Acoustic positioning device, method, and program

Abstract

To enable accurate positioning by implementing SSBL method calculation with high accuracy using an iterative gradient method, which has a little computational amount, even in situations where there are interfering waves such as reflected waves.SOLUTION: An envelope detection section 2031 performs pulse compression processing, square processing, and low-pass filtering processing on acoustic signals received by respective hydrophones 201 to create an envelope. A calculation range setting section 2032 sets a part of a front side of a positioning pulse part A in the envelope of each channel to a calculation range in which an arrival time difference is obtained by applying an iterative gradient method. An arrival time difference calculation section 2033 obtains an arrival time difference Δd1 of received signals between channels 0 and 1 by applying the iterative gradient method to signals in a calculation range ΔT0 of a channel 0 and a calculation range ΔT1 of a channel 1. Similarly, by applying the iterative gradient method to signals in the calculation range ΔT0 of the channel 0 and a calculation range ΔT2 of a channel 2, an arrival time difference Δd2 of the received signals between the channels 0 and 2 is obtained.SELECTED DRAWING: Figure 3

Description

Technical Field

[0001] The present invention relates to an acoustic positioning device, method, and program for positioning the position of a pinger or an object equipped with the pinger based on an acoustic signal. In particular, the present invention relates to an acoustic positioning device, method, and program that receive an acoustic signal transmitted by a pinger with a plurality of receivers and perform positioning by calculating a delay difference using a gradient method.

Background Art

[0002] Non-Patent Document 1 discloses a "water-air combined drone" in which an aerial drone holds a submersible drone and flies, separates and submerges the submersible drone after landing in the target water area, and recovers and takes off from the water after the work is completed, as shown in FIG. 13.

[0003] One feature of the water-air combined drone is that it is equipped with a technology of "acoustic positioning" in which an acoustic signal is transmitted from a transmitter (pinger) attached to the submersible drone, received by an underwater microphone (hydrophone) on the aerial drone side, and automatically analyzed to calculate the position of the submersible drone. The acoustic positioning technology is disclosed, for example, in P.236-245 of Non-Patent Document 2.

[0004] In the water-air combined drone, the SSBL (Super Short Base Line) method is adopted for acoustic positioning. First, the delay difference of acoustic signals received by three or more hydrophones is obtained, and the position of the transmission source is calculated from this.

[0005] Non-Patent Document 3 discloses a case of the SSBL acoustic positioning method mounted on a water-air combined drone. In the SSBL acoustic positioning method, a set of acoustic pulses is transmitted within a certain period (for example, 1 second). A set of pulses consists of a certain number (for example, two pulses), and positioning is performed by receiving these with a hydrophone and performing signal processing.

Prior Art Documents

Patent Documents

[0006] [Patent Document 1] Patent application No. 2022-145518 [Patent Document 2] Patent application No. 2022-195588 [Non-patent literature]

[0007] [Non-Patent Document 1] KDDI / KDDI Research Institute / ProDrone, News Release: "World's First Aero-Water Combined Drone Successfully Conducts Remote Underwater Photography ~Enables Safe and Efficient Inspection of Offshore Wind Power Generation Facilities Without Launching a Boat~", December 14, 2021 https: / / news.kddi.com / kddi / corporate / newsrelease / 2021 / 12 / 14 / 5593.html [Non-Patent Document 2] The Japan Society for Marine Acoustics, "Fundamentals and Applications of Marine Acoustics," Seizando Shoten, pp. 236-245. [Non-Patent Document 3] Kawada, Nishitani, Kojima: "Acoustic Positioning System for Water-Air Combined Drones," Proceedings of the Japan Society for Marine Acoustics, No. 22-2, pp. 3-4 (2022) [Non-Patent Document 4] Kawada, Nishitani, Kojima: "Acoustic Positioning System for Aquatic-Air Combined Drones," IEICE Technical Report, EA2022-73, pp.72-77 (December 2022) [Overview of the project] [Problems that the invention aims to solve]

[0008] Calculating the cross-correlation function typically requires a Fourier transform (FFT). As is well known, this process is computationally intensive, and in the case of aerial-underwater drones, these signal processing needs to be done within the receiver attached to the aerial drone. If we anticipate needing to simultaneously position a large number of underwater drones and divers in the future, it is highly likely that a weak CPU will not be able to keep up with performing FFT processing for that many drones.

