Target detection device
The target detection device adjusts thresholds based on lateral position and dynamic factors to ensure accurate target identification across different sensors, addressing the challenge of targets moving relative to the vehicle's direction.
Patent Information
- Application Number
- PCT/JP2024/014968
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2024-04-15
- Publication Date
- 2025-10-23
AI Technical Summary
Existing target detection systems struggle to accurately determine that targets detected by different sensors, such as cameras and millimeter-wave radars, are the same when the difference in left-right positions becomes large due to targets moving relative to the vehicle's direction of travel.
A target detection device that adjusts the threshold for determining if targets detected by different sensors are the same based on the lateral position, relative speed, and other dynamic factors, such as lateral acceleration and direction changes, to ensure accurate identification.
Enables accurate determination of targets as the same even when they move left or right relative to the vehicle's direction, improving the reliability of target fusion and vehicle control systems.
Smart Images

Figure JP2024014968_23102025_PF_FP_ABST
Abstract
Description
Target detection device
[0001] The present invention relates to a target detection device that detects surrounding targets such as automobiles and pedestrians.
[0002] In recent years, automobiles have been equipped with target detection devices for the purposes of accident prevention, damage mitigation, driving support, etc. These target detection devices use sensors such as cameras that process images to detect targets and millimeter-wave radars that detect targets using reflected millimeter waves to detect people, other automobiles, etc. around the vehicle and use the information for vehicle control.
[0003] Furthermore, to improve the detection performance of target detection devices, technologies such as target fusion have been developed that detect targets by combining the detection results of multiple different sensors rather than just a single sensor. One conventional target fusion method first aligns the coordinate systems of the target information detected by each sensor to synchronize the time. Next, the forward / backward and left / right positions of the target detected by each sensor are compared to determine whether they are the same target. If they are determined to be the same target, the detection results of each sensor are combined to calculate target information.
[0004] When target fusion is performed, differences in the characteristics of each sensor cause differences in target position information. For example, when a target is located far away from the vehicle, the difference in the longitudinal position between the target detected by the camera and the target detected by the radar tends to be large due to the difference in longitudinal detection accuracy between the camera and the radar. Therefore, in Patent Document 1, when performing target fusion to determine whether the target is the same target, the threshold for the difference in longitudinal position used to determine whether the target is the same target is changed depending on whether the distance between the target and the vehicle is equal to or greater than a certain value, making it easier to determine whether the target is the same target.
[0005] Japanese Patent Application Laid-Open No. 2006-292475
[0006] In the method of Patent Document 1, by switching the threshold value used for determining the forward / backward direction, it becomes easier to determine that targets at long distances are the same, but the threshold value used for determining the left / right direction is fixed. Therefore, when a target moves left / right relative to the traveling direction of the host vehicle, such as when a vehicle traveling ahead of the host vehicle changes lanes or when a vehicle crosses in front of the host vehicle, the difference in the left / right positions of the targets detected by each of the multiple sensors becomes large, and it is not possible to correctly determine that the targets are the same.
[0007] The present invention has been made in consideration of the above circumstances, and its object is to provide a target detection device that can easily determine that the target is the same target even when the difference in the left-right positions of the targets detected by each sensor becomes large as the target moves left or right relative to the direction of travel of the vehicle.
[0008] In order to solve the above-mentioned problems, a target detection device according to the present invention acquires target information around the host vehicle detected by a first sensor and target information around the host vehicle detected by a second sensor, and, for targets determined to be the same in a processing frame prior to a predetermined processing frame, determines that the target detected by the first sensor and the target detected by the second sensor are the same target if the lateral position of the target detected by the second sensor is within a predetermined threshold from the lateral position of the target detected by the first sensor based on the target information detected by the first sensor and the target information detected by the second sensor in the predetermined processing frame, and changes the predetermined threshold for determining that the target is the same target.
[0009] According to the present invention, even when the difference in the left-right positions of the target detected by each sensor becomes large due to the target moving left or right relative to the vehicle's direction of travel, it becomes easier to determine that the target is the same target.
