Information processing device, object tracking device, tracking method, and program

Abstract

To track an object with good accuracy when vehicle speed is not abnormal, and to prevent object tracking accuracy from deteriorating when the vehicle speed is abnormal.SOLUTION: An information processing device 3 is mounted on a vehicle, and includes an observation value detection unit 21, a tracking unit 27, and an abnormality determination unit 22. The abnormality determination unit determines whether a vehicle speed abnormality is occurring. When it is determined that the vehicle speed abnormality is not occurring, the tracking unit performs a first process of using a predicted value of relative speed based on a detection result of its own vehicle speed and calculates a current estimated value. When it is determined that vehicle speed is abnormal, the tracking unit performs a second process different from the first process and calculates a current estimated value.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present disclosure relates to an information processing device, an object tracking device, a tracking method, and a program for tracking an object. [Background technology]

[0002] When an on-vehicle radar device recognizes an object around the vehicle, the state of the object is estimated using the result of frequency analysis of the observation signal obtained from the on-vehicle radar device (hereinafter, also referred to as the observation value) in a predetermined processing cycle. In other words, a technique is known in which the state of the object is estimated in a time series to track the object.

[0003] The observed values ​​may include distance, direction, and relative speed. For example, Patent Document 1 below proposes a technique for tracking an object with high accuracy using distance, direction, and relative speed as observed values. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2018-25492 A Summary of the Invention [Problem to be solved by the invention]

[0005] In technology for tracking objects, for example, observed values ​​are associated with predicted values ​​for the above-mentioned relative speed, distance, direction, etc., and an estimate indicating the current state of each object is calculated based on the associated observed values ​​and predicted values.

[0006] For example, when calculating the predicted value of the relative speed, a method of calculating the predicted value of the relative speed based on the vehicle speed is conceivable. In this case, for example, if the vehicle speed changes between the previous processing cycle and the current processing cycle (i.e., before and after the processing cycle), the predicted value of the relative speed can be calculated with high accuracy as a value corresponding to the change in the vehicle speed. For example, if the change in the vehicle speed before and after the processing cycle is 1 km / hour, the change in the vehicle speed (i.e., 1 km / hour) is reflected in the magnitude of the predicted value of the relative speed in the current processing cycle. This is believed to enable the prediction value of the relative velocity, and therefore the estimated value of the relative velocity, to be calculated with high accuracy. Accurate estimation of the relative velocity results in improved accuracy in tracking the object (e.g., accuracy of the estimated value of the object's position based on the relative velocity, the estimated value of the object's ground speed based on the relative velocity, etc.).

[0007] However, this does not apply when a state occurs in which the vehicle speed differs from the actual vehicle speed due to, for example, wheel spinning, etc. In the method of calculating the predicted value of the relative speed based on the vehicle speed as described above, the error in the predicted value of the relative speed may be large due to a large error in the vehicle speed.

[0008] In other words, when the vehicle speed is not abnormal, the current estimated value of the object can be calculated with high accuracy using the accurate predicted value of the relative speed, but when the vehicle speed is abnormal, the accuracy of the predicted value of the relative speed decreases, and the accuracy of the current estimated value of the object decreases. As a result, when the vehicle speed is abnormal, there is a risk that the accuracy of tracking the object will decrease.

[0009] One aspect of the present disclosure provides a technique for tracking an object with high accuracy when there is no abnormality in the vehicle speed, and suppressing a decrease in the accuracy of tracking the object when an abnormality in the vehicle speed is detected. [Means for solving the problem]

[0010] One aspect of the present disclosure is an information processing device (3) mounted on a vehicle, comprising an observation value detection unit (21), a tracking unit (27), and an abnormality determination unit (22). The observation value detection unit is configured to acquire an observation signal observed by a sensor (2) that transmits and receives radar waves, and detect at least one observation value for at least one target around the vehicle from the observation signal.

[0011] The tracking unit tracks the target by calculating, for each target, a current predicted value from a past estimated value that is an estimate indicating a state of the target, and calculating a current estimated value from a current observed value and a current predicted value in a predetermined processing cycle. The abnormality determination unit determines whether or not a vehicle speed abnormality has occurred. When it is determined that a vehicle speed abnormality has not occurred, the tracking unit executes a first process that uses a predicted value of a relative speed based on a detection result of the host vehicle speed to calculate a current estimated value. When it is determined that a vehicle speed abnormality has occurred, the tracking unit executes a second process different from the first process to calculate a current estimated value.

[0012] In the present disclosure, based on the result of the determination of the vehicle speed abnormality, the process is switched to either a first process (i.e., a process that obtains a highly accurate predicted value and thus an estimated value of the relative speed based on the detection result of the vehicle speed) or a second process (i.e., a process different from the first process).Therefore, the second process is a process in which the estimated value of the relative speed is less affected by errors in the vehicle speed, and when the vehicle speed is determined to be abnormal, the second process is executed instead of the first process, thereby suppressing the effect of errors in the vehicle speed when estimating the relative speed. As a result, when there is no abnormality in the vehicle speed, it is possible to track objects with high accuracy, and when an abnormality in the vehicle speed is detected, it is possible to prevent a decrease in the accuracy of tracking the target (i.e., object). The above-mentioned information processing device may be provided as an object tracking device equipped with a radar device. Such an object tracking device can provide the same effect. The procedure performed by the above-mentioned information processing device may be provided as a tracking method. Such a tracking method can provide the same effect. A program may be configured to operate a computer as the above-mentioned information processing device. By operating a computer according to such a program, the same effect can be provided. [Brief description of the drawings]

[0013] [Figure 1] 1 is a block diagram showing a configuration of an object tracking device including an information processing device according to a first embodiment. [Diagram 2] FIG. 4 is an explanatory diagram showing an example of a detection area when a radar device is mounted in front of a vehicle. [Diagram 3] FIG. 11 is an explanatory diagram showing an example of a detection area when a radar device is installed in a location other than the front of the vehicle. [Figure 4] FIG. 1 is a block diagram functionally showing a configuration of an information processing device. [Diagram 5] 11 is a flowchart showing a tracking process. [Figure 6] 11 is a flowchart showing a prediction process. [Figure 7] 13 is a flowchart showing an association process. [Figure 8] FIG. 11 is an explanatory diagram illustrating an example of determining an observed value to be associated with a predicted value and calculating an estimated value. [Figure 9] FIG. 11 is an explanatory diagram illustrating another example of determining an observed value to be associated with a predicted value and calculating an estimated value. [Figure 10] 13 is a flowchart showing an estimation process. [Figure 11] 4 is a flowchart showing a vehicle speed abnormality detection process. [Figure 12] 11 is a flowchart showing a stationary object prediction residual process. [Figure 13] 10 is a flowchart showing a prediction process according to a second embodiment. [Figure 14]FIG. 11 is an explanatory diagram for explaining a process for calculating a current estimated value of an object from a previous estimated value of the object in a case where the position of the object is used as an estimated value in another embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0014] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. [1. First embodiment] [1. Overall composition] As shown in Fig. 1, the configuration of an object tracking device 1 will be described with reference to Fig. 1. The object tracking device 1 is mounted on a vehicle (hereinafter, also referred to as a host vehicle JV), and includes a radar device 2, an information processing device 3, and a detection unit 5.

[0015] As shown in Fig. 2, the radar device 2 may be mounted in the front center of the host vehicle JV (e.g., in the center of the front bumper), and the surroundings of the host vehicle JV, specifically the area in the front center of the host vehicle JV, may be set as the detection area Rd. Also, as shown in Fig. 3, the radar device 2 may be mounted on each of the left front side and right front side of the host vehicle JV (e.g., the left end and right end of the front bumper), and the surroundings of the host vehicle JV, specifically the areas in the left front and right front of the host vehicle JV, may be set as the detection area Rd. The radar devices 2 may be mounted on the left and right rear sides of the host vehicle JV (for example, the left and right ends of the rear bumper), and the surroundings of the host vehicle JV, specifically, the left and right rear areas of the host vehicle JV, may be set as the detection areas Rd. The number and mounting positions of the radar devices 2 mounted on the host vehicle JV may be appropriately selected.

[0016] The radar device 2 is a millimeter wave radar that transmits and receives radio waves. The radar device 2 includes a transmitting array antenna composed of multiple antenna elements and a receiving array antenna composed of multiple antenna elements. The radar device 2 irradiates a transmission wave to a detection area Rd in each processing cycle that arrives at a predetermined period Tcy.

[0017] The radar device 2 receives reflected waves (i.e., received waves) generated when the transmitted waves are reflected at a reflecting point of the target. Examples of targets include vehicles, road surfaces, roadside objects, etc. The radar device 2 generates beat signals by mixing the transmitted waves and the reflected waves, samples the beat signals, and outputs the generated signals to the information processing device 3.

[0018] The signal output from the radar device 2 is called an observation signal. Note that, here, a signal generated by sampling a beat signal is output as the observation signal, but the present disclosure is not limited to this. Note that, here, the radar device 2 is assumed to be of the FMCW type, but the present disclosure is not limited to this. For example, any modulation method such as a multi-frequency CW type or FCM may be used. FCM is an abbreviation for Fast-Chirp Modulation.

[0019] 1, the information processing device 3 is mainly configured with a well-known microcomputer (i.e., a microcomputer) having a CPU 11, a ROM 13, a RAM 15, a flash memory 17, etc. The CPU 11 realizes various functions by executing a program stored in the ROM 13. By executing the program, a method corresponding to the program is executed.

[0020] Incidentally, the memories 19, such as the ROM 13, the RAM 15, and the flash memory 17, are non-transient tangible recording media. The information processing device 3 may also include a coprocessor that executes fast Fourier transform processing (i.e., FFT processing) and the like. The number of microcomputers constituting the information processing device 3 may be one or more. Furthermore, the method of realizing the various functions possessed by the information processing device 3 is not limited to software, and some or all of the elements may be realized using one or more pieces of hardware. For example, when the above functions are realized by electronic circuits that are hardware, the electronic circuits may be realized by digital circuits including a large number of logic circuits, or analog circuits, or a combination of these.

