Object detection device

Abstract

This object detection device detects an obstacle present in the surroundings of the moving body on the basis of time-series changes in the intensity of reflected waves which are produced by transmission waves transmitted from a moving body being reflected by an object. The object detection device is provided with: a threshold value computation unit that calculates, on the basis of the moving average value of the intensity of reflected waves, a variable threshold value which fluctuates according to the movement of the moving body; and a determination unit that determines the presence or absence of an obstacle on the basis of whether the intensity of the reflected waves is greater than the variable threshold value. The threshold value computation unit calculates, on the basis of at least one past moving average value, a smoothed moving average value obtained by smoothing the current moving average value, and calculates, on the basis of the smoothed moving average value, the variable threshold value.

Description

Object detection device

[0001] The present invention relates to an object detection device.

[0002] Mobile objects such as vehicles use object detection devices that detect obstacles around the mobile object based on information such as time of flight (TOF) and Doppler shift obtained by transmitting and receiving ultrasonic waves. Such object detection devices utilize CFAR (Constant False Alarm Rate) processing as a technique for suppressing the influence of noise caused by reflected waves from objects that are not the target of detection (e.g., road surfaces). CFAR processing optimizes a threshold value for determining whether an object that reflects a reflected wave is an obstacle based on a moving average value of the intensity of the reflected wave, which changes over time, in accordance with the condition of the road surface, etc.

[0003] Japanese Patent Application Laid-Open No. 2016-206011

[0004] While the CFAR processing described above can basically reduce the influence of noise and improve the accuracy of obstacle detection, it may decrease the detection accuracy depending on the situation. For example, when an obstacle is present within the detection range, the moving average value of the intensity of the reflected wave increases, so the threshold set by the CFAR processing becomes higher, increasing the possibility of overlooking the obstacle.

[0005] The present invention has been made in view of the above, and provides an object detection device that can reduce overlooking of obstacles and improve the accuracy of obstacle detection.

[0006] An object detection device according to one aspect of the present invention is an object detection device that detects obstacles present in the vicinity of a moving body based on time-series changes in the intensity of reflected waves generated when a transmission wave transmitted from the moving body is reflected by an object, and is equipped with a threshold calculation unit that calculates a variable threshold that varies in accordance with the movement of the moving body based on a moving average value of the intensity of the reflected wave, and a judgment unit that judges the presence or absence of an obstacle based on whether the intensity of the reflected wave is greater than the variable threshold, and the threshold calculation unit calculates a smoothed moving average value by smoothing the current moving average value based on at least one past moving average value, and calculates the variable threshold value based on the smoothed moving average value.

[0007] According to the present invention, it is possible to provide an object detection device that can reduce overlooking of obstacles and improve the accuracy of obstacle detection.

[0008] FIG. 1 is a diagram illustrating an example of the configuration of a vehicle according to the first embodiment. FIG. 2 is a diagram illustrating an example of the configuration of a vehicle control system according to the first embodiment. FIG. 3 is a diagram illustrating an example of a distance calculation method using the TOF method according to the first embodiment. FIG. 4 is a diagram illustrating an example of the functional configuration of an object detection device according to the first embodiment. FIG. 5 is a diagram illustrating an example of the functional configuration included in a threshold calculation unit according to the first embodiment. FIG. 6 is a diagram illustrating an example of processing for calculating a CFAR threshold according to the first embodiment. FIG. 7 is a diagram illustrating an example of the relationship between a current moving average value and a past moving average value according to the first embodiment. FIG. 8 is a diagram illustrating an example of a CFAR threshold based on a current moving average value according to the first embodiment. FIG. 9 is a diagram illustrating an example of a CFAR threshold based on a smoothed moving average value according to the first embodiment. FIG. 10 is a flowchart illustrating an example of processing in the object detection device according to the first embodiment. FIG. 11 is a diagram illustrating an example of the relationship between a CFAR threshold based on a current moving average value according to the first embodiment and a fixed threshold. FIG. 12 is a flowchart illustrating an example of processing in the object detection device according to the second embodiment.