[0009] Non-patent document 4 discloses a positioning method that uses an iterative gradient method, which is less computationally intensive than calculating the cross-correlation function, and its basic performance has already been verified by computer simulation.

[0010] However, as shown in Figure 14, the received acoustic signal is not necessarily limited to the positioning pulse being measured. In real-world environments, due to sound wave reflection from surfaces such as water, the reflected wave portion B may be received with a delay after the positioning pulse portion A, as shown in Figure 15. Furthermore, if the arrival time of the reflected wave portion B is early, the trailing portion of the positioning pulse portion A and the leading portion of the reflected wave portion B may overlap, as shown in Figure 16, and be received as a superimposed wave C.

[0011] Even in this case, with the cross-correlation method, selecting the largest peak among several peaks in the time domain often gives the desired time difference. In contrast, the iterative gradient method uses the entire waveform (which is distorted by interference) for calculation, which leads to a problem of larger errors.

[0012] The object of the present invention is to solve the above technical problems and to provide an acoustic positioning device, method, and program that can accurately perform SSBL calculations using a computationally intensive iterative gradient method even when interference waves such as reflected waves are received, thereby enabling accurate positioning. [Means for solving the problem]

[0013] To achieve the above objective, the present invention is characterized in that it comprises the following configuration in an acoustic positioning device that receives an acoustic signal including a positioning pulse transmitted by an object underwater using a plurality of hydrophones and determines the position of the object based on the difference in arrival times of each positioning pulse.

[0014] (1) The system comprises means for processing each acoustic signal and converting it into an envelope, and means for calculating the difference in arrival times between parts of the forward positioning pulse portion of each envelope using the gradient method.

[0015] (2) It is equipped with means for setting the calculation range of the gradient method based on the waveform pattern of the reflected wave (interference wave) part of each envelope line.

[0016] In addition, the present invention can be realized not only as an acoustic positioning device having the above-described characteristic configuration, but also as an acoustic positioning method in which the processing based on such a characteristic structure is taken as a procedure, or as an acoustic positioning program for causing a computer to execute such a characteristic procedure.

Effects of the Invention

[0017] (1) Even when the received acoustic signal includes a positioning pulse and its reflected wave, since it is possible to perform the calculation of the gradient method while suppressing the influence of interference by the reflected wave or the like, it becomes possible to perform the SSBL method calculation accurately and with low weight.

[0018] (2) Since the calculation range of the gradient method is limited to the front part of the positioning pulse part, it becomes possible to suppress the influence of interference by the subsequent reflected wave or the like to a low level.

[0019] (3) If the calculation range of the gradient method is adaptively limited based on the waveform pattern of the reflected wave, the calculation range can be optimized according to the positioning environment, so that the SSBL method calculation can be performed more accurately.