[0021] In the information processing device 3, the CPU 11 executes a program, and thereby, specifically, as shown by solid lines in Fig. 4, the CPU 11 realizes the functions of the sensor unit 21, the abnormality detection unit 22, the switching unit 23, the prediction unit 24, the association unit 25, and the estimation unit 26, and executes the tracking process. Note that the tracking unit 27 referred to below includes the functions of the prediction unit 24, the association unit 25, and the estimation unit 26.

[0022] Although details will be described later, the information processing device 3 performs tracking processing based on the observation signal generated by the radar device 2, and estimates the state of the target at the current time. To represent the state of the target, for example, the distance from the host vehicle JV to the target, the direction of the target relative to the host vehicle JV, and the relative speed of the target relative to the host vehicle JV may be used. Furthermore, based on these, the position of the target relative to the host vehicle JV calculated from the distance and direction, and the ground speed of the target calculated from the relative speed of the target relative to the host vehicle JV and the speed of the host vehicle JV (hereinafter also referred to as the host vehicle speed) may be used to represent the state of the target.

[0023] In the following, the distance from the host vehicle JV to the target, the direction of the target relative to the host vehicle JV, and the relative speed of the target relative to the host vehicle JV will be simply referred to as the distance, direction, and relative speed, respectively. The information processing device 3 may calculate, for example, estimated values ​​of the target's current direction, distance, relative speed, etc., by executing the tracking process, and may calculate, for example, an estimated value of the target's current position and an estimated value of the target's ground speed based on these, and output them to a driving support device, etc. The driving support device refers to various devices that realize driving support, although not shown.

[0024] 1, the detection unit 5 includes various detection devices other than the radar device 2. The detection devices include at least a wheel speed sensor 9. The wheel speed sensors 9 are provided, for example, on each of the four wheels on the front, rear, left and right sides of the host vehicle JV, and output signals indicating the rotational speeds of the corresponding wheels (hereinafter referred to as wheel speed signals). The wheel speed signals from each wheel speed sensor 9 are input to the information processing device 3. As will be described later, the information processing device 3 can detect the rotational speed of each wheel based on the wheel speed signals from each wheel speed sensor 9. Then, from the detection results, it can be determined, for example, whether or not the wheels are spinning (i.e., slipping).

[0025] [2. Processing] [2-1. Tracking process] Next, the tracking process executed by the information processing device 3 of the first embodiment will be described with reference to the flowchart in Fig. 5. The information processing device 3 repeatedly executes this tracking process at a predetermined period (i.e., period Tcy). A series of processes from S10 to S120 repeatedly executed at each period Tcy is also called a processing cycle. The period Tcy may be, for example, several msec to several hundred msec.

[0026] First, in S10, the sensor unit 21 detects the observation value of each target object existing around the host vehicle JV. For example, in S10, the sensor unit 21 first causes the radar device 2 to emit a transmission wave. Next, the sensor unit 21 acquires an observation signal generated based on the reflected wave received by the radar device 2 from a reflection point. Then, the sensor unit 21 detects the observation value of each target object existing around the host vehicle JV at the current time (i.e., the current processing cycle) from the observation signal acquired from the radar device 2.

[0027] For example, in S10, the distance from the host vehicle JV to the target, the direction of the target relative to the host vehicle JV, and the relative speed of the target relative to the host vehicle JV are detected as the observation values. Note that, for example, when the radar device 2 is configured to detect each observation value of each target from the observation signal, the sensor unit 21 may be configured to acquire each observation value of each target from the radar device 2.

[0028] Next, in S20, the abnormality detection unit 22 detects the vehicle speed from the wheel speed signals acquired from each wheel speed sensor 9, and acquires the detected vehicle speed as the observed value of the vehicle speed in the current processing cycle. For example, the abnormality detection unit 22 may calculate the vehicle speed based on the average value of the rotation speeds of each wheel. Note that, when the vehicle speed is detected by a configuration other than the abnormality detection unit 22 (for example, any of the wheel speed sensors 9 or other devices), the abnormality detection unit 22 may be configured to acquire the detected vehicle speed.

[0029] Next, in S30, the abnormality detection unit 22 detects whether or not a vehicle speed abnormality has occurred. An abnormal vehicle speed means that the host vehicle JV is in a state where an abnormal vehicle speed may be detected by the wheel speed sensor 9 (i.e., a state where the detected host vehicle speed is unreliable). For example, a state where the wheels of the host vehicle JV are slipping corresponds to an abnormal vehicle speed. Note that hereinafter, a state where the wheels of the JV are slipping is also simply referred to as a state where the host vehicle JV is slipping. Specifically, the abnormality detection unit 22 executes a subroutine shown in FIG. 11 (hereinafter also referred to as an abnormal vehicle speed detection process). The abnormal vehicle speed detection process will be described later.

[0030] Next, in S40, the switching unit 23 determines whether or not the vehicle speed is abnormal based on the detection result by the abnormality detection unit 22 in S30. Here, if the switching unit 23 determines that the vehicle speed is not abnormal (i.e., a normal vehicle speed is detected), it sets the first tracking process as the process (hereinafter also referred to as the processing mode) to be executed by the information processing device 3 in S50, and shifts the processing to S70. On the other hand, if the switching unit 23 determines that the vehicle speed is abnormal (i.e., a normal vehicle speed is not detected), it sets the second tracking process as the processing mode in S60, and shifts the processing to S70.

[0031] In S70, the tracking unit 27 determines whether or not there is unprocessed target information. The target information is stored in the memory 19 for each target. The target information indicates the past state of the target. That is, the target information includes past estimated values. In detail, the tracking unit 27 determines whether or not there is a target for which the subsequent processes of S80-S100 have not been executed among the registered targets. If it is determined in S70 that there is an unprocessed target, the tracking unit 27 shifts the process to S80 and executes the processes of S80-S100 for the selected target. On the other hand, if it is determined that there is no unprocessed target, the tracking unit 27 shifts the process to S110.

[0032] In S80, the prediction unit 24 calculates a predicted value of a target at the current time for one of the unprocessed targets based on an estimated value of the target in the past. For example, the estimated value of the target in the past refers to an estimated value of the target in the previous processing cycle. For example, the predicted value of the target at the current time refers to a predicted value of the target in the current processing cycle. The predicted value of the target includes distance, direction, and relative speed as elements, similar to the observed value. The predicted value of the target may include the ground speed of the target as an element. Specifically, the prediction unit 24 executes a subroutine shown in FIG. 6 (hereinafter, also referred to as prediction processing).

[0033] First, in S210, the prediction unit 24 calculates a predicted value of the target in the current processing cycle (hereinafter, also simply referred to as the current cycle) from the estimated value of the target in the previous processing cycle (hereinafter, also simply referred to as the previous cycle) for elements other than the relative speed among the predicted values ​​of the target. That is, the current predicted value is calculated from the previous estimated value for the distance of the target, and the current predicted value is calculated from the previous estimated value for the azimuth of the target.

[0034] Next, in S220-S250, the prediction unit 24 calculates the current predicted value for the relative speed of the target in a different manner depending on whether the processing mode is the first tracking process or not, in other words, whether a vehicle speed abnormality has occurred or not.

[0035] Here, if the processing mode is the first tracking process (that is, the vehicle speed is not abnormal), in S230-S240, a predicted value of the relative speed is calculated using the host vehicle speed detected in S20.

[0036] Specifically, in S230, the prediction unit 24 calculates a predicted value of the target's ground speed in the current processing cycle from the estimated value of the target's ground speed calculated in the previous processing cycle. During the calculation, it is assumed that the target moves at a constant speed in a short period such as the period from the previous processing cycle to the current processing cycle (i.e., the period Tcy). Then, for the target, the previous estimated value of the ground speed is used as the predicted value of the current ground speed.

[0037] In the next S240, the prediction unit 24 calculates a predicted value of the current relative speed of the target. Specifically, based on the predicted value of the current ground speed of the target and the vehicle speed detected in S20, the prediction unit 24 calculates the predicted value of the current relative speed of the target (the predicted value of the current ground speed of the target minus the vehicle speed) as the predicted value of the current relative speed. Then, the prediction unit 24 ends this subroutine.

[0038] On the other hand, when the processing mode is the second tracking process (i.e., the vehicle speed is abnormal), in S250, the predicted value of the current relative speed is calculated without using the vehicle speed detected in S20. Specifically, it is assumed that the movement of the vehicle JV and the target is a uniform linear motion in a short period such as the period from the previous processing cycle to the current processing cycle (i.e., the period Tcy). Then, the estimated value of the previous relative speed of the target is used as the predicted value of the current relative speed. Then, the prediction unit 24 ends this subroutine.

[0039] Next, in S90, the associating unit 25 sets a prediction gate for one of the unprocessed targets based on at least one element of the predicted value calculated in S80. The prediction gate refers to a range in which the current observation value is estimated to be acquired. In this embodiment, the associating unit 25 sets a prediction gate for three elements, namely, distance, direction, and relative speed. Furthermore, the associating unit 25 calculates an association cost. The association cost is an index indicating the degree of deviation between the predicted value and the observed value. In this embodiment, the smaller the value of the association cost, the higher the association between the predicted value and the observed value. In other words, the larger the value of the association cost, the lower the association between the predicted value and the observed value.

[0040] Then, the associating unit 25 determines the observed value with the lowest association cost among the observed values ​​within the prediction gate as the observed value to be associated with the predicted value. Specifically, the associating unit 25 executes a subroutine shown in FIG.

[0041] First, in S310, the association unit 25 sets a prediction gate, which is a range in which observation values ​​are estimated to be obtained in the current processing cycle, based on each of the predicted values ​​of elements other than relative velocity (i.e., distance, direction, relative velocity) calculated in S80.

[0042] The observed value detected from the same target as the predicted value should be close to the predicted value. Therefore, the range of observed values ​​estimated to be detected from the same target as the predicted value, centered on the predicted value calculated in S80, is set as a prediction gate. For example, as shown in FIG. 8, if a distance prediction value Rp is calculated in S80, a range of ±ΔR is set as a distance prediction gate (hereinafter also referred to as a prediction gate GR) for the distance prediction value Rp.

