JP7119626B2 - Target object detection method and device for vehicle - Google Patents

Description

The present invention relates to technology for detecting a target existing on the travel route of a vehicle.

2. Description of the Related Art In recent years, vehicle-to-vehicle communication, in which wireless communication is performed between a plurality of vehicles traveling on a road, has attracted attention. Vehicle-to-vehicle communication is used, for example, at an intersection with poor visibility so that vehicles entering the intersection exchange information such as the position and speed of each vehicle with each other. This makes it possible to determine the possibility of a collision and perform various controls for collision avoidance (for example, outputting an alarm, applying automatic braking, etc.).

As a technique for ensuring the communication quality of the vehicle-to-vehicle communication, for example, the technique disclosed in Patent Document 1 below is known. Specifically, in Patent Document 1, a stable diversity effect is achieved in a vehicle that employs a space diversity communication technology in which a plurality of antennas for receiving radio waves from other vehicles are arranged and signals received by each antenna are synthesized. An evaluation model has been proposed to design the optimal antenna placement that is obtained.

JP 2016-23951 A

Here, in a vehicle capable of receiving electromagnetic waves arriving from the outside, such as the vehicle to which vehicle-to-vehicle communication is applied, the profile of the received signal, particularly the angular profile that is the angular distribution characteristic of the received intensity, is It changes according to the road environment. Therefore, in such a vehicle, it is proposed to detect obstacles around the own vehicle from the angular profile of the received signal on the premise that the communication quality is sufficiently ensured. However, in Patent Document 1, the focus is placed on designing an antenna arrangement that optimizes communication quality, and the idea of detecting an obstacle from the angle profile of a received electromagnetic wave signal is not reached. .

SUMMARY OF THE INVENTION The present invention has been made in view of the circumstances as described above. It is an object to provide a detection method and apparatus.

In order to solve the above-described problems, the present invention is provided on a travel route of a vehicle, including a receiving section for receiving electromagnetic waves and a processing section for performing analysis based on the electromagnetic waves received by the receiving section . A vehicle target detection method for detecting a target, wherein the receiving unit receives an electromagnetic wave arriving at the vehicle, and the processing unit specifies an angular profile, which is an angular distribution characteristic of the intensity of the received electromagnetic wave. a second step in which the processing unit predicts an angular profile based on pre-obtained road environment information about the host vehicle; a third step of calculating the difference between these two angular profiles by subtracting the angular profile predicted in the second step from the angular profile of the second step; When it is confirmed that a difference element larger than the threshold of exists, the processing unit performs a fourth step of examining a time change of the difference element, and a time change of the difference element in the fourth step and a fifth step in which the processing unit determines that a new static obstacle exists and changes the road environment information when it is confirmed that the obstacle does not substantially occur. (Claim 1).

Even if an actual angle profile (hereinafter also referred to as an actual angle profile) is specified based on the received electromagnetic wave signal, it is not possible to obtain accurate information about various obstacles existing around the vehicle by simply examining the actual angle profile. is difficult to know. In contrast, in the present invention, an angle profile is predicted from road environment information obtained in advance, and the difference between this predicted angle profile (hereinafter also referred to as a predicted angle profile) and the actual angle profile is obtained. , the presence or absence of a new obstacle can be determined with relatively high accuracy by examining the difference (expected actual difference).

For example, the presence or absence of a new obstacle and its properties can be estimated based on information such as the presence or absence of a difference element exceeding a predetermined threshold and the intensity and angle of a peak corresponding to the difference element. Specifically, in the present invention, when a difference element exceeding a predetermined threshold exists, that is, when the presence of some obstacle that was not included in the road environment information is estimated, the temporal change of the difference element is examined. , the obstacle is determined to be a static obstacle (for example, a building, a fence, a guardrail, etc.) when it is confirmed that the time change does not substantially occur, so the obstacle is not moving. A static obstacle (an obstacle that does not move) can be properly detected by utilizing the fact that the difference element does not change with time unless the difference element is present.

In addition, when a static obstacle is detected, the road environment information is changed, so the next time the vehicle travels through the same point, the angle profile is predicted based on the changed road environment information. can do. This means that when the difference between the predicted angular profile and the actual angular profile is taken, the difference element due to the detected static obstacle is no longer generated. This eliminates the need for processing such as checking the temporal change of the difference element, and can reduce the processing load. Furthermore, if another new static obstacle exists during the next and subsequent runs, the new obstacle is calculated based on the difference between the predicted angle profile based on the road environment information after the change and the actual angle profile. Static obstacles can be detected with high accuracy.

Preferably, in the target object detection method, in the sixth step, the processing unit determines that a dynamic obstacle exists when it is confirmed in the fourth step that the difference element changes with time. and if the existence of the dynamic obstacle is confirmed in the sixth step, but the static obstacle is not confirmed in the fifth step, the processing unit determines whether the road The environment information is not changed (claim 2).

