Obstacle recognition device, obstacle recognition method, and program

Abstract

The invention relates to an obstacle recognition device, an obstacle recognition method, and a program. The obstacle recognition device acquires, from a millimeter-wave radar mounted on a vehicle, information relating to a target object present in front of the vehicle detected by the millimeter-wave radar, and performs a first determination and a second determination on the basis of the information relating to the target object. In the first determination, it is determined whether or not a target is present in a first region in which the distance from the preceding vehicle is less than a predetermined value, and in the second determination, it is determined whether or not the target that is determined to be present in the first region in the first determination is detected by the millimeter-wave radar when the target is present in a second region in which the distance from the preceding vehicle is equal to or greater than the predetermined value. If it is determined in the first determination that the target is present in the first region and if it is determined in the second determination that the target is not detected by the millimeter-wave radar when present in the second region, information relating to the target is discarded.

Description

Technical Field

[0001] This invention relates to an obstacle recognition device, obstacle recognition method, and procedure. Background Technology

[0002] Japanese Patent No. 6473685 discloses an autonomous vehicle that travels at a predetermined speed along a predetermined path within a mine. Furthermore, Japanese Patent No. 6473685 describes the characteristic of millimeter-wave radar having difficulty distinguishing whether detected objects are road imperfections, fallen rocks, or manned vehicles. Therefore, the autonomous vehicle described in Japanese Patent No. 6473685 utilizes both millimeter-wave radar and lidar (LiDAR). Summary of the Invention

[0003] LiDAR is less effective in harsh environments such as sandstorms, rain, fog, and backlighting. Therefore, to ensure environmental robustness, it is preferable to improve the recognition accuracy of individual millimeter-wave radar units (i.e., to improve obstacle recognition accuracy using only millimeter-wave radar without using LiDAR).

[0004] On the other hand, when using millimeter-wave radar alone without LiDAR, the millimeter-wave radar identifies reflected waves from road surface irregularities, small stones, etc., near unpaved roads as targets (obstacles). This is because the reflected waves from nearby road surface irregularities, small stones, etc., are large. Furthermore, without LiDAR, millimeter-wave radar cannot distinguish between road surface irregularities, small stones, etc., which vehicles do not need to avoid colliding with, and targets such as rocks of a certain height, other vehicles, or pedestrians, which vehicles need to avoid colliding with. This is because millimeter-wave radar has low height resolution. Therefore, if information is based solely on targets detected by millimeter-wave radar in front of the vehicle, it may be impossible to avoid collisions with obstacles and allow the vehicle to drive appropriately.

[0005] In view of the above, the purpose of the present invention is to provide an obstacle recognition device, obstacle recognition method and program, which can distinguish between targets that the vehicle does not need to avoid colliding with and targets that the vehicle needs to avoid colliding with, based on information related to targets existing in front of the vehicle detected by millimeter-wave radar.

[0006] (1) One aspect of the present invention is an obstacle recognition device, which comprises:

[0007] The acquisition unit acquires information related to a target object detected by the millimeter-wave radar mounted on the vehicle that exists in front of the vehicle.

[0008] The determination unit performs a first determination and a second determination based on information related to the target object acquired by the acquisition unit. The first determination determines whether the target object exists in a first region (a region where the distance from the moving vehicle is less than a predetermined value). The second determination determines whether the target object, determined to exist in the first region in the first determination, is detected by the millimeter-wave radar if it exists in a second region (a region where the distance from the moving vehicle is greater than or equal to the predetermined value).

[0009] The output unit discards information related to the target object when the first determination determines that the target object exists in the first region and the second determination determines that the target object is not detected by the millimeter-wave radar when it exists in the second region; and outputs information related to the target object when the first determination determines that the target object exists in the first region and the second determination determines that the target object is detected by the millimeter-wave radar when it exists in the second region.

[0010] (2) In the obstacle recognition device of (1), the information related to the target object includes the target object location information and tracking information.