[0010] 1 is a block diagram of a driving assistance system according to an embodiment of the present invention. FIG. 1 is a flowchart illustrating target fusion processing. FIG. 2 is a diagram illustrating the differences in the longitudinal and lateral positions of a target detected by a camera and a millimeter-wave radar. FIG. 3 is a schematic diagram illustrating an example of a situation in which an embodiment of the present invention is effective. FIG. 4 is a diagram illustrating the change in the lateral position of a target when the target moves leftward in front of the host vehicle. FIG. 5 is a diagram illustrating the change in the lateral relative velocity of a target when the target moves leftward in front of the host vehicle. FIG. 6 is a flowchart illustrating an example of identical target determination processing according to an embodiment of the present invention. FIG. 7 is a schematic diagram before switching of the lateral position threshold in the identical target determination processing according to an embodiment of the present invention. FIG. 8 is a schematic diagram after switching of the lateral position threshold in the identical target determination processing according to an embodiment of the present invention. FIG. 9 is a diagram illustrating the change in the lateral relative acceleration of a target when the target moves leftward in front of the host vehicle. FIG. 10 is a diagram illustrating the change in the direction of a target as seen from the host vehicle when the target moves leftward in front of the host vehicle.
[0011] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.
[0012] [Embodiment 1] Embodiment 1 of the present invention will be described below. Fig. 1 shows an example of the configuration of a driving assistance system for implementing the present invention.
[0013] The driving assistance system 1 is composed of a camera 11 and a millimeter wave radar 12 as external sensors that detect surrounding targets, an ECU 13 that performs processing such as target fusion and calculation of vehicle control information, and a vehicle control unit 16 that controls the vehicle such as the accelerator, brake, and steering.
[0014] The camera 11, the millimeter wave radar 12, the ECU 13, and the vehicle control unit 16 transmit and receive data via CAN communication.
[0015] The camera 11 is attached inside the vehicle, for example, at the center of the windshield, and detects targets ahead of the vehicle. The camera 11 calculates target information such as the position and relative speed of the target based on the captured image, and transmits the target information to the ECU 13.
[0016] The millimeter-wave radar 12 is attached to, for example, the four corners of the vehicle and detects targets all around the vehicle. The millimeter-wave radar 12 calculates target information such as the position and relative speed of the target using millimeter waves and transmits the target information to the ECU 13.
[0017] The ECU 13 performs processes such as target object fusion and calculation of vehicle control information based on information obtained from various external sensors attached to the vehicle. The ECU 13 has, as internal functions, a target object fusion unit 14 that performs target object fusion processing and a control information calculation unit 15 that calculates vehicle control information.
[0018] The target fusion unit 14 is made up of a target data preprocessing unit 14a, a same target determination unit 14b, and a target fusion output value calculation unit 14c.
[0019] The control information calculation unit 15 includes an ACC control unit 15a for performing ACC (Adaptive Cruise Control) control, which causes the vehicle to travel at a constant speed when there is no preceding vehicle, and automatically maintains a safe distance from the preceding vehicle when there is a preceding vehicle, and an AEB control unit 15b for performing collision damage mitigation brake control.
[0020] The target fusion processing by the target fusion unit 14 will be explained with reference to FIG.
[0021] The processing flow of FIG. 2 is periodically executed by the target fusion unit 14.
[0022] First, the target data preprocessing unit 14a receives target information detected by the camera 11 and the millimeter wave radar 12 via CAN communication (S101).
[0023] Next, the target data pre-processing unit 14a unifies the coordinate systems of the received target information from the camera 11 and the target information from the millimeter-wave radar 12, and synchronizes the times. At this time, as shown in Fig. 3, the target data pre-processing unit 14a sets the coordinate system so that the center of the front end of the host vehicle 20 is the origin, the traveling direction of the host vehicle 20 is the Z axis, and the rightward direction as seen from the host vehicle 20 is the X axis. The target data pre-processing unit 14a also converts the target information so that the coordinates indicating the target position are the rear end of the target in the front-rear direction and the center of the target in the left-right direction, and calculates a camera target position 21 (hereinafter also referred to as a camera target) as the position of the target detected by the camera 11 and a radar target position 22 (hereinafter also referred to as a radar target) as the position of the target detected by the millimeter-wave radar 12 (S102).
[0024] Next, the same target determination unit 14b determines whether the targets detected by the camera 11 and the millimeter wave radar 12, which were determined to be the same in the previous processing frame before the current processing frame, are also the same in the current processing frame (S103).