According to this configuration, when it is confirmed that the differential element (the differential element exceeding the predetermined threshold) changes over time, the obstacle is not a static obstacle but a dynamic obstacle (for example, vehicle or pedestrian), dynamic obstacles and static obstacles can be clearly distinguished and recognized.

However, even if a dynamic obstacle is detected as described above, the presence of this dynamic obstacle does not change the road environment information (the road environment information is changed only when a static obstacle is detected). Therefore, it is possible to prevent the road environment information, which is the source of the angular profile prediction, from being unnecessarily changed by the presence of dynamic obstacles that just happen to be present around the vehicle.

In the above configuration, more preferably, in the target detection method, when it is determined in the sixth step that the dynamic obstacle exists, the processing unit detects the actual object identified in the first step. a seventh step of identifying the intensity of the peak corresponding to the difference element exceeding the threshold in the angular profile of the above, and the processing unit uses the intensity of the peak identified in the seventh step as a predetermined reference intensity and an eighth step of comparing and identifying the type of the dynamic obstacle based on the comparison result, wherein in the eighth step , the processing unit determines that the intensity of the peak is higher than the reference intensity If it is large, it is determined that the dynamic obstacle is a vehicle, and if the intensity of the peak is equal to or less than the reference intensity, it is determined that the dynamic obstacle is a pedestrian (claim 3).

According to this configuration, when a dynamic obstacle is newly detected, its type can be properly determined. That is, it is known that, as a characteristic of pedestrians and vehicles representing dynamic obstacles, the reflectance of electromagnetic waves reflected by vehicles is much higher than the reflectance of electromagnetic waves reflected by pedestrians. On the other hand, in the above configuration, if the intensity of the peak corresponding to the difference element exceeding the threshold in the actual angle profile is greater than the reference intensity, it is determined to be a vehicle, and if it is equal to or less than the reference intensity, it is determined to be a pedestrian. Therefore, it is possible to easily and properly determine whether the dynamic obstacle is a pedestrian or a vehicle by using the above-described difference in reflectance.

Preferably, in the target detection method, when it is determined that the static obstacle exists in the fifth step, the processing unit detects the actual angular profile identified in the first step. A ninth step of identifying angular spread of peaks corresponding to differential elements exceeding the threshold and estimating the size of the static obstacle based on the identified angular spread (claim 4).

According to this configuration, when a static obstacle is newly detected, the size of the static obstacle is determined based on the angular spread of the peak corresponding to the differential element exceeding the threshold in the actual angular profile. It can be estimated properly.

Preferably, the target object detection method includes a tenth step in which the processing unit determines whether or not the reception of the electromagnetic waves arriving at the own vehicle is likely to be obstructed; and an eleventh step of switching the travel route of the vehicle being guided to another route when it is determined that reception of electromagnetic waves is likely to be obstructed (claim 5).

According to this configuration, the own vehicle can be guided to a route that ensures the reception quality of the electromagnetic waves reaching the own vehicle, and the obstacle detection accuracy using the electromagnetic waves can be maintained satisfactorily. .

The present invention also provides a vehicle target detection device for detecting a target existing on a travel route of a vehicle, comprising: a receiver for receiving electromagnetic waves arriving at the vehicle; and a processing unit that identifies an obstacle based on the first processing that identifies an angular profile that is an angular distribution characteristic of the intensity of the electromagnetic wave received by the receiving unit, based on the electromagnetic wave received by the receiving unit. , a second process of predicting an angular profile based on the road environment information around the own vehicle obtained in advance; By subtracting the angular profile, it is confirmed that there is a third process of calculating the difference between these two angular profiles, and that a difference element larger than a predetermined threshold exists in the difference calculated by the third process. a fourth process of examining a time change of the differential element, and a new static failure when it is confirmed in the fourth process that the time change of the differential element does not substantially occur and a fifth process of determining that an object exists and changing the road environment information (claim 6).

According to this target detection device, as in the invention of the target detection method described above, an angle profile (actual angle profile) specified from the received electromagnetic wave signal and an angle profile predicted from the road environment information (predicted angle profile) are provided. ), the obstacles around the vehicle can be properly detected.

As described above, according to the present invention, obstacles around the own vehicle can be properly detected based on the angle profile of the received electromagnetic wave signal.