[0009] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. The configuration of the embodiment described below and the actions and effects brought about by the configuration are merely examples, and the present invention is not limited to the following description.

[0010] (First embodiment) Fig. 1 is a diagram showing an example of the configuration of a vehicle 1 according to a first embodiment. The vehicle 1 is an example of a moving body on which an object detection device according to this embodiment is mounted. The object detection device according to this embodiment is a device that detects obstacles present around the vehicle 1 based on information such as time of flight (TOF) and Doppler shift acquired by transmitting and receiving ultrasonic waves.

[0011] The object detection device according to this embodiment includes a plurality of transceivers 21A to 21L. Hereinafter, when there is no need to distinguish between the plurality of transceivers 21A to 21L, they may be referred to as a transceiver 21. Each transceiver 21 is installed on a vehicle body 2 that serves as the exterior of the vehicle 1, transmits ultrasonic waves toward the outside of the vehicle body 2, and receives ultrasonic waves generated when the ultrasonic waves are reflected by an object present outside the vehicle body 2. Hereinafter, ultrasonic waves transmitted from the transceiver 21 may be referred to as a transmitted wave, and ultrasonic waves generated when the transmitted wave is reflected by an object may be referred to as a reflected wave.

[0012] 1, four transceivers 21A to 21D are arranged at the front end of the vehicle body 2, four transceivers 21E to 21H are arranged at the rear end, two transceivers 21I and 21J are arranged on the right side, and two transceivers 21K and 21L are arranged on the left side. Note that the number and installation positions of the transceivers 21 are not limited to this example.

[0013] 2 is a diagram showing an example of the configuration of the vehicle control system 10 according to the first embodiment. The vehicle control system 10 performs processing for controlling the vehicle 1 based on information output from the object detection device 11. The vehicle control system 10 according to this embodiment includes the object detection device 11 and an ECU 12.

[0014] The object detection device 11 includes multiple transceivers 21 and a control unit 22. Each transceiver 21 includes a vibrator 31 configured using a piezoelectric element or the like, an amplifier, and the like, and transmits and receives ultrasonic waves through the vibration of the vibrator 31. Specifically, each transceiver 21 transmits ultrasonic waves generated in response to the vibration of the vibrator 31 as a transmission wave and detects the vibration of the vibrator 31 caused by the reflected waves of the transmission wave reflected by an object such as an obstacle O or a road surface G. The obstacle O is an object that is present around the vehicle 1 and is to be detected, and may be, for example, another vehicle, a roadside object, a wall, a person, a curb, a bollard, etc. The vibration of the vibrator 31 is converted into an electrical signal, and based on the electrical signal, the time of flight (TOF) corresponding to the distance from the transceiver 21 to the obstacle O, a Doppler shift corresponding to the relative velocity between the vehicle 1 and the obstacle O, and the like can be obtained.

[0015] 2 illustrates a configuration in which both transmission of the transmission wave and reception of the reflected wave are performed using a single oscillator 31, but the configuration of the transmitter / receiver 21 is not limited to this. For example, the transmitter and receiver may be separated, such as a configuration in which an oscillator for transmitting the transmission wave and an oscillator for receiving the reflected wave are separately provided.

[0016] The control unit 22 includes an input / output device 41, a storage device 42, and a processor 43. The input / output device 41 is an interface device that enables transmission and reception of information between the control unit 22 and external devices (such as the transceiver unit 21 and the ECU 12). The storage device 42 includes a main storage device such as a read-only memory (ROM) or a random access memory (RAM), and an auxiliary storage device such as a hard disk drive (HDD) or a solid state drive (SSD). The processor 43 is an integrated circuit that executes various processes to realize the functions of the control unit 22, and may be configured using, for example, a central processing unit (CPU), an application specific integrated circuit (ASIC), or a field programmable gate array (FPGA) that operates according to a program. The processor 43 executes various arithmetic and control processes by reading and executing programs stored in the storage device 42.