[0011] (3) In the obstacle recognition device of (1), the determination area information is adjusted according to the usage environment of the vehicle. The determination area information is information that defines the first area where the target object may exist, corresponding to the information discarded by the output unit.

[0012] (4) One aspect of the present invention is an obstacle recognition method, which comprises the following steps:

[0013] In the acquisition step, the obstacle recognition device acquires information related to the target object in front of the vehicle detected by the millimeter-wave radar mounted on the vehicle.

[0014] In the determination step, the obstacle recognition device performs a first determination and a second determination based on the information related to the target object acquired in the acquisition step. The first determination determines whether the target object exists in a first region (distance from the moving vehicle less than a predetermined value). The second determination determines whether the target object, determined to exist in the first region in the first determination, is detected by the millimeter-wave radar if it exists in a second region (distance from the moving vehicle greater than the predetermined value).

[0015] In the output step, if the obstacle recognition device determines in the first determination that the target object exists in the first area and in the second determination that the target object is not detected by the millimeter-wave radar when it exists in the second area, it discards information related to the target object; if the obstacle recognition device determines in the first determination that the target object exists in the first area and in the second determination that the target object is detected by the millimeter-wave radar when it exists in the second area, it outputs information related to the target object.

[0016] (5) One aspect of the present invention is a program that causes a processor to perform the following steps:

[0017] The acquisition step involves acquiring information related to a target object detected by the millimeter-wave radar mounted on the vehicle that exists in front of the vehicle.

[0018] The determination step involves performing a first determination and a second determination based on the information related to the target object acquired in the acquisition step. The first determination determines whether the target object exists in a first region (a region where the distance from the moving vehicle is less than a predetermined value). The second determination determines whether the target object, determined to exist in the first region in the first determination, is detected by the millimeter-wave radar if it exists in a second region (a region where the distance from the moving vehicle is greater than the predetermined value).

[0019] In the output step, if the target is determined to exist in the first region in the first determination and is not detected by the millimeter-wave radar when it exists in the second region in the second determination, information related to the target is discarded; if the target is determined to exist in the first region in the first determination and is detected by the millimeter-wave radar when it exists in the second region in the second determination, information related to the target is output.

[0020] According to the present invention, it is possible to distinguish between targets that the vehicle does not need to avoid colliding with and targets that the vehicle needs to avoid colliding with, based on information related to targets present in front of the vehicle detected by millimeter-wave radar. Attached Figure Description

[0021] Hereinafter, the features, advantages, and technical and industrial significance of exemplary embodiments of the present invention will be described with reference to the accompanying drawings, in which the same reference numerals denote the same elements, and wherein:

[0022] Figure 1 This is a diagram showing an example of a vehicle 1 to which the obstacle recognition device 15 of the first embodiment is applied.

[0023] Figure 2 It means Figure 1 The diagram shows an example of data flow within vehicle 1.

[0024] Figure 3 This is a flowchart illustrating an example of the processing performed by the processor 153 of the obstacle recognition device 15 of the first embodiment.

[0025] Figure 4A It is used for explanation Figure 3 The diagram shows a specific example of the processing.

[0026] Figure 4B It is used for explanation Figure 3 The diagram shows another specific example of the processing. Detailed Implementation

[0027] Hereinafter, with reference to the accompanying drawings, embodiments of the obstacle recognition device, obstacle recognition method and program of the present invention will be described.

[0028] <First Embodiment>

[0029] Figure 1 This is a diagram showing an example of a vehicle 1 to which the obstacle recognition device 15 of the first embodiment is applied. Figure 2 It means Figure 1 The diagram shows an example of data flow within vehicle 1.

[0030] exist Figure 1 and Figure 2 In the example shown, vehicle 1 includes millimeter-wave radar 11, human-machine interface (HMI) 12, vehicle status sensor 13, location information acquisition device 14, obstacle recognition device 15, vehicle control device 16, steering actuator 16A, braking actuator 16B, and drive actuator 16C.

