Information processing device, control method, program, and storage medium

Abstract

In this information processing device 1, a controller 13 primarily functions as an acquisition means, a surface detection means, an orientation estimation means, and an overall area estimation means. The acquisition means acquires measurement data output by a measurement device. The surface detection means detects at least one surface of an object from the measurement data. The orientation estimation means estimates the orientation of the object on the basis of the detected surface. The overall area estimation means estimates the overall area of the object on the basis of the estimated orientation and the measurement data.

Description

Technical Field

[0004] , , , ,

[0005] , , ,

[0001] This disclosure relates to the processing of measured data.

Background Art

[0002] Conventionally, a lidar device that irradiates a pulse of laser light onto a detection space and detects an object in the detection space based on the level of the reflected light has been known. For example, in Patent Document 1, a lidar that scans the surrounding space by appropriately controlling the emission direction (scanning direction) of repeatedly emitted light pulses and generates point cloud data representing information such as distance and reflectance, which are information regarding objects existing in the surroundings, by observing the return light is disclosed. Further, in Patent Document 2, a technique for detecting an object frame surrounding an object based on point cloud data and performing object tracking processing by associating the object frames in time series is disclosed.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0004] When determining a representative position of an object from point cloud data obtained by measuring the object, there has been a problem in that the portion of the object measured by the measuring device changes temporally depending on how the object appears to the measuring device, and as a result, the representative position of the object estimated from the point cloud data varies on the object.

[0005] This disclosure has been made to solve the above problems, and a main object thereof is to provide an information processing device capable of stably determining a representative position of an object.

Means for Solving the Problems

[0006] The invention according to the claims is a surface detection means for detecting at least one surface of an object from the measurement data, an orientation estimation means for estimating the orientation of the object based on the surface, A size estimation means for estimating the size of an object based on a segment representing the object detected from the measurement data, The object's size history estimated based on the detected measurement data, its orientation, and its surface and an overall area estimation means for estimating the overall area of the object based on and includes 、 The overall area estimation means estimates the overall area by excluding the size of the top predetermined number of items from the history. an information processing apparatus. Furthermore, the invention described in the claims is, A means for acquiring measurement data output by a measuring device, A surface detection means for detecting at least one surface of an object from the aforementioned measurement data, An orientation estimation means for estimating the orientation of the object based on the aforementioned surface, A size estimation means for estimating the size of an object based on a segment representing the object detected from the measurement data, A whole-area estimation means for estimating the whole-area of ​​an object based on the size that is the largest among the size history estimated based on the measurement data in which the object was detected, and the orientation of the object, This is an information processing device equipped with [a specific feature / feature].

[0007] Also, the invention according to the claims is a computer acquiring measurement data output by a measurement device, detecting at least one surface of an object from the measurement data, estimating the orientation of the object based on the surface, Based on the segment representing the object detected from the measurement data, the size of the object is estimated. The object's size history estimated based on the detected measurement data, its orientation, and its surface estimating the overall area of the object based on 、 From the history, the size of the top predetermined number of items is excluded to estimate the overall area. a control method. Furthermore, the invention described in the claims is, Computers The measurement data output by the measuring device is acquired, From the aforementioned measurement data, at least one surface of the object is detected. The orientation of the object is estimated based on the aforementioned surface. Based on the segment representing the object detected from the measurement data, the size of the object is estimated. Based on the measurement data in which the object was detected, the largest size among the size histories and the orientation of the object are used to estimate the entire area of ​​the object. This is a control method.

[0008] Also, the invention according to the claims is to acquire measurement data output by a measurement device, to detect at least one surface of an object from the measurement data, to estimate the orientation of the object based on the surface, Based on the segment representing the object detected from the measurement data, the size of the object is estimated. The object's size history estimated based on the detected measurement data, its orientation, and its surface to estimate the entire region of the object based on 、 The overall region is estimated by excluding the size of the top predetermined number of items from the history. [[ID=1 ]] Furthermore, the invention described in the claims is, The measurement data output by the measuring device is acquired, From the aforementioned measurement data, at least one surface of the object is detected. The orientation of the object is estimated based on the aforementioned surface. Based on the segment representing the object detected from the measurement data, the size of the object is estimated. This program causes a computer to perform a process to estimate the entire area of ​​an object based on the largest size among the size histories estimated based on the measurement data in which the object was detected, and the orientation of the object.