1 is a block diagram showing an electrical configuration example of a vehicle to which a target object detection method (or device) of the present invention is applied; FIG. 4 is a flowchart showing the first half of a target (static obstacle or dynamic obstacle) detection process executed while the vehicle is running; 4 is a flowchart showing the second half of the target detection process; 3 is a graph showing an example of an angular profile predicted in step S2 of FIG. 2 (predicted angular profile); 5 is a schematic perspective view of the environmental conditions under which the predicted angular profile of FIG. 4 is obtained; FIG. 5 is a graph showing an actual angular profile based on a signal received at an antenna superimposed on the predicted angular profile of FIG. 4; FIG. 6 is an explanatory diagram showing the detected static obstacle superimposed on the perspective view of FIG. 5; FIG. 4 is a schematic plan view illustrating a situation in which a large number of other vehicles are present on the travel path of the host vehicle;

FIG. 1 is a block diagram showing an electrical configuration example of a vehicle to which the target object detection method (or device) of the present invention is applied. As shown in the figure, the vehicle of this embodiment includes an antenna device 1 , a front camera 2 , a navigation device 3 and a controller 10 . A vehicle is a vehicle equipped with a vehicle-to-vehicle communication function, and can mutually exchange information such as the position and speed of each vehicle with other vehicles equipped with the same function. Hereinafter, for clarity, the vehicle of this embodiment may be referred to as own vehicle, and the vehicle to be the other party of vehicle-to-vehicle communication may be referred to as partner vehicle.

The antenna device 1 is a receiver that receives radio waves for inter-vehicle communication transmitted from the other vehicle, and includes a plurality of antennas arranged at predetermined intervals. Specifically, the antenna device 1 is a space diversity antenna of the maximum ratio combining method, and by matching and combining the phases of the signals received by the plurality of antennas, the correlation of the radio wave strength received by each antenna is reduced. is configured to prevent deterioration of the received signal due to fading. In the space diversity antenna, the smaller the correlation of the reception strength of each antenna, the greater the diversity effect. On the other hand, in this embodiment, the arrangement of the plurality of antennas in the antenna device 1 is optimized so as to obtain a stable diversity effect under various assumed road environments. As a method for determining the optimum antenna arrangement, for example, the method of Patent Document 1 (Japanese Patent Application Laid-Open No. 2016-23951) described above can be used.

The front camera 2 is a camera for imaging the front of the own vehicle. The front camera 2 is provided, for example, on the rear surface of the upper edge of the windshield of the vehicle (the surface inside the vehicle interior), and is configured to capture an image of a relatively wide field of view in front of the vehicle through the windshield. It is

The navigation device 3 is a device that provides route guidance to a destination while specifying the travel position of the own vehicle. The navigation device 3 includes a GPS receiver 3a that receives signals for specifying positional information transmitted from GPS satellites, a map data storage 3b that stores road map data, traffic control information for each road, and the like. When a user sets a destination, a route guidance unit 3c that calculates and presents an optimal travel route (recommended travel route) to the destination to the user, and a display that displays information on the recommended travel route together with a road map. 3d.

The controller 10 is a microprocessor for controlling each part of the own vehicle, and is composed of a well-known CPU, RAM, ROM and the like. The controller 10 is electrically connected to the antenna device 1, the front camera 2, and the navigation device 3, and transmits and receives various signals to and from each of these elements. For example, the controller 10 sequentially receives radio wave signals received by the antenna device 1 and image data captured by the front camera 2 . In addition, the controller 10 extracts necessary information from the map data in the navigation device 3 in order to know the surrounding environment of the own vehicle, and also outputs control signals for correcting operations such as route guidance by the navigation device 3 as necessary. Output.

The controller 10 functionally has a processing unit 11 and a storage unit 12 . The processing unit 11 performs various analyzes and judgments based on the data input from the antenna device 1, the front camera 2, and the navigation device 3, and controls each part of the vehicle based on the results of the analysis/judgment. It is. The storage unit 12 stores various data necessary for processing by the processing unit 11 . The data stored in the storage unit 12 includes road environment correction information necessary for the processing of FIGS. 2 and 3, which will be described later.

FIGS. 2 and 3 are flowcharts showing specific procedures of target (static or dynamic obstacle) detection processing executed by the controller 10 as described above. When the process shown in FIG. 2 starts, the processing unit 11 of the controller 10, in step S1, based on the map data stored in the map data storage unit 3b of the navigation device 3 and the image captured by the front camera 2, Identify the road environment information around the current travel position of the own vehicle. The road environment information is information such as the position and size of a large fixed structure that affects the arrival route of radio waves to the own vehicle. It should be noted that the road environment information does not include information on mobile objects such as other vehicles.

That is, the processing unit 11 selects the map data stored in the map data storage unit 3b in advance from the current running position of the own vehicle (mainly in the vicinity of the front and side of the traveling direction) of buildings, stores, etc. Extract building information. In addition, the processing unit 11 extracts the building information extracted from the map data in this way, and adds information such as other buildings, fences, guardrails, etc. specified from the image captured by the front camera 2, that is, information existing in the map data. Then, information of a large fixed structure specified only from the image is added, and the information after this addition is specified as the road environment information at the current location.

Next, the processing unit 11 proceeds to step S2, and based on the road environment information obtained in step S1, the angle profile of the radio waves arriving at the vehicle, that is, the received power (strength ), the incoming wave angular profile is predicted. This prediction of the incoming wave angle profile is performed using an arithmetic expression based on a predetermined model. The outline of this prediction method will be described below.

