Object detection device

Abstract

An object detection device (50) is provided with: a speed calculation unit (113) that calculates the absolute movement speed of each of a plurality of measurement points corresponding to point cloud data on the basis of the relative movement speed of the measurement points detected by a lidar (5) and the absolute movement speed of the host vehicle; a classification unit (114) that classifies the point cloud data into moving point cloud data corresponding to the measurement points at which the absolute value of the absolute movement speed is equal to or greater than a prescribed speed, and stationary point cloud data other than the moving point cloud data; an extraction unit (115) that extracts, from the stationary point cloud data, changed point cloud data in which the amount of movement during a predetermined time is equal to or greater than a predetermined value; and a detection unit (116) that identifies a moving object on the basis of the change point cloud data. The detection unit (116) extracts a candidate moving object, which is a candidate for a moving object, from among objects detected by the lidar (5) on the basis of the change point cloud data, and determines whether or not the candidate moving object is a moving object on the basis of the moving speed and / or the moving direction of the candidate moving object.

Description

Technical Field

[0001] The present invention relates to an object detection device for detecting objects around a vehicle. Background Art

[0002] In recent years, there has been a growing demand for vehicle control systems that can improve traffic safety and contribute to the sustainable development of transportation systems. One such system is known, which uses machine learning to detect moving objects and the like from three-dimensional point cloud data acquired by laser radar (see, for example, Patent Document 1).

[0003] However, in a method using machine learning, such as the device described in Patent Document 1, since detection accuracy is affected by the reliability of the learning model, there is a possibility that sufficient detection accuracy of a moving object cannot be guaranteed.

[0004] Prior art literature

[0005] Patent Literature

[0006] Patent Document 1: Japanese Patent Application Publication No. 2023-027736 (JP 2023-027736 A). Summary of the Invention

[0007] An object detection device according to one embodiment of the present invention comprises: a detector mounted on a moving body, irradiating electromagnetic waves around the moving body and detecting an external condition around the moving body based on the reflected waves; a data acquisition unit, which acquires point cloud data detected by the detector at each predetermined time, the point cloud data including position information of a measurement point on the surface of the object where the reflected waves are obtained and first speed information indicating a relative moving speed of the measurement point; a speed acquisition unit, which acquires second speed information indicating an absolute moving speed of the moving body; and a calculation unit, which calculates a plurality of measurement points corresponding to the point cloud data based on the first speed information and the second speed information. The invention also includes a method for determining the absolute movement speed of each point cloud data item on the surface of the object, wherein the absolute value of the absolute movement speed calculated by the calculation unit is greater than or equal to a predetermined speed, and a method for determining the absolute movement speed of each point cloud data item. The method further includes a classification unit for classifying the point cloud data acquired by the data acquisition unit into first point cloud data corresponding to measurement points on the surface of the object for which the absolute value of the absolute movement speed calculated by the calculation unit is greater than or equal to a predetermined speed, and second point cloud data excluding the first point cloud data. The method further includes an extraction unit for calculating the movement amount of the measurement points corresponding to the second point cloud data over a predetermined period of time, and extracting, from the second point cloud data, predetermined point cloud data corresponding to measurement points for which the movement amount is greater than or equal to the predetermined value. The method further includes a classification unit for classifying the point cloud data acquired by the data acquisition unit into first point cloud data corresponding to measurement points on the surface of the object for which the absolute movement speed calculated by the calculation unit is greater than or equal to a predetermined speed, and a classification unit for classifying the point cloud data into second point cloud data excluding the first point cloud data. The method further includes a classification unit for classifying the point cloud data into first point cloud data corresponding to measurement points corresponding to the second point cloud data, and a classification unit for classifying the point cloud data into second point cloud data excluding the first point cloud data. The method further includes a classification unit for classifying the point cloud data into a candidate mobile object, and a classification unit for classifying the second ... BRIEF DESCRIPTION OF THE DRAWINGS

[0008] The objects, features and advantages of the present invention will be further clarified through the following description of the embodiments in conjunction with the accompanying drawings.

[0009] Figure 1 is a block diagram showing a main part configuration of a vehicle control device including an object detection device according to an embodiment of the present invention;

[0010] Figure 2 This figure is used to illustrate the relationship between the relative moving speed of a moving object measured by a laser radar and the moving direction of the moving object;

[0011] Figure 3 It is shown by Figure 1 A flowchart of an example of a process executed by the CPU of the controller;

[0012] Figure 4 FIG is a diagram schematically showing an example of point cloud data acquired by a laser radar;

[0013] Figure 5A Schematically shows an example of stationary point cloud data of the previous frame;

[0014] Figure 5B is a diagram schematically showing an example of stationary point cloud data of the current frame;

[0015] Figure 6A Schematically illustrates a state in which the static point cloud data of the previous frame and the static point cloud data of the current frame are aligned;

[0016] Figure 6B 1 is a diagram showing the stationary point cloud data of the previous frame and the stationary point cloud data of the current frame after alignment;

[0017] Figure 7 FIG2 is a diagram schematically showing another example of point cloud data acquired by a laser radar;

[0018] Figure 8 1 is a diagram showing a state of extracting change points from stationary point cloud data;

[0019] Figure 9 is a diagram for illustrating the motion vector of a moving object;

[0020] Figure 10A is a diagram for explaining misdetection of a moving object;

[0021] Figure 10B is a diagram for explaining misdetection of a moving object;

[0022] Figure 11A FIG. 1 is a diagram for explaining an example of a method for identifying an erroneously detected moving object;

[0023] Figure 11B FIG. 1 is a diagram showing a state in which an erroneously detected moving object is determined;

[0024] Figure 12 FIG. 1 is a diagram for explaining another example of a method for identifying an erroneously detected moving object;

[0025] Figure 13A A diagram illustrating the degree of possibility of the host vehicle approaching or colliding with a moving object;

[0026] Figure 13B A diagram illustrating the degree of possibility of the host vehicle approaching or colliding with a moving object;

[0027] Figure 13C This is a diagram for explaining the degree of possibility of the host vehicle approaching or colliding with a moving object. DETAILED DESCRIPTION

[0028] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. The object detection device of the embodiment of the present invention can be applied to a vehicle with an automatic driving function, i.e., an automatic driving vehicle. It should be noted that the vehicle to which the object detection device of the present embodiment is applied is sometimes distinguished from other vehicles and referred to as the present vehicle. The present vehicle may be any one of an engine vehicle having an internal combustion engine (engine) as a driving source, an electric vehicle having a driving motor as a driving source, and a hybrid vehicle having an engine and a driving motor as driving sources. The present vehicle can travel not only in an automatic driving mode that does not require the driver's driving operation, but also in a manual driving mode in which the driver performs the driving operation.

[0029] When operating in autonomous driving mode (hereinafter referred to as automated driving or autonomous driving), an autonomous vehicle identifies the surrounding environment based on data from onboard sensors such as cameras and LiDAR (Light Detection and Ranging). Based on this identification, the autonomous vehicle generates a driving trajectory (target trajectory) extending from the current point in time to a predetermined time, and controls its driving actuators to ensure the vehicle follows the target trajectory.