Brief Description of the Drawings

[0009] [Figure 1] It is a schematic configuration of a lidar unit according to an embodiment. [Figure 2] An installation example of a lidar for measuring a vehicle on a road is shown. [Figure 3] It is a block diagram showing an example of the hardware configuration of an information processing device. [Figure 4] (A) It is a side view showing the state of a certain vehicle, the detected vehicle segment, and the representative position in the comparative example. (B) It is a top view showing the state of another vehicle, the detected vehicle segment, and the representative position in the comparative example. [Figure 5] It is a top view showing the vehicle segments in time series. [Figure 6] It is a diagram in which the entire region of the vehicle estimated at each time is indicated by a dashed frame. [Figure 7] It is a diagram in which the representative position based on the entire region of the vehicle estimated at each time is indicated. [Figure 8]This diagram shows an overview of the process for generating and updating past size information. [Figure 9] This is an example of a flowchart illustrating the procedure for vehicle detection. [Modes for carrying out the invention]

[0010] According to a preferred embodiment of the present invention, the information processing device includes acquisition means for acquiring measurement data output by a measuring device, surface detection means for detecting at least one surface of an object from the measurement data, orientation estimation means for estimating the orientation of the object based on the surface, and overall area estimation means for estimating the overall area of ​​the object based on the orientation and the measurement data. In this embodiment, the information processing device can accurately estimate the overall area of ​​an object which is necessary to stably determine a representative position of the object.

[0011] In one embodiment of the above-described information processing device, the information processing device further includes a size estimation means for estimating the size of an object based on a segment representing the object detected from the measurement data, and the overall area estimation means for estimating the overall area based on the size history estimated from the measurement data in which the object was detected, the orientation, and the surfaces. In this embodiment, the information processing device can grasp the size of an object and accurately estimate the overall area of ​​the object.

[0012] In another embodiment of the information processing device described above, the overall area estimation means estimates the overall area by excluding the size of the top predetermined number of items from the history. In this embodiment, the information processing device can accurately estimate the overall area of ​​an object while taking noise into consideration.

[0013] In another embodiment of the information processing device described above, the object is a vehicle, and the surface detection means detects at least one surface of the vehicle from the measurement data. In a preferred example, the surface detection means detects data representing the vehicle's reflector from the measurement data and, based on this data, detects the rear surface of the vehicle. This embodiment enables the information processing device to accurately estimate the entire area of ​​the vehicle.

[0014] In another embodiment of the information processing device described above, the measuring device generates measurement data including reflection intensity values ​​for each measurement direction by receiving reflected light emitted from the light, and the surface detection means detects the surface based on the reflection intensity values. In this embodiment, the information processing device can accurately detect the surface of an object that has a characteristic reflectivity.

[0015] In another embodiment of the information processing apparatus described above, the information processing apparatus further comprises representative position determination means for determining a representative position of the object based on the overall area. In a preferred example, the information processing apparatus further comprises tracking means for tracking the object based on the transition of the representative position over time. This embodiment enables the information processing apparatus to accurately determine the representative position of the object and to track the object with high precision.

[0016] In another embodiment of the above-described information processing device, the information processing device further includes segment detection means for detecting segments representing the object from the measurement data, and the surface detection means for detecting the surface based on the segments. In this embodiment, the information processing device can accurately detect the surface of the object.

[0017] According to another preferred embodiment of the present invention, a control method performed by a computer includes acquiring measurement data output by a measuring device, detecting at least one face of an object from the measurement data, estimating the orientation of the object based on the face, and estimating the overall area of ​​the object based on the orientation and the measurement data. By performing this control method, the information processing device can accurately estimate the overall area of ​​the object necessary to stably determine a representative position of the object.

[0018] According to another preferred embodiment of the present invention, the present invention provides a program that causes a computer to perform the following processes: acquire measurement data output by a measuring device, detect at least one face of an object from the measurement data, estimate the orientation of the object based on the face, and estimate the overall area of ​​the object based on the orientation and the measurement data. By executing this program, the computer can accurately estimate the overall area of ​​the object necessary to stably determine a representative position of the object. Preferably, the above program is stored in a storage medium. [Examples]

[0019] Preferred embodiments of the present invention will be described below with reference to the drawings.

[0020] (1) Rider Unit Overview Figure 1 shows a schematic configuration of a Lidar unit 100 according to an embodiment. The Lidar unit 100 includes an information processing device 1 that processes data generated by a sensor group 2, and a sensor group 2 that includes at least a Lidar (Light Detection and Ranging, or Laser Illuminated Detection And Ranging) 3. Figure 2 shows an example of the installation of a Lidar 3 for measuring vehicles on a road. The Lidar 3 shown in Figure 2 is installed so that the road is included within the field of view "Rv", which is the range in which the Lidar 3 can measure distance. In reality, the field of view Rv will have a shape corresponding to the maximum distance measurement distance of the Lidar 3. The Lidar unit 100 then detects vehicles on the road and outputs information regarding the vehicle detection result. Note that the Lidar 3 may be installed so that a parking lot or the like is included in the field of view Rv instead of the road.