[0031] The millimeter-wave radar 11 is, for example, disposed at the front of the vehicle 1. The millimeter-wave radar 11 detects targets TG1 to TG6 present in front of the vehicle 1 (see reference). Figure 4A and Figure 4B And send the information (sensor data) related to the target objects TG1 to TG6 to the obstacle recognition device 15.

[0032] In their in-depth research, the inventors discovered that when vehicle 1 is traveling on an unpaved road, millimeter-wave radar 11 detects a large rock that vehicle 1 needs to avoid colliding with (i.e., that requires control of steering actuator 16A and braking actuator 16B) as a target object TG5 (see reference). Figure 4A and Figure 4BFurthermore, the inventors discovered that TG6, which detects other vehicles as targets (see reference...). Figure 4A and Figure 4B Furthermore, the inventors discovered that small bumps, pebbles, etc., on the road surface where the vehicle 1 does not need to avoid collisions are detected as target objects TG1 to TG4 (see reference). Figure 4A and Figure 4B (That is, receiving reflected waves from small bumps, pebbles, etc.). Vehicle 1 does not need to avoid collisions, meaning there is no need to control the steering actuator 16A or the brake actuator 16B. Furthermore, the inventors discovered that small bumps, pebbles, etc., on the road surface exist in areas where the distance from the moving vehicle 1 is less than a predetermined value, namely, the first area AR1 (see reference). Figure 4A and Figure 4B Within this area, the millimeter-wave radar 11 detected objects TG1 to TG4. Furthermore, the inventors discovered that small bumps and pebbles on the road surface exist in an area, namely the second area AR2 (see reference), at a distance of more than a predetermined value from the moving vehicle 1. Figure 4A and Figure 4B When within the range of vehicle 1, the vehicles will not be detected as targets TG1 to TG4 by the millimeter-wave radar 11. The first area is the area near vehicle 1, and the second area is the area away from vehicle 1.

[0033] Therefore, in Figure 1 and Figure 2 In the example shown, in order to distinguish between targets TG1 to TG4 that vehicle 1 does not need to avoid colliding with and targets TG5 and TG6 that vehicle 1 needs to avoid colliding with, the countermeasures described later were implemented.

[0034] exist Figure 1 and Figure 2 In the example shown, the human-machine interface 12 has the function of receiving various operations from the user of vehicle 1 and sending signals representing the user's operations to the vehicle control device 16, etc. The operations received by the human-machine interface 12 from the user of vehicle 1 include, for example, instructing the vehicle control device 16 to perform automatic driving of vehicle 1, or switching automatic driving of vehicle 1 to manual driving. The vehicle status sensor 13 includes, for example, a vehicle speed sensor.

[0035] The vehicle status sensor 13 sends information indicating the status of the vehicle 1 (such as vehicle speed) to the obstacle recognition device 15, the vehicle control device 16, etc.

[0036] The location information acquisition device 14 acquires information indicating the location of vehicle 1. The location information acquisition device 14 may include, for example, a Global Positioning System (GPS) device for measuring the location of vehicle 1. The location information acquisition device 14 can perform known self-position estimation processing (localization) to improve the accuracy of the information indicating the location of vehicle 1. The location information acquisition device 14 then transmits the information indicating the location of vehicle 1 to the obstacle recognition device 15, the vehicle control device 16, etc.

[0037] The vehicle control device 16 is, for example, composed of a vehicle control electronic control unit (ECU). The vehicle control device 16 controls the steering actuator 16A, brake actuator 16B, and drive actuator 16C based on information (signals) sent from the human-machine interface 12, vehicle status sensor 13, position information acquisition device 14, obstacle recognition device 15, etc. The vehicle control device 16 has the function of performing automatic driving of the vehicle 1.

[0038] The obstacle recognition device 15 is composed of a microcomputer with a communication interface (I / F) 151, a memory 152 and a processor 153.

[0039] The communication interface 151 has an interface circuit for connecting the obstacle recognition device 15 to the millimeter-wave radar 11, the human-machine interface 12, the vehicle status sensor 13, the location information acquisition device 14, the vehicle control device 16, etc.