FIG. 4 is a graph showing an example of the incoming wave angle profile predicted in step S2, and FIG. 5 is a perspective view schematically showing environmental conditions under which the prediction result of FIG. 4 is obtained. As can be understood from FIG. 5, the arrival wave angle profile in FIG. An example of an angle profile that changes according to the surrounding structure of the intersection and the like is shown.

In the perspective view of FIG. 5, the other vehicle that transmits radio waves is the transmitting vehicle Tx, and the own vehicle that receives the radio waves transmitted from the transmitting vehicle Tx is the receiving vehicle Rx. In the graph of FIG. 4, the horizontal axis represents the incoming wave angle in the global coordinate system, and the vertical axis represents 1 when the incoming wave angle is integrated in the range of 0 to 2π (rad). represents the received power normalized as follows. The global coordinate system has the position of the receiving point of the receiving vehicle Rx (the position of the receiving antenna) as the origin, and the direction from this origin to the intersection (in other words, from the rear of the receiving vehicle Rx in the traveling direction direction) is assumed to be 0 rad, and the sign of the angle is assumed to be positive in the counterclockwise direction.

As shown in FIG. 5, the radio waves transmitted from the transmitting vehicle Tx are reflected by the side surfaces of structures on the sides of the transmitting vehicle Tx and the receiving vehicle Rx, or diffracted by the edges of structures near intersections. As a result, the receiving vehicle Rx is reached from different angles via a plurality of routes. For this reason, the angular profile of the incoming wave at the receiving point when vehicle-to-vehicle communication is performed at an intersection in an urban area is due to the angular profile due to the diffracted wave (first arriving wave) and the reflected wave (second arriving wave). It can be treated as a combination of the resulting angular profile. That is, the angular profile due to the diffracted wave (first arriving wave) and the angular profile due to the reflected wave (second arriving wave) are expressed as Laplace distributions, respectively, and these two Laplace distributions are superimposed. can obtain a model of the incoming wave angle profile peculiar to vehicle-to-vehicle communication.

The arrival wave angle profile S (Φ) in the vehicle-to-vehicle communication environment modeled by the above method is the center angle of the received power peak of each of the diffracted waves and reflected waves reaching the antenna of the receiving vehicle Rx, the spread of the peak, and Using the peak power as a parameter, it can be calculated by the following equations (1) and (2).

Here, Φ is the incoming wave angle at the antenna position (receiving point) of the receiving vehicle Rx. Also, as shown in _Fig . ₄ , A1 is the peak power of the diffracted wave, _A2 is the peak power of the reflected wave, σ1 is the angular spread of the peak of the diffracted wave, and _σ2 is the peak power of the reflected wave. is the angular spread of the peak, _Φ1 is the central angle of the diffracted wave, and _Φ2 is the central angle of the reflected wave.

In step S2, based on the road environment information obtained in step S1, each parameter used in the above equations (1) and (2) is determined, and by substituting the determined parameters into each equation, prediction Calculate the incoming wave angle profile. Note that the incoming wave angle profile illustrated in FIG. 4, that is, the incoming wave angle profile predicted using the above equations (1) and (2) has a central angle of about 2.8 rad and about 4.2 rad. Two peaks are included, of which the peak of about 2.8 rad is due to the diffracted wave and the peak of about 4.2 rad is due to the reflected wave.

As described above, in order to predict the incoming wave angle profile, it is necessary to determine the values of each parameter in the above equations (1) and (2) from road environment information (information on large fixed structures located around intersections). There is Various methods of determining the parameters are conceivable. For example, the following method of determining the parameters is conceivable.

First, based on the positional relationship between the own vehicle (receiving vehicle Rx) and the other vehicle (transmitting vehicle Tx), the diffracted wave and the reflected wave reaching the own vehicle are selected from an area of a predetermined radius around the intersection. Identify areas that can be considered to have a large impact as research areas. Next, based on the specified investigation target area and the road environment information obtained in step S1, the area/position of the reflecting surface of the large fixed structure existing in the investigation target area, and the number of edges/ Locate. Then, from the areas/positions of the reflecting surfaces and the number/positions of edges in the survey target area specified in this way, the above formula (1) ( 2) Determine each parameter.

Returning to FIG. 2, the processing unit 11 executes processing for generating an incoming wave angle profile based on the actual received signal of the radio wave in parallel with the prediction processing of the incoming wave angle profile described above.

That is, in step S3, the processing unit 11 acquires the signal received by the antenna device 1, that is, the signal of the radio wave transmitted from the opponent vehicle having the vehicle-to-vehicle communication function and received by the antenna device 1 of the own vehicle.