[0043] Also, for example, if the predicted azimuth value θp is calculated in S80, a range of ±Δθ for the predicted azimuth value θp is set as a prediction gate of the azimuth (hereinafter also referred to as a prediction gate Gθ).

[0044] Next, in S320-S340, the association unit 25 sets a prediction gate in a different manner for the relative speed depending on whether the processing mode is the first tracking process or not, in other words, whether a vehicle speed abnormality has occurred or not. In S320, the association unit 25 determines whether the processing mode is the first tracking process or not, and if it is determined that the processing mode is the first tracking process, the processing proceeds to S330, and if it is determined that the processing mode is the second tracking process, the processing proceeds to S340.

[0045] Here, when the processing mode is the first tracking process (i.e., the vehicle speed is not abnormal), in S330, the associating unit 25 sets, as a prediction gate, a range of observed values ​​of the relative speed estimated to be detected from the same target as the predicted value of the relative speed, centered on the predicted value of the relative speed calculated in S80. The set prediction gate is called the first prediction gate of the relative speed.

[0046] 8, for example, if the predicted value Vrp of the relative speed is calculated in S80, a range of ±ΔVr1 is set as a first prediction gate of the relative speed (hereinafter, also referred to as a first prediction gate Gv1) for the predicted value Vrp of the relative speed. The first prediction gate Gv1 (i.e., the range of ±ΔVr1) may be set to a predetermined range, or may be set variably according to, for example, the vehicle speed at the current time.

[0047] On the other hand, when the processing mode is the second tracking process (i.e., the vehicle speed is abnormal), in S340, the associating unit 25 sets a prediction gate that is a range of observed values ​​of the relative speed estimated to be detected from the same target as the predicted value of the relative speed, centered on the predicted value of the relative speed calculated in S80. The set prediction gate is called a second prediction gate of the relative speed.

[0048] 8, for example, if the predicted value Vrp of the relative speed is calculated in S80, a range of ±ΔVr2 is set as a second prediction gate of the relative speed (hereinafter, also referred to as a second prediction gate Gv2) for the predicted value Vrp of the relative speed. The second prediction gate Gv2 (i.e., the range of ±ΔVr2) may be set to a predetermined range, or may be set variably according to, for example, the vehicle speed at the current time. However, the second prediction gate Gv2 is set to a range wider than that of the first prediction gate.

[0049] In the next step S350, the degree to which elements (i.e., distance and direction) other than the relative speed among the predicted values ​​(i.e., distance, direction, and relative speed) contribute to the calculation of the association cost (hereinafter, also referred to as the contribution degree). For example, α r , α as the contribution of the orientation θ is set. α r , α θ is a positive value.

[0050] Next, in S360-S380, the association unit 25 sets the contribution of the relative speed when calculating the association cost in different manners depending on whether the processing mode is the first tracking process or not, in other words, depending on whether a vehicle speed abnormality has occurred or not. In S360, the association unit 25 determines whether the processing mode is the first tracking process or not, and if it is determined that the processing mode is the first tracking process, transitions to S370, and if it is determined that the processing mode is the second tracking process, transitions to S380.

[0051] Here, if the processing mode is the first tracking process (i.e., the vehicle speed is not abnormal), in S370, the associating unit 25 sets the first contribution degree α v1 Set. On the other hand, if the processing mode is the second tracking process (i.e., the vehicle speed is abnormal), in S380, the associating unit 25 sets the second contribution degree α v2 Set the first contribution α v1 and the second contribution rate α v2 is a positive value, and the second contribution α v2 is the first contribution α v1 For example, the second contribution rate α v2 is the first contribution α v1 For example, it may be set to a value less than 1, such as 1 / 100, 1 / 1000, or the like. In the next step S390, the associating unit 25 calculates an association cost. The association cost is expressed by, for example, equation (1). Note that each of the observed value and predicted value is a scalar quantity.

[0052]

number

[0053] where d a is the difference between the predicted distance and the observed distance, and d b is the difference between the predicted and observed orientations, and d c is the difference between the predicted value of the relative velocity and the observed value of the relative velocity. The difference here refers to the magnitude of the difference in the scalar quantity (i.e., the absolute value). In addition, in the case of the first tracking process, the contribution degree α v1 is used, and in the case of the second tracking process, the contribution of the relative velocity α v2 is used.

[0054] The associating unit 25 determines the observed value within the prediction gate that has the smallest calculated association cost as the observed value to be associated with the predicted value. Then, the associating unit 25 ends this subroutine.

[0055] Next, the operation of the association unit 25 will be described with reference to Fig. 8 and Fig. 9. For example, in Fig. 8, for distance, the observed value A1 and the observed value A2 are detected within the prediction gate GR based on the current predicted value Rp. However, the difference d between the predicted value Rp and the observed value A1 is a1 > The difference between the predicted value Rp and the observed value A2, d a2 Regarding the azimuth, the observation value B1 and the observation value B2 are detected within the prediction gate Gθ based on the current predicted value θp. However, the difference d between the predicted value θp and the observation value B1 b1 > Difference between predicted value θp and observed value B2 d b2 It is.

[0056] Regarding the relative velocity, the observed values ​​C1 and C2 are detected within the first prediction gate Gv1 based on the current relative velocity Vrp. The difference d between the predicted value Vrp and the observed value C1 c1 > Difference between predicted value Vrp and observed value C2 d c2 It is.

[0057] The observation values ​​A1, B1, and C1 are detection values ​​from the same target (e.g., the first target), and the observation values ​​A2, B2, and C2 are detection values ​​from the same target (e.g., the second target) different from the first target. In addition, when the vehicle speed is not abnormal, the contributions of the distance, the direction, and the relative speed when calculating the association cost are set to be equal (i.e., α r = α θ = α v1 ).

[0058] Here, for example, when the vehicle speed is not abnormal, the observed values ​​A2, B2, and C2 of the second target having a small association cost are determined as the observed values ​​to be associated with the current predicted values ​​Rp, θp, and Vrp, respectively, based on the calculated association cost. On the other hand, when the vehicle speed is abnormal, the observed values ​​A2, B2, and C2 of the second target having a small association cost are determined as the observed values ​​to be associated with the current predicted values ​​Rp, θp, and Vrp, respectively, based on the calculated association cost, similarly to the case when the vehicle speed is not abnormal in the example of FIG.

[0059] On the other hand, the example shown in FIG. 9 is similar to the example shown in FIG. 8 in terms of distance and direction, but in terms of relative speed, the observed value C1 is detected within the first prediction gate Gv1 based on the current relative speed Vrp, and the observed value C2 is not detected within the first prediction gate Gv1. However, the observed value C2 is detected within the second prediction gate Gv2 based on the current relative speed Vrp. In other words, when the vehicle speed is abnormal, both the observed values ​​A1, B1, and C1 detected from the first target and the observed values ​​A2, B2, and C2 detected from the second target are listed as candidates for the observed values ​​to be associated with the current predicted values ​​Rp, θp, and Vrp, respectively. Note that the difference d between the predicted value Vrp and the observed value C1 is c1 < Difference between predicted value Vrp and observed value C2 d c2 It is.

[0060] Here, for example, when the vehicle speed is not abnormal, the observed values ​​A1, B1, and C1 of the first target are determined as observed values ​​to be associated with the current predicted values ​​Rp, θp, and Vrp, respectively, based on the calculated association cost. a1 , d a2, d b1 , d b2 , d c1 , d c2 (i.e., d c1 <d c2 ) contribute equally. On the other hand, when the vehicle speed is abnormal, in the example of FIG. 9, the second association contribution degree α v2 (That is, 0<α v2 The association cost is calculated based on ≪1). As a result, the observed values ​​A2, B2, and C2 of the second target are determined as the observed values ​​to be associated with the current predicted values ​​Rp, θp, and Vrp, respectively. In calculating the association cost, d c1 , d c2 (i.e., d c1 <d c2 ) contribution is greatly reduced, and d a1 , d a2 (i.e., d a1 >d a2 ), d b1 , d b2 (i.e., d b1 >d b2 ) is dominant.

[0061] In this way, in the example of Figure 9, when the vehicle speed is abnormal, the contribution of the relative speed to the association cost is reduced, so that the observed values ​​A2 and B2 that are closer to the predicted values ​​in terms of elements other than the relative speed are determined to be the observed values ​​to be associated.

[0062] When the association process shown in FIG. 7 is completed, the process proceeds to S100 in FIG. In S100, the estimation unit 26 calculates an estimate value in the current processing cycle from the predicted value calculated in S80 and the observed value determined as the association target in S90, for example, by various filter processes. The estimated value of the target includes distance, direction, and relative speed as elements, similar to the observed value and predicted value. Specifically, the estimation unit 26 executes a subroutine shown in FIG. 10 (hereinafter, also referred to as estimation process).

[0063] First, in S410, the estimation unit 26 sets the degree to which the observed value contributes to the calculation of the estimated value of the target for elements other than the relative speed (hereinafter, also referred to as gain). The gain is a numerical value less than 1. The degree to which the predicted value contributes to the calculation of the estimated value of the target is calculated as "1-gain". For example, a gain β is set as the gain of the distance. r , Gain β as azimuth gain θ is set. Gain β r , β θ is a number less than 1.

[0064] Next, in S420-S440, the estimation unit 26 sets a gain for the relative speed in a different manner depending on whether the processing mode is the first tracking process or not, in other words, depending on whether a vehicle speed abnormality has occurred or not. As described above, the gain is a numerical value less than 1 that indicates the degree to which the observed value contributes to the calculation of the estimated value of the target. In S420, the estimation unit 26 determines whether the processing mode is the first tracking process or not, and if it is determined that the processing mode is the first tracking process, the estimation unit 26 transitions the processing to S430, and if it is determined that the processing mode is the second tracking process, the estimation unit 26 transitions the processing to S440.