[0040] Memory 152 stores programs and various data used in the processing executed by processor 153. The data stored in memory 152 includes, for example, decision region information (see reference). Figure 2 The determination area information is the first area AR1 (reference) where there are targets TG1 to TG4 that do not require collision avoidance with vehicle 1 (the road surface does not need targets). Figure 4A and Figure 4B Information regarding parameters (e.g., size) is specified. The range within which the millimeter-wave radar 11 can detect targets TG1 to TG4 that the vehicle 1 does not need to avoid colliding with (and the road surface does not need targets (e.g., small bumps, pebbles, etc. on unpaved roads)) varies depending on the condition or material of the road surface on which the vehicle 1 is traveling. Therefore, parameters related to the first region AR1 are adjusted, for example, by the user of the vehicle 1 based on the usage environment of the vehicle 1. That is, the memory 152 stores, for example, the determination region information adjusted by the user of the vehicle 1 based on the usage environment of the vehicle 1.

[0041] The processor 153 functions as an acquisition unit 3A, a determination unit 3B, and an output unit 3C.

[0042] Acquisition unit 3A acquires targets TG1 to TG6 (reference) in front of vehicle 1 from millimeter-wave radar 11 and detected by millimeter-wave radar 11. Figure 4A and Figure 4B Information related to targets TG1 to TG6 (specifically, time-series data). Information related to targets TG1 to TG6 includes target position information (information indicating the relative position of targets TG1 to TG6 relative to vehicle 1) and tracking information. Tracking information is information that can distinguish whether targets TG1 to TG6 output from millimeter-wave radar 11 in a time-series format are the same. (Further details will follow.) Figure 4A and Figure 4B In the example shown, it is possible to identify in Figure 4A The targets TG5 and TG6, detected by millimeter-wave radar 11 at past points in time, are shown below. Figure 4B The targets TG5 and TG6 detected by the millimeter-wave radar 11 at the current time point are identical. This identification is based on the tracking information output from the millimeter-wave radar 11.

[0043] The determination unit 3B determines the target objects TG1 to TG6 (reference) obtained by the acquisition unit 3A. Figure 4B Information related to the execution of targets TG1 to TG6, including whether they exist in region AR1 (see reference). Figure 4B The determination within the first region AR1 (reference 1) is further performed by the determination unit 3B. Figure 4B Targets TG1 to TG5 within (reference) Figure 4B ) exists in region 2 AR2 (reference) Figure 4A The determination (second determination) is based on whether the target TG1 to TG6 (reference) acquired by the acquisition unit 3A is detected by the millimeter-wave radar 11. Figure 4A and Figure 4B (This is based on relevant information.)

[0044] Output unit 3C discards information related to target objects TG1 to TG4. This process is determined to be target objects TG1 to TG4 in the first determination performed by determination unit 3B (see reference). Figure 4B It exists in region 1, AR1 (reference). Figure 4B Within the scope of the second determination performed by the determination unit 3B, the objects TG1 to TG4 are determined to be targets (see reference). Figure 4A ) exists in region 2 AR2 (reference) Figure 4AIf targets TG1 to TG4 are not detected by millimeter-wave radar 11 within a certain time period, the output unit 3C outputs information related to target TG5 to vehicle control device 16. This processing is determined to be target TG5 (see reference) in the first determination performed by determination unit 3B. Figure 4B It exists in region 1, AR1 (reference). Figure 4B Within and in the second determination performed by the determination unit 3B, it is determined to be the target object TG5 (refer to...). Figure 4A ) exists in region 2 AR2 (reference) Figure 4A This is carried out when the target TG5 is detected by the millimeter-wave radar 11.

[0045] Figure 3 This is a flowchart illustrating an example of the processing performed by the processor 153 of the obstacle recognition device 15 of the first embodiment.

[0046] Figure 3 The process shown is performed, for example, while vehicle 1 is in motion (specifically, while moving forward).

[0047] exist Figure 3 In the example shown, in S10, the acquisition unit 3A acquires information (time series data) related to the target object existing in front of the vehicle 1 detected by the millimeter-wave radar 11 from the millimeter-wave radar 11.