[0021] The information processing device 1 is electrically connected to the sensor group 2 by wire or wireless means and processes the data output by the various sensors included in the sensor group 2. In this embodiment, the information processing device 1 performs a process to detect vehicles on the road (also called "vehicle detection processing") based on the point cloud data output by the rider 3. The vehicle detection processing includes processes for estimating the state of the vehicle, such as estimating the vehicle's position, and tracking the vehicle. The information processing device 1 is fixedly installed, for example, in a housing together with the rider 3. The information processing device 1 may be provided as an electronic control device for the rider 3 and integrated with the rider 3, or it may be provided at a location separate from the rider 3 while being able to communicate with the rider 3 for data.

[0022] LIDA 3 discretely measures the distance to objects in the external environment by emitting a pulsed infrared laser while changing its angle within a predetermined angular range in the horizontal and vertical directions. In this case, LIDA 3 has an irradiation unit that irradiates laser light while changing the irradiation direction (i.e., scanning direction), a light receiving unit that receives reflected light (scattered light) of the irradiated laser light, and an output unit that outputs data based on the received signal output by the light receiving unit. LIDA 3 then generates point cloud data that shows the distance to the object irradiated by the pulsed laser (measured distance) and the received intensity of the reflected light (reflection intensity value) for each measurement direction (i.e., the direction of pulsed laser emission), with LIDA 3 as the reference point. In this case, LIDA 3 calculates the time from the emission of the pulsed laser until the light receiving unit detects the reflected light as the time of flight of light, and determines the measured distance corresponding to the calculated time of flight. Hereafter, the point cloud data obtained from one measurement across the entire field of view Rv will be considered as one frame of point cloud data.

[0023] Here, the point cloud data can be considered as an image where each measurement direction is represented by a pixel, and the measured distance and reflectance value for each measurement direction are used as pixel values. In this case, the pulse laser emission direction differs in elevation and depression angles for the vertical arrangement of pixels, and the pulse laser emission direction differs in horizontal angles for the horizontal arrangement of pixels. Then, for each pixel, coordinate values ​​in a three-dimensional coordinate system based on the lidar 3 are determined based on the corresponding pair of emission direction and measured distance.

[0024] Note that LIDA 3 is not limited to the scanning type LIDA described above, but may also be a flash type LIDA that generates 3D data by diffusing laser light into the field of view of a 2D array sensor. Hereafter, the point measured by the pulsed laser emitted from the irradiation unit (and its measurement data) will also be referred to as the "measured point". LIDA 3 is an example of a "measurement device" in the present invention.

[0025] Sensor group 2 may include various external and / or internal sensors in addition to the lidar 3. For example, sensor group 2 may include a GNSS (Global Navigation Satellite System) receiver necessary for generating location information.

[0026] (2) Configuration of an information processing device Figure 3 is a block diagram showing an example of the hardware configuration of the information processing device 1. The information processing device 1 mainly consists of an interface 11, a memory 12, and a controller 13. Each of these elements is interconnected via a bus line.

[0027] Interface 11 performs interface operations related to the exchange of data between the information processing device 1 and external devices. In this embodiment, interface 11 acquires output data from the sensor group 2, such as the lidar 3, and supplies it to the controller 13. Interface 11 may be a wireless interface such as a network adapter for wireless communication, or it may be a hardware interface for connecting to external devices via cables, etc. Interface 11 may also perform interface operations with various peripheral devices such as input devices, display devices, and sound output devices.

[0028] Memory 12 is composed of various volatile and non-volatile memories such as RAM (Random Access Memory), ROM (Read Only Memory), hard disk drive, and flash memory. Memory 12 stores programs for the controller 13 to execute predetermined processes. Note that the programs executed by the controller 13 may be stored in storage media other than memory 12.

[0029] Furthermore, memory 12 stores information necessary for the controller 13 to perform predetermined processing. For example, in this embodiment, memory 12 stores the vehicle list LV. The vehicle list LV is a list of detected vehicles generated in the vehicle detection process and is generated for each frame period. For example, each detected vehicle is assigned identification information (detected vehicle ID), and in the vehicle list LV, the detected vehicles are listed by their detected vehicle ID. If the same detected vehicle exists at different processing times in the vehicle detection process, the detected vehicle is managed by the same detected vehicle ID by utilizing the vehicle tracking information from a later stage.

[0030] Furthermore, memory 12 stores information regarding the history of the size of the detected vehicle (e.g., length and width of the vehicle) estimated based on the detection results of the detected vehicle. In this case, for example, memory 12 stores information representing the largest size in the history of previously estimated sizes (also called "past size information") for each detected vehicle ID used in the vehicle list LV. Past size information is generated by controller 13. In addition to past size information, memory 12 may also store prior information regarding the size of the vehicle (also called "prior size information"). In this case, the prior size information may be information indicating the general size of the vehicle, or information indicating the size for each vehicle classification. In addition to prior information regarding the size of the vehicle, memory 12 may also store prior information regarding the distribution of the reflectivity intensity of objects. This prior information may be information indicating, for example, that there is a part with strong reflectivity (due to the reflector) on the rear of the vehicle.