[0065] Here, if the processing mode is the first tracking process (i.e., the vehicle speed is not abnormal), in S430, the estimation unit 26 sets the first gain β v1 Set. On the other hand, if the processing mode is the second tracking process (i.e., the vehicle speed is abnormal), in S440, the estimation unit 26 sets the second gain β v2 Set the second gain β v2 is the first gain β v1 For example, the second gain β v2 is the first gain β v1 Alternatively, the second gain β may be set to a value sufficiently larger than 1. v2 may be set to 1.

[0066] In the next step S450, the estimation unit 26 updates the filter. The filter is a filter that calculates an estimated value based on the formulas (2)-(4). Updating the filter means calculating an estimated value using each predicted value and observed value in the current processing cycle. Specifically, the estimation unit 26 calculates an estimated value in the current processing cycle based on the formulas (2)-(4) using each gain set in S410, S430, or S440.

[0067]

number

[0068] Setting the gain value small reduces the contribution of the observed value when calculating the estimated value, while setting the gain value large increases the contribution of the observed value when calculating the estimated value. In other words, setting the gain value small increases the calculated value of "1-gain," so that the contribution of the predicted value when calculating the estimated value increases, while setting the gain value large decreases the contribution of the predicted value when calculating the estimated value. The contribution of the predicted value when calculating an estimated value based on the observed value and the predicted value is a value obtained by subtracting the gain from 1. For example, the contribution of the predicted value of the relative speed when calculating an estimated value of the relative speed is (1-β v1 ), and if the vehicle speed is abnormal, (1-β v2 ). Regarding the relative speed, when the vehicle speed is abnormal, the degree to which the observed value contributes to the estimated value is increased (in other words, the degree to which the predicted value contributes is decreased) when calculating the estimated value of the target, and the second gain β v2 In other words, for the relative speed, the degree to which the predicted value contributes to the estimated value is greater when the vehicle speed is not abnormal (i.e., (1-β v1 )>(1-β v2 For this reason, in this embodiment, as described above, the gain of the relative velocity is set to 0<first gain β v1 <<Second gain β v2 < 1. Or, the second gain β v2 is set to 1.

[0069] 8, for example, for the distance, an estimated value K1 of the distance is calculated based on the predicted value Rp of the distance and the observed value A2 determined as the observed value of the distance to be associated. For example, for the direction, an estimated value L1 of the direction is calculated based on the predicted value θp of the direction and the observed value B2 determined as the observed value of the direction to be associated.

[0070] For example, for the relative speed, if the vehicle speed is not abnormal, an estimated value M1 of the relative distance is calculated based on the predicted value Vrp of the relative speed and the observed value C2 determined as the observed value of the relative speed to be associated. If the vehicle speed is abnormal, an estimated value M2 of the relative speed is calculated as a value closer to the observed value C2 based on the predicted value Vrp of the relative speed and the observed value C2 determined as the observed value of the relative speed to be associated.

[0071] Also, for example, in the example of Figure 9 described above, when the vehicle speed is abnormal, an estimated value M2 of the relative speed is calculated as a value close to the observed value C2 based on the predicted value Vrp of the relative speed and the observed value C2 determined as the associated observed value of the relative speed.

[0072] In addition, in S450, the estimation unit 26 may calculate an estimate of the target position based on the estimated distance K1 and the estimated direction L1 determined as described above. Also, in S450, the estimation unit 26 may calculate the ground speed of the target based on the estimated relative speed M1 or M2 determined as described above and the vehicle speed acquired in S20 (i.e., the estimated ground speed of the target = the estimated relative speed + the detected vehicle speed). The calculated estimated value of the target is stored in the memory 19. The estimation unit 26 then ends this subroutine.

[0073] When the estimation process shown in Fig. 10 is completed, the process proceeds to S70 in the object tracking process shown in Fig. 5. Then, while unprocessed target information exists, the processes of S70-S100 are repeatedly executed. On the other hand, when there is no unprocessed target information and the tracking unit 27 determines in S70 that there is no unprocessed target information, the process proceeds to S110.

[0074] In S110, the tracking unit 27 determines whether or not there is an unused observed value among the observed values ​​detected in S10. That is, it determines whether or not there is an observed value that is not associated with any predicted value among the observed values ​​detected in S10. If the tracking unit 27 determines that there is no unused observed value, it ends this process. On the other hand, if the tracking unit 27 determines that there is an unused observed value, it shifts the process to S120.

[0075] In S120, the tracking unit 27 registers the unused observation value (i.e., the target for which the unused observation value was detected) as a new target. After that, the process returns to S110, and while there is an unused observation value for which the processes of S110-S120 have not been performed, the process of S110-S120 is repeatedly performed. Then, the tracking unit 27 ends this tracking process.

[0076] [2-2. Vehicle speed abnormality detection processing] Next, the vehicle speed abnormality detection process executed by the information processing device 3 of the first embodiment in S30 of the tracking process will be described with reference to the flowchart of FIG.

[0077] First, in S500, the abnormality detection unit 22 determines whether the acceleration of the host vehicle JV is equal to or greater than a predetermined acceleration threshold value. In the above-mentioned S20, the abnormality detection unit 22 detects the rotation speed of each wheel based on the wheel speed signal from the wheel speed sensor 9 provided on each wheel, detects the speed of the host vehicle JV based on the detected rotation speed of each wheel, and stores the detection result in the memory 19.

[0078] The abnormality detection unit 22 may calculate the acceleration of the host vehicle JV based on the host vehicle speed detected in S20, for example, by calculating the difference between the host vehicle speed detected in the current processing cycle and the host vehicle speed detected in the previous processing cycle. The acceleration threshold may be set to a value that is large enough to determine whether the host vehicle JV is slipping or not. For example, the acceleration threshold may be set to a value smaller than the acceleration of the host vehicle JV that can be measured when the vehicle is slipping. The acceleration threshold and the host vehicle speeds detected in the current and previous cycles are stored in the memory 19. The abnormality detection unit 22 ends this process when the acceleration of the host vehicle JV is less than the acceleration threshold value, and transitions to S510 when the acceleration of the host vehicle JV is equal to or greater than the acceleration threshold value. Note that the various threshold values ​​described below are stored in the memory 19 in advance.

[0079] Next, in S510, the abnormality detection unit 22 determines whether the number of stationary objects is equal to or greater than a predetermined threshold (hereinafter also referred to as the stationary object threshold). The number of stationary object thresholds may be set to a predetermined value, such as 1 to several tens. In this embodiment, the number of stationary object thresholds is an integer equal to or greater than 2. For example, based on the estimated value of the relative speed of the target in the previous processing cycle, the abnormality detection unit 22 determines that the target is a stationary object when the estimated value of the relative speed is equal to and has an opposite sign to the host vehicle speed. The host vehicle speed refers to the host vehicle speed detected in S20. Here, the abnormality detection unit 22 shifts the process to S540 when the number of stationary objects (hereinafter also referred to as the number of targets) is less than a predetermined threshold (hereinafter also referred to as the stationary object threshold). Then, in S540, the abnormality detection unit 22 determines that the vehicle speed is normal and ends this subroutine. On the other hand, the abnormality detection unit 22 shifts the process to S520 when the number of stationary objects is equal to or greater than the stationary object threshold, and from S520 onwards, further calculates the prediction residual of the relative speed of each stationary object to determine whether the vehicle speed is abnormal.

[0080] In S520, the anomaly detection unit 22 calculates a prediction residual of the relative speed for each stationary object identified in S510. However, the calculation of the prediction residual of the relative speed in S520 does not necessarily have to be performed for all stationary objects detected in S510. For example, the process from S520 onwards may be performed only for each of the stationary objects corresponding to the number of stationary object targets described later. The prediction residual is the difference between the predicted value and the observed value (i.e., predicted value - observed value). Specifically, the anomaly detection unit 22 executes a subroutine shown in Fig. 12 (hereinafter also referred to as stationary object prediction residual process).

[0081] First, in S600, the abnormality detection unit 22 selects a predetermined number of stationary objects (hereinafter, referred to as the number of stationary objects) from among the stationary objects identified in S510. For example, the number of stationary objects may be set to a value equal to or less than the above-mentioned stationary object threshold, such as 1 to several tens of stationary objects. In this embodiment, the number of stationary objects is an integer equal to or greater than 2. Furthermore, the abnormality detection unit 22 may select the number of stationary objects from among the stationary objects identified in S510 in order of the distance from the host vehicle JV to the stationary object number.

[0082] Next, in S610, the abnormality detection unit 22 judges whether or not there is an unprocessed stationary object. More specifically, it judges whether or not there is a stationary object for which the subsequent processes of S620-S640 have not been executed among the stationary objects corresponding to the number of stationary object targets selected in S600. If it is judged in S610 that there is an unprocessed stationary object, the abnormality detection unit 22 selects one of the unprocessed stationary objects, proceeds to the process of S620, and executes the processes of S620-S640 for the relative velocity of the selected stationary object. On the other hand, if it is judged that there is no unprocessed stationary object, the abnormality detection unit 22 shifts the process to S650.

[0083] In the next step S620, the abnormality detection unit 22 selects one stationary object from the unprocessed stationary objects, and calculates a predicted value for this stationary object at the current time (for example, the current processing cycle) from the estimated value in the previous processing cycle. For example, the abnormality detection unit 22 calculates the current predicted value in the same manner as in the first tracking process in S80 described above.

[0084] Next, in S630, the anomaly detection unit 22 performs association in the same manner as the first tracking process of S90 described above based on the predicted value of the relative velocity calculated in S620, and determines the observation value with the smallest relative velocity association cost among the observation values ​​within the prediction gate as the observation value of the relative velocity to be associated with the predicted value of the relative velocity.

[0085] In the next step S640, the difference in relative velocity between the observed value that results in the smallest association cost determined in S630 and the predicted value calculated in S620 (i.e., the absolute value of the relative velocity predicted value - the relative velocity observed value) is calculated as the prediction residual of the relative velocity for stationary objects.

[0086] Then, the abnormality detection unit 22 shifts the process to S610, and repeats the processes of S620-S640 while there are unprocessed stationary objects. On the other hand, when there are no more unprocessed stationary objects and it is determined that there are no unprocessed stationary objects, the abnormality detection unit 22 shifts the process to S650.