[0048] In S11, the determination unit 3B, based on the information related to the target object obtained in S10 (specifically, the data at the current time point in the time series data), determines whether the target object exists in the first region AR1 (see reference). Figure 4B The decision within (first decision). If "yes", proceed to S12; if "no", proceed to S14.

[0049] In S12, the determination unit 3B, based on the information related to the target object obtained in S10, executes the determination in S11 that the target object, which was determined to exist in the first region AR1, exists in the second region AR2 (see reference). Figure 4A The determination of whether the target was detected by the millimeter-wave radar 11 within the specified time (second determination). The information related to the target acquired in S10 is specifically time-series data including data at the current time point and data at past time points compared to the current time point. If "yes", proceed to S14; if "no", proceed to S13.

[0050] In S13, the output unit 3C discards information related to the target object.

[0051] In S14, the output unit 3C outputs information related to the target object to the vehicle control device 16.

[0052] Figure 4A and Figure 4B It is used for explanation Figure 3 The diagram shows a specific example of the processing. In detail, Figure 4A This example illustrates the positional relationship between vehicle 1, targets TG1 to TG6, region AR1, and region AR2 at a past point in time. Figure 4B This example illustrates the positional relationship between vehicle 1, targets TG1 to TG6, region AR1, and region AR2 at the current time.

[0053] exist Figure 4A and Figure 4B In the example shown, Figure 3 In S10, the acquisition unit 3A acquires data from millimeter-wave radar 11, for example, from... Figure 4A The past time point shown Figure 4B The data shown represents the time series data of targets TG1 to TG6 detected by millimeter-wave radar 11 at the current time point. This data is acquired as information related to targets TG1 to TG6 located in front of vehicle 1, as detected by millimeter-wave radar 11.

[0054] exist Figure 3 In S11, the determination unit 3B determines the millimeter-wave radar 11 based on the time-series data acquired in S10. Figure 4B The data of targets TG1 to TG6 detected at the current time point are shown, and targets TG1 to TG5 are determined to exist in the first area AR1. Furthermore, the determination unit 3B determines that target TG6 (another vehicle) does not exist in the first area AR1.

[0055] exist Figure 4A and Figure 4B In the example shown, Figure 4B At the current point in time, there is no risk of collision between vehicle 1 and target TG6 (another vehicle). Furthermore, unnecessary control actions by the steering actuator 16A or braking actuator 16B, used to avoid a collision between vehicle 1 and target TG6, can be prevented. Therefore, in Figure 3 In S14, information related to the target TG6 is not discarded but output to the vehicle control device 16.

[0056] Furthermore, in Figure 4A and Figure 4B In the example shown, Figure 3 In S12, the determination unit 3B determines that it is determined to be in Figure 4B The target TG5 (large rock) that exists in region AR1 at the current time point is shown as being in region AR2. Figure 4A The past time points shown were also detected by millimeter-wave radar 11. This determination is based on the data acquired in S10 from... Figure 4A The past time point shown Figure 4B The current time point shown is based on the time series data of targets TG1 to TG6 detected by the millimeter-wave radar 11.

[0057] In the example shown in Figure 4, Figure 4B At the current point in time, there is a risk of collision between vehicle 1 and target object TG5 (large rock). Therefore, control of the steering actuator 16A and / or braking actuator 16B is required to avoid a collision between vehicle 1 and target object TG5. Therefore, in Figure 3 In S14, information related to the target TG5 is not discarded but output to the vehicle control device 16.

[0058] The vehicle control unit 16 executes control to activate the steering actuator 16A and / or the brake actuator 16B based on the information related to the target object TG5 output from the output unit 3C, so as to avoid a collision between the vehicle 1 and the target object TG5.