[0030] Figure 1 This is a block diagram showing the main components of a vehicle control device 100, including an object detection device. The vehicle control device 100 includes a controller 10, a communication unit 1, a positioning unit 2, an internal sensor group 3, a camera 4, a laser radar 5, and actuators AC for driving. Furthermore, the vehicle control device 100 includes an object detection device 50, which constitutes a portion of the vehicle control device 100. Object detection device 50 detects objects around the vehicle based on detection data from the laser radar 5.

[0031] The communication unit 1 communicates with various servers not shown in the figure through a network including a wireless communication network represented by the Internet, a mobile phone network, etc., and obtains map information, driving history information, traffic information, etc. from the server regularly or at any time. The network includes not only public wireless communication networks, but also closed communication networks set up in each specified management area, such as wireless LAN, Wi-Fi (registered trademark), Bluetooth (registered trademark), etc. The acquired map information is output to the storage unit 12, and the map information is updated. The positioning unit (GNSS unit) 2 has a positioning sensor that receives positioning signals sent from positioning satellites. Positioning satellites are artificial satellites such as GPS satellites and quasi-zenith satellites. The positioning unit 2 uses the positioning information received by the positioning sensor to measure the current position (latitude, longitude, altitude) of the vehicle.

[0032] The internal sensor group 3 is a collective term for multiple sensors (internal sensors) that detect the vehicle's driving state. For example, the internal sensor group 3 includes a speed sensor that detects the vehicle's speed, acceleration sensors that detect the vehicle's front-to-back acceleration and left-to-right acceleration (lateral acceleration), a rotational speed sensor that detects the rotational speed of the vehicle's driving source, and a yaw rate sensor that detects the angular velocity of the vehicle's center of gravity around its vertical axis. Sensors that detect driver operations in manual driving mode, such as those on the accelerator pedal, brake pedal, and steering wheel, are also included in the internal sensor group 3.

[0033] The camera 4 includes imaging elements such as a CCD (charge-coupled device) and a CMOS (complementary metal oxide semiconductor) and captures the vehicle's surroundings (front, rear, and sides). The laser radar 5 radiates electromagnetic waves (reflected waves) into the three-dimensional space surrounding the vehicle and detects the surrounding external conditions based on the reflected waves. More specifically, the electromagnetic waves (laser, etc.) emitted by the laser radar 5 reflect off a specific point on the surface of an object (the measurement point) and return. This allows the distance from the laser source to that point, the intensity of the reflected electromagnetic waves, and the relative speed of the object at that measurement point to be measured. The laser radar 5, mounted at a predetermined position (the front) of the vehicle, emits electromagnetic waves that scan the vehicle's surroundings (the front) horizontally and vertically, detecting the position, shape, and relative speed of objects in front of the vehicle (moving objects such as other vehicles, and stationary objects such as the road surface and structures). It should be noted that the three-dimensional space described above is represented by the X-axis, which corresponds to the vehicle's travel direction; the Y-axis, which corresponds to the vehicle's width; and the Z-axis, which corresponds to the vehicle's height. Therefore, the above three-dimensional space is sometimes referred to as XYZ space.

[0034] Actuators AC are driving actuators used to control the vehicle's travel. If the driving source is an engine, actuators AC include a throttle actuator that adjusts the engine's throttle valve opening (throttle opening). If the driving source is a travel motor, the travel motor is included in actuators AC. Actuators AC also include a brake actuator that activates the vehicle's brakes and a steering actuator that drives the steering system.

[0035] The controller 10 includes an electronic control unit (ECU). More specifically, the controller 10 includes a computer having a computing unit 11 such as a CPU (microprocessor), a storage unit 12 such as a ROM (read-only memory) and a RAM (random access memory), and other peripheral circuits not shown in the figure, such as an I / O (input / output) interface. It should be noted that multiple ECUs with different functions, such as an engine control ECU, a travel motor control ECU, and a brake device ECU, can be set separately, but for convenience, Figure 1 Controller 10 is shown in FIG as a collection of these ECUs.

[0036] The storage unit 12 stores high-precision, detailed map information (referred to as high-precision map information). This high-precision map information includes information on road locations, road shapes (such as curvature), road slopes, intersections, and forks, the number of lanes (driving lanes), lane widths, and the location of each lane (including lane center positions and lane boundary lines), the location of landmarks (such as traffic lights, signs, and buildings) that mark the map, and road surface contour information such as surface roughness. The storage unit 12 also stores various control programs, information such as thresholds used in these programs, and settings for on-board sensors such as the lidar 5.

[0037] The computing unit 11 includes a data acquisition unit 111, an estimation unit 112, a speed calculation unit 113, a classification unit 114, an extraction unit 115, a detection unit 116, a determination unit 117, a vector calculation unit 118, a removal unit 119, and a travel control unit 120 as functional structures. Figure 1 As shown, the data acquisition unit 111, the estimation unit 112, the velocity calculation unit 113, the classification unit 114, the extraction unit 115, the detection unit 116, the determination unit 117, the vector calculation unit 118, and the removal unit 119 are included in the object detection device 50. The data acquisition unit 111, the estimation unit 112, the velocity calculation unit 113, the classification unit 114, the extraction unit 115, the detection unit 116, the determination unit 117, the vector calculation unit 118, and the removal unit 119 included in the object detection device 50 will be described in detail later.

[0038] In the automatic driving mode, the driving control unit 120 generates a target trajectory based on the external conditions around the vehicle, including the size, position, relative moving speed, etc. of the object detected by the object detection device 50. Specifically, the driving control unit 120 generates a target trajectory in a manner that avoids collision, contact, or following the object based on the size, position, relative moving speed, etc. of the object detected by the object detection device 50. The driving control unit 120 controls the actuator AC so that the vehicle travels along the target trajectory. Specifically, the actuator AC is controlled along the target trajectory to adjust the accelerator opening, drive the brake device, or drive the steering device. It should be noted that in the manual driving mode, the driving control unit 120 controls the actuator AC based on the driving instructions (steering operation, etc.) from the driver obtained by the internal sensor group 3.

[0039] The object detection device 50 will be described in detail. As described above, the object detection device 50 includes the data acquisition unit 111, the estimation unit 112, the speed calculation unit 113, the classification unit 114, the extraction unit 115, the detection unit 116, the determination unit 117, the vector calculation unit 118, and the removal unit 119. The object detection device 50 also includes the laser radar 5.

[0040] The data acquisition unit 111 acquires four-dimensional data (hereinafter referred to as point cloud data) as detection data from the laser radar 5. This four-dimensional data includes position information indicating the three-dimensional position coordinates of a measurement point on the surface of an object from which the laser radar 5 reflects the reflected wave, and speed information indicating the relative movement speed of the measurement point. The point cloud data is acquired by the laser radar 5 in frames, specifically, at predetermined time intervals (time intervals determined by the frame rate of the laser radar 5).