[0017] The ECU 12 is a unit that executes various processes for controlling the vehicle 1 based on information acquired from the object detection device 11 and the like. The ECU 12 includes an input / output device 51, a storage device 52, and a processor 53. The input / output device 51 is an interface device that enables transmission and reception of information between the ECU 12 and external mechanisms (such as the object detection device 11, drive mechanism, braking mechanism, steering mechanism, transmission mechanism, in-vehicle display, speakers, and various sensors). The storage device 52 includes a main storage device such as a ROM or RAM, and an auxiliary storage device such as an HDD or SSD. The processor 53 is an integrated circuit that executes various processes for realizing the functions of the ECU 12, and may be configured using, for example, a CPU, an ASIC, an FPGA, or the like. The processor 53 reads programs stored in the storage device 52 and executes various arithmetic and control processes.

[0018] Fig. 3 is a diagram showing an example of a distance calculation method using the TOF method according to the first embodiment. Fig. 3 illustrates an envelope curve L that indicates changes over time in the intensity (signal level) of ultrasonic waves transmitted and received by the transmitter-receiver 21. In the graph shown in Fig. 3, the horizontal axis corresponds to time (TOF), and the vertical axis corresponds to the intensity of ultrasonic waves transmitted and received by the transmitter-receiver 21 (the magnitude of vibration of the transducer 31).

[0019] The envelope L shows the change over time in the intensity, which indicates the magnitude of the vibration of the vibrator 31. From the envelope L shown in Fig. 3, it can be seen that the vibrator 31 is driven to vibrate for a period of time Ta from time t0, completing the transmission of the transmission wave at time t1, and then the vibration of the vibrator 31 due to inertia continues while attenuating for a period of time Tb until time t2. Therefore, in the graph shown in Fig. 3, the period of time Tb corresponds to the so-called reverberation time.

[0020] The envelope L reaches a peak at time t4, a time Tp after the start of transmission of the transmission wave at time t0, when the magnitude of the vibration of the vibrator 31 reaches or exceeds a threshold value Th. This threshold value Th is a value set to distinguish whether the vibration of the vibrator 31 is caused by reception of a wave reflected from the obstacle O or reception of a wave reflected from an object other than the obstacle O (e.g., the road surface G). Note that although the threshold value Th is shown as a constant value here, the threshold value Th may also be a variable value that changes depending on the situation. Vibrations having a peak equal to or greater than the threshold value Th can be considered to be caused by reception of a wave reflected from the obstacle O.

[0021] In this example, the envelope curve L indicates that the vibration of the vibrator 31 is attenuated after the timing t4. Therefore, the timing t4 corresponds to the timing at which the reception of the reflected wave from the obstacle O is completed, in other words, the timing at which the transmitted wave last transmitted at the timing t1 returns as a reflected wave.

[0022] Furthermore, in envelope L, timing t3, which is the start point of the peak at timing t4, corresponds to the timing when reception of the reflected wave from obstacle O begins, in other words, the timing when the transmitted wave first transmitted at timing t0 returns as a reflected wave. Therefore, the time ΔT between timing t3 and timing t4 is equal to the time Ta, which is the transmission time of the transmitted wave.

[0023] From the above, in order to use TOF to find the distance from the transmitter / receiver 21, which is the source of the ultrasonic waves, to the obstacle O, it is necessary to find the time Tf between the time t0 when the transmission of the transmitted wave begins and the time t3 when the reflected wave begins to be received. This time Tf can be found by subtracting the time ΔT, which is equal to the time Ta as the transmission time of the transmitted wave, from the time Tp, which is the difference between the time t0 and the time t4 when the intensity of the reflected wave exceeds the threshold Th and reaches its peak.

[0024] The time t0 when the transmission of the transmission wave begins can be easily identified as the time when the object detection device 200 starts operating, and the time Ta as the transmission time of the transmission wave is determined in advance by settings, etc. Therefore, by identifying the time t4 when the intensity of the reflected wave reaches a peak equal to or greater than the threshold value Th, the distance from the vehicle 1 (the transmitter / receiver 21, which is the source of the transmission and reception of the ultrasonic waves) to the obstacle O can be calculated. Note that the above calculation method is an example, and the distance from the vehicle 1 to the obstacle O may be calculated using any known or new method as appropriate.