[0031] The controller 13 includes one or more processors such as a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), and a TPU (Tensor Processing Unit), and controls the entire information processing device 1. In this case, the controller 13 executes various processes described later by running programs stored in memory 12, etc. Functionally, the controller 13 has a segment detection unit 15, a representative position determination unit 16, and a vehicle information output unit 17.

[0032] The segment detection unit 15 extracts adjacent data sets (also called "segments") from the point cloud data generated in the frame period corresponding to the current processing time, and detects segments (also called "vehicle segments") that are estimated to represent vehicles from the extracted segments. Then, based on the vehicle segment detection results, the segment detection unit 15 generates a vehicle list LV, which is a list of detected vehicles.

[0033] In this case, the segment detection unit 15 extracts one or more segments from the point cloud data based on any clustering technique, such as Euclidean distance clustering, and detects segments that satisfy predetermined conditions as vehicle segments from the extracted segments. The predetermined conditions mentioned above are, for example, conditions relating to the size or shape of the segments. For example, the segment detection unit 15 detects segments as vehicle segments if the number of measured points constituting the segment is greater than or equal to a predetermined number. The segment detection unit 15 may also perform vehicle segment detection using a vehicle detection model trained based on deep learning (neural network). In this case, the vehicle detection model is, for example, a model based on instance segmentation, and is pre-trained to output vehicle segments when point cloud data represented in a predetermined tensor format is input.

[0034] The segment detection unit 15 may further determine the classification of the detected vehicle based on its size, etc. (for example, whether it is a large vehicle or a regular passenger car). In this case, for example, the segment detection unit 15 recognizes vehicle segments that satisfy predetermined conditions (for example, the number of measurement points is greater than or equal to a predetermined number) as large vehicle segments, and the others as regular passenger car segments. In another example, the segment detection unit 15 may classify vehicle segments based on the output results of the vehicle detection model described above, which has been trained to output classification results according to the size of the detected vehicle.

[0035] The representative position determination unit 16 first performs a process to associate the vehicle segment with the tracking object. In this case, the tracking object is the vehicle that was detected and is being tracked in the past time. In this case, the representative position determination unit 16 performs tracking based on the transition of the representative position of the center point of the vehicle segment over time. In this case, for example, the representative position determination unit 16 uses the center point of the vehicle segment and an arbitrary object tracking model to determine whether the vehicle segments detected in the time series over consecutive frame periods represent the same vehicle. The object tracking model may be a model based on a Kalman filter or a model based on deep learning.

[0036] Next, the representative position determination unit 16 detects at least one face of the detected vehicle (also called the "vehicle face") and the orientation of the detected vehicle based on the vehicle segment. Then, the representative position determination unit 16 estimates the size (e.g., width and length) of the detected vehicle based on the vehicle segment obtained for each frame period. The representative position determination unit 16 then updates the past size information to be included in the vehicle list LV based on the estimated size (e.g., width and length) of the detected vehicle. In this case, for detected vehicles to which past size information has not been linked, the representative position determination unit 16 generates past size information indicating the estimated size of the vehicle. On the other hand, for detected vehicles to which past size information has already been linked, the representative position determination unit 16 updates the past size information with the estimated size of the vehicle if the estimated size of the detected vehicle is larger than the size of the detected vehicle indicated by the past size information.

[0037] The representative position determination unit 16 then determines the representative position of each detected vehicle in the vehicle list LV based on the vehicle surface and orientation of the detected vehicle and the latest past size information. Here, the "representative position" is the position of a predetermined representative point of the vehicle, which may be the center position of the vehicle, the center position of the front of the vehicle, or any other predetermined position on the vehicle. In this case, the representative position determination unit 16 estimates the overall area of ​​the detected vehicle based on the detected vehicle surface, the orientation of the detected vehicle identified based on the vehicle surface, and the past size information, and estimates the representative position from the estimated overall area. Note that if past size information for the target detected vehicle does not exist, the representative position determination unit 16 may use prior size information instead of past size information to estimate the overall area. Details of the representative position estimation method will be described later.

[0038] Then, the representative position determination unit 16 updates the estimated position of each detected vehicle using the determined representative position of each detected vehicle as an observed value.

[0039] The vehicle information output unit 17 outputs information about vehicles detected by the rider 3. In this case, the vehicle information output unit 17 may, for example, output information about the number of vehicles detected within a predetermined period based on the processing results of the representative position determination unit 16, or it may output information about the position and / or speed of each detected vehicle. In this case, as one form of output, the vehicle information output unit 17 may store the above information in the memory 12, or transmit it via the interface 11 to a server device that collects information about vehicles traveling on each road.