[0041] As a method for detecting moving objects from the detection data (point cloud data) of the laser radar 5, there is a method of performing clustering processing on the point cloud data to classify the point cloud data into measurement points corresponding to the moving object and other measurement points to detect the moving object. However, the point cloud data also includes information on measurement points corresponding to stationary objects. Therefore, when the point cloud data obtained from the laser radar 5 is directly used for clustering processing, not only the measurement point group corresponding to the moving object but also the measurement point group corresponding to the stationary object is classified, and the computational load in the clustering processing may increase. Therefore, in order to reduce the computational load, a method can be considered to convert the relative movement speed represented by the speed information of each measurement point into an absolute speed (hereinafter referred to as the absolute movement speed) and classify the point cloud data into stationary point cloud data corresponding to the stationary object and moving point cloud data corresponding to the moving object based on the absolute movement speed, and then perform clustering processing on the classified moving point cloud data.

[0042] However, depending on the moving direction of the moving object, its relative moving speed may not be measured by the laser radar 5. Figure 2 : is a diagram for explaining the relationship between the relative moving speed of a moving object measured by the laser radar 5 and the moving direction of the moving object. Figure 2 In the figure, hollow arrows indicate the movement of mobile objects MO1 and MO2 in the directions of the arrows. Solid arrows schematically represent the electromagnetic waves emitted by the laser radar 5 and reflected off the surfaces of the mobile objects MO1 and MO2 and returned. Dashed line CL1 is a virtual line connecting points at equal distances from the laser radar 5 mounted on the vehicle 101.

[0043] The laser radar 5 cannot detect the speed in the direction perpendicular to the projection angle. Figure 2When moving on the dotted line CL1, its moving direction is perpendicular to the projection angle of the laser radar 5, so the relative moving speed of the moving objects MO1 and MO2 measured by the laser radar 5 remains unchanged at zero. As a result, the moving objects MO1 and MO2 are detected as stationary objects rather than moving objects, and the measurement points corresponding to the moving objects MO1 and MO2 are classified as stationary point cloud data. Therefore, in the method of performing clustering processing only on the moving point cloud data, the moving objects MO1 and MO2 moving in the direction perpendicular to the projection angle of the laser radar 5 may be excluded from the objects of clustering processing. In this case, the moving objects MO1 and MO2 will be missed, and the moving objects MO1 and MO2 cannot be tracked with high precision. Therefore, in this embodiment, clustering processing of the point cloud data is performed as follows to reduce the calculation load and detect the moving objects with high precision.

[0044] Figure 3 It is shown by Figure 1 This flowchart illustrates an example of processing executed by the CPU of controller 10 according to a predetermined program. The processing shown in this flowchart is repeated at a predetermined interval while object detection device 50 is activated. More specifically, it is repeated at a predetermined interval each time data acquisition unit 111 acquires detection data from laser radar 5.

[0045] When the detection data (point cloud data) of the laser radar 5 is obtained, the point cloud data is first classified into moving point cloud data and stationary point cloud data (step S1). Next, the detection processing of the moving object on the moving point cloud data (step S2) and the detection processing of the moving object on the stationary point cloud data (steps S31 to S33) are performed in parallel. It should be noted that the detection processing (step S2) and the detection processing (steps S31 to S33) may not be performed in parallel, but after executing one side, the other side is executed. Next, for the detected moving object, the processing of identifying the same object between frames (between the current frame and the previous frame) is performed (step S4), and then the processing of calculating the motion vector of the moving object identified as the same object is performed (step S5). It should be noted that in the detection processing of the moving object on the stationary point cloud data (steps S31 to S33), the following use sometimes occurs Figure 10A and Figure 10B Therefore, in order to further improve the detection accuracy of the moving object, the process of removing the above-mentioned false detection is finally performed (S6).

[0046] In the mobile object detection process (step S2) performed on the moving point cloud data, clustering processing is performed on the mobile point cloud data. In the mobile object detection process (steps S31 to S33) performed on the stationary point cloud data, first, in step S31, a predetermined scan matching process is performed to overlap the stationary point cloud data of the previous frame with the stationary point cloud data of the current frame, thereby estimating the azimuth angle difference and movement vector of the vehicle 101. The movement vector represents the direction of movement of the representative point (center of gravity, etc.) of the vehicle 101 between frames and the speed of movement in that direction. The azimuth angle difference is the angular difference between the azimuth of the vehicle 101 in the current frame and the azimuth (direction of travel) in the previous frame. Hereinafter, the axis along the direction of travel of the vehicle 101 is referred to as the X-axis, and the axes in the lateral and height directions relative to the direction of travel are referred to as the Y-axis and the Z-axis, respectively. It should be noted that the movement vector can be two-dimensional (X, Y) or three-dimensional (X, Y, Z). In addition, the azimuth angle difference can be the angle of one axis (Z-axis rotation angle), the angle of two axes (X-axis rotation angle, Z-axis rotation angle), or the angle of three axes (X-axis rotation angle, Y-axis rotation angle, Z-axis rotation angle). In the scan matching process, ICP (Iterative Closest Point), NDT (Normal Distributions Transform), or other methods can be used. Next, in step S32, the distance difference between each corresponding measurement point in the previous frame and each corresponding measurement point in the current frame detected by the overlap in the scan matching process of step S31 (the overlap of the static point cloud data of the previous frame and the static point cloud data of the current frame) is calculated. Then, the measurement points whose distance difference is greater than a specified threshold are extracted as change points. Finally, in step S33, clustering processing is performed on the extracted measurement point group (change point group). It should be noted that the azimuth angle difference and movement vector of the vehicle 101 estimated in step S31 are accumulated, and a own position estimation process (not shown) is performed to estimate the own position (traveling position of the vehicle 101) based on the accumulated azimuth angle difference and movement vector.

[0047] right Figure 3 The processing in each step is described in detail.

[0048] <Point Cloud Classification (S1)>

[0049] The estimating unit 112 estimates the absolute moving speed of the host vehicle 101 based on the point cloud data acquired by the data acquiring unit 111. Here, the estimation of the absolute moving speed of the host vehicle 101 by the estimating unit 112 will be described.

[0050] The estimation unit 112 extracts point cloud data, excluding information on measurement points corresponding to three-dimensional objects, from the point cloud data acquired by the data acquisition unit 111. Specifically, this point cloud data corresponds to the road surface surrounding the vehicle 101 (hereinafter referred to as road surface point cloud data). It should be noted that the road surface point cloud data can be extracted using a planar approximation method or other methods. The estimation unit 112 calculates a unit vector ei indicating the direction of the relative movement velocity vi based on the position coordinates (xi, yi, zi) contained in the four-dimensional data (xi, yi, zi, vi) of the measurement points Pi (i = 1, 2, ..., n) corresponding to the road surface in the road surface point cloud data. Specifically, the estimation unit 112 calculates the unit vector ei using the following equation (i).