[0025] 4 is a diagram showing an example of the functional configuration of the object detection device 11 according to the first embodiment. The object detection device 11 (control unit 22) according to this embodiment includes an echo information generation unit 101, a threshold calculation unit 102, a determination unit 103, and an output unit 104. These functional units can be realized, for example, by cooperation between hardware and software (programs, etc.) of the object detection device 11 as shown in FIG. 2. Furthermore, at least some of these functional units may be realized by dedicated hardware (circuits).

[0026] The echo information generating unit 101 generates echo information indicating time-series changes in the intensity of the reflected wave based on information acquired from the transmitting / receiving unit 21. The echo information may include, for example, data on an envelope L as illustrated in FIG.

[0027] The threshold calculation unit 102 performs CFAR processing on the echo information to calculate a CFAR threshold (an example of a variable threshold) that varies in accordance with the movement of the vehicle 1. The threshold calculation unit 102 calculates a moving average value of the intensity of the reflected wave based on the echo information, and calculates the CFAR threshold based on the moving average value. The method of calculating the CFAR threshold will be described later.

[0028] The determination unit 103 determines whether or not an obstacle is present around the vehicle 1 based on the echo information and the CFAR threshold calculated by the threshold calculation unit 102. In this embodiment, the determination unit 103 determines that an obstacle is present when the intensity of the reflected wave is greater than the CFAR threshold. If the determination unit 103 determines that an obstacle is present, it generates obstacle information related to the obstacle. The obstacle information may include, for example, the distance from the vehicle 1 (transmitter / receiver 21) to the obstacle, the relative speed of the obstacle, etc.

[0029] The output unit 104 outputs the obstacle information generated by the determination unit 103 to a predetermined mechanism (for example, the ECU 12, etc.).

[0030] The threshold calculation unit 102 of this embodiment calculates a smoothed moving average value by smoothing the current moving average value based on at least one past moving average value, and calculates a CFAR threshold value based on the smoothed moving average value.

[0031] 5 is a diagram showing an example of the functional configuration included in the threshold calculation unit 102 according to the first embodiment. The threshold calculation unit 102 according to this embodiment includes a moving average calculation unit 201, a storage unit 202, a smoothing processing unit 203, and an adjustment processing unit 204.

[0032] The moving average calculation unit 201 calculates a moving average value of the intensity of the reflected wave based on echo information indicating time-series changes in the intensity of the reflected wave. The moving average calculation unit 201 calculates the moving average value for each predetermined detection period, for example. The detection period is a period from when a transmitted wave is transmitted once to when a reflected wave of the transmitted wave is awaited to be received.

[0033] The storage unit 202 stores the moving average value calculated by the moving average calculation unit 201 for each detection period. That is, the storage unit 202 stores a plurality of past moving average values ​​in chronological order.

[0034] The smoothing processing unit 203 executes a smoothing process to smooth the current moving average value calculated by the moving average calculation unit 201, based on at least one past moving average value stored in the storage unit 202. The specific method of the smoothing process should be determined appropriately depending on the usage situation, etc., but may be, for example, a process to calculate the average of the current moving average value and at least one past moving average value as a smoothed moving average value.

[0035] The adjustment processing unit 204 calculates the CFAR threshold by performing a predetermined adjustment process on the smoothed moving average value calculated by the smoothing processing unit 203. The adjustment process may be, for example, a multiplication process or an addition process. The integration process is a process of integrating a predetermined constant with respect to the moving average value (in this embodiment, the smoothed moving average value). The addition process is a process of adding a predetermined constant with respect to the moving average value.