Next, the processing unit 11 shifts to step S4 and generates an incoming wave angle profile by analyzing the received signal (arriving radio wave) obtained in step S3. For example, when receiving radio signals at two receiving points, if the distance between the receiving points is constant, the correlation (correlation characteristic) of the temporal fluctuations of the signals received at both receiving points and the angle of arrival of the wave It is known that there is a certain relationship between the characteristics. Therefore, in the above step S4, based on the difference between the reception power and the reception time of each of the plurality of antennas included in the antenna device 1, the correlation characteristics between the antennas are obtained, and the distance between the obtained correlation characteristics and the antennas is calculated. From the distance, through predetermined arithmetic processing (using a predetermined arithmetic expression reflecting the relationship between the correlation characteristic and the angle characteristic), the incoming wave angle profile is calculated.

When the actual wave-of-arrival angle profile based on the received signal of the antenna device 1 is generated as described above, the processing unit 11 proceeds to step S5 to generate the actual wave-of-arrival angle profile (hereinafter referred to as the actual angle profile). by subtracting the incoming wave angle profile predicted by the calculation using the model formula in step S2 (hereinafter also referred to as the predicted angle profile) from the difference between these two angle profiles, that is, the expected difference of the angle profiles Calculate

Next, the processing unit 11 proceeds to step S6, and determines whether or not there is a difference element larger than a predetermined threshold α in the difference (expected actual difference) calculated in step S5. .

FIG. 6 is a graph illustrating a profile when there is a difference element larger than the threshold value α, showing an actual angle profile (one-dot chain line) superimposed on a predicted angle profile (solid line). The actual angular profile shown in this figure includes a peak Q centered at angle Z (approximately 1.5 rad), but a similar peak is not included in the predicted angular profile, resulting in A relatively large difference element ΔX (>α) corresponding to the peak Q is generated. That is, let X be the intensity (normalized power) of the peak Q included in the actual angle profile, let Y be the intensity of the corresponding predicted angle profile (intensity at the peak center angle Z), and subtract the latter from the former ( =XY) is a difference element ΔX, the difference element ΔX is larger than a predetermined threshold value α.

Incidentally, when comparing the actual angle profile and the predicted angle profile, it is considered that in addition to the difference element ΔX described above, various difference elements due to minute differences in waveforms actually occur. However, in the example of FIG. 6, there are no other large difference elements that exceed the threshold α. Therefore, in FIG. 6, only the difference element ΔX (corresponding to the peak Q) is illustrated, and other minute difference elements are omitted.

If the determination in step S6 is YES and it is confirmed that there is a difference element ΔX exceeding the threshold value α, the processing unit 11 proceeds to step S7 and calculates the time change of the difference element ΔX. That is, the processing unit 11 compares the most recently calculated difference element ΔX with the difference element ΔX calculated after the lapse of a predetermined specified time (for example, 0.1 to 0.5 seconds). is calculated as the time change.

Next, the processing unit 11 shifts to step S8 to determine the time change of the difference element ΔX calculated in step S7, more specifically, the time change of the magnitude of ΔX itself and the position (angle) of ΔX on the horizontal axis. is equal to or less than a predetermined value. The predetermined value is a threshold value for determining whether or not the difference element ΔX changes with time, and the fact that the time change of the difference element ΔX is equal to or less than the predetermined value means that the difference element ΔX does not particularly change with time. not (substantially no change over time).

If the determination in step S8 is YES and it is confirmed that the difference element ΔX does not substantially change with time, the processing unit 11 proceeds to step S9 to generate a new value that has not been recognized so far. Determine that static obstacles (eg, buildings, fences, guardrails, etc.) are present.

That is, the occurrence of a large difference element ΔX exceeding the threshold value α between the actual angle profile and the expected angle profile indicates that the road environment information on which the prediction is based (i.e., the road environment information acquired in step S1) ) is different from the actual one, which means that there is a new obstacle that is not recognized on the map data or the captured image. Moreover, the fact that the difference element ΔX does not substantially change over time means that the new obstacle has not substantially moved (fixed). Under these circumstances, when there is a difference element ΔX exceeding the threshold α and it does not change with time (that is, when the determination in step S8 is YES), a new static obstacle is generated around the host vehicle. It is determined that the object exists (step S9 above). For example, in step S9, as shown in FIG. 7, it is determined that a new static obstacle W exists at one corner of the intersection where the presence of a fixed structure has not been recognized. The position of this new static obstacle W can be specified based on the position (angle) on the horizontal axis of the differential element ΔX in the graph of FIG.

On the other hand, when the differential element ΔX exceeding the threshold value α changes over time, it means that the new obstacle is moving. Therefore, when it is determined to be NO in step S8 and it is confirmed that the difference element ΔX changes with time, the processing unit 11 proceeds to step S10 to determine whether there are any dynamic obstacles around the host vehicle. (For example, another vehicle or a pedestrian) is determined to exist.

When the property of the obstacle is specified as described above, the processing unit 11 proceeds to the processing shown in the flowchart of FIG. 3 below. In FIG. 3, different steps are taken depending on whether the new obstacle recognized on the basis of the expected angular profile difference is a static obstacle or a dynamic obstacle.