[0051]

[0052] Next, the estimation unit 112 estimates the moving speed (absolute moving speed) Vself of the vehicle 101. Specifically, the estimation unit 112 sets the conversion formula for converting the relative moving speed vi of the measurement point Pi corresponding to the road surface into the absolute moving speed as the objective function L, and solves the optimization problem of optimizing the objective function L to be close to zero. The measurement point Pi is a measurement point on the road surface, so the absolute speed of each measurement point should be zero. Therefore, by optimizing the objective function L to be close to zero, the correct Vself can be estimated. Vself is shown in the following formula (ii), which is expressed by the speed components in the X, Y, and Z axis directions. The objective function L is expressed by the following formula (iii). By solving the above optimization problem, Vself that makes the right side of formula (iii) zero can be searched. It should be noted that Vself can be set to zero as the initial value, or it can be set to the Vself estimated in the previous frame.

[0053] V self =(v x , v y ,v z ) (ii)

[0054] L(V,f(A,V self ))=V+A·V self T (iii)

[0055] In equation (iii), A is the matrix of unit vectors ei at the n measurement points corresponding to the road surface, as expressed by equation (iv). Also in equation (iii), V is a 1×n matrix representing the velocity components (relative moving speeds) at the n measurement points Pi corresponding to the road surface, as expressed by equation (v). The estimating unit 112 obtains Vself, obtained by solving the aforementioned optimization problem, as the estimated value of the absolute moving speed of the host vehicle 101 in the current frame.

[0056]

[0057] V=[v1,v2,...,v n ] T (v)

[0058] The speed calculation unit 113 calculates the absolute moving speeds of all measurement points, more specifically, all measurement points including those corresponding to three-dimensional objects, based on the absolute moving speed Vself of the host vehicle 101 estimated by the estimation unit 112. The calculated absolute moving speed is negative when approaching the host vehicle and positive when moving away from the host vehicle.

[0059] The classification unit 114 classifies the point cloud data acquired by the data acquisition unit 111 into moving point cloud data corresponding to measurement points whose absolute moving speed calculated by the speed calculation unit 113 is greater than or equal to the predetermined speed Th_V, and stationary point cloud data corresponding to measurement points whose absolute moving speed is less than the predetermined speed Th_V.

[0060] <Estimation of Motion Vector (S31)>

[0061] Figure 4 、 Figure 5A 、 Figure 5B 、 Figure 6A as well as Figure 6B 1 is a diagram for explaining the process (step S31) in which the estimating unit 112 estimates the movement vector and the heading angle of the vehicle 101 between frames. Figure 4 In FIG, the arrow on the vehicle 101 indicates the moving direction (traveling direction) of the vehicle 101, that is, the X-axis direction. Figure 4 ) and height directions ( Figure 4 The directions from the depth to the front in the figure represent the Y-axis direction and the Z-axis direction, respectively. It should be noted that the definitions of the X-axis direction, the Y-axis direction, and the Z-axis direction are the same in other figures, and therefore, the labels of the X-axis, Y-axis, and Z-axis are omitted in those figures. Figure 4An example of a schematic diagram when observing the point cloud data obtained by the laser radar 5 at a past time point (time t1) from above (Z-axis direction) is shown. Areas N1 to N6 schematically represent the measurement point group corresponding to the stationary object, and more specifically, the position and size of the measurement point group. Stationary objects include the road surface of the road on which the vehicle 101 is traveling, walls set on the side of the road, structures such as separation strips, other vehicles parked on the shoulder of the road, etc. Areas M31 and M32 schematically represent the measurement point group corresponding to the moving object, and more specifically, the position and size of the measurement point group. It should be noted that the objects detected by the laser radar 5 will be referred to as objects below, including people. Therefore, in addition to moving vehicles such as cars and bicycles, moving objects also include moving people (pedestrians, etc.). The arrows marked on areas M31 and M32 indicate the moving direction of the moving object.

[0062] exist Figure 5A It is schematically shown in Figure 4 Point cloud data is the static point cloud data classified from the point cloud data obtained by the laser radar 5 at a past time point (time t1). Figure 5B Schematically shows stationary point cloud data classified from point cloud data acquired by the laser radar 5 at a current time point (time t2) after a predetermined time point in the past (time t1).

[0063] The estimation unit 112 first uses the static point cloud data of the previous frame ( Figure 5A ) and the static point cloud data of the current frame ( Figure 5B ) performs alignment and estimates the azimuth angle difference and movement vector of the vehicle 101 within a specified time (frame interval). Figure 6A and Figure 6B The figure schematically shows the static point cloud data of the previous frame ( Figure 5A ) and the static point cloud data of the current frame ( Figure 5B ) is aligned. Figure 6A In FIG, the stationary point cloud data of the previous frame is shown by a dotted line, and the stationary point cloud data of the current frame is shown by a solid line. The estimation unit 112 searches for (estimates) the azimuth angle difference and motion vector of the host vehicle 101 such that the measurement point clusters N1 to N6 of the previous frame, shown by the dotted lines, overlap (or coincide) with the measurement point clusters N1 to N6 of the current frame, shown by the solid lines. Figure 6B The static point cloud data of the previous frame and the static point cloud data of the current frame after alignment are shown in FIG. Figure 6BIn the figure, the angle MA represents the azimuth angle difference of the vehicle 101 between frames. The hollow arrow MV represents the motion vector of the vehicle 101 between frames. The white circle in the figure schematically represents the center of gravity of the vehicle 101. The estimation unit 112 solves the optimization problem of minimizing the deviation (error) between the positions of the measurement point group N1 to N6 of the previous frame after alignment and the positions of the measurement point group N1 to N6 of the current frame, and outputs the final search result (estimation result) of the azimuth angle difference and the motion vector of the vehicle 101. It should be noted that the three-dimensional point cloud data can also be converted into two-dimensional point cloud data represented by the XY coordinate system before performing the above-mentioned alignment to reduce the computational load.

[0064] <Extraction of Change Points (S32)>

[0065] Figure 7 and Figure 8 It is a diagram for explaining the change point extraction process (step S32). Figure 7 2 shows another example of a schematic diagram when the point cloud data acquired by the laser radar 5 at a past time point (time t1) is viewed from above (Z-axis direction). Figure 7 The point cloud data shown is the same as Figure 4 The point cloud data shown is the same, but Figure 7 The illustrated point cloud data includes a measurement point group (region M33 ) corresponding to a moving object moving in a direction perpendicular to the light projection angle of the laser radar 5 . Figure 7 The shaded area in the middle represents the measurement point group classified as the stationary point cloud data. As mentioned above, the laser radar 5 cannot detect the speed in the direction perpendicular to the projection angle. Figure 7 As shown in FIG. 1 , the object corresponding to region M33 is mistakenly detected as a stationary object rather than a moving object. Therefore, the extraction unit 115 calculates the amount of change (movement) in the position of each measurement point in the stationary point cloud data between frames (predetermined time) and extracts measurement points where this change exceeds a predetermined threshold as change points, thereby detecting the object corresponding to region M33 as a moving object. Figure 8 The state of extracting change points from static point cloud data is shown in FIG. Figure 8 In the figure, the dotted line shows the static point cloud data of the previous frame ( Figure 7 ), the solid line shows the static point cloud data of the current frame. Figure 8 The figure shows a state where the point cloud data of the previous frame and the point cloud data of the current frame are aligned using the movement vector and the azimuth angle difference of the vehicle estimated by the estimation unit 112. Figure 8As shown, the positions of measurement point cluster M33, corresponding to a moving object mistakenly detected as a stationary object, vary significantly between frames compared to the other measurement point clusters N1 to N6. Therefore, the extraction unit 115 extracts each measurement point in measurement point cluster M33 as a change point. Hereinafter, the point cloud data corresponding to the change points (change point cluster) extracted by the extraction unit 115 is referred to as change point cloud data.