[0036] 6 is a diagram illustrating an example of a process for calculating a CFAR threshold according to the first embodiment. In the process illustrated here, first, a received signal indicating the intensity of a reflected wave is sampled at a predetermined time interval. Then, a first sum Σ1 is calculated, which is the sum of the intensity values ​​of N samples of the reflected wave received during a first period Δt1 before a certain detection time t. Then, a second sum Σ2 is calculated, which is the sum of the intensity values ​​of N samples of the reflected wave received during a second period Δt2 after the detection time t. Then, the sum Σ obtained by adding together the first sum Σ1 and the second sum Σ2 is divided by 2N, which is the sum of the number of samples N in the first period Δt1 and the number of samples N in the second period Δt2, to calculate a current moving average value (the moving average value corresponding to the detection time t).

[0037] Then, a smoothed moving average value is calculated by smoothing the current moving average value based on the current moving average value and at least one past moving average value (a moving average value corresponding to a detection timing prior to the detection timing t). The smoothed moving average value may be, for example, the average of the current moving average value and the past moving average value. A CFAR threshold value is then calculated by performing a predetermined integration process, addition process, or the like on the smoothed moving average value. Then, based on a differential signal indicating the difference between the CFAR threshold value and the intensity value I of the reflected wave corresponding to the detection timing t, it is determined whether or not an obstacle is present at the distance corresponding to the detection timing t.

[0038] Note that the current method of calculating the moving average value is not limited to the above. For example, the above describes CA (Cell Averaging)-CFAR processing in which the moving average value is the average of all intensity values ​​corresponding to periods Δt1 and Δt2 before and after the detection timing t, but the CFAR threshold Thc may also be calculated by GO (Greatest Of)-CFAR processing or SO (Smallest Of)-CFAR processing, etc. GO-CFAR processing is processing in which the moving average value is calculated using the larger value of the first sum Σ1 and the second sum Σ2. SO-CFAR processing is processing in which the moving average value is calculated using the smaller value of the first sum Σ1 and the second sum Σ2.

[0039] 7 is a diagram showing an example of the relationship between the current moving average value Ac and the past moving average values ​​Ap1 and Ap2 according to the first embodiment. In FIG. 7, the current moving average value Ac, the previous moving average value Ap1 (an example of a past moving average value), and the moving average value Ap2 before the previous (an example of a past moving average value) are illustrated.

[0040] The current moving average value Ac is the moving average value (moving average line) of the envelope curve Lc corresponding to the current detection cycle. The previous moving average value Ap1 is the moving average value of the envelope curve Lp1 corresponding to the detection cycle prior to the current detection cycle (e.g., the cycle immediately preceding). The moving average value Ap2 before the previous is the moving average value of the envelope curve Lp2 corresponding to the detection cycle prior to the detection cycle corresponding to the previous moving average value Ap1 (e.g., the cycle immediately preceding the current detection cycle). Note that although an example is shown here in which two moving average values ​​are used as past moving average values, three or more past moving average values ​​may also be used.

[0041] 7, the peak Pc in the current envelope Lc, the peak Pp1 in the previous envelope Lp1, and the peak Pp2 in the envelope Lp2 before last are all caused by the same obstacle. That is, the obstacle was located at a distance Dp2 in the detection cycle before last, at a distance Dp1 in the previous detection cycle, and at a distance Dc in the current detection cycle (Dc<Dp1<Dp2).

[0042] In such a case, as shown in Fig. 7 , the height (intensity) of each of the peaks Pc, Pp1, and Pp2 caused by the obstacle increases as the distance from the vehicle 1 to the obstacle decreases. Therefore, the influence of the reflected waves from the obstacle on the current moving average value Ac is greater than the influence of the past moving average values ​​Ap1 and Ap2. Therefore, when an obstacle is present, the current moving average value Ac is likely to be higher than the past moving average values ​​Ap1 and Ap2 due to the influence of the reflected waves from the obstacle.

[0043] 8 is a diagram illustrating an example of the CFAR threshold Th based on the current moving average value Ac according to the first embodiment. As shown in FIG. 8, the CFAR threshold Th based on the current moving average value Ac that has not been subjected to smoothing processing is set relatively high due to the strong influence of waves reflected from obstacles. In such a case, a peak Pc appearing in the envelope Lc due to an obstacle is unlikely to exceed the CFAR threshold Th, increasing the possibility that the obstacle will not be detected.