[0025] Here, a method for the identical target determining unit 14b to determine that the targets detected by the camera 11 and the millimeter wave radar 12 are the same target will be described with reference to FIG.
[0026] The same target determination unit 14b compares the longitudinal positions and lateral positions of the camera target position 21 and the radar target position 22 as viewed from the host vehicle 20. The same target determination unit 14b determines that the target detected by the camera 11 and the target detected by the millimeter-wave radar 12 are the same when the difference ΔZ in the longitudinal positions between the camera target position 21 and the radar target position 22 is equal to or less than the longitudinal threshold value and the difference ΔX in the lateral positions is equal to or less than the lateral threshold value.
[0027] Here, consider a scene as shown in Fig. 4, in which a preceding vehicle 31, which is a target traveling in front of the vehicle 20 in the right lane, changes lanes ahead of the vehicle 20. Fig. 5 is a diagram showing the change in the left-right positions of the target detected by the camera 11 and the target detected by the millimeter-wave radar 12 in the lane change scene.
[0028] As can be seen from FIG. 5, when the preceding vehicle is changing lanes from the adjacent lane to the right into the own lane, there is a portion where the difference in the left-right positions of the camera target and the radar target becomes large.
[0029] 6 is a diagram showing the change in the relative lateral speed of each of the targets detected by the camera 11 and the millimeter-wave radar 12 in a scene where the preceding vehicle 31, which is a target traveling in front of the host vehicle 20 in the lane adjacent to the right, changes lanes ahead of the host vehicle 20 as described above. Comparing the relative lateral speeds of the camera target and the radar target from FIG. 6 reveals that there is a time period during which the difference in the relative lateral speeds also becomes large while the preceding vehicle is changing lanes from the lane adjacent to the right to the host vehicle's lane.
[0030] Therefore, in this embodiment, in S103 (FIG. 2), the same target determination unit 14b changes the left and right threshold values used for same target determination according to the processing flow shown in FIG. 7 using the difference in the left and right relative speeds.
[0031] First, the same target determination unit 14b determines whether the difference in the relative speed in the left and right directions as seen from the vehicle is less than a predetermined value for the targets detected by the camera 11 and the millimeter-wave radar 12 that were determined to be the same in the previous processing frame before the current processing frame (S201).
[0032] If the difference in the relative lateral speed between the target detected by the camera 11 and the target detected by the millimeter-wave radar 12 is less than a predetermined value, the same target determination unit 14b sets a predetermined threshold value (TH_X) as the threshold value for the difference in lateral position used for same target determination, as shown in FIG. 8A (S202).
[0033] If the difference in the left-right relative speed between the target detected by the camera 11 and the target detected by the millimeter-wave radar 12 is equal to or greater than a predetermined value, the same target determination unit 14b sets a threshold value (TH_Xw(>TH_X)) that is larger than the default threshold value (TH_X) as the left and right threshold values used for same target determination, as shown in FIG. 8B (S203).
[0034] Then, the same target determination unit 14b determines that the target detected by the camera 11 and the target detected by the millimeter-wave radar 12 are the same if the difference ΔZ in the forward / backward direction positions between the target detected by the camera 11 and the target detected by the millimeter-wave radar 12 is equal to or less than the forward / backward threshold and the difference ΔX in the left / right direction positions is equal to or less than the left / right threshold set in S202 or S203 (S204). In other words, the same target determination unit 14b determines that the target detected by the camera 11 and the target detected by the millimeter-wave radar 12 are the same target if the forward / backward direction position of the target detected by the millimeter-wave radar 12 is within the forward / backward threshold from the forward / backward direction position of the target detected by the camera 11 and the left / right direction position of the target detected by the millimeter-wave radar 12 is within the left / right threshold set in S202 or S203 from the left / right direction position of the target detected by the camera 11 (S204).
[0035] Finally, the target fusion output value calculation unit 14c calculates fusion target information such as the position information in the forward / backward and left / right directions of the target and the relative speed based on the result of the determination by the same target determination unit 14b (S104), and outputs the fusion target information to the control information calculation unit 15 of the ECU 13 (S105).
[0036] The ACC control unit 15a and the AEB control unit 15b of the control information calculation unit 15 calculate vehicle control information for the host vehicle based on the fusion target information (fusion result) output from the target fusion unit 14.