[0087] In S650, the abnormality detection unit 22 calculates the average value of the prediction residuals of the relative speed of the stationary objects. That is, the average value of the prediction residuals for the number of stationary object targets is calculated. Hereinafter, the average value of the prediction residuals of the relative speed of the stationary objects is also simply referred to as the prediction residual of the relative speed of the stationary objects. The abnormality detection unit 22 then ends this subroutine.

[0088] When the subroutine shown in FIG. 12 ends, the abnormality detection unit 22 shifts the process to S530. In S530, the abnormality detection unit 22 judges whether the prediction residual of the relative speed of the stationary object is equal to or greater than a predetermined threshold (hereinafter also referred to as the prediction residual threshold). When the target is a stationary object, it is considered that there is no deviation between the predicted value and the observed value of the relative speed. In other words, it is considered that the prediction residual is approximately 0. Therefore, for example, the prediction residual threshold may be set to a positive value larger than 0 and close to 0.

[0089] If the prediction residual of the relative speed of the stationary object is less than the prediction residual threshold, the abnormality detection unit 22 shifts the process to S540. In S540, the abnormality detection unit 22 determines that the vehicle speed is not abnormal, that is, that the detected vehicle speed is normal, and ends this subroutine.

[0090] On the other hand, if the prediction residual of the stationary object relative speed is equal to or greater than the prediction residual threshold, the abnormality detection unit 22 shifts the process to S550. Then, in S550, the abnormality detection unit 22 determines that the vehicle speed is abnormal, and ends this subroutine. When this subroutine (i.e., the stationary object prediction residual process) ends, the process shifts to S40.

[0091] [1-3. Effects] According to the first embodiment described above, the following effects can be obtained. (1a) When it is determined that no abnormality in vehicle speed has occurred, a current estimated value of the target is calculated by a first tracking process using a predicted value of the relative speed based on the detection result of the vehicle speed. On the other hand, when it is determined that an abnormality in vehicle speed has occurred, a current estimated value of the target is calculated by a second tracking process different from the first tracking process.

[0092] This allows the process to be switched to either the first tracking process or the second tracking process based on the result of the vehicle speed abnormality determination. For example, the second tracking process is a process in which the estimated value of the relative speed is less susceptible to the effect of errors in the vehicle speed, and when the vehicle speed is determined to be abnormal, the second tracking process is executed instead of the first tracking process, thereby suppressing the effect of errors in the vehicle speed when estimating the relative speed.

[0093] As a result, when the vehicle speed is not abnormal, the target (i.e., object) can be tracked (e.g., the object's position, ground speed, etc.) with high accuracy by reflecting the vehicle speed, and when the vehicle speed abnormality is detected, the accuracy of the target tracking can be prevented from decreasing. Note that tracking of a target means repeating an estimated value (e.g., position, ground speed, etc.) indicating the state of the target in a time series (i.e., with the passage of time), that is, repeating the current state based on the past state.

[0094] (1b) For each calculated current predicted value of the object, a prediction gate is set, which is a range within which the current observed value is estimated to be obtained, based on the predicted value. Also, from at least one detected observed value, an observed value to be associated with the predicted value is determined from among the observed values ​​within the prediction gate. Then, an estimated value of the current object is calculated based on the determined observed value and predicted value. In this way, by repeatedly calculating estimated values ​​based on the observed values ​​and predicted values ​​and obtaining estimated values ​​in a time series, it is possible to track the target more accurately than when only observed values ​​or only predicted values ​​are used.

[0095] (1c) If it is determined that the vehicle speed is not abnormal, the current predicted value of the relative speed is calculated using the vehicle speed. For example, if the vehicle JV accelerates or decelerates between the previous processing cycle and the current processing cycle (i.e., before and after the processing cycle), the vehicle speed reflecting this acceleration or deceleration is detected by the wheel speed sensor 9. Then, by subtracting the detected vehicle speed from the estimated value of the ground speed of the target, the predicted value of the relative speed is accurately predicted using the vehicle speed. Since an accurate predicted value of the relative speed, and therefore an accurate estimated value of the relative speed, can be obtained, the state of the target, such as the target position, the target ground speed, etc., can be accurately calculated. As a result, when the vehicle speed is not abnormal, the target can be tracked accurately.

[0096] However, the vehicle speed is not always detected accurately. For example, when the wheels spin (i.e., slip), the vehicle speed may be detected as a value different from the actual speed. Therefore, the presence or absence of an abnormality in the vehicle speed (i.e., the certainty of the vehicle speed) is determined, and if it is determined that the vehicle speed is abnormal (i.e., the vehicle speed may be uncertain), the current predicted value of the relative speed is calculated without using the vehicle speed. For example, the same value as the previous estimated value of the relative speed is calculated as the current predicted value of the relative speed. This makes it possible to reduce the influence of errors in the vehicle speed in the calculation of the predicted value of the relative speed, compared to calculating the predicted value of the relative speed using the vehicle speed when the vehicle speed is abnormal. As a result, it is possible to suppress a decrease in the accuracy of tracking the target when an abnormality in the vehicle speed is detected.

[0097] (1d) The prediction gate when it is determined that the vehicle speed is abnormal (i.e., the second prediction gate Gv2) is set larger (i.e., wider) than the prediction gate when it is determined that the vehicle speed is not abnormal (i.e., the first prediction gate Gv1). As a result, even if a deviation occurs in the predicted value of the relative speed, the prediction gate is set large, so that it is possible to appropriately associate the relative speed with the observed value even within the prediction gate centered on the predicted value of the relative speed where a deviation occurs. In other words, it is possible to suppress the influence of errors in the host vehicle speed in the calculation of the estimated value of the relative speed. As a result, it is possible to suppress a decrease in the accuracy of tracking of the target when a vehicle speed abnormality is detected.

[0098] (1e) When it is determined that the vehicle speed is abnormal, the contribution of the relative speed in the calculation of the association cost, which is an index showing the degree of deviation between the predicted value and the observed value, is made smaller than when it is determined that the vehicle speed is not abnormal. As a result, even if there is a deviation in the predicted value of the relative speed, the association cost is calculated mainly based on the predicted values ​​and observed values ​​of elements other than the relative speed (e.g., distance, direction), and it becomes possible to perform association based on this association cost. As a result, it is possible to suppress a decrease in the accuracy of tracking of targets when an abnormal vehicle speed is detected.

[0099] (1f) When it is determined that the vehicle speed is abnormal, the contribution of the predicted value to the estimated value is reduced when calculating an estimated value of the relative speed based on the observed value and the predicted value, compared to when it is determined that the vehicle speed is not abnormal. This makes the observed value more likely, and reduces the influence of the prediction error of the relative speed on the estimated value. As a result, it is possible to suppress a decrease in the accuracy of tracking the target when an abnormal vehicle speed is detected.

[0100] (1g) Based on the magnitude of the acceleration of the host vehicle JV, if the acceleration is equal to or greater than a predetermined acceleration threshold, it is determined that the vehicle speed is abnormal. This makes it possible to determine the vehicle speed abnormality using a relatively simple method.

[0101] (1h) When the prediction residual of the relative speed of the target is equal to or greater than a predetermined prediction residual threshold, it may be determined that the vehicle speed is abnormal. For example, in a short period of time such as before or after a processing cycle, whether the target is a stationary or moving object, if the vehicle speed is not abnormal, the prediction residual of the relative speed of the target is considered to be within a predetermined range. This makes it possible to detect the vehicle speed abnormality based on the prediction residual. In particular, in this embodiment, when the acceleration of the host vehicle JV is equal to or greater than a predetermined acceleration threshold and the prediction residual of the relative speed of the target is equal to or greater than a prediction residual threshold, it is determined that the vehicle speed is abnormal. This makes it possible to accurately determine the vehicle speed abnormality based on multiple conditions.

[0102] (1i) When the prediction residual of the relative speed of a stationary object among the targets is equal to or greater than the prediction residual threshold, it is determined that the vehicle speed is abnormal. For example, when the target is a stationary object, the predicted value and the observed value are considered to be equal (i.e., the prediction residual is approximately 0), so the prediction residual threshold may be set to a value close to 0. Since a stationary object is stationary and does not change its moving speed like a moving object, by determining the vehicle speed abnormality using the prediction residual of the relative speed of the stationary object, it is possible to perform a more accurate determination than when using the prediction residual of the relative speed of a moving object.

[0103] (1j) When the number of stationary objects is equal to or greater than the stationary object threshold, at least one stationary object is selected, and when the prediction residual of the relative speed of the selected stationary object is equal to or greater than the prediction residual threshold, it is determined that a vehicle speed abnormality exists. For example, a number of stationary objects equal to the number of stationary object targets are selected, an average value of the prediction residual of the relative speed of the stationary objects is calculated, and when the average value of the prediction residual is equal to or greater than the prediction residual threshold, it is determined that a vehicle speed abnormality exists. This makes it possible to increase the accuracy of determining a vehicle speed abnormality compared to when a single stationary object is selected and a determination is made based on the prediction residual.

[0104] [1-4. Correspondence of Wording] Note that the object tracking device 1 corresponds to the object tracking device, the radar device 2 corresponds to the sensor, the information processing device 3 corresponds to the information processing device, the sensor unit 21 corresponds to the observation value detection unit, the abnormality detection unit 22 corresponds to the abnormality determination unit, the prediction unit 24 corresponds to the prediction unit, the association unit 25 corresponds to the association unit, the estimation unit 26 corresponds to the estimation unit, and the tracking unit 27 corresponds to the tracking unit. The host vehicle JV corresponds to the vehicle. Also, the first tracking process (i.e., the process executed when the processing mode is the first tracking process) corresponds to the first process. Specifically, S230-S240, S330, S370, and S430 correspond to the first process.