[0044] Therefore, in this embodiment, a smoothed moving average value is calculated by smoothing the current moving average value Ac based on the past moving average values ​​Ap1 and Ap2, and the CFAR threshold is calculated based on the smoothed moving average value.

[0045] 9 is a diagram illustrating an example of a CFAR threshold value Ths based on the smoothed moving average value As according to the first embodiment. The smoothed moving average value As illustrated here is the average value of the current moving average value Ac, the previous moving average value Ap1, and the moving average value Ap2 before that illustrated in FIG.

[0046] 9, the smoothed moving average value As is smoother than the current moving average value Ac (see FIG. 8) by reducing the influence of peaks Pc that appear in the envelope Lc due to obstacles. This makes it possible to avoid setting the CFAR threshold value Ths calculated based on the smoothed moving average value As excessively high, thereby reducing the possibility of overlooking an obstacle.

[0047] 10 is a flowchart showing an example of processing in the object detection device 11 of the first embodiment. In step S101, the transmitter / receiver 21 starts transmitting and receiving ultrasonic waves. In step S102, the echo information generator 101 generates echo information indicating time-series changes in the intensity of reflected waves from objects present around the vehicle 1.

[0048] In step S103, the threshold calculation unit 102 calculates the current moving average value Ac based on the echo information. In step S104, the threshold calculation unit 102 calculates the smoothed moving average value As based on the current moving average value Ac and the past moving average values ​​Ap1 and Ap2. In step S105, the threshold calculation unit 102 calculates the CFAR threshold value Ths based on the smoothed moving average value As.

[0049] In step S106, the determination unit 103 determines whether the intensity of the current reflected wave is equal to or greater than the CFAR threshold Ths. If the intensity of the reflected wave is not equal to or greater than the CFAR threshold Ths (S106: No), the determination unit 103 determines that no obstacle is present, and the routine ends. If the intensity of the reflected wave is equal to or greater than the CFAR threshold Ths (S106: Yes), the determination unit 103 determines that an obstacle is present and generates obstacle information related to the obstacle. In step S107, the output unit 104 outputs the obstacle information to a mechanism such as the ECU 12.

[0050] As described above, according to this embodiment, the CFAR threshold Ths is set based on the smoothed moving average value As smoothed using the past moving average values ​​Ap1 and Ap2, so it is possible to avoid setting the CFAR threshold too high due to the influence of waves reflected from obstacles, etc. This makes it possible to suppress overlooking of obstacles and improve the accuracy of obstacle detection.

[0051] Other embodiments will be described below with reference to the drawings, but descriptions of parts that are the same as or similar to the first embodiment will be omitted as appropriate.

[0052] (Second embodiment) The object detection device 11 according to the second embodiment differs from the first embodiment in that, when the CFAR threshold Th based on the current moving average value Ac is greater than a predetermined fixed threshold, the object detection device 11 calculates a smoothed moving average value Ths based on at least one past moving average value Ap1, Ap2 without using the current moving average value Ac.

[0053] FIG. 11 is a diagram illustrating an example of the relationship between the CFAR threshold Th based on the current moving average value Ac and the fixed threshold Th0 according to the first embodiment. The fixed threshold Th0 illustrated here is a fixed value uniquely determined for each distance from the vehicle 1. The fixed threshold Th0 may be pre-stored in an appropriate storage device (e.g., the storage device 42) mounted on the vehicle 1. The specific method for setting the fixed threshold Th0 should be determined appropriately depending on the specifications of the vehicle 1, the performance of the transceiver 21, and the like. However, the fixed threshold Th0 may be set, for example, based on a statistical value (e.g., a lower limit value or an average value) of the strength of the reflected wave from an obstacle for each distance. Because the strength of the reflected wave from an obstacle decreases as the distance from the vehicle 1 increases, the fixed threshold Th0 is typically set to decrease as the distance from the vehicle 1 increases. Note that, although the fixed threshold Th0 illustrated here decreases linearly with increasing distance, the form of the fixed threshold Th0 is not limited thereto.