[0066] <Clustering Process (S2, S33)>

[0067] In step S2, the detection unit 116 performs clustering processing on the moving point cloud data (the moving point cloud data classified from the point cloud data of the current frame in step S1). In addition, in step S33, the detection unit 116 performs clustering processing on the change point cloud data extracted from the static point cloud data (the static point cloud data classified from the point cloud data of the current frame in step S1). As a result, the bounding boxes (circumscribed areas) corresponding to the measurement point groups M31 and M32, respectively, are detected from the moving point cloud data, and the bounding box corresponding to the measurement point group M33 is detected from the change point cloud data. The detection unit 116 detects the position and size of the detected bounding boxes as the position and size of the moving object. In this way, the moving object contained in the three-dimensional space around the vehicle 101 is detected. The detection unit 116 outputs information (image information, etc.) indicating the detection result of the moving object to a display device, etc., not shown. It should be noted that the clustering process performed by the detection unit 116 may use any method such as DBSCAN (Density-based spatial clustering of applications with noise) or K-means (k-means clustering algorithm).

[0068] <Identification of the same object (S4)>

[0069] The determination unit 117 determines whether the moving object detected from the previous frame and the moving object detected from the current frame are the same object.

[0070] Specifically, first, the determination unit 117 performs an offset rotation process to offset (parallel move) and rotate each bounding box detected by the clustering process performed by the detection unit 116 in the previous frame based on the motion vector and azimuth angle difference between the frames of the host vehicle 101 estimated in step S31. In the offset rotation process, the determination unit 117 first offsets each bounding box detected in the previous frame according to the above-mentioned motion vector and rotates the above-mentioned azimuth angle difference. Next, the determination unit 117 further offsets each offset and rotated bounding box according to the motion vector of the corresponding moving object. Specifically, the bounding boxes of the measurement point groups M31, M32, and M33 are further offset by the motion amount obtained by multiplying the vector physical quantity of the motion vector of the moving object corresponding to the measurement point groups M31, M32, and M33 by the frame time (prescribed time). The motion vector of the moving object will be explained below.

[0071] As described above, determination unit 117 overlaps the bounding boxes of the previous frame, which has undergone offset rotation processing based on the motion vector and azimuth angle difference of host vehicle 101, and offset processing based on the motion vector of the moving object, with the current frame. As a result of this overlap, if a bounding box in the current frame overlaps with a bounding box from the previous frame that has undergone offset rotation and offset processing, determination unit 117 determines that the moving objects corresponding to the overlapping bounding boxes are the same. It should be noted that the overlap determination of the bounding boxes can be based on whether the overlap ratio is above a predetermined threshold, or based on whether the distance between the centers of gravity of the bounding boxes is less than a predetermined distance.

[0072] <Calculation of Movement Vector (S5)>

[0073] The vector calculation unit 118 calculates the motion vector of the moving object based on the determination result of the determination unit 117. Figure 9 The calculation of the motion vector of a moving object is explained. Figure 9 This is a diagram for explaining the motion vector of a moving object. Figure 9 In the figure, the dotted line represents the point cloud data of the previous frame, and the solid line represents the point cloud data of the current frame. Figure 9The figure shows a state where the point cloud data of the previous frame and the point cloud data of the current frame, which have undergone offset rotation processing based on the motion vector and azimuth angle difference of the vehicle 101, are overlapped. White circles G1, G2, and G3 represent the centroids of the measurement point clusters M31, M32, and M33. More specifically, they represent the centroids of the bounding boxes of the measurement point clusters M31, M32, and M33 detected by the clustering processing of the detection unit 116. The vector calculation unit 118 calculates the motion vector of the mobile object corresponding to the measurement point clusters M31, M32, and M33 based on the positional relationship between the centroids G1, G2, and G3 between frames (between the previous frame and the current frame). The vector calculation unit 118 stores the calculated motion vector of the mobile object in the storage unit 12 along with information (identifier) that can identify the mobile object. The motion vector of the mobile object stored in the storage unit 12 is used for the next offset processing based on the motion vector of the mobile object, that is, for the offset processing in step S4 performed on the next frame. As a result, in the next frame, the same object can be identified with high precision, and the position, moving direction, and moving speed of the moving object can be tracked well.

[0074] <Removal of erroneously detected moving objects (S6)>

[0075] Figure 10A and Figure 10B This is a diagram for explaining erroneous detection of a moving object in a detection process of a moving object based on change point cloud data. Figure 10A ] shows an example of a schematic diagram of a space in front of the vehicle 101 at a past time point (time t1) when viewed from above (Z-axis direction). Figure 10A The objects OB1 to OB6 are schematically represented by Figure 7 The stationary object corresponding to the measurement point group N1 to N6. Object OB33 schematically represents Figure 7 The moving object corresponding to the measurement point group M33.

[0076] like Figure 10A As shown in the figure, if there are objects OB2 and OB3 between object OB1 and the vehicle 101, object OB1 is hidden behind objects OB2 and OB3, and the electromagnetic waves emitted by the laser radar 5 can only reach a part of the area of object OB1. In this case, the measurement point group corresponding to the part of object OB1, specifically the area reached by the electromagnetic waves emitted by the laser radar 5, is actually obtained, rather than the measurement point group corresponding to the part of object OB1, specifically the area reached by the electromagnetic waves emitted by the laser radar 5. Figure 7 The measurement point group N1 corresponding to the entire object OB1 is the measurement point group corresponding to the object OB1. Figure 10A In order to simplify the drawing, only the measurement point fp1 is illustrated as the measurement point group corresponding to the object OB1 obtained at this time. Figure 10AIn order to simplify the drawing, the depiction of the measurement point group corresponding to objects other than the object OB1 is omitted.

[0077] Figure 10B 1 shows an example of a schematic diagram of the space in front of the vehicle 101 when viewed from above (Z-axis direction) at the current time point (time t2) which has passed a predetermined time from a predetermined time point in the past (time t1). The vehicle 101 moves between time t1 and time t2, and thus Figure 10B As shown in FIG. 1 , the electromagnetic waves emitted by the laser radar 5 reach the area hidden behind the objects OB2 and OB3 at time t1. As a result, Figure 10B As shown in FIG, at time t2, measurement points fp1_1 to fp1_5 are acquired as a measurement point group corresponding to the object OB1. Figure 10B The measurement point fp1_1 and Figure 10A Corresponding to the measurement point fp1.