[0059] On the other hand, Figure 4A and Figure 4B In the example shown, Figure 3 In S12, the determination unit 3B determines that it is determined to be in Figure 4B The targets TG1 to TG4, which exist in region AR1 at the current time, also exist in region AR2. Figure 4A The past time points shown were not detected by millimeter-wave radar 11. This determination is based on the data acquired in S10 from... Figure 4A The past time point shown Figure 4B The current time point shown is determined by the time sequence data of targets TG1 to TG6 detected by millimeter-wave radar 11. Targets TG1 to TG4 are small bumps and pebbles on the unpaved road surface. That is, the determination unit 3B determines that targets TG1 to TG4 existing in the first region AR1 at the current time point are small bumps and pebbles on the unpaved road surface, and there is no need to avoid collision between vehicle 1 and targets TG1 to TG4.

[0060] That is, in Figure 4A and Figure 4B In the example shown, control of the steering actuator 16A and / or brake actuator 16B for avoiding collisions between vehicle 1 and targets TG1 to TG4 is not required. Therefore, information related to targets TG1 to TG4 is... Figure 3 It is discarded in S13 and is not output to the vehicle control unit 16.

[0061] As a result, the vehicle control unit 16 does not perform unnecessary controls to activate the steering actuator 16A and / or the brake actuator 16B in order to avoid collisions between the vehicle 1 and the targets TG1 to TG4.

[0062] That is, such as Figure 4B As shown, when vehicle 1 is traveling on an unpaved road (moving forward), millimeter-wave radar 11 receives reflected waves from small bumps, small stones, etc. (targets TG1 to TG4) on the road surface near vehicle 1, and detects targets TG1 to TG4. The vicinity of vehicle 1 is within the first area AR1.

[0063] like Figure 4B As shown, it is necessary to distinguish and identify the large rock (target TG5) near vehicle 1 that needs to be avoided from colliding with vehicle 1, and the small bumps and pebbles on the road surface (targets TG1 to TG4) that do not need to be avoided from colliding with vehicle 1. The vicinity of vehicle 1 is within the first area AR1.

[0064] Therefore, in Figure 4A and Figure 4B In the example shown, small bumps, pebbles, etc. (targets TG1 to TG4) on the road surface are detected by the millimeter-wave radar 11 when they are located within the first region AR1, which is less than a predetermined distance from the moving vehicle 1. However, the feature and tracking information that they are not detected by the millimeter-wave radar 11 when they are located within the second region AR2, which is more than the predetermined distance from the moving vehicle 1, are also used.

[0065] As a result, it is able to distinguish between large rocks (target TG5) near vehicle 1 that require avoiding collisions with vehicle 1, and small bumps and pebbles on the road surface (targets TG1 to TG4) that do not require avoiding collisions with vehicle 1. The vicinity of vehicle 1 is within the first area AR1.

[0066] <Second Implementation>

[0067] The obstacle recognition device 15 of the second embodiment is configured in the same way as the obstacle recognition device 15 of the first embodiment, except for the points described later.

[0068] As described above, the obstacle recognition device 15 of the first embodiment is provided on the vehicle 1.

[0069] On the other hand, the obstacle recognition device 15 of the second embodiment can be disposed outside the vehicle 1, such as a control unit that controls the movement of the vehicle 1.

[0070] <Third Implementation>

[0071] The vehicle 1 using the obstacle recognition device 15 of the third embodiment is configured in the same way as the vehicle 1 using the obstacle recognition device 15 of the first embodiment, except for the points described later.

[0072] As described above, in the vehicle 1 (autonomous driving vehicle) to which the obstacle recognition device 15 of the first embodiment is applied, the vehicle control device 16 performs control to activate the steering actuator 16A and / or the braking actuator 16B to avoid a collision between the vehicle 1 and the target TG5. This control is performed based on information related to the target TG5 output from the output unit 3C.

[0073] On the other hand, in the vehicle 1 equipped with the obstacle recognition device 15 of the third embodiment, a warning indicating that an operation is required to avoid a collision between the vehicle 1 and the target object TG5 is output to the human-machine interface 12. This output is based on information related to the target object TG5 output from the output unit 3C by the vehicle control device 16.