If the new obstacle is a static obstacle (for example, a building, a fence, a guardrail, etc.), the processing unit 11 proceeds to step S11, and the peak Q corresponding to the difference element ΔX described above, that is, the threshold value α The waveform of the peak Q of the actual angular profile that caused the difference element ΔX exceeding σ is examined, and the angular spread σx (FIG. 6) of the peak Q is calculated. This angular spread σx is a value having the same meaning as σ ₁ and σ ₂ in the above equations (1) and (2).

Next, the processing unit 11 proceeds to step S12 and estimates the size of the static obstacle based on the angular spread σx calculated in step S11. That is, it is estimated that the larger the angular spread σx, the larger the size of the static obstacle.

Next, the processing unit 11 proceeds to step S13, and stores the information such as the position and size of the static obstacle specified from the estimation result of step S12 as road environment correction information in the storage unit of the controller 10. Store in 12. The road environment correction information stored in the storage unit 12 is used as the road environment information (information specified in step S1) specified from the map data and the captured image when the host vehicle enters the same intersection from the next time onward. used to correct. In this way, the newly stored road environment correction information leads to a change in the road environment information from the next time onwards, so the process of step S13 is synonymous with changing (updating) the road environment information. .

On the other hand, if the new obstacle is a dynamic obstacle (for example, another vehicle or a pedestrian), the processing unit 11 proceeds to step S14, and the peak Q corresponding to the difference element ΔX described above, that is, the threshold value The peak Q of the actual angular profile that caused the differential element ΔX exceeding α is examined, and its intensity X (FIG. 6) is identified.

Next, the processing unit 11 proceeds to step S15 and determines whether or not the peak intensity X obtained in step S14 is greater than a predetermined reference intensity β.

If the determination in step S15 is YES and it is confirmed that the peak intensity X is greater than the reference intensity β (X>β), the processing unit 11 proceeds to step S16, and the dynamic obstacle is a vehicle. (another vehicle). On the other hand, if it is determined NO (X≦β) in step S15, the processing unit 11 proceeds to step S17 and determines that the dynamic obstacle is a pedestrian. Such determination is possible because the reflectance of radio waves reflected by a vehicle is significantly higher than the reflectance of radio waves reflected by a pedestrian.

After any one of steps S13, S16, and S17 is completed, or after determination of NO is made in step S6 (that is, when no difference element exceeding the threshold value α is found), the processing unit 11 Proceeding to step S20, it is determined whether or not a destination has been set for the navigation device 3, that is, whether or not route guidance to the destination is being performed by the route guidance section 3c. .

When it is determined as YES in step S20 and it is confirmed that the route guidance is in progress, the processing unit 11 proceeds to step S21, and a predetermined number or more of vehicles (other vehicle) is present.

For example, as shown in FIG. 8, it is assumed that there are more than a predetermined number of other vehicles V traveling in the oncoming lane, and their presence is recognized by the front camera 2 or other means. In such a case, the determination in step S21 is YES. Then, the processing unit 11 proceeds to the next step S22 and determines whether or not there is an appropriate route passing through a road other than the road on which the vehicle is currently traveling. For example, as shown in FIG. 8, there is another traveling path L2 that branches off from the current traveling path L1 on which the host vehicle (receiving vehicle Rx) is traveling and that is expected to have few other vehicles. Further, it is determined whether or not the destination can be reached without any trouble even if the other travel path L2 is passed.

If it is determined as YES in step S22 and it is confirmed that another suitable route exists, the processing unit 11 proceeds to step S23, route) to the destination. As a result, the driver can be guided to the destination along the road L2 with few other vehicles.

As described above, in this embodiment, as the angle profile of the radio waves arriving at the vehicle through inter-vehicle communication, the actual angle profile based on the signal received by the antenna device 1 and the road environment information (radio wave (Information such as the position and size of a large fixed structure that affects the arrival path of ) is generated. It is determined whether or not there is a large difference element ΔX (FIG. 6). Then, when the presence of the difference element ΔX larger than the threshold value α is confirmed, the time change of the difference element ΔX is further examined, and if the time change does not substantially occur, a new static obstacle ( (buildings, fences, guardrails, etc.) are determined to exist, and the road environment information is changed. According to such a configuration, there is an advantage that an obstacle around the own vehicle can be properly detected based on the signal received by the antenna device 1 .

In other words, even if the actual angular profile is specified from the actual received signal, it is difficult to obtain accurate information about various obstacles existing around the own vehicle simply by examining the actual angular profile. In contrast, in the above embodiment, the angle profile is predicted from the road environment information specified from the map data and the captured image, and the difference between this predicted angle profile and the actual angle profile is obtained. Presence/actual difference) can be used to determine the presence or absence of new obstacles with relatively high accuracy.