[0037] As in this embodiment, by determining that a target is moving in the left-right direction as viewed from the vehicle based on the difference in the relative left-right speed between the target detected by the camera 11 and the target detected by the millimeter-wave radar 12 and changing (increasing) the left and right threshold values used for determining the same target, it becomes easier to correctly determine that the target is the same regardless of whether the target is moving in the left-right direction.
[0038] In this embodiment, the camera 11 is a stereo camera or a monocular camera.
[0039] In addition, in this embodiment, the same target determination unit 14b determines whether or not the targets detected by the camera 11 and the millimeter-wave radar 12 that were determined to be the same in the previous processing frame before the current processing frame are also the same in the current processing frame, but it may also subsequently determine whether or not combinations that were determined to be not the same in the previous processing frame are the same targets.
[0040] In this embodiment, as shown in Figure 4, an example is given of a situation in which a preceding vehicle 31, which is a target traveling in front of the host vehicle 20 in the right lane, changes lanes in front of the host vehicle 20. However, the present invention can also be applied to other situations, such as when another vehicle crosses in front of the host vehicle or when the host vehicle turns right or left, causing the target to move left or right relative to the host vehicle, and further when the relative position of the target changes due to the influence of the road surface while driving on a rough road.
[0041] In this embodiment, the same target determination unit 14b performs determination using the relative lateral velocities of the target detected by the camera and the target detected by the millimeter-wave radar for each processing frame, and switches the left and right threshold values. However, once the enlarged threshold value (TH_Xw) is set as the left and right threshold values, it is also possible to use the enlarged threshold value (TH_Xw) for a certain number of processing frames thereafter without relying on the above determination.
[0042] In this embodiment, the same target determination unit 14b switches between two types of left and right thresholds used for same target determination: a default threshold (TH_X) and an expanded threshold (TH_Xw). However, it is also possible to switch between three types of thresholds, such as a threshold (TH_X) when the difference in left-right relative speed is less than A, a threshold (TH_Xw1) when the difference is A or more but less than B (>A), and a threshold (TH_Xw2) when the difference is B or more. It is also possible to switch between four or more types of thresholds.
[0043] Furthermore, the same target determination unit 14b can vary the threshold value, for example, by calculating the threshold value from target information detected by the camera 11 and target information detected by the millimeter-wave radar 12. For example, by utilizing the fact that the difference in the left-right position of the target can be calculated by integrating the difference in the relative left-right speed of the target over time, the same target determination unit 14b can calculate and set the threshold value from the difference in the relative left-right speed of the target detected by the camera 11 and the target detected by the millimeter-wave radar 12.
[0044] [Example 2] In Example 1, the same target determination unit 14b switches between the left and right thresholds used for the same target determination using the difference in the relative lateral speed between the target detected by the camera 11 and the target detected by the millimeter-wave radar 12. However, it is also possible to switch between the left and right thresholds used for the same target determination using the difference in the left and right positions of the targets (see FIG. 5) instead of the difference in the relative lateral speed of the targets.
[0045] [Example 3] In Example 1, the same target determination unit 14b switches between the left and right thresholds used for same target determination using the difference in the relative lateral speeds of the targets detected by the camera 11 and the targets detected by the millimeter-wave radar 12. However, it is also possible to switch between the left and right thresholds used for same target determination using the difference in the relative lateral accelerations of the targets instead of the difference in the relative lateral speeds of the targets.
[0046] 9 is a diagram showing changes in the lateral relative acceleration of the camera target position 21 (camera target) and the radar target position 22 (radar target) in a scene where a preceding vehicle, which is a target traveling in front of the host vehicle in the right lane, changes lanes in front of the host vehicle. From FIG. 9 , it can be seen that when the lateral relative accelerations of the target detected by the camera 11 and the target detected by the millimeter-wave radar 12 are compared, there are situations where the difference in lateral relative acceleration becomes large while the target is moving in the lateral direction. Therefore, the same target determination unit 14b can use the difference in the lateral relative accelerations of the targets to determine whether they are the same target and switch the left and right thresholds used for same target determination.
