于基于所述跳变点对所述多个测量点进行分割聚类得到至少一个点簇；
    识别模块，用于从所述至少一个点簇中识别与所述货台匹配的第一目标点簇。
  10. 根据权利要求9所述的装置，还包括：
    位置获取模块，用于确定与所述传感器的位置信息关联的预设感知范围；
    筛选模块，用于基于所述预设感知范围，从所述初始测量点序列中确定出目标测量点序列，所述目标测量点序列包括多个目标测量点，所述目标测量点为在所述预设感知范围内的测量点，
    其中，所述确定模块还用于：
    根据所述第一位置信息和所述序号信息从所述多个目标测量点中确定出跳变点。
  11. 根据权利要求9或10所述的装置，其中，所述确定模块包括：
    选择单元，用于从所述多个测量点中选出所述序号信息连续且依次排列的第一点、第二点和第三点；
    计算单元，用于基于所述第一位置信息和所述传感器的位置信息，计算所述第二点到所述传感器之间的距离与所述第三点到所述传感器之间的距离的差值，以及所述第一点和所述第二点的连线与所述第二点和所述第三点的连线之间的夹角；
    确定单元，用于在所述差值大于预设第一阈值，和/或所述夹角大于预设角度值的情况下，确定所述第三点为跳变点。
  12. 根据权利要求9所述的装置，其中，所述识别模块还用于：
    从所述至少一个点簇中识别与所述货台的预设属性匹配的第一目标点簇，所述预设属性包括长度、水平重心和最大残差中的至少一项。
  13. 根据权利要求9所述的装置，其中，所述识别模块包括：
    初选单元，用于从所述至少一个点簇中确定出至少一个初选点簇，所述初选点簇中的测量点的数量满足预设数量条件；
    识别单元，用于从所述至少一个初选点簇中识别与所述货台匹配的第一目标点簇。
  14. 根据权利要求9所述的装置，还包括：
    橡胶块识别模块，用于从所述至少一个点簇中识别与所述货台配设的橡胶块匹配的第二目标点簇。
  15. 根据权利要求9所述的装置，还包括：
    停靠模块，用于根据第二位置信息和所述传感器的位置信息，生成车辆行驶路径，所述第二位置信息指示所述货台的位置，所述车辆行驶路径用于指示车辆行驶至所述货台。
  16. 根据权利要求9所述的装置，其中，所述传感器为单线激光雷达。
  17. 一种电子设备，包括：处理器以及存储有计算机程序指令的存储器；
    所述处理器执行所述计算机程序指令时实现如权利要求1-8任意一项所述的货台识别方法。
  18. 一种计算机可读存储介质，所述计算机可读存储介质上存储有计算机程序指令，所述计算机程序指令被处理器执行时实现如权利要求1-8任意一项所述的货台识别方法。
PCT/CN2022/072754 2021-01-22 2022-01-19 货台识别方法、装置、电子设备及计算机可读存储介质 WO2022156711A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110089750.0A CN114814780A (zh) 2021-01-22 2021-01-22 货台识别方法、装置、电子设备及计算机存储介质
CN202110089750.0 2021-01-22

Publications (1)

Publication Number Publication Date
WO2022156711A1 true WO2022156711A1 (zh) 2022-07-28

Family

ID=82525206

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/072754 WO2022156711A1 (zh) 2021-01-22 2022-01-19 货台识别方法、装置、电子设备及计算机可读存储介质

Country Status (2)