[0040] The controller 13 functions as an "acquisition means," a "size estimation means," a "segment detection means," a "representative position determination means," a "tracking means," and a computer that executes a program.

[0041] Furthermore, the processing performed by the controller 13 is not limited to being implemented by software through a program, but may also be implemented by a combination of hardware, firmware, and software. Additionally, the processing performed by the controller 13 may be implemented using a user-programmable integrated circuit, such as an FPGA (Field-Programmable Gate Array) or a microcontroller. In this case, the program executed by the controller 13 in this embodiment may be implemented using this integrated circuit.

[0042] (3) Estimation of representative position Next, the method for estimating the representative position of the detected vehicle by the representative position determination unit 16 will be described. In general terms, the representative position determination unit 16 detects the vehicle surface, orientation, and size of the detected vehicle based on the vehicle segment, then estimates the overall area of ​​the detected vehicle based on the detection results, and determines the representative position from the estimated overall area. In this way, the representative position determination unit 16 accurately determines the representative position of the detected vehicle regardless of the detection status of the detected vehicle by the rider 3.

[0043] First, we will explain the challenges when adopting the example (comparative example) of defining the center of gravity of a vehicle segment as the representative position of the vehicle. Figure 4(A) is a side view showing the vehicle segment of vehicle A and the representative position determined based on the comparative example at times ta and tb, and Figure 4(B) is an overhead view showing the vehicle segment of vehicle B and the representative position determined based on the comparative example at times ta and tb. In Figures 4(A) and 4(B), the direction in which the vehicle traveled between time ta and time tb is clearly indicated by arrows.

[0044] In the example shown in Figure 4(A), at time ta, the entire top and rear of vehicle A are detected by RIDA 3, and a vehicle segment including these measured points is generated. On the other hand, at time tb, the entire rear of vehicle A and a portion of the top are detected by RIDA 3, and a vehicle segment including these measured points is generated. As a result, in the method for calculating the representative position in the comparative example, the representative position of vehicle A is near the center of vehicle A at time ta, and near the rear seat of vehicle A at time tb.

[0045] Similarly, in the example in Figure 4(B), at time ta, the entire left side of vehicle B is detected by lidar 3, and a vehicle segment including these measured points is generated. At time tb, the entire left side and the entire rear of vehicle B are detected by lidar 3, and a vehicle segment including these measured points is generated. As a result, in the method for calculating the representative position in the comparative example, the representative position of vehicle B is near the center of the left side of vehicle B at time ta, and near the center of vehicle B at time tb.

[0046] Thus, in the comparative example, there is a problem in that the representative position of the detected object is shifted depending on how the object appears to the lidar 3. Taking the above into consideration, in this embodiment, the representative position determination unit 16 determines the representative position after estimating the entire area of ​​the detected vehicle, including the part that is not visible to the lidar 3. As a result, the representative position determination unit 16 accurately detects the same location on the detected vehicle as the representative position of the detected vehicle.

[0047] Next, the details of the processing performed by the representative position determination unit 16 will be explained using a specific example shown in Figure 5. Figure 5 is an overview view showing the vehicle segment in time series from time t0 to t3. In Figure 5, at time t0, a vehicle segment including the entire top surface of the detected vehicle is generated. Subsequently, as the vehicle moves over time, the detection area of ​​the top surface of the detected vehicle decreases, and at time t3, the top surface of the detected vehicle is almost no longer included in the vehicle segment.

[0048] Figure 6 is a diagram in which the estimated total area of ​​the vehicle at each time point is clearly indicated by a dashed line frame in the specific example shown in Figure 5. In addition, in Figure 6, the flow of the estimation process for the total area of ​​the detected vehicle at time t1 is shown by a callout 60. In this example, the representative position determination unit 16 assumes that the side of the vehicle segment on the rider 3 side coincides with the side of the detected vehicle on the rider 3 side, and assumes that the detected vehicle, based on past size information, is at that position.

[0049] First, the method for detecting the vehicle surface at time t1 will be specifically explained. At time t1, the representative position determination unit 16 detects the vehicle surface of the detected vehicle based on the vehicle segment of the target detected vehicle acquired by the segment detection unit 15 at time t1. Here, as an example, the representative position determination unit 16 detects the back of the detected vehicle, which has a pair of reflectors made of retroreflective material, as the vehicle surface. Note that a pair of reflectors is provided on the back of a typical vehicle.

[0050] In this case, for example, the representative position determination unit 16 considers that the reflectance intensity value of the point of measurement of the reflector is sufficiently higher than the reflectance intensity values ​​of other points of measurement in the vehicle segment, and detects a cluster of points of measurement within the vehicle segment whose reflectance intensity value is equal to or greater than a predetermined value as a reflector region. Then, the representative position determination unit 16 detects a surface containing the pair of reflector regions (for example, a point of measurement located near a plane obtained by the least squares method or the like from a point of measurement belonging to the reflector region) as a vehicle surface. Note that the representative position determination unit 16 may also detect a specific vehicle surface (not necessarily the rear) of the detected vehicle based on any method other than the rear-side detection method of the detected vehicle based on the reflector described above.