[0047] [Example 4] In Example 1, the same target determination unit 14b switches between the left and right thresholds used for the same target determination using the difference in the relative velocities in the left and right directions between the target detected by the camera 11 and the target detected by the millimeter-wave radar 12. However, it is also possible to switch between the left and right thresholds used for the same target determination using the difference in the direction of the target as viewed in a polar coordinate system centered on the vehicle itself, instead of the difference in the relative velocities in the left and right directions between the targets.
[0048] 10 is a diagram showing the change in the direction of the target detected by the camera 11 and the direction of the target detected by the millimeter-wave radar 12 as seen from the vehicle itself, in a scene where a preceding vehicle traveling in front of the vehicle in the right lane changes lanes ahead of the vehicle itself. From FIG. 10, it can be seen that when the directions of the target detected by the camera 11 and the directions of the target detected by the millimeter-wave radar 12 as seen from the vehicle itself are compared, there are situations where the difference in the direction as seen from the vehicle itself becomes large while the target is moving left or right. Therefore, the same target determination unit 14b can use the difference in the direction of the target as seen from the vehicle itself to determine whether the target is the same or different from the left or right threshold value used for the same target determination.
[0049] In addition, the same target determination unit 14b can also use the angular velocity and angular acceleration of the target detected by the camera 11 and the target detected by the millimeter-wave radar 12 as seen from the vehicle to determine whether the left and right thresholds are switched.
[0050] Fifth Embodiment In the first embodiment, the combination of external sensors is a camera and a millimeter wave radar, but this may be a combination of a camera and a LiDAR, for example.
[0051] [Summary] As described above, the target detection device (ECU 13) of this embodiment acquires target information around the vehicle detected by a first sensor and target information around the vehicle detected by a second sensor (target data pre-processing unit 14a), and for targets determined to be identical in a processing frame prior to a predetermined processing frame, determines whether the second sensor is located within a predetermined threshold (TH_X) from the left-right position of the target detected by the first sensor based on the target information detected by the first sensor and the target information detected by the second sensor in the predetermined processing frame. In the target detection device (ECU 13), which determines that the target detected by the first sensor and the target detected by the second sensor are the same target when the target detected by the sensors has a different lateral position (identical target determination unit 14b), the ECU 13 determines that the target is moving in the lateral direction as seen from the vehicle based on target information detected by the first sensor and the second sensor in the predetermined processing frame, and changes the predetermined threshold value (TH_X) for determining that the targets are the same target (enlarged to TH_Xw(>TH_X)).
[0052] Furthermore, the target detection device (ECU 13) compares the left-right positions of the targets as seen from the vehicle detected by the first sensor and the second sensor in the predetermined processing frame for targets determined to be the same in a processing frame prior to the predetermined processing frame, and if the difference in the left-right positions of the targets is equal to or greater than a predetermined value, determines that the targets are moving in the left-right direction as seen from the vehicle, and changes the predetermined threshold value (TH_X) for determining that the targets are the same (enlarges it to TH_Xw(>TH_X)).
[0053] Furthermore, the target detection device (ECU 13) compares the relative velocities in the left and right directions of the target as seen from the host vehicle, which are detected by the first sensor and the second sensor in the predetermined processing frame, with respect to targets determined to be the same in a processing frame prior to the predetermined processing frame, and if the difference in the relative velocities in the left and right directions of the target is equal to or greater than a predetermined value, determines that the target is moving in the left and right directions as seen from the host vehicle, and changes the predetermined threshold value (TH_X) for determining that the targets are the same (enlarges it to TH_Xw(>TH_X)).
[0054] Furthermore, the target detection device (ECU 13) compares the lateral relative accelerations of the target as seen from the host vehicle, which are detected by the first sensor and the second sensor in the predetermined processing frame, for targets determined to be the same in a processing frame prior to the predetermined processing frame, and if the difference between the lateral relative accelerations of the target is equal to or greater than a predetermined value, determines that the target is moving in the lateral direction as seen from the host vehicle, and changes the predetermined threshold value (TH_X) for determining that the targets are the same (enlarges it to TH_Xw(>TH_X)).