Country Link
CN (1) CN114814780A (zh)
WO (1) WO2022156711A1 (zh)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102087512A (zh) * 2009-12-07 2011-06-08 厦门雅迅网络股份有限公司 使用金属接近传感器实现混凝土搅拌运输车远程监控的方法
WO2016020347A1 (de) * 2014-08-05 2016-02-11 Valeo Schalter Und Sensoren Gmbh Verfahren zum erkennen eines objekts in einem umgebungsbereich eines kraftfahrzeugs mittels eines ultraschallsensors, fahrerassistenzsystem sowie kraftfahrzeug
CN108152823A (zh) * 2017-12-14 2018-06-12 北京信息科技大学 一种基于视觉的无人驾驶叉车导航系统及其定位导航方法
CN108320563A (zh) * 2017-11-06 2018-07-24 梁崇彦 一种基于汽车电子标识的平面车位检测装置
CN109581389A (zh) * 2017-09-28 2019-04-05 上海汽车集团股份有限公司 一种识别泊车车位边界的方法和装置
CN110146910A (zh) * 2019-05-15 2019-08-20 重庆大学 一种基于gps与激光雷达数据融合的定位方法及装置

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102087512A (zh) * 2009-12-07 2011-06-08 厦门雅迅网络股份有限公司 使用金属接近传感器实现混凝土搅拌运输车远程监控的方法
WO2016020347A1 (de) * 2014-08-05 2016-02-11 Valeo Schalter Und Sensoren Gmbh Verfahren zum erkennen eines objekts in einem umgebungsbereich eines kraftfahrzeugs mittels eines ultraschallsensors, fahrerassistenzsystem sowie kraftfahrzeug
CN109581389A (zh) * 2017-09-28 2019-04-05 上海汽车集团股份有限公司 一种识别泊车车位边界的方法和装置
CN108320563A (zh) * 2017-11-06 2018-07-24 梁崇彦 一种基于汽车电子标识的平面车位检测装置
CN108152823A (zh) * 2017-12-14 2018-06-12 北京信息科技大学 一种基于视觉的无人驾驶叉车导航系统及其定位导航方法
CN110146910A (zh) * 2019-05-15 2019-08-20 重庆大学 一种基于gps与激光雷达数据融合的定位方法及装置

Also Published As

Publication number Publication date
CN114814780A (zh) 2022-07-29

Similar Documents

Publication Publication Date Title
KR101900228B1 (ko) 주차 지원 방법 및 장치
CN109829351B (zh) 车道信息的检测方法、装置及计算机可读存储介质
Hata et al. Road marking detection using LIDAR reflective intensity data and its application to vehicle localization
US20210350149A1 (en) Lane detection method and apparatus,lane detection device,and movable platform
CN110794406B (zh) 多源传感器数据融合系统和方法
KR20180094055A (ko) 주차 지원 방법 및 장치
WO2021207954A1 (zh) 一种目标识别的方法和装置
US10996678B2 (en) Obstacle avoidance method and system for robot and robot using the same
CN112455430B (zh) 无车位线的斜列车位的检测方法、泊车方法及泊车系统
JP2011511281A (ja) センサにより検出されたオブジェクトとマップマッチングする方法
CN113297881B (zh) 一种目标检测方法及相关装置
US10846546B2 (en) Traffic signal recognition device
WO2021075210A1 (ja) センサ性能評価システム及び方法、並びに、自動運転システム
CN110341621B (zh) 一种障碍物检测方法及装置
CN112613424A (zh) 铁轨障碍物检测方法、装置、电子设备和存储介质
WO2019065970A1 (ja) 車両用外界認識装置
US20220126836A1 (en) Travel road recognition device
US10970870B2 (en) Object detection apparatus
US20220171975A1 (en) Method for Determining a Semantic Free Space
US11054245B2 (en) Image processing apparatus, device control system, imaging apparatus, image processing method, and recording medium
CN116533998B (zh) 车辆的自动驾驶方法、装置、设备、存储介质及车辆
CN115995163B (zh) 一种车辆碰撞预警方法及系统
WO2022156711A1 (zh) 货台识别方法、装置、电子设备及计算机可读存储介质
CN112183157A (zh) 道路几何识别方法及装置
JP5533323B2 (ja) 移動量推定装置及び移動量推定方法

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: 22742190

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 22742190

Country of ref document: EP

Kind code of ref document: A1