[0105] The second tracking process (i.e., the process executed when the processing mode is the second tracking process) corresponds to the second process. Specifically, S250, S340, S380, and S440 correspond to the second process. S10 corresponds to the process of the observation value detection unit, S20-S30 correspond to the process of the anomaly determination unit, S80 corresponds to the process of the prediction unit, S90 corresponds to the process of the association unit, S100 corresponds to the process of the estimation unit, and S70-S120 correspond to the process of the tracking unit. The transmitted wave and the reflected wave correspond to the radar wave, the estimated value of the target in the previous processing cycle corresponds to the past estimated value, and the relative speed calculated in S240 corresponds to the relative speed calculated using the host vehicle speed. The prediction gate, the first prediction gate, and the second prediction gate correspond to the prediction range, and the first contribution of the relative velocity α v1 , the second contribution of the relative velocity α v2corresponds to the contribution of the relative speed when calculating the association cost. (1-first gain β v1 ), (1-second gain β v2 ) corresponds to the contribution of the predicted value of the relative velocity when calculating the estimated value of the relative velocity.

[0106] [2. Second embodiment] [2-1. Differences from the first embodiment] The second embodiment has a basic configuration similar to that of the first embodiment, and therefore differences will be described below. Note that the same reference numerals as those in the first embodiment indicate the same configuration, and the preceding description will be referred to.

[0107] In the first embodiment described above, when the vehicle speed is not abnormal, the prediction unit 24 calculates a predicted value of the current relative speed by subtracting the current detection value of the vehicle speed from the previous estimated value of the ground speed of the target using the current detection value of the vehicle speed. In contrast, the second embodiment differs from the first embodiment in that the predicted value of the current relative speed is calculated using the difference between the previous detection value of the vehicle speed and the current detection value of the vehicle speed.

[0108] [2-2. Processing] Next, a process executed by the information processing device 3 of the second embodiment in place of the prediction process of the first embodiment (i.e., FIG. 6) will be described with reference to the flowchart of FIG. 13. Note that the processes of S210-220 and S250 in FIG. 13 are similar to those in FIG. 6, and therefore the description thereof will be partially simplified.

[0109] In S235, to which the process proceeds when it is determined in S220 that the first tracking process is in progress (i.e., the vehicle speed is not abnormal), the prediction unit 24 calculates a predicted value of the relative speed. The prediction unit 24 calculates the difference between the vehicle speed acquired in S20 in the current processing cycle and the vehicle speed acquired in S20 in the previous processing cycle (i.e., the current vehicle speed - the previous vehicle speed). That is, the difference in the vehicle speed before and after the processing cycle (i.e., the speed change amount of the vehicle JV before and after the processing cycle) is calculated. The prediction unit 24 calculates a value obtained by subtracting the speed change amount of the vehicle JV before and after the processing cycle from the estimated value of the previous relative speed as the predicted value of the current relative speed. The prediction unit 24 then ends the subroutine of the prediction process.

[0110] [2-3. Effects] According to the second embodiment described above in detail, it is possible to obtain the effects (1a)-(1j) of the first embodiment described above, and further, the following effect is obtained.

[0111] (2a) The value obtained by subtracting the speed change amount of the host vehicle JV before and after the processing cycle from the previous estimated value of the relative speed is calculated as the current predicted value of the relative speed. As a result, when acceleration or deceleration of the host vehicle JV is detected based on the change in the host vehicle speed, this acceleration or deceleration amount (i.e., the speed change amount) is reflected in the relative speed, and the relative speed can be predicted with high accuracy. For example, when the host vehicle JV is accelerating, a value obtained by subtracting the magnitude of the acceleration amount from the previous estimated value of the relative speed is calculated as the predicted value of the relative speed, and when the host vehicle JV is decelerating, a value obtained by increasing the magnitude of the deceleration amount from the previous estimated value of the relative speed is calculated as the predicted value of the relative speed.

[0112] Since it is possible to obtain an accurate prediction value of the relative speed, and therefore an accurate estimation value of the relative speed, it is possible to accurately calculate the state of the object, such as the object's position, the object's ground speed, etc. As a result, as in the above-mentioned (1c), when the vehicle speed is not abnormal, it is possible to accurately track the target.

[0113] In the above embodiment, S235 corresponds to the first process, and the relative speed calculated in S235 corresponds to the relative speed calculated using the host vehicle speed. 3. Other embodiments Although the embodiments of the present disclosure have been described above, the present disclosure is not limited to the above-described embodiments and can be implemented in various modified forms.

[0114] (3a) In the information processing device 3 described in the present disclosure, the abnormality detection unit 22 determines whether or not the vehicle speed is abnormal by using the average value of the prediction residuals of the relative speeds of a plurality of stationary objects. However, the present disclosure is not limited to this. For example, the prediction residuals of the relative speeds of one stationary object may be used to determine whether or not the vehicle speed is abnormal. For example, one stationary object that is closest to the vehicle JV may be selected, and the prediction residuals of this stationary object may be used to determine whether or not the vehicle speed is abnormal. In this case, for example, S510-S520 may be deleted in the processing of FIG. 11.

[0115] (3b) In the information processing device 3 described in the present disclosure, the smaller the association cost value, the higher the correlation between the predicted value and the observed value, and the larger the association cost value, the lower the correlation between the predicted value and the observed value; however, the present disclosure is not limited to this. For example, the larger the association cost value, the higher the correlation between the predicted value and the observed value, and the smaller the association cost value, the lower the correlation between the predicted value and the observed value. In this case, for example, instead of the difference between the predicted value and the observed value, the reciprocal of the difference between the predicted value and the observed value may be used to calculate the association cost. Second association contribution degree α of relative speed v2 is the first association contribution degree α of the relative velocity, similarly to the above-described information processing device 3. v1 It is sufficient that the number is set to a number less than 1, which is sufficiently smaller than the

[0116] (3c) In the information processing device 3 described in the present disclosure, the prediction unit 24 may be configured to always calculate the relative speed with respect to the host vehicle JV using the host vehicle speed regardless of the presence or absence of a vehicle speed abnormality. That is, the prediction unit 24 may be configured to delete the processes of S220 and S250. In this case, it is sufficient that at least one of the setting of the prediction gate by the association unit 25, the setting of the association contribution degree by the association unit 25, and the setting of the gain by the estimation unit 26 is configured to execute a process (i.e., S340, S380, S440) in which the processing mode is set to the second tracking process when the vehicle speed is abnormal. This reduces the influence of the detection error of the host vehicle speed in the tracking process when the vehicle speed abnormality is detected, and as a result, it is possible to suppress a decrease in the accuracy of tracking of the target.

[0117] (3d) Similarly, in the information processing device 3 described in the present disclosure, the association unit 25 may be configured to set the prediction gate in the same manner regardless of the presence or absence of a vehicle speed abnormality. That is, the association unit 25 may be configured to delete the processes of S320 and S340. In this case, it is sufficient that at least one of the calculation of the predicted value of the relative speed by the prediction unit 24, the setting of the association contribution degree by the association unit 25, and the setting of the gain by the estimation unit 26 is configured to execute a process (i.e., S250, S380, S440) of setting the processing mode to the second tracking process when the vehicle speed is abnormal.

[0118] (3e) Similarly, in the information processing device 3 described in the present disclosure, the associating unit 25 may be configured to set the association contribution degree in the same manner regardless of the presence or absence of a vehicle speed abnormality. That is, the associating unit 25 may be configured to delete the processes of S360 and S380. In this case, it is sufficient that at least one of the calculation of the predicted value of the relative speed by the predicting unit 24, the setting of the prediction gate by the associating unit 25, and the setting of the gain by the estimating unit 26 is configured to execute a process (i.e., S250, S340, S440) of setting the processing mode to the second tracking process when the vehicle speed is abnormal.

[0119] (3f) Similarly, in the information processing device 3 described in the present disclosure, the estimation unit 26 may be configured to set the gain in the same manner regardless of the presence or absence of a vehicle speed abnormality. That is, the estimation unit 26 may be configured to delete the processes of S420 and S430. In this case, it is sufficient that the estimation unit 26 is configured to execute a process (i.e., S250, S340, S380) of setting the processing mode to the second tracking process when the vehicle speed is abnormal in at least one of the calculation of the predicted value of the relative speed by the prediction unit 24, the setting of the prediction gate by the association unit 25, and the setting of the association contribution degree by the association unit 25.

[0120] (3g) In the information processing device 3 described in the present disclosure, the estimated value for indicating the state of the target includes at least distance, direction, and relative speed as elements, but the elements included in the estimated value are not limited to these. For example, the estimated value may include the position of the object as an element. Specifically, the estimated value may include the X-axis coordinate value Cx and the Y-axis coordinate value Cy as elements. The X-axis is an axis along the width direction of the host vehicle JV, and the Y-axis is an axis perpendicular to the X-axis and along the longitudinal direction of the host vehicle JV. For these, a predicted value, an observed value, and an estimated value may be calculated, respectively.

[0121] 14, first, a predicted value Pp of the object position in the current processing cycle is calculated from an estimated value Pe in the previous processing cycle. Next, among the observed values ​​of the object position observed by the radar device 2, observed values ​​D1 and D2 within a prediction gate Gp related to the position set with the predicted value Pp at the center are detected as observed values ​​that may be associated with the predicted value Pp.

[0122] If the distance between the predicted value Pp and the observed value is used as the association cost, the observed value D1 is determined as the observed value associated with the predicted value Pp. Then, the associated observed value D1 and predicted value Pp are used to calculate the estimated value N of the position in the current processing cycle based on the gain related to the position.

[0123] In the information processing device 3 described in the present disclosure, the estimated value may have the same elements as the observed value or the predicted value, or may have elements different from the observed value or the predicted value.

[0124] (3h) The information processing device 3 and the method thereof described in the present disclosure may be realized by a special-purpose computer provided by configuring a processor and a memory programmed to execute one or more functions embodied in a computer program. Alternatively, the information processing device 3 and the method thereof described in the present disclosure may be realized by a special-purpose computer provided by configuring a processor with one or more dedicated hardware logic circuits.

[0125] Alternatively, the information processing device 3 and the method thereof described in the present disclosure may be realized by one or more dedicated computers configured by combining a processor and memory programmed to perform one or more functions with a processor configured by one or more hardware logic circuits.