[0055] Furthermore, the target detection device (ECU 13) compares the directions of the targets as seen from the host vehicle detected by the first sensor and the second sensor in the predetermined processing frame for targets determined to be the same in a processing frame prior to the predetermined processing frame, and if the difference in the directions of the targets as seen from the host vehicle is equal to or greater than a predetermined value, determines that the target is moving in the left / right direction as seen from the host vehicle, and changes the predetermined threshold value (TH_X) for determining that the targets are the same (enlarges it to TH_Xw(>TH_X)).
[0056] Furthermore, for targets determined to be the same in a processing frame prior to the predetermined processing frame, the target detection device (ECU 13) compares the left-right positions of the targets as seen from the vehicle detected by the first sensor and the second sensor in the predetermined processing frame, determines that the targets are moving in the left-right direction as seen from the vehicle based on the difference in the left-right positions of the targets, and switches the predetermined threshold value for determining that the targets are the same (switching between TH_X and TH_Xw).
[0057] Furthermore, for targets determined to be the same in a processing frame prior to the predetermined processing frame, the target detection device (ECU 13) compares the relative velocities in the left and right directions of the targets as seen from the host vehicle detected by the first sensor and the second sensor in the predetermined processing frame, determines that the targets are moving in the left and right directions as seen from the host vehicle based on the difference between the relative velocities in the left and right directions of the targets, and switches the predetermined threshold value for determining that the targets are the same (switching between TH_X and TH_Xw).
[0058] Furthermore, for targets determined to be the same in a processing frame prior to the predetermined processing frame, the target detection device (ECU 13) compares the lateral relative accelerations of the targets as seen from the vehicle detected by the first sensor and the second sensor in the predetermined processing frame, determines that the targets are moving in the lateral direction as seen from the vehicle based on the difference between the lateral relative accelerations of the targets, and switches the predetermined threshold value for determining that the targets are the same (switching between TH_X and TH_Xw).
[0059] Furthermore, for targets determined to be the same in a processing frame prior to the predetermined processing frame, the target detection device (ECU 13) compares the directions of the targets as seen from the host vehicle detected by the first sensor and the second sensor in the predetermined processing frame, determines that the targets are moving in the left / right direction as seen from the host vehicle based on the difference in the directions of the targets as seen from the host vehicle, and switches the predetermined threshold value for determining that the targets are the same (switching between TH_X and TH_Xw).
[0060] In other words, when a target is moving in the left-right direction, the target detection device (ECU13) of this embodiment determines whether the target is moving in the left-right direction as seen from the vehicle, and if it is moving, increases the threshold value for the difference in left-right position at which it is determined that the targets detected by the sensors are the same, since differences in the left-right positions of the target detected by different sensors may make it impossible to correctly determine whether the targets detected by each sensor are the same.
[0061] According to this embodiment, even when the difference in the left-right positions of the target detected by each sensor becomes large due to the target moving left or right relative to the vehicle's direction of travel, it is possible to easily determine that the target is the same target.
[0062] It should be noted that the present invention is not limited to the above-described embodiment, and includes various modifications. For example, the above-described embodiment has been described in detail to clearly explain the present invention, and the present invention is not necessarily limited to an embodiment having all of the described configurations.
[0063] Furthermore, the above-described configurations, functions, processing units, processing means, etc. may be partially or entirely implemented in hardware, for example, by designing them as integrated circuits. The above-described configurations, functions, etc. may also be implemented in software, with a processor interpreting and executing a program that implements each function. Information such as the programs, tables, and files that implement each function can be stored in a memory, a recording device such as a hard disk or SSD (Solid State Drive), or a recording medium such as an IC card, SD card, or DVD.
[0064] In addition, the control lines and information lines shown are those that are considered necessary for the explanation, and do not necessarily show all the control lines and information lines in the product. In reality, it can be assumed that almost all components are interconnected.
[0065] REFERENCE SIGNS LIST 1 Driving assistance system 11 Camera (first sensor) 12 Millimeter wave radar (second sensor) 13 ECU (target detection device) 14 Target fusion unit 15 Control information calculation unit 16 Vehicle control unit 20 Vehicle 21 Target detected by camera (camera target position) 22 Target detected by millimeter wave radar (radar target position) 31 Leading vehicle as target
Claims
1. A target detection device that acquires target information around the vehicle detected by a first sensor and target information around the vehicle detected by a second sensor, and determines that the target detected by the first sensor and the target detected by the second sensor are the same target when the left-right position of the target detected by the second sensor is within a predetermined threshold from the left-right position of the target detected by the first sensor based on the target information detected by the first sensor and the target information detected by the second sensor in the predetermined processing frame, characterized in that the target detection device determines that the target is moving in the left-right direction as seen from the vehicle based on the target information detected by the first sensor and the second sensor in the predetermined processing frame, and changes the predetermined threshold for determining that the targets are the same.