[0078] In the scan matching process performed in step S31, Figure 10A The measurement point fp1 should be Figure 10B However, if Figure 10B As shown, when the measurement points fp1_2 to fp1_5 that were not obtained at time t1 are newly obtained due to the movement of the vehicle 101, the measurement point fp1 may establish a correspondence with any one of the measurement points fp1_2 to fp1_5 instead of establishing a correspondence with the measurement point fp1_1. In this case, it is erroneously determined that the position of the measurement point fp1 has changed between frames (predetermined time). In addition, when the amount of change (movement amount) is greater than the prescribed threshold, in step S32, any one of the measurement points fp1_2 to fp1_5 is erroneously detected as a change point. In this way, when the measurement point (in the field of view) corresponding to the area newly included in the field of view (FOV) of the laser radar 5 due to the movement of the vehicle 101 is newly obtained. Figure 10B In the example, when the measurement points fp1_2 to fp1_5) are erroneously extracted as change points, the object OB1, which is a stationary object, is erroneously detected as a moving object corresponding to the change point group in step S33.

[0079] Therefore, the removal unit 119 identifies the mobile objects that have been erroneously detected as described above from the mobile objects detected in step S33 (hereinafter referred to as candidate mobile objects) as follows. The removal unit 119 then removes the erroneously detected mobile objects. Specifically, the removal unit 119 deletes the information on the erroneously detected mobile objects (the mobile object identifier and motion vector stored in the storage unit 12 in step S5) from the storage unit 12.

[0080] Figure 11A and Figure 11BThis is a diagram for explaining an example of a method for determining a moving object that is misdetected. The removal unit 119 determines whether the angle formed by the motion vector of the moving object corresponding to the change point group calculated in the motion vector calculation (S5) and the line segment connecting the laser radar 5 and the representative point (e.g., center of gravity) of the moving object is within the angle range AR. Figure 11A As shown, the angle range AR is the range of angles d1 to d2 relative to the line segment L connecting the laser radar 5 and the representative point of the object OB being determined. Angle d1 is, for example, 60°, and angle d2 is, for example, 120°. It should be noted that angles d1 and d2 may also have other values. When the angle formed by the motion vector of object OB and line segment L is within the angle range AR, the removal unit 119 determines that the object OB has moved in a direction perpendicular to the projection angle of the laser radar 5, and therefore the measurement point cluster corresponding to object OB was extracted as a change point cluster in step S32. In this case, the removal unit 119 determines that object OB is a true moving object. On the other hand, if the angle is outside the angle range AR, the removal unit 119 determines that the measurement point cluster corresponding to object OB was incorrectly extracted as a change point cluster in step S32 due to the movement of the host vehicle 101, and object OB is determined to be an erroneously detected moving object. More specifically, when the angle is outside the angle range AR, as long as the object is actually moving and the relative moving speed of the object in step S1 is above the threshold, the measurement point group corresponding to the object should be classified as moving point cloud data. Therefore, an object that is determined to be stationary point cloud data in step S1 and has the angle outside the angle range AR can be determined as a moving object that has been misdetected. Figure 11B In the example shown, the angle formed by the line segment connecting the host vehicle 101 (laser radar 5) and the objects OB11 and OB12 and the motion vectors of the objects OB11 and OB12 is within the angle range AR, so the objects OB11 and OB12 are determined to be real moving objects. On the other hand, the angle formed by the line segment connecting the host vehicle 101 (laser radar 5) and the object OB13 and the motion vector of the object OB13 is outside the angle range AR, so the object OB13 is determined to be an erroneously detected moving object. In this way, the erroneously detected moving object is determined from the candidate moving objects. It should be noted that the dotted line in the figure is a virtual line connecting points that are equidistant from the laser radar 5 mounted on the host vehicle 101. In addition, the hollow arrows in the figure schematically represent the motion vectors of each object.

[0081] Figure 12This figure illustrates another example of a method for identifying an erroneously detected moving object. The removal unit 119 can also determine whether object OB is an erroneously detected moving object based on the component Vcosθ of the magnitude (velocity) V of the motion vector of the object OB being determined, as measured on line segment L. Angle θ is the angle formed between line segment L and the motion vector. When the absolute value of Vcosθ is greater than a predetermined velocity Th_V, the removal unit 119 determines that the measurement point group corresponding to object OB has already been classified as moving point cloud data in step S1, and therefore determines that object OB is an erroneously detected moving object.

[0082] It should be noted that the removal unit 119 can not only remove the mobile objects that are mistakenly detected in step S6, but also remove the mobile objects that have a low possibility of approaching or colliding with the vehicle 101. Specifically, the removal unit 119 determines the possibility of the mobile object (candidate mobile object) detected in step S33 approaching or colliding with the vehicle 101 based on the moving direction and position of the mobile object. When the possibility of approaching or colliding is lower than a specified level, the removal unit 119 can also remove the mobile object from the candidate mobile objects. Specifically, the information of the mobile object (the identifier and movement vector of the mobile object stored in the storage unit 12 in step S5) can also be deleted from the storage unit 12. Figure 13A 、 Figure 13B 、 Figure 13C This is a diagram for explaining the degree of possibility of the host vehicle 101 approaching or colliding with a moving object.

[0083] Figure 13A The dotted lines BL and BR shown are straight lines along the length direction of the vehicle 101, and are a pair of left and right straight lines having the same width as the road width (lane width) or a pre-set width. The dotted lines DL and DR are boundary lines used to determine whether a moving object is to be removed. The removal unit 119 determines whether the degree of possibility of approaching or colliding with the vehicle 101 is above a specified degree based on the boundary lines DL and DR, and selects moving objects whose degree of possibility is lower than the specified degree as removal objects. Specifically, the removal unit 119 determines that a moving object that exists outside the boundary lines DL and DR and whose movement vector is not toward the inside of the boundary lines DL and DR, that is, a moving object that does not enter the area sandwiched by the boundary lines DL and DR when moving in the direction of the movement vector, is a moving object whose degree of possibility of approaching or colliding with the vehicle 101 is lower than the specified degree, and selects it as a removal object. The hollow arrows in the figure schematically represent the movement vectors of each object. As shown Figure 13A As shown in FIG. 1 , when the vehicle 101 is traveling straight, the boundary lines DL and DR are set by providing a margin of a predetermined angle FA from the dotted lines BL and BR relative to the traveling direction of the vehicle 101. Figure 13AIn the example shown, among the candidate moving objects MO71 to MO76, moving objects MO71 and MO75 are selected as removal targets and removed from the candidate moving objects. The blocks in the figure schematically represent the situation where moving objects MO71 and MO75 are selected as removal targets.