[0126] The computer program may be stored in a computer-readable non-transitory tangible recording medium as instructions executed by a computer. The method for realizing the functions of each unit included in the information processing device 3 does not necessarily need to include software, and all of the functions may be realized using one or more pieces of hardware. (3i) The information processing device 3 described in the present disclosure may be configured on a single chip.

[0127] (3j) Multiple functions possessed by one component in the above-mentioned embodiments may be realized by multiple components, or one function possessed by one component may be realized by multiple components. Also, multiple functions possessed by multiple components may be realized by one component, or one function realized by multiple components may be realized by one component. Also, part of the configuration of the above-mentioned embodiments may be omitted. Also, at least part of the configuration of the above-mentioned embodiments may be added to or substituted for the configuration of another of the above-mentioned embodiments.

[0128] (3k) In addition to the above-mentioned information processing device 3, the present disclosure can be realized in various forms, such as the CPU 11 of the information processing device 3, the object tracking device 1 including the information processing device 3 as a component, a program for causing the information processing device 3 to function, a program for causing the CPU 11 of the information processing device 3 to function, a non-transient tangible recording medium such as a semiconductor memory on which this program is recorded, and a tracking method realized by this program. Also, the present disclosure can be realized in various forms, such as a method realized by the information processing device 3, a method realized by the CPU 11 of the information processing device 3, and a tracking method of the object tracking device 1. [Explanation of symbols]

[0129] 1... object tracking device, 2... radar device, 3... information processing device, 21... sensor unit, 22... abnormality detection unit, 27... tracking processing unit.

Claims

1. An information processing device (3) mounted on a vehicle, an observation value detection unit (21) configured to acquire an observation signal observed by a sensor (2) that transmits and receives radar waves, and detect at least one observation value for at least one target around the vehicle from the observation signal; a tracking unit (27) configured to track the target by calculating, for each target, an estimated value indicating a state of the target from a past estimated value, and calculating the current estimated value from the current observed value and the current predicted value in a predetermined processing cycle; an abnormality determination unit (22) configured to determine whether or not a vehicle speed abnormality exists as an abnormality in detection of the vehicle speed; Equipped with The tracking unit is When it is determined that the vehicle speed is not abnormal, a first process using the predicted value of the relative speed based on a detection result of the vehicle speed is executed to calculate the current estimated value, and when it is determined that the vehicle speed is abnormal, a second process different from the first process is executed to calculate the current estimated value, a prediction unit (24) configured to calculate, for each target, the current predicted value from the past estimated value; an association unit (25) configured to set, for each of the calculated predicted values, a prediction range within which the observed value is estimated to be acquired this time based on the predicted value, determine an observed value within the prediction range to be associated with the predicted value from the at least one detected observed value, and determine the observed value to be associated with the calculated predicted value; an estimator (26) configured to calculate the estimated value of the current target based on the determined observed value and the predicted value; Equipped with When it is determined that the vehicle speed is not abnormal, the prediction unit calculates the current predicted value of the relative speed by using the host vehicle speed in the first process, and when it is determined that the vehicle speed is abnormal, the prediction unit calculates the predicted value of the relative speed without using the host vehicle speed in the second process. Information processing device.

2. 2. The information processing device according to claim 1, The association unit sets the prediction range to be larger in the second process when it is determined that the vehicle speed is abnormal than in the first process when it is determined that the vehicle speed is not abnormal. Information processing device.

3. An information processing device (3) mounted on a vehicle, an observation value detection unit (21) configured to acquire an observation signal observed by a sensor (2) that transmits and receives radar waves, and detect at least one observation value for at least one target around the vehicle from the observation signal; a tracking unit (27) configured to track the target by calculating, for each target, an estimated value indicating a state of the target from a past estimated value, and calculating the current estimated value from the current observed value and the current predicted value in a predetermined processing cycle; an abnormality determination unit (22) configured to determine whether or not a vehicle speed abnormality exists as an abnormality in detection of the vehicle speed; Equipped with The tracking unit is When it is determined that the vehicle speed is not abnormal, a first process using the predicted value of the relative speed based on a detection result of the vehicle speed is executed to calculate the current estimated value, and when it is determined that the vehicle speed is abnormal, a second process different from the first process is executed to calculate the current estimated value, a prediction unit (24) configured to calculate, for each target, the current predicted value from the past estimated value; an association unit (25) configured to set, for each of the calculated predicted values, a prediction range within which the observed value is estimated to be acquired this time based on the predicted value, determine an observed value within the prediction range to be associated with the predicted value from the at least one detected observed value, and determine the observed value to be associated with the calculated predicted value; an estimator (26) configured to calculate the estimated value of the current target based on the determined observed value and the predicted value; Equipped with The association unit sets the prediction range to be larger in the second process when it is determined that the vehicle speed is abnormal than in the first process when it is determined that the vehicle speed is not abnormal. Information processing device.

4. The information processing device according to any one of claims 1 to 3, In the second process when it is determined that the vehicle speed is abnormal, the associating unit reduces a contribution of the relative speed when calculating an association cost that is an index indicating a degree of deviation between the predicted value and the observed value, compared to the first process when it is determined that the vehicle speed is not abnormal. Information processing device.

5. An information processing device (3) mounted on a vehicle, an observation value detection unit (21) configured to acquire an observation signal observed by a sensor (2) that transmits and receives radar waves, and detect at least one observation value for at least one target around the vehicle from the observation signal; a tracking unit (27) configured to track the target by calculating, for each target, an estimated value indicating a state of the target from a past estimated value, and calculating the current estimated value from the current observed value and the current predicted value in a predetermined processing cycle; an abnormality determination unit (22) configured to determine whether or not a vehicle speed abnormality exists as an abnormality in detection of the vehicle speed; Equipped with The tracking unit is When it is determined that the vehicle speed is not abnormal, a first process using the predicted value of the relative speed based on a detection result of the vehicle speed is executed to calculate the current estimated value, and when it is determined that the vehicle speed is abnormal, a second process different from the first process is executed to calculate the current estimated value, a prediction unit (24) configured to calculate, for each target, the current predicted value from the past estimated value; an association unit (25) configured to set, for each of the calculated predicted values, a prediction range within which the observed value is estimated to be acquired this time based on the predicted value, determine an observed value within the prediction range to be associated with the predicted value from the at least one detected observed value, and determine the observed value to be associated with the calculated predicted value; an estimator (26) configured to calculate the estimated value of the current target based on the determined observed value and the predicted value; Equipped with In the second process when it is determined that the vehicle speed is abnormal, the associating unit reduces a contribution of the relative speed when calculating an association cost that is an index indicating a degree of deviation between the predicted value and the observed value, compared to the first process when it is determined that the vehicle speed is not abnormal. Information processing device.

6. 6. The information processing device according to claim 1, The estimation unit, in the second process when it is determined that the vehicle speed is abnormal, reduces a contribution rate of the predicted value of the relative speed when calculating the estimated value of the relative speed compared to the first process when it is determined that the vehicle speed is not abnormal. Information processing device.

7. An information processing device (3) mounted on a vehicle, an observation value detection unit (21) configured to acquire an observation signal observed by a sensor (2) that transmits and receives radar waves, and detect at least one observation value for at least one target around the vehicle from the observation signal; a tracking unit (27) configured to track the target by calculating, for each target, an estimated value indicating a state of the target from a past estimated value, and calculating the current estimated value from the current observed value and the current predicted value in a predetermined processing cycle; an abnormality determination unit (22) configured to determine whether or not a vehicle speed abnormality exists as an abnormality in detection of the vehicle speed; Equipped with The tracking unit is When it is determined that the vehicle speed is not abnormal, a first process using the predicted value of the relative speed based on a detection result of the vehicle speed is executed to calculate the current estimated value, and when it is determined that the vehicle speed is abnormal, a second process different from the first process is executed to calculate the current estimated value, a prediction unit (24) configured to calculate, for each target, the current predicted value from the past estimated value; an association unit (25) configured to set, for each of the calculated predicted values, a prediction range within which the observed value is estimated to be acquired this time based on the predicted value, determine an observed value within the prediction range to be associated with the predicted value from the at least one detected observed value, and determine the observed value to be associated with the calculated predicted value; an estimator (26) configured to calculate the estimated value of the current target based on the determined observed value and the predicted value; Equipped with The estimation unit, in the second process when it is determined that the vehicle speed is abnormal, reduces a contribution rate of the predicted value of the relative speed when calculating the estimated value of the relative speed compared to the first process when it is determined that the vehicle speed is not abnormal. Information processing device.

8. The information processing device according to any one of claims 1 to 7, The abnormality determination unit determines that the vehicle speed is abnormal when the acceleration is equal to or greater than a predetermined acceleration threshold based on a magnitude of the acceleration of the host vehicle. Information processing device.

9. An information processing device (3) mounted on a vehicle, an observation value detection unit (21) configured to acquire an observation signal observed by a sensor (2) that transmits and receives radar waves, and detect at least one observation value for at least one target around the vehicle from the observation signal; a tracking unit (27) configured to track the target by calculating, for each target, an estimated value indicating a state of the target from a past estimated value, and calculating the current estimated value from the current observed value and the current predicted value in a predetermined processing cycle; an abnormality determination unit (22) configured to determine whether or not a vehicle speed abnormality exists as an abnormality in detection of the vehicle speed; Equipped with the tracking unit is configured to, when it is determined that the vehicle speed is not abnormal, execute a first process using the predicted value of the relative speed based on a detection result of the vehicle speed to calculate the current estimated value, and when it is determined that the vehicle speed is abnormal, execute a second process different from the first process to calculate the current estimated value; The abnormality determination unit determines that the vehicle speed is abnormal when the acceleration is equal to or greater than a predetermined acceleration threshold based on a magnitude of the acceleration of the host vehicle. Information processing device.

10. The information processing device according to claim 9, The tracking unit is a prediction unit (24) configured to calculate, for each target, the current predicted value from the past estimated value; an association unit (25) configured to set, for each of the calculated predicted values, a prediction range within which the observed value is estimated to be acquired this time based on the predicted value, determine an observed value within the prediction range to be associated with the predicted value from the at least one detected observed value, and determine the observed value to be associated with the calculated predicted value; an estimator (26) configured to calculate the estimated value of the current target based on the determined observed value and the predicted value; An information processing device comprising:

11. The information processing device according to any one of claims 1 to 10, The abnormality determination unit determines that the vehicle speed is abnormal when a prediction residual, which is a difference between the predicted value and the observed value of the relative speed, is equal to or greater than a predetermined prediction residual threshold value. Information processing device.