[0051] Furthermore, the representative position determination unit 16 identifies the orientation of the detected vehicle based on the normal direction perpendicular to the calculated vehicle surface. Specifically, the representative position determination unit 16 identifies the direction away from the rider 3, which is the normal direction mentioned above, as the orientation of the detected vehicle.

[0052] The representative position determination unit 16 then estimates the overall area of ​​the detected vehicle based on past size information associated with the detected vehicle. At time t1, past size information indicating the size of the detected vehicle estimated at time t0 has been generated, and the representative position determination unit 16 estimates the overall area of ​​the detected vehicle by referring to this past size information. Here, as an example, in order to determine the representative position of the detected vehicle on a two-dimensional coordinate system representing its position on the horizontal plane, the representative position determination unit 16 considers the width and length as the size of the detected vehicle, and estimates a rectangular area with the back as one side and the specified orientation as the longitudinal direction (length direction) as the overall area of ​​the detected vehicle. Alternatively, the representative position determination unit 16 may estimate the overall area of ​​the detected vehicle on a three-dimensional coordinate system in order to determine the representative position of the detected vehicle on a three-dimensional coordinate system. In this case, the past size information includes width, length, and height representing the size of the detected vehicle, and the representative position determination unit 16 estimates the entire area of ​​the detected vehicle by using the back of the detected vehicle as one face of a rectangular parallelepiped and the specified orientation as the length direction of the rectangular parallelepiped.

[0053] Furthermore, the representative position determination unit 16 performs the same processing as at time t1 at times t2 and t3 to estimate the overall area of ​​the detected vehicle. The representative position determination unit 16 may also estimate the overall area of ​​the detected vehicle at time t0 using past size information or prior size information generated at the same time. The estimated overall area of ​​the detected vehicle at time t0 is a rectangular area that roughly coincides with the outer edge of the vehicle segment.

[0054] Figure 7 is a diagram in which, in the specific example shown in Figure 5, the representative position based on the estimated overall area of ​​the vehicle at each time point is indicated by an "x" mark. Here, the representative position of the detected vehicle is, for example, the center position of the front of the vehicle. The representative position determination unit 16 recognizes the center position of the front of the vehicle at each time point based on the estimated overall area at each time point, and sets the recognized position as the representative position of the detected vehicle. In this way, the representative position determination unit 16 can stably determine a predetermined position on the detected vehicle as the representative position over time. The representative position of the detected vehicle determined by the representative position determination unit 16 is then suitably used in the tracking process.

[0055] Figure 8 shows an overview of the process for generating and updating past size information in the specific example shown in Figure 5. Here, an overview of the method for generating past size information at time t0 is shown.

[0056] The representative position determination unit 16 considers the detected vehicle as a rectangle with its longitudinal direction in a two-dimensional coordinate space viewed from above (i.e., with the coordinate representing height removed), and calculates the width and length of the rectangle. In this case, the representative position determination unit 16, for example, defines the smallest rectangle containing the vehicle segment in the aforementioned two-dimensional coordinate space with the detected vehicle's orientation as the longitudinal direction, as the overall area of ​​the detected vehicle. The representative position determination unit 16 then calculates the width and length of the rectangle as the width and length of the detected vehicle. Alternatively, the representative position determination unit 16 may define the smallest rectangular parallelepiped containing the vehicle segment in a three-dimensional coordinate space with the detected vehicle's orientation as the overall area of ​​the detected vehicle. In this case, the representative position determination unit 16 calculates the width, length, and height of the rectangular parallelepiped as the width, length, and height of the detected vehicle. The representative position determination unit 16 may also use an orientation based on the normal of the vehicle surface as the orientation of the detected vehicle, or it may specify the direction of movement of the detected vehicle, which is determined based on the transition of the representative position, as the orientation of the detected vehicle. Furthermore, if map information is stored in memory 12, the representative position determination unit 16 may determine the orientation of the detected vehicle based on the direction of the lane in which the detected vehicle is located.

[0057] Then, at time t0, the representative position determination unit 16 generates past size information indicating the width and length (and height) of the detected vehicle calculated from the vehicle segment obtained at the same time using the method described above, and stores it in memory 12 in association with the detected vehicle ID of the target detected vehicle. Furthermore, at each time from t1 to t3, the representative position determination unit 16 calculates the size (width and length (and height)) of the detected vehicle based on the vehicle segment obtained at the same time, and updates the past size information with the calculated size if the calculated size is larger than the size indicated by the stored past size information. However, at each time from t1 to t3, the calculated size is smaller than the size indicated by the stored past size information, so the representative position determination unit 16 does not update the past size information.