For example, the presence or absence of a new obstacle and its properties can be estimated based on information such as the presence or absence of a difference element exceeding a predetermined threshold and the intensity and angle of a peak corresponding to the difference element. Specifically, in the above embodiment, when there is a difference element ΔX exceeding the threshold value α, that is, when it is estimated that there is an obstacle that was not included in the road environment information, the time change of the difference element ΔX is The obstacle is determined to be a static obstacle (building, fence, guardrail, etc.) when it is checked and it is confirmed that the time change does not substantially occur, so the obstacle does not move. Static obstacles (obstacles that do not move; for example, W in FIG. 8) can be properly detected by utilizing the fact that the difference element basically does not change with time unless there is.

Further, when a static obstacle is detected, the road environment information is changed (stored in the storage unit 12 as road environment correction information). An angular profile can be predicted based on this changed road environment information. This means that when the difference between the predicted angular profile and the actual angular profile is taken, the difference element (ΔX in FIG. 6) due to the static obstacle that has already been detected will no longer occur. This eliminates the need for processing such as checking the temporal change of the difference element, and can reduce the processing load. Furthermore, if another new static obstacle exists during the next and subsequent runs, based on the difference between the predicted angle profile based on the road environment information after the change and the actual angle profile, the new Static obstacles can be detected with high accuracy.

On the other hand, contrary to the above case, when the difference element changes with time, the obstacle is considered to be moving. Utilizing this, in the above-described embodiment, when the presence of the difference element ΔX exceeding the threshold value α is confirmed and the difference element ΔX is confirmed to change with time, the obstacle is determined to be a dynamic obstacle (vehicle or pedestrian), it is possible to clearly distinguish and recognize a dynamic obstacle and a static obstacle.

However, even if a dynamic obstacle is detected as described above, the presence of this dynamic obstacle will not change the road environment information (the road environment information will only change when a static obstacle is detected). Therefore, it is possible to prevent unnecessary rewriting of the road environment information, which is the source of the angle profile prediction, by dynamic obstacles that just happen to exist around the own vehicle.

Further, in the above embodiment, when a new dynamic obstacle is detected (that is, when a differential element ΔX exceeding the threshold α is detected and it changes over time), the threshold α is exceeded The intensity X (FIG. 6) of the peak Q corresponding to the differential element ΔX is compared with the reference intensity β. If the peak intensity X is greater than the reference intensity β, it is determined that the dynamic obstacle is a vehicle, If the strength is equal to or less than the reference strength β, the dynamic obstacle is determined to be a pedestrian. According to such a configuration, by utilizing the fact that the reflectance of radio waves reflected by a vehicle (another vehicle) is significantly higher than the reflectance of radio waves reflected by a pedestrian, the vehicle can be easily and properly detected as a pedestrian. A vehicle can be distinguished.

Further, in the above embodiment, when a new static obstacle is detected (that is, when a difference element ΔX exceeding the threshold α is detected and it does not change over time), the threshold α is exceeded The angular spread σx of the peak Q corresponding to the differential element ΔX is specified, and the size of the static obstacle is estimated based on the specified angular spread σx. It is possible to estimate the size of a target obstacle.

Further, in the above-described embodiment, while the navigation device 3 is providing route guidance to the destination, it is assumed that a predetermined number or more of other vehicles (for example, the vehicle V in FIG. 8) exist on the travel path of the host vehicle. is confirmed by the front camera 2, etc., the driving route being guided is switched to another route, so the vehicle is set on a route that ensures the reception quality (communication quality) of the radio waves that reach the vehicle. By guiding, it is possible to maintain good obstacle detection accuracy using the radio waves. That is, if there are many other vehicles on the travel path of the own vehicle, the radio wave (indicated by the solid polygonal line arrow in FIG. 8) from the other vehicle (transmitting vehicle Tx in FIG. ) is likely to be blocked by the large number of other vehicles, it tends to be difficult to ensure sufficient communication quality. On the other hand, according to the above-described embodiment, in which the own vehicle is guided to a different route when there are more than a predetermined number of other vehicles on the traveling road, the above-mentioned obstruction of radio wave reception by the other vehicles is resolved, Since sufficient communication quality can be ensured, it is possible to maintain good obstacle detection accuracy using received radio waves.

Although the preferred embodiments of the present invention have been described above, the present invention should not be construed as being limited to these embodiments, and various modifications can be made without departing from the scope of the present invention.

For example, in the above-described embodiment, when a static obstacle is detected based on the difference between the actual angle profile and the predicted angle profile, the information about the static obstacle is used as the road environment correction information and stored in the storage unit 12 of the controller 10. , and using this road environment correction information to correct the road environment information specified from the map data or the like stored in the navigation device 3 from the next time onwards. The map data may be stored in advance, and the map data in the storage unit 12 may be directly changed (updated) when a static obstacle is detected.