[0054] 11 illustrates a state in which a portion of the CFAR threshold Th based on the current moving average value Ac is greater than the fixed threshold value Th0. In such a case, in this embodiment, the smoothed moving average value Ths (see FIG. 9 ) is calculated based on at least one past moving average value without using the current moving average value Ac. The specific method for calculating the smoothed moving average value Ths in this case may be determined appropriately depending on the usage situation, etc. For example, the smoothed moving average value Ths may be the average value of multiple past moving average values ​​(e.g., past moving average values ​​Ap1, Ap2, etc. illustrated in FIG. 7 ), or one past moving average value selected from multiple past moving average values.

[0055] 12 is a flowchart showing an example of processing in the object detection device 11 according to the second embodiment. In step S201, the transmitter / receiver 21 starts transmitting and receiving ultrasonic waves. In step S202, the echo information generator 101 generates echo information indicating time-series changes in the intensity of reflected waves from objects present around the vehicle 1.

[0056] In step S203, the threshold calculation unit 102 calculates a current moving average value Ac based on the echo information. In step S204, the threshold calculation unit 102 calculates a CFAR threshold value Th based on the current moving average value Ac. In step S205, the threshold calculation unit 102 determines whether the CFAR threshold value Th based on the current moving average value Ac is greater than a fixed threshold value Th0.

[0057] In step S205, if the CFAR threshold Th based on the current moving average value Ac is not greater than the fixed threshold value Th0 (S205: No), in step S206, the threshold calculation unit 102 calculates the smoothed moving average value As based on the current moving average value Ac and the past moving average values ​​Ap1 and Ap2.

[0058] In step S205, if the CFAR threshold Th based on the current moving average value Ac is greater than the fixed threshold value Th0 (S205: Yes), in step S207, the threshold calculation unit 102 calculates the smoothed moving average value As based on the past moving average values ​​Ap1 and Ap2 without using the current moving average value Ac. The method of calculating the smoothed moving average value As based on the past moving average values ​​Ap1 and Ap2 may be determined appropriately depending on the usage situation, etc. For example, the smoothed moving average value As may be the average value of multiple past moving average values ​​Ap1 and Ap2, or one past moving average value selected from multiple past moving average values ​​(e.g., either the past moving average values ​​Ap1 or Ap2) may be used as the smoothed moving average value As.

[0059] In step S208, the threshold calculation unit 102 calculates a smoothed CFAR threshold Ths based on the smoothed moving average value As calculated in step S206 or step S207. In step S209, the determination unit 103 determines whether the current intensity of the reflected wave is equal to or greater than the smoothed CFAR threshold Ths. If the intensity of the reflected wave is not equal to or greater than the smoothed CFAR threshold Ths in step S209 (S209: No), the determination unit 103 determines that no obstacle is present, and this routine ends. If the intensity of the reflected wave is equal to or greater than the smoothed CFAR threshold Ths in step S209 (S209: Yes), the determination unit 103 determines that an obstacle is present in step S210 and generates obstacle information related to the obstacle, and the output unit 104 outputs the obstacle information to a mechanism such as the ECU 12.

[0060] As described above, according to this embodiment, when the CFAR threshold Th based on the current moving average value Ac is greater than the predetermined fixed threshold value Th0, the smoothed moving average value As is calculated based on at least one of the past moving average values ​​Ap1 and Ap2 without using the current moving average value Ac. The presence or absence of an obstacle is then determined based on the CFAR threshold value Ths calculated based on the smoothed moving average value As. This makes it possible to exclude the current moving average value Ac, which is strongly affected by waves reflected from obstacles, from the elements used to calculate the smoothed moving average value As, thereby enabling the CFAR threshold value Ths to be set more appropriately.

[0061] In the above-described embodiment, an example has been given of a configuration in which ultrasonic waves are used as the waves transmitted and received to detect obstacles, but the configuration of the object detection device is not limited to this. For example, a configuration in which millimeter-wave radar, a LiDAR (Light Detection and Ranging) sensor, or the like is used may also be used.