[0058] Furthermore, the representative position determination unit 16 may, with consideration for noise immunity, avoid estimating the entire area of ​​the detected vehicle based on an estimated size that has suddenly become large. For example, the representative position determination unit 16 stores the history of previously estimated sizes for each detected vehicle as past size information. When the representative position determination unit 16 estimates the entire area of ​​the detected vehicle by referring to the past size information, it considers the top N sizes (where N is an integer greater than or equal to 1) from the past estimated sizes as noise and does not use them in the estimation of the entire area. The representative position determination unit 16 then estimates the entire area based on the (N+1)th largest estimated size. This allows the representative position determination unit 16 to estimate the entire area of ​​the detected vehicle more accurately.

[0059] (4) Processing flow Figure 9 is an example flowchart showing the procedure for vehicle detection processing performed by the information processing device 1. The information processing device 1 repeatedly executes the processing shown in this flowchart at predetermined processing cycles (for example, the frame cycle of the rider 3).

[0060] First, the controller 13 of the information processing device 1 acquires point cloud data measured by the rider 3 via the interface 11 (step S10). Then, based on the point cloud data acquired in step S10, the controller 13 detects vehicle segments and, based on the detection results, generates a vehicle list LV, which is a list of detected vehicles to be used in the current processing cycle (step S11).

[0061] Next, the controller 13 selects the vehicle to be processed in steps S14 to S17 (target vehicle) from the vehicle list LV (step S12). Then, the controller 13 performs a process to associate the vehicle segment of the target vehicle with the tracked object (step S13). Then, the controller 13 detects the vehicle surface, orientation, and size of the target vehicle based on the vehicle segment (step S14). In this case, the controller 13 detects a pair of reflector regions based on the reflectance intensity value of the vehicle segment corresponding to the target vehicle, and detects the back of the target vehicle including the reflector region as the vehicle surface. The controller 13 also detects the orientation of the target vehicle based on the normal of the detected vehicle surface. Furthermore, the controller 13 estimates the width and length of the target vehicle by approximating the vehicle segment with a rectangle. Finally, the controller 13 updates the past size information based on the processing result of step S14.

[0062] Then, the controller 13 estimates the entire area of ​​the target vehicle based on the processing result of step S14 (step S15).

[0063] Then, the controller 13 estimates a representative position of the target vehicle based on the estimated overall area (step S16). In this case, the controller 13 can stably determine a predetermined position on the target vehicle as the representative position. Then, the controller 13 updates the estimated position of the target vehicle based on the representative position estimated in step S16 (step S17).

[0064] Next, the controller 13 determines whether all vehicles in the vehicle list LV have already been designated as target vehicles (step S18). If all vehicles in the vehicle list LV have already been designated as target vehicles (step S18; Yes), the controller 13 terminates the flowchart process. On the other hand, if there are vehicles in the vehicle list LV that have not been designated as target vehicles (step S18; No), the controller 13 returns to step S12 and selects the vehicles that have not been designated as target vehicles as target vehicles.

[0065] (5) Variation The Rider Unit 100 may be mounted on a vehicle. In this case, the Rider 3 is provided on the vehicle, and the information processing device 1 is an on-board unit of the vehicle or an electronic control unit (ECU) built into the vehicle. In this case, the information processing device 1 detects vehicles in the vicinity of the vehicle on which the Rider Unit 100 is mounted based on point cloud data generated by the Rider 3. Even in this case, the information processing device 1 can accurately perform tracking of vehicles in the vicinity of the vehicle on which the Rider Unit 100 is mounted.

[0066] Furthermore, the object to be tracked is not limited to vehicles; it may be any moving object. In this case, the segment detection unit 15 detects segments of the moving object to be tracked, and the representative position determination unit 16 performs a process to determine the representative position of the moving object to be tracked based on those segments.

[0067] As described above, the controller 13 of the information processing device 1 according to this embodiment mainly functions as an acquisition means, a surface detection means, an orientation estimation means, and an overall area estimation means. The acquisition means acquires measurement data output by the measuring device. The surface detection means detects at least one surface of the object from the measurement data. The orientation estimation means estimates the orientation of the object based on the detected surface. The overall area estimation means estimates the overall area of ​​the object based on the estimated orientation and the measurement data. As a result, even when only a part of the object can be measured by the measuring device, the information processing device 1 can suitably estimate the overall area of ​​the object and stably determine the representative position of the object.

[0068] In the above-described embodiment, the program can be stored using various types of non-transitory computer-readable medium and supplied to a computer, such as a controller. Non-transitory computer-readable medium includes various types of tangible storage medium. Examples of non-transitory computer-readable medium include magnetic storage medium (e.g., flexible disks, magnetic tapes, hard disk drives), magneto-optical storage medium (e.g., magneto-optical disks), CD-ROM (Read Only Memory), CD-R, CD-R / W, and semiconductor memory (e.g., mask ROM, PROM (Programmable ROM), EPROM (Erasable PROM), flash ROM, RAM (Random Access Memory)).