Brief Description of the Drawings

[0020] [Figure 1] It is a functional block diagram of an acoustic positioning system to which the present invention is applied. [Figure 2] It is a diagram showing a method of transmitting an acoustic signal in the present invention. [Figure 3] It is a functional block diagram of a first embodiment of a direction calculation unit. [Figure 4] It is a diagram showing an arrangement example of three hydrophones. [Figure 5] It is a diagram showing an example of compressing a received acoustic signal. [Figure 6] It is a diagram showing an example of detecting an envelope line of a received acoustic signal. [Figure 7]This figure shows an example (part 1) of setting the calculation range to which the iterative gradient method is applied. [Figure 8] This figure shows an example (part 2) of setting the calculation range to which the iterative gradient method is applied. [Figure 9] This is a functional block diagram of the second embodiment of the direction calculation unit. [Figure 10] This figure shows an example of setting the calculation range to which the iterative gradient method is applied in the second embodiment. [Figure 11] This is a functional block diagram of the third embodiment of the direction calculation unit. [Figure 12] This figure shows an example of setting the calculation range to which the iterative gradient method is applied in the third embodiment. [Figure 13] This diagram illustrates an example of how a water-air combined drone can be used. [Figure 14] This figure shows an ideal example of a received acoustic signal. [Figure 15] This figure shows an example where the received acoustic signal includes a positioning pulse and its reflected wave. [Figure 16] This figure shows an example where the reflected wave of a positioning pulse is superimposed behind the positioning pulse in a received acoustic signal. [Modes for carrying out the invention]

[0021] Embodiments of the present invention will be described in detail below with reference to the drawings. Figure 1 is a functional block diagram showing the configuration of an acoustic positioning system 1 to which the present invention is applied, and its main components are a plurality of pingers 10 that transmit acoustic signals and a receiver 20 that calculates the position of each pinger 10 based on positioning pulses. When the present invention is applied to an aerial-water combined drone, the receiver 20 is mounted on the aerial drone and at least one of the pingers 10 is mounted on the underwater drone.

[0022] Each pinger 10 mainly consists of a pressure sensor 101, a depth measurement unit 102, a pulse generation unit 103, and a pulse transmission unit 104, and transmits pulse signals according to the detection and measurement results from the pressure sensor 101 and the depth measurement unit 102 at predetermined time slot periods.

[0023] The receiver 20 primarily consists of a pulse receiving unit 202 equipped with at least three hydrophones 201 and a positioning unit 203. The positioning unit 203 includes a direction calculation unit 203a and a depth calculation unit 203b, which calculate the direction and depth of each pinger 10 based on the delay difference of the acoustic signals received by each hydrophone 201, and position each pinger 10 based on the calculation results.

[0024] Such a receiver 20 can be configured by implementing an application (program) that realizes each of the functions detailed below on a general-purpose computer or server equipped with a CPU, ROM, RAM, bus, interface, etc. Alternatively, it can be configured as a dedicated machine or single-function machine in which part of the application is implemented in hardware or software.

[0025] Figure 2 shows the method of transmitting acoustic signals by the pinger 10, where each pinger 10 transmits a positioning pulse AP1 and a depth pulse AP2 as an acoustic signal AP in that order for each time slot. The positioning pulse AP1 is transmitted to the receiver 20 for the purpose of causing it to calculate the direction of the pinger 10 based on the difference in arrival times (delay difference) Δd to the three hydrophones 201. The depth pulse AP2 is transmitted to the receiver 20 for the purpose of causing it to calculate the depth of the pinger 10 based on the difference in arrival times with the positioning pulse AP1.

[0026] The duration of both the positioning pulse AP1 and the depth pulse AP2 is set to 1.6 ms. In this embodiment, the transmission interval and time slot length of each pulse AP1 and AP2 are set based on the maximum arrival time of the acoustic signal, which is assumed to be 50 m for the maximum distance from the pinger 10 to the receiver 20 and 1500 m / s for the speed of sound, resulting in 50 / 1500 = 0.033 s = 33 ms.

[0027] The pulse generation unit 103 generates a positioning pulse AP1 with a duration of 1.6 ms for each time slot. The pulse generation unit 103 further generates a depth pulse AP2 at a time when 50 ms has elapsed since the transmission of the positioning pulse AP1 (which is defined as depth 0 m) plus a time interval proportional to the depth measured by the depth measurement unit 102 based on the output of the pressure sensor 101.