[0062] Furthermore, the threshold calculation unit 102 (smoothing processing unit 203) may limit the past moving average values ​​used in calculating the smoothed moving average value to past moving average values ​​corresponding to detection periods in which it was determined that no obstacle was present. This configuration more effectively prevents the CFAR threshold from being set too high due to the influence of waves reflected from obstacles.

[0063] Furthermore, the threshold calculation unit 102 (smoothing processing unit 203) may change the number of past moving average values ​​used to calculate the smoothed moving average value, depending on the distance to an obstacle determined to be present by the determination unit 103. The relationship between the distance to an obstacle and the number of past moving average values ​​should be determined appropriately depending on the usage situation, etc. For example, the closer the distance from the vehicle 1 to the obstacle, the greater the number of past moving average values ​​used to calculate the smoothed moving average value. This is because the influence of reflected waves from an obstacle becomes greater the closer the distance from the vehicle 1 to the obstacle. With this configuration, the CFAR threshold can be more appropriately set depending on the distance from the vehicle 1 to the obstacle.

[0064] A program that causes a computer (such as the processor 43) to execute processing for realizing the functions of the object detection device described above can be provided by being recorded in an installable or executable file on a computer-readable recording medium such as a CD (Compact Disc)-ROM, a flexible disk (FD), a CD-R (Recordable), or a DVD (Digital Versatile Disk). The program may also be provided or distributed via a network such as the Internet.

[0065] Although the embodiments of the present invention have been described above, the above-described embodiments and their modifications are merely examples and are not intended to limit the scope of the invention. The novel embodiments and modifications described above can be implemented in various forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. The above-described embodiments and modifications are included within the scope and spirit of the invention, and are also included in the invention and its equivalents as defined in the claims.

[0066] 1...vehicle (moving body), 11...object detection device, 21, 21A to 21L...transmitter / receiver unit, 102...threshold calculation unit, 103...determination unit, Ac...current moving average value, Ap1, Ap2...past moving average value, As...smoothed moving average value, Th...(before smoothing) CFAR threshold (variation threshold), Ths...(after smoothing) CFAR threshold (variation threshold), Th0...fixed threshold.

Claims

1. An object detection device that detects obstacles present in the vicinity of a moving body based on time-series changes in the intensity of a reflected wave generated when a transmission wave transmitted from the moving body is reflected by an object, the object detection device comprising: a threshold calculation unit that calculates a variable threshold that varies in accordance with the movement of the moving body based on a moving average value of the intensity of the reflected wave; and a judgment unit that judges the presence or absence of the obstacle based on whether the intensity of the reflected wave is greater than the variable threshold, wherein the threshold calculation unit calculates a smoothed moving average value by smoothing a current moving average value based on at least one past moving average value, and calculates the variable threshold value based on the smoothed moving average value.

2. The object detection device according to claim 1, wherein, when the fluctuation threshold based on the current moving average value is greater than a predetermined fixed threshold value, the threshold calculation unit calculates the smooth moving average value based on at least one of the past moving average values ​​without using the current moving average value.

3. The object detection device according to claim 1 or 2, wherein the threshold calculation unit limits the past moving average value used in calculating the smooth moving average value to the past moving average value corresponding to the detection period in which it was determined that the obstacle was not present.

4. The object detection device according to claim 1 or 2, wherein the threshold calculation unit changes the number of past moving average values ​​used in calculating the smooth moving average value depending on the distance of the obstacle determined to exist by the determination unit.

5. An object detection device as described in claim 1 or 2, wherein the threshold calculation unit calculates a moving average value of the intensity of the reflected wave for each predetermined detection period and stores the calculated moving average value in chronological order, and the detection period is a period from transmitting the transmission wave once to waiting for reception of the reflected wave of the transmission wave.

6. The object detection device according to claim 1 or 2, wherein the threshold calculation unit calculates an average value of the current moving average value and at least one of the past moving average values ​​as the smooth moving average value.

7. The object detection device according to claim 4, wherein the threshold calculation unit increases the number of past moving average values ​​used in calculating the smoothed moving average value as the distance from the moving body to the obstacle becomes shorter.

Images