[0074] As described above, embodiments of the obstacle recognition device, obstacle recognition method, and program of the present invention have been explained with reference to the accompanying drawings. However, the obstacle recognition device, obstacle recognition method, and program of the present invention are not limited to the above embodiments, and appropriate modifications can be made without departing from the spirit of the present invention. The structures of the various examples of the above embodiments can be appropriately combined. In the examples of the above embodiments, the processing performed in the obstacle recognition device 15 has been described as software processing performed by executing a program. However, the processing performed in the obstacle recognition device 15 may also be hardware processing. Alternatively, the processing performed in the obstacle recognition device 15 may be a combination of software and hardware processing. Furthermore, the program stored in the memory 152 of the obstacle recognition device 15 may be recorded in a computer-readable storage medium such as a semiconductor memory, magnetic recording medium, or optical recording medium for provision and distribution. This program is a program that implements the functions of the processor 153 of the obstacle recognition device 15.

Claims

1. An obstacle recognition device, characterized in that, have: The acquisition unit acquires information related to a target object detected by the millimeter-wave radar mounted on the vehicle that exists in front of the vehicle. The determination unit performs a first determination and a second determination based on information related to the target object acquired by the acquisition unit. The first determination determines whether the target object exists in a first region where the distance from the moving vehicle is less than a predetermined value. The second determination determines whether the target object, which was determined to exist in the first region in the first determination, is detected by the millimeter-wave radar when it exists in a second region where the distance from the moving vehicle is greater than or equal to the predetermined value. and The output unit discards information related to the target object when the first determination determines that the target object exists in the first region and the second determination determines that the target object is not detected by the millimeter-wave radar when it exists in the second region; and outputs information related to the target object when the first determination determines that the target object exists in the first region and the second determination determines that the target object is detected by the millimeter-wave radar when it exists in the second region.

2. The obstacle recognition device according to claim 1, characterized in that, Information related to the target object includes the target object's location information and tracking information.

3. The obstacle recognition device according to claim 1, characterized in that, The determination area information is adjusted according to the vehicle's usage environment. The determination area information is information that defines the first area where the target object may exist, corresponding to the information discarded by the output unit.

4. An obstacle recognition method, characterized in that, Includes the following steps: In the acquisition step, the obstacle recognition device acquires information related to the target object in front of the vehicle detected by the millimeter-wave radar mounted on the vehicle. In the determination step, the obstacle recognition device performs a first determination and a second determination based on the information related to the target object acquired in the acquisition step. The first determination determines whether the target object exists in a first region where the distance from the moving vehicle is less than a predetermined value. The second determination determines whether the target object, which was determined to exist in the first region in the first determination, is detected by the millimeter-wave radar when it exists in a second region where the distance from the moving vehicle is greater than or equal to the predetermined value. and In the output step, if the obstacle recognition device determines in the first determination that the target object exists in the first area and in the second determination that the target object is not detected by the millimeter-wave radar when it exists in the second area, it discards information related to the target object; if the obstacle recognition device determines in the first determination that the target object exists in the first area and in the second determination that the target object is detected by the millimeter-wave radar when it exists in the second area, it outputs information related to the target object.

5. A program that causes a processor to perform the following steps: The acquisition step involves acquiring information related to a target object detected by the millimeter-wave radar mounted on the vehicle that exists in front of the vehicle. The determination step involves performing a first determination and a second determination based on the information related to the target object obtained in the acquisition step. The first determination determines whether the target object exists in a region (i.e., a first region) where the distance from the moving vehicle is less than a predetermined value. The second determination determines whether the target object, which was determined to exist in the first region in the first determination, is detected by the millimeter-wave radar when it exists in a region (i.e., a second region) where the distance from the moving vehicle is greater than or equal to the predetermined value. and In the output step, if the target is determined to exist in the first region in the first determination and is not detected by the millimeter-wave radar when it exists in the second region in the second determination, information related to the target is discarded; if the target is determined to exist in the first region in the first determination and is detected by the millimeter-wave radar when it exists in the second region in the second determination, information related to the target is output.

Images