[0069] The present invention has been described above with reference to the examples, but the present invention is not limited to the above examples. Various modifications to the structure and details of the present invention can be made as understood by those skilled in the art within the scope of the present invention. That is, the present invention includes the full disclosure, including the claims, and of course, various modifications and alterations that those skilled in the art could make in accordance with the technical idea. Furthermore, the above-mentioned patent and non-patent documents and other disclosures cited herein are incorporated by reference. [Explanation of Symbols]

[0070] 1. Information Processing Device 2 Sensor Groups 3 Riders 100 Rider Unit

Claims

1. A means for acquiring measurement data output by a measuring device, A surface detection means for detecting at least one surface of an object from the aforementioned measurement data, An orientation estimation means for estimating the orientation of the object based on the aforementioned surface, A size estimation means for estimating the size of an object based on a segment representing the object detected from the measurement data, A whole-area estimation means for estimating the entire area of ​​an object based on the size history estimated from the measurement data in which the object was detected, the orientation, and the surface, Equipped with, The overall area estimation means is an information processing device that estimates the overall area by excluding the largest predetermined number of items of the size from the history.

2. A means for acquiring measurement data output by a measuring device, A surface detection means for detecting at least one surface of an object from the aforementioned measurement data, An orientation estimation means for estimating the orientation of the object based on the aforementioned surface, A size estimation means for estimating the size of an object based on a segment representing the object detected from the measurement data, A whole-area estimation means for estimating the whole-area of ​​an object based on the size that is the largest among the size history estimated based on the measurement data in which the object was detected, and the orientation of the object, An information processing device equipped with the following features.

3. The information processing apparatus according to claim 2, wherein the overall area estimation means excludes a predetermined number of the largest sizes from the history, and estimates the overall area based on the largest size among the history after exclusion.

4. The aforementioned object is a vehicle, The surface detection means detects at least one surface of the vehicle from the measurement data, as described in any one of claims 1 to 3.

5. The information processing apparatus according to claim 4, wherein the surface detection means detects data representing the reflector of the vehicle from the measurement data and detects the rear surface of the vehicle based on the data.

6. The measuring device generates the measurement data, including the reflection intensity value for each measurement direction, by receiving the reflected light of the emitted light. The information processing apparatus according to any one of claims 1 to 5, wherein the surface detection means detects the surface based on the reflection intensity value.

7. Representative position determination means for determining a representative position of the object based on the overall area. The information processing apparatus according to any one of claims 1 to 6, further comprising:

8. Tracking means for tracking the object based on the transition of the representative position over time. The information processing apparatus according to claim 7, further comprising:

9. The system further includes a segment detection means for detecting a segment representing the object from the measurement data, The information processing apparatus according to any one of claims 1 to 8, wherein the surface detection means detects the surface based on the segment.

10. Computers The measurement data output by the measuring device is acquired, From the aforementioned measurement data, at least one surface of the object is detected. The orientation of the object is estimated based on the aforementioned surface. Based on the segment representing the object detected from the measurement data, the size of the object is estimated. Based on the size history estimated from the measurement data in which the object was detected, the orientation, and the surface, the entire area of ​​the object is estimated. From the history, the size of the top predetermined number of items is excluded to estimate the overall area. Control method.

11. Computers The measurement data output by the measuring device is acquired, From the aforementioned measurement data, at least one surface of the object is detected. The orientation of the object is estimated based on the aforementioned surface. Based on the segment representing the object detected from the measurement data, the size of the object is estimated. Based on the measurement data in which the object was detected, the largest size among the size histories and the orientation of the object are used to estimate the entire area of ​​the object. Control method.

12. The measurement data output by the measuring device is acquired, From the aforementioned measurement data, at least one surface of the object is detected. The orientation of the object is estimated based on the aforementioned surface. Based on the segment representing the object detected from the measurement data, the size of the object is estimated. Based on the size history estimated from the measurement data in which the object was detected, the orientation, and the surface, the entire area of ​​the object is estimated. A program that causes a computer to perform a process of estimating the total area by excluding the top predetermined number of items of size from the history.

13. The measurement data output by the measuring device is acquired, From the aforementioned measurement data, at least one surface of the object is detected. The orientation of the object is estimated based on the aforementioned surface. Based on the segment representing the object detected from the measurement data, the size of the object is estimated. A program that causes a computer to perform a process to estimate the entire area of ​​an object based on the largest size among the size histories estimated based on the measurement data in which the object was detected, and the orientation of the object.

14. A storage medium storing the program according to claim 12 or 13.

Images