12. The information processing device according to claim 11, The abnormality determination unit determines that the vehicle speed is abnormal when the prediction residual for the relative speed of a stationary object among the targets is equal to or greater than the predetermined prediction residual threshold. Information processing device.

13. An information processing device (3) mounted on a vehicle, an observation value detection unit (21) configured to acquire an observation signal observed by a sensor (2) that transmits and receives radar waves, and detect at least one observation value for at least one target around the vehicle from the observation signal; a tracking unit (27) configured to track the target by calculating, for each target, an estimated value indicating a state of the target from a past estimated value, and calculating the current estimated value from the current observed value and the current predicted value in a predetermined processing cycle; an abnormality determination unit (22) configured to determine whether or not a vehicle speed abnormality exists as an abnormality in detection of the vehicle speed; Equipped with the tracking unit is configured to, when it is determined that the vehicle speed is not abnormal, execute a first process using the predicted value of the relative speed based on a detection result of the vehicle speed to calculate the current estimated value, and when it is determined that the vehicle speed is abnormal, execute a second process different from the first process to calculate the current estimated value; The abnormality determination unit determines that the vehicle speed is abnormal when a prediction residual, which is a difference between the predicted value and the observed value, for the relative speed is equal to or greater than a predetermined prediction residual threshold, and determines that the vehicle speed is abnormal when the prediction residual for the relative speed of a stationary object among the targets is equal to or greater than the predetermined prediction residual threshold. Information processing device.

14. The information processing device according to claim 13, The tracking unit is a prediction unit (24) configured to calculate, for each target, the current predicted value from the past estimated value; an association unit (25) configured to set, for each of the calculated predicted values, a prediction range within which the observed value is estimated to be acquired this time based on the predicted value, determine an observed value within the prediction range to be associated with the predicted value from the at least one detected observed value, and determine the observed value to be associated with the calculated predicted value; an estimator (26) configured to calculate the estimated value of the current target based on the determined observed value and the predicted value; An information processing device comprising:

15. The information processing device according to any one of claims 12 to 14, The abnormality determination unit determines that the vehicle speed is abnormal when the number of the stationary objects is equal to or greater than a predetermined number of stationary object targets and the prediction residual for the relative speed of at least one of the stationary objects is equal to or greater than the predetermined prediction residual threshold value. Information processing device.

16. An object tracking device (1) mounted on a vehicle, A sensor (2) for transmitting and receiving radar waves; an observation value detection unit (21) configured to acquire an observation signal observed by the sensor and detect at least one observation value for at least one target around the vehicle from the observation signal; a tracking unit (27) configured to track the target by calculating, for each target, an estimated value indicating a state of the target from a past estimated value, and calculating the current estimated value from the current observed value and the current predicted value in a predetermined processing cycle; an abnormality determination unit (22) configured to determine whether or not a vehicle speed abnormality exists as an abnormality in detection of the vehicle speed; Equipped with The tracking unit is When it is determined that the vehicle speed is not abnormal, a first process using the predicted value of the relative speed based on a detection result of the vehicle speed is executed to calculate the current estimated value, and when it is determined that the vehicle speed is abnormal, a second process different from the first process is executed to calculate the current estimated value, a prediction unit (24) configured to calculate, for each target, the current predicted value from the past estimated value; an association unit (25) configured to set, for each of the calculated predicted values, a prediction range within which the observed value is estimated to be acquired this time based on the predicted value, determine an observed value within the prediction range to be associated with the predicted value from the at least one detected observed value, and determine the observed value to be associated with the calculated predicted value; an estimator (26) configured to calculate the estimated value of the current target based on the determined observed value and the predicted value; Equipped with The association unit sets the prediction range to be larger in the second process when it is determined that the vehicle speed is abnormal than in the first process when it is determined that the vehicle speed is not abnormal. Object tracking device.

17. An object tracking device (1) mounted on a vehicle, A sensor (2) for transmitting and receiving radar waves; an observation value detection unit (21) configured to acquire an observation signal observed by the sensor and detect at least one observation value for at least one target around the vehicle from the observation signal; a tracking unit (27) configured to track the target by calculating, for each target, an estimated value indicating a state of the target from a past estimated value, and calculating the current estimated value from the current observed value and the current predicted value in a predetermined processing cycle; an abnormality determination unit (22) configured to determine whether or not a vehicle speed abnormality exists as an abnormality in detection of the vehicle speed; Equipped with the tracking unit is configured to, when it is determined that the vehicle speed is not abnormal, execute a first process using the predicted value of the relative speed based on a detection result of the vehicle speed to calculate the current estimated value, and when it is determined that the vehicle speed is abnormal, execute a second process different from the first process to calculate the current estimated value; The abnormality determination unit determines that the vehicle speed is abnormal when the acceleration is equal to or greater than a predetermined acceleration threshold based on a magnitude of the acceleration of the host vehicle. Object tracking device.

18. A tracking method for an information processing device mounted on a vehicle, comprising: an observation value detection step of acquiring an observation signal observed by a sensor that transmits and receives radar waves, and detecting at least one observation value for at least one target around the vehicle from the observation signal; a tracking step of tracking the target by calculating, for each target, an estimated value indicating a state of the target from a past estimated value, and calculating the current estimated value from the current observed value and the current predicted value in a predetermined processing cycle; an abnormality determination step of determining whether or not the abnormality in the detection of the vehicle speed is a vehicle speed abnormality; Equipped with The tracking step includes: When it is determined that the vehicle speed is not abnormal, a first process is executed using the predicted value of the relative speed based on the detection result of the vehicle speed to calculate the current estimated value, and when it is determined that the vehicle speed is abnormal, a second process different from the first process is executed to calculate the current estimated value. a prediction step (24) adapted to calculate, for each target, the current predicted value from the past estimated values; an association step (25) configured to set, for each of the calculated predicted values, a prediction range in which the observed value is estimated to be acquired this time based on the predicted value, determine an observed value within the prediction range to be associated with the predicted value from the at least one detected observed value, and determine the observed value to be associated with the calculated predicted value; an estimation step (26) configured to calculate the estimate of the current target based on the determined observed value and the predicted value; Equipped with In the associating step, the second process when it is determined that the vehicle speed is abnormal is performed to set the prediction range larger than the first process when it is determined that the vehicle speed is not abnormal. How to track.

19. A tracking method for an information processing device mounted on a vehicle, comprising: an observation value detection step of acquiring an observation signal observed by a sensor that transmits and receives radar waves, and detecting at least one observation value for at least one target around the vehicle from the observation signal; a tracking step of tracking the target by calculating, for each target, an estimated value indicating a state of the target from a past estimated value, and calculating the current estimated value from the current observed value and the current predicted value in a predetermined processing cycle; an abnormality determination step of determining whether or not the abnormality in the detection of the vehicle speed is a vehicle speed abnormality; Equipped with the tracking step, when it is determined that the vehicle speed is not abnormal, executes a first process using the predicted value of the relative speed based on a detection result of the vehicle speed to calculate the current estimated value, and when it is determined that the vehicle speed is abnormal, executes a second process different from the first process to calculate the current estimated value; The abnormality determination step determines that the vehicle speed is abnormal when the acceleration is equal to or greater than a predetermined acceleration threshold based on a magnitude of the acceleration of the host vehicle. How to track.

20. a program for causing a computer constituting an information processing device mounted on a vehicle to function as: an observation value detection unit configured to acquire an observation signal observed by a sensor that transmits and receives radar waves, and detect at least one observation value for at least one target around the vehicle from the observation signal; a tracking unit that tracks the target by calculating, for each target in a predetermined processing cycle, an estimated value indicating a state of the target from a past estimated value, and calculating the current estimated value from the current observed value and the current predicted value; and an abnormality determination unit that determines whether or not a vehicle speed abnormality has occurred as an abnormality in the detection of the vehicle speed, The tracking unit is When it is determined that the vehicle speed is not abnormal, a first process using the predicted value of the relative speed based on a detection result of the vehicle speed is executed to calculate the current estimated value, and when it is determined that the vehicle speed is abnormal, a second process different from the first process is executed to calculate the current estimated value, a prediction unit (24) configured to calculate, for each target, the current predicted value from the past estimated value; an association unit (25) configured to set, for each of the calculated predicted values, a prediction range within which the observed value is estimated to be acquired this time based on the predicted value, determine an observed value within the prediction range to be associated with the predicted value from the at least one detected observed value, and determine the observed value to be associated with the calculated predicted value; an estimator (26) configured to calculate the estimated value of the current target based on the determined observed value and the predicted value; Equipped with The association unit sets the prediction range to be larger in the second process when it is determined that the vehicle speed is abnormal than in the first process when it is determined that the vehicle speed is not abnormal. program.

21. a program for causing a computer constituting an information processing device mounted on a vehicle to function as: an observation value detection unit configured to acquire an observation signal observed by a sensor that transmits and receives radar waves, and detect at least one observation value for at least one target around the vehicle from the observation signal; a tracking unit that tracks the target by calculating, for each target in a predetermined processing cycle, an estimated value indicating a state of the target from a past estimated value, and calculating the current estimated value from the current observed value and the current predicted value; and an abnormality determination unit that determines whether or not a vehicle speed abnormality has occurred as an abnormality in the detection of the vehicle speed, the tracking unit is configured to, when it is determined that the vehicle speed is not abnormal, execute a first process using the predicted value of the relative speed based on a detection result of the vehicle speed to calculate the current estimated value, and when it is determined that the vehicle speed is abnormal, execute a second process different from the first process to calculate the current estimated value; The abnormality determination unit determines that the vehicle speed is abnormal when the acceleration is equal to or greater than a predetermined acceleration threshold based on a magnitude of the acceleration of the host vehicle. program.

Images