[0084] Figure 13B The vehicle 101 is shown as traveling while turning left. Figure 13B As shown in FIG. 1 , when the vehicle 101 turns left, the removal unit 119 changes the left boundary line DL. More specifically, the removal unit 119 sets the boundary line DL according to the curvature of the vehicle 101. Figure 13B The curvature of the turning can be calculated based on the target trajectory of the vehicle 101 or based on the sensor values of the internal sensor group 3 (yaw rate sensor, vehicle speed sensor, steering angle sensor, etc.). The removal unit 119 selects the moving objects that exist outside the boundary line DL and whose movement vectors are not directed to the inside of the boundary lines DL and DR as removal targets. Figure 13B In the example shown, it is determined that the possibility of the moving object MO81 among the moving objects MO81 to MO85 as candidate moving objects approaching or colliding with the host vehicle 101 is lower than a predetermined level, and the moving object MO81 is selected as a removal target.

[0085] Figure 13C FIG. 1 shows a situation in which the vehicle 101 is turning right. Figure 13C As shown in FIG. 1 , the removal unit 119 changes the right boundary line DR. Similar to the left boundary line DL, the removal unit 119 sets the right boundary line DR according to the curvature of the vehicle 101. Figure 13C In the example, it is determined that the possibility of the moving object MO95 among the moving objects MO91 to MO95 as candidate moving objects approaching or colliding with the host vehicle 101 is lower than a predetermined level, and the moving object MO95 is selected as the removal target.

[0086] The above-described embodiment has the following effects.

[0087] (1) The object detection device 50 comprises: a laser radar 5 as a detector, which is mounted on the vehicle 101 as a moving body and irradiates electromagnetic waves around the vehicle 101 to detect the external conditions around the vehicle 101 based on the reflected waves; a data acquisition unit 111, which acquires point cloud data at each specified time, the point cloud data including position information of a measurement point on the surface of the object from which the reflected waves are detected by the laser radar 5 and first speed information indicating the relative moving speed of the measurement point; an estimation unit 112, which, as a speed acquisition unit, acquires second speed information indicating the absolute moving speed of the vehicle 101; and a speed calculation unit 113, which calculates a plurality of measurement points corresponding to the point cloud data based on the first speed information and the second speed information. Respective absolute moving speeds; a classification unit 114, which classifies the point cloud data acquired by the data acquisition unit 111 into moving point cloud data as first point cloud data corresponding to the measurement point on the surface of the object whose absolute value of the absolute moving speed calculated by the speed calculation unit 113 is greater than the specified speed, and stationary point cloud data as second point cloud data other than the first point cloud data; an extraction unit 115, which calculates the movement amount of the measurement point corresponding to the stationary point cloud data in a specified time, and extracts the change point cloud data as the specified point cloud data corresponding to the measurement point whose movement amount is greater than the specified value from the stationary point cloud data; and a detection unit 116, which determines the moving object among the objects detected by the laser radar 5 based on the moving point cloud data and the change point cloud data. The moving object includes the first moving object determined based on the moving point cloud data (for example, Figure 7 The mobile object corresponding to the measurement point group M31, M32) and the second mobile object determined according to the change point cloud data (for example, Figure 7 The detection unit 116 extracts candidate mobile objects that are candidates for the second mobile object from the objects detected by the laser radar 5 based on the change point cloud data, and determines whether the candidate mobile object is the second mobile object based on at least one of the moving speed and moving direction of the measurement point corresponding to the candidate mobile object. This prevents the misdetection of stationary objects as moving objects due to changes in the FOV of the laser radar 5. Furthermore, it prevents the misdetection of moving objects that are moving in a direction perpendicular to the projection angle of the laser radar 5 as stationary objects. As a result, moving objects can be detected with high accuracy.

[0088] (2) The detection unit 116 determines that the candidate moving object is the second moving object when the moving direction of the candidate moving object extracted from the change point cloud data and the angle formed by the line segment connecting the candidate moving object and the laser radar 5 are within the predetermined angle range AR. Thus, even when the FOV of the laser radar 5 changes, the moving object can be identified with high accuracy.

[0089] (3) The detection unit 116 determines the likelihood of an object approaching or colliding with the host vehicle 101 based on the moving direction and position of an object detected as a candidate mobile object based on the specified point cloud data, and removes objects whose likelihood is lower than a specified level from the candidate mobile objects. This reduces the number of candidate mobile objects and thus reduces the processing load for false detection of mobile objects.

[0090] (4) The estimation unit 112 estimates the absolute moving speed of the vehicle 101 based on the position information and speed information of the representative measurement point (road surface point cloud data) extracted from the point cloud data obtained by the data acquisition unit 111, and obtains the estimation result as speed information. The representative measurement point is selected from the measurement points remaining after removing the measurement points corresponding to the three-dimensional object from the plurality of measurement points. In this way, the absolute moving speed of the vehicle 101 is estimated with the measurement points corresponding to the road surface as a reference. As a result, the absolute moving speed of the vehicle 101 can be estimated with high accuracy. In addition, since the absolute moving speed of the vehicle (mobile body) 101 is estimated and obtained without using sensor values such as vehicle speed sensors, the present invention can also be applied to self-propelled robots that do not have vehicle speed sensors.

[0091] The above embodiment can be modified in various ways. The following describes variations. In the above embodiment, the laser radar 5 serving as a detector is mounted on a vehicle, irradiating electromagnetic waves into the three-dimensional space surrounding the vehicle and detecting the surrounding external conditions based on the reflected waves. However, the detector can also be a device other than a laser radar, such as a radar. Furthermore, the mobile object on which the detector is mounted can also be a mobile object other than a vehicle, such as a self-propelled robot.

[0092] Furthermore, in the above-described embodiment, the estimating unit 112, serving as the speed acquisition unit, selects a representative measurement point Pi from the remaining measurement points after removing the measurement point corresponding to the three-dimensional object from the plurality of measurement points. Based on the position information and speed information of the representative measurement point extracted from the point cloud data acquired by the data acquisition unit 111, the estimating unit estimates the absolute moving speed of the host vehicle 101 and acquires this estimation result as the second speed information. However, the speed acquisition unit may also acquire the measurement result of the absolute moving speed of the host vehicle 101 acquired by a measuring device included in the internal sensor group 3 as the second speed information. In this case, the object detection device 50 includes at least a vehicle speed sensor in the internal sensor group 3 as the measuring device. Alternatively, the speed acquisition unit may calculate and acquire the absolute moving speed of the host vehicle 101 based on the current position of the host vehicle 101 measured by the positioning unit 2. In this case, the object detection device 50 includes the positioning unit 2.