Further, in the above embodiment, the actual angle profile specified from the received signal received by the antenna device 1 is compared with the predicted angle profile predicted from the road environment information specified from map data or the like, and both profiles are compared. It is determined whether or not there is a relatively large difference element ΔX such that the intensity difference (XY) of exceeds the threshold α, and if the existence of such a difference element ΔX (>α) is confirmed, Although it is determined that a new static or dynamic obstacle exists based on the presence or absence of that change over time, other information can also be estimated from the difference between the actual angle profile and the predicted angle profile. . For example, when the intensity of the corresponding peak in the actual angle profile is much smaller than the intensity of the peak in the predicted angle profile, part of the fixed structure included in the road environment information has disappeared. , and compared to the angle spread of the peaks present in the predicted angle profile (values corresponding to σ ₁ and σ ₂ in FIG. 6), the angle spread of the peaks of the corresponding actual angle profile was significantly different. In this case, it can be estimated that some of the fixed structures included in the road environment information have been replaced with other fixed structures.

In the above embodiment, the radio waves transmitted from the other vehicle for inter-vehicle communication are received by the antenna device 1, and the angular distribution characteristics of the intensity of the received radio waves are specified as the actual angle profile. By analyzing some electromagnetic waves arriving at the own vehicle (for example, by receiving and analyzing electromagnetic waves transmitted from a specific base station), the angular distribution characteristics of the intensity of the electromagnetic waves may be specified as an actual angle profile. good.

1 Antenna device (receiving unit)
11 processing unit Q peak (peak corresponding to the difference element exceeding the threshold)
X (peak) intensity ΔX Differential element

Claims (6)

A vehicle target detection method for detecting a target existing on a travel route of the own vehicle , comprising a receiving unit for receiving electromagnetic waves and a processing unit for performing analysis based on the electromagnetic waves received by the receiving unit. ,
a first step in which the receiving unit receives an electromagnetic wave arriving at the host vehicle, and the processing unit specifies an angular profile that is an angular distribution characteristic of the intensity of the received electromagnetic wave;
a second step in which the processing unit predicts an angle profile based on pre-obtained road environment information around the own vehicle;
a third step in which the processing unit calculates the difference between the two angular profiles by subtracting the angular profile predicted in the second step from the actual angular profile identified in the first step; ,
a fourth step in which, when it is confirmed that a difference element larger than a predetermined threshold exists in the difference calculated in the third step, the processing unit examines a time change of the difference element; ,
When it is confirmed in the fourth step that the difference element does not substantially change with time, the processing unit determines that a new static obstacle exists and changes the road environment information. and a fifth step of: In the vehicle target object detection method according to claim 1,
Further comprising a sixth step in which the processing unit determines that a dynamic obstacle exists when it is confirmed in the fourth step that the difference element changes over time,
Even if the presence of the dynamic obstacle is confirmed in the sixth step, if the static obstacle is not confirmed in the fifth step, the processing unit does not change the road environment information. A vehicle target object detection method characterized by: In the vehicle target object detection method according to claim 2,
When it is determined in the sixth step that the dynamic obstacle exists, the processing unit causes the peak corresponding to the difference element exceeding the threshold in the actual angular profile identified in the first step to a seventh step of determining the intensity of
an eighth step in which the processing unit compares the intensity of the peak identified in the seventh step with a predetermined reference intensity and identifies the type of the dynamic obstacle based on the comparison result; including
In the eighth step , the processing unit determines that the dynamic obstacle is a vehicle if the intensity of the peak is greater than the reference intensity, and if the intensity of the peak is equal to or less than the reference intensity and determining that the dynamic obstacle is a pedestrian. In the vehicle target object detection method according to any one of claims 1 to 3,
When it is determined in the fifth step that the static obstacle exists, the processing unit generates peaks corresponding to differential elements exceeding the threshold among the actual angular profiles identified in the first step and estimating the size of the static obstacle based on the determined angular spread. In the vehicle target object detection method according to any one of claims 1 to 4,
a tenth step in which the processing unit determines whether or not there is a situation in which reception of the electromagnetic waves arriving at the own vehicle is likely to be hindered;
and an eleventh step, wherein, when it is determined in the tenth step that the reception of the electromagnetic wave is likely to be hindered, the processing unit switches the traveling route of the vehicle being guided to another route. A target object detection method for a vehicle, comprising: A target object detection device for a vehicle that detects a target existing on a travel route of the own vehicle,
a receiving unit for receiving electromagnetic waves arriving at the own vehicle;
A processing unit that identifies obstacles based on the electromagnetic waves received by the receiving unit,
The processing unit is
A first process of identifying an angular profile, which is an angular distribution characteristic of the intensity of the electromagnetic wave, based on the electromagnetic wave received by the receiving unit;
a second process of predicting an angle profile based on pre-obtained road environment information around the own vehicle;
a third process of subtracting the angular profile predicted in the second process from the actual angular profile identified in the first process to calculate the difference between these two angular profiles;
a fourth process of examining a time change of the difference element when it is confirmed that a difference element larger than a predetermined threshold exists in the difference calculated in the third process;
A fifth process for determining that a new static obstacle exists and changing the road environment information when it is confirmed in the fourth process that the difference element does not substantially change with time. A target object detection device for a vehicle, characterized by:

Images