2. The target detection device according to claim 1, characterized in that for targets determined to be the same in a processing frame prior to the predetermined processing frame, the target detection device compares the left-right positions of the targets as seen from the vehicle detected by the first sensor and the second sensor in the predetermined processing frame, and if the difference in the left-right positions of the targets is equal to or greater than a predetermined value, determines that the targets are moving in the left-right direction as seen from the vehicle, and changes the predetermined threshold for determining that the targets are the same.
3. The target detection device according to claim 1, characterized in that, for targets determined to be the same in a processing frame prior to the predetermined processing frame, the target detection device compares the relative lateral speeds of the targets as seen from the vehicle detected by the first sensor and the second sensor in the predetermined processing frame, and if the difference in the relative lateral speeds of the targets is equal to or greater than a predetermined value, determines that the targets are moving in the left or right direction as seen from the vehicle, and changes the predetermined threshold for determining that the targets are the same.
4. The target detection device according to claim 1, characterized in that for targets determined to be the same in a processing frame prior to the predetermined processing frame, the target detection device compares the lateral relative acceleration of the target as seen from the vehicle detected by the first sensor and the second sensor in the predetermined processing frame, and if the difference in the lateral relative acceleration of the target is equal to or greater than a predetermined value, determines that the target is moving in the lateral direction as seen from the vehicle, and changes the predetermined threshold for determining that the targets are the same.
5. The target detection device according to claim 1, characterized in that, for targets determined to be the same in a processing frame prior to the predetermined processing frame, the target detection device compares the directions of the targets as seen from the vehicle detected by the first sensor and the second sensor in the predetermined processing frame, and if the difference in the directions of the targets as seen from the vehicle is equal to or greater than a predetermined value, determines that the targets are moving in the left / right direction as seen from the vehicle, and changes the predetermined threshold value for determining that the targets are the same.
6. The target detection device according to claim 1, characterized in that, for targets determined to be the same in a processing frame prior to the predetermined processing frame, the target detection device compares the left-right positions of the targets as seen from the vehicle detected by the first sensor and the second sensor in the predetermined processing frame, determines that the targets are moving in the left-right direction as seen from the vehicle based on the difference in the left-right positions of the targets, and switches the predetermined threshold value for determining that the targets are the same.
7. The target detection device according to claim 1, characterized in that, for targets determined to be the same in a processing frame prior to the predetermined processing frame, the target detection device compares the relative lateral speeds of the targets as seen from the vehicle detected by the first sensor and the second sensor in the predetermined processing frame, determines that the targets are moving in the left-right direction as seen from the vehicle based on the difference in the relative lateral speeds of the targets, and switches the predetermined threshold value for determining that the targets are the same.
8. The target detection device according to claim 1, characterized in that, for targets determined to be the same in a processing frame prior to the predetermined processing frame, the target detection device compares the lateral relative acceleration of the target as seen from the vehicle detected by the first sensor and the second sensor in the predetermined processing frame, determines that the target is moving in the lateral direction as seen from the vehicle based on the difference in the lateral relative acceleration of the target, and switches the predetermined threshold for determining that the targets are the same.
9. The target detection device according to claim 1, characterized in that, for targets determined to be the same in a processing frame prior to the predetermined processing frame, the target detection device compares the directions of the targets as seen from the vehicle detected by the first sensor and the second sensor in the predetermined processing frame, and determines that the targets are moving in the left / right direction as seen from the vehicle based on the difference in the directions of the targets as seen from the vehicle, and switches the predetermined threshold value for determining that the targets are the same.
10. The target detection device according to claim 1, wherein the first sensor is a camera and the second sensor is a millimeter wave radar.
11. The target detection device according to claim 1, wherein the first sensor is a camera and the second sensor is a LiDAR.
Citation Information
Patent Citations
Collision determination device and collision determination method
US20150206435A1
Object detection device
WO2013108664A1