[0093] In addition, in the above-described embodiment, the detection unit 116 performs a predetermined clustering process on the three-dimensional point cloud data (movement point cloud data and change point cloud data) to detect moving objects from the three-dimensional point cloud data. However, the detection unit may also generate velocity-added data (XYV data) by adding the absolute movement speed of each measurement point calculated by the velocity calculation unit 113 to the two-dimensional point cloud data obtained by projecting each measurement point corresponding to the three-dimensional point cloud data onto the XY plane, and perform the predetermined clustering process on this velocity-added data. In this way, clustering processing is performed that takes into account the position and movement speed of the moving objects. As a result, it is possible to prevent multiple moving objects that are close to each other, such as two moving objects passing each other, from being detected as a single moving object, thereby further improving the accuracy of moving object detection. It should be noted that when accuracy of cluster size in three-dimensional space (XYZ space) is required, the detection unit may also generate velocity-added data (XYZV data) by adding the absolute movement speed of each measurement point calculated by the velocity calculation unit 113 to the three-dimensional point cloud data, and perform the predetermined clustering process on this velocity-added data.

[0094] Furthermore, in the above-described embodiment, objects detected as candidate mobile objects by the detection unit 116 as the mobile object identification unit based on the specified point cloud data are excluded from the candidate mobile objects if their probability of approaching or colliding with the host vehicle 101 is less than a specified level. However, the mobile object identification unit may also exclude objects detected as candidate mobile objects based on the specified point cloud data if their width and height do not meet specified criteria. The mobile object identification unit determines that the width and height of an object do not meet the specified criteria when the width of the object is less than a first reference value and the height is greater than a second reference value. The first and second reference values are pre-set to exclude mobile objects that are unlikely to actually exist on or near the road from the candidate mobile objects. For example, the first reference value is 1.5 meters, and the second reference value is 2.5 meters. It should be noted that the first and second reference values are not limited to these values. This can further reduce the number of candidate mobile objects and further reduce the processing load for false detection of mobile objects.

[0095] In the above embodiment, the estimation unit 112 uses the static point cloud data ( Figure 5A ) and the static point cloud data of the current frame ( Figure 5B) performs alignment, thereby estimating the azimuth angle difference and movement vector of the vehicle 101 at a specified time (between frames). However, the estimation unit may also perform the above-mentioned alignment after converting the three-dimensional stationary point cloud data into two-dimensional stationary point cloud data represented by an XY coordinate system. In addition, the estimation unit may also use not only the stationary point cloud data of a past time point (time t1) but also the stationary point cloud data of multiple time points in the past, that is, not only the stationary point cloud data of the previous frame but also the stationary point cloud data of multiple frames in the past for the above-mentioned alignment. In this way, by using the stationary point cloud data of multiple frames in the past, even when the stationary object in front of the vehicle 101 is temporarily blocked by other vehicles, etc., the above-mentioned alignment can be performed well, which can improve the robustness.

[0096] In addition, in the above-described embodiment, the driving control unit 120 controls the driving of the host vehicle 101 so as to avoid collision or contact with an object detected by the detection unit 116. However, the driving control unit 120 may also function as a notification unit to predict the possibility of collision or contact with a moving object based on the size, position, and moving speed of the moving object detected by the detection unit 116. Furthermore, when the possibility of collision or contact with a moving object exceeds a predetermined level, information (visual information, audio information) warning of collision or contact with a moving object detected by the detection unit 116 may be notified to the occupants of the host vehicle 101 via a display or speaker (not shown) included in the vehicle control device 100.

[0097] Furthermore, in the above embodiment, object detection device 50 is applied to an autonomous vehicle. However, object detection device 50 can also be applied to vehicles other than autonomous vehicles. For example, object detection device 50 can also be applied to manually driven vehicles equipped with ADAS (Advanced Driver-Assistance Systems).

[0098] The above description is merely an example, and the above embodiment and modifications do not limit the present invention unless the characteristics of the present invention are impaired. One or more of the above embodiment and modifications can be arbitrarily combined, and modifications can also be combined with each other.

[0099] By adopting the present invention, a moving object can be detected with high precision.

[0100] The present invention has been described above with reference to preferred embodiments. However, it should be understood by those skilled in the art that various modifications and changes can be made without departing from the scope of the claims.

Claims

1. An object detection device, characterized in that: have: A detector (5) is mounted on a moving body, irradiates electromagnetic waves around the moving body, and detects external conditions around the moving body based on reflected waves; a data acquisition unit (111) for acquiring point cloud data at each predetermined time, the point cloud data including position information of a measurement point on the surface of the object from which the reflected wave is obtained, detected by the detector (5), and first speed information indicating a relative moving speed of the measurement point; a speed acquisition unit (112) for acquiring second speed information indicating the absolute moving speed of the moving object; a calculation unit (113) for calculating the absolute moving speed of each of the plurality of measurement points corresponding to the point cloud data based on the first speed information and the second speed information; a classification unit (114) for classifying the point cloud data acquired by the data acquisition unit (111) into first point cloud data corresponding to the measurement point on the surface of the object at which the absolute value of the absolute moving speed calculated by the calculation unit (113) is greater than a predetermined speed, and second point cloud data other than the first point cloud data; an extraction unit (115) for calculating a movement amount of the measurement point corresponding to the second point cloud data during the prescribed time, and extracting prescribed point cloud data corresponding to the measurement point whose movement amount is greater than or equal to a prescribed value from the second point cloud data; as well as a moving object determining unit (116) for determining a moving object among the objects detected by the detector (5) based on the first point cloud data and the prescribed point cloud data; The moving object includes a first moving object determined according to the first point cloud data and a second moving object determined according to the specified point cloud data. The moving object determination unit (116) extracts a candidate moving object that is a candidate for the second moving object from the objects detected by the detector (5) based on the specified point cloud data, and determines whether the candidate moving object is the second moving object based on at least one of the moving speed and the moving direction of the measurement point corresponding to the candidate moving object.

2. The object detection device according to claim 1, wherein The moving object determination unit (116) determines that the candidate moving object is the second moving object when the moving direction of the candidate moving object extracted from the specified point cloud data and the angle formed by the line segment connecting the candidate moving object and the detector (5) are included in the specified angle range.

3. The object detection device according to claim 1, wherein The moving object determination unit (116) removes the object from the candidate moving objects when the absolute value of the component of the moving speed of the candidate moving object extracted from the specified point cloud data on the line segment connecting the candidate moving object and the detector is greater than a specified speed.

4. The object detection device according to claim 1, wherein The moving object determination unit (116) determines the possibility of the object approaching or colliding with the moving body based on the moving direction and position of the object extracted as the candidate moving object based on the specified point cloud data, and removes the objects whose possibility is lower than a specified level from the candidate moving objects.

5. The object detection device according to claim 1, wherein: The moving object determination unit (116) removes the objects detected by the detector (5) whose width and height do not meet a predetermined standard from the candidate moving objects.

6. The object detection device according to claim 1, wherein: The speed acquisition unit (112) estimates the absolute moving speed of the moving object based on the position information of the representative measurement point extracted from the point cloud data acquired by the data acquisition unit (111) and the first speed information, and acquires the estimated result as the second speed information. The representative measurement point is selected from the measurement points remaining after removing the measurement point corresponding to the three-dimensional object from the plurality of measurement points.

7. The object detection device according to any one of claims 1 to 6, characterized in that: The detector (5) is a laser radar.

Images