[0028] For example, if the measurement depth is 10m, the depth pulse AP2 is generated 50+10=60ms after the transmission time of the positioning pulse AP1. If the measurement depth is the maximum depth of 50m, the depth pulse AP2 is generated 50+50=100ms after the transmission time. The pulse transmission unit 104 transmits the positioning pulse AP1 and the depth pulse AP2 at their respective generation times.

[0029] In this embodiment, since the time slot length is set to 150ms, even if the position of the pinger 10 is at the maximum distance and depth from the receiver 20, the transmission time from transmission to reception of each pulse signal AP1, AP2 can be contained within the time slot.

[0030] In the receiver 20, the three hydrophones 201 receive the acoustic signal AP transmitted by each pinger 10. In this embodiment, before starting positioning, for example, the transmission timing of the acoustic signal is learned (recorded) for each pinger 10 when the pinger 10 and the receiver 20 are in contact at a distance of zero. As a result, the receiver 20 can predict the time slots in which it is possible to receive the acoustic signal AP from each pinger 10, and can process the acoustic signal received by each pinger 10 separately.

[0031] In the positioning unit 203, the direction calculation unit 203a calculates the direction of the pinger 10 based on the arrival time difference Δd of the positioning pulse AP1 contained in each acoustic signal received by each hydrophone 201. The depth calculation unit 203b calculates the depth of the pinger 10 based on the arrival time difference between the positioning pulse AP1 and the depth pulse AP2. The positioning unit 203 determines the position of the pinger 10 based on the above-mentioned calculation results for direction and depth.

[0032] In the example above, the time slot length was set to 150ms to transmit two acoustic signals (positioning pulse AP1 and depth pulse AP2) in each time slot. However, for example, the depth pulse AP2 could be omitted and only the positioning pulse AP1 could be transmitted. In that case, the time slot length could be shortened to about 50ms (assuming a maximum distance of about 50 meters).

[0033] Figure 3 is a block diagram showing the configuration of the first embodiment of the direction calculation unit 203a, and its main components are the envelope detection unit 2031, the calculation range setting unit 2032, and the arrival time difference calculation unit 2033 which calculates the arrival time difference Δd of each positioning pulse AP1 using the iterative gradient method.

[0034] In this embodiment, as shown in Figure 4, three hydrophones 201 are arranged at the vertices of a right-angled isosceles triangle. The output system of the hydrophone 201 located at the right-angled vertex is referred to as channel 0, the output system of the hydrophone 201 in channel 0 and the hydrophone 201 arranged parallel to the water surface is referred to as channel 1, and the output system of the hydrophone 201 in channel 0 and the hydrophone 201 arranged perpendicular to the water surface is referred to as channel 2.

[0035] The envelope detection unit 2031 processes the acoustic signals received by each hydrophone 201 and converts them into an envelope. The received acoustic signal includes a positioning pulse portion A corresponding to the positioning pulse AP1, a reflected wave portion (interference wave portion) B thereof, and an overlapping portion C where the rear portion of positioning pulse portion A and the front portion of reflected wave portion B overlap.

[0036] In this embodiment, such an acoustic signal is first subjected to convolution processing using a positioning pulse AP1, as disclosed in Non-Patent Document 4, to obtain pulse-compressed signal waveforms for each channel, as shown in Figure 5. Next, these compressed signal waveforms are squared and a low-pass filter is applied to obtain the envelopes shown in Figure 6 for each channel.

[0037] The calculation range setting unit 2032 sets a portion of the forward side of the positioning pulse portion A in the envelope of each channel as a calculation range for determining the arrival time difference Δd by applying the iterative gradient method. The positioning pulse portion A can be defined, for example, as a predetermined time width centered on the peak position.

[0038] In this embodiment, as shown in Figure 7, the calculation range is set to a certain range ΔT of the positioning pulse portion A that starts around time 0.123 seconds and precedes the peak position. Alternatively, as shown in Figure 8, the calculation range ΔT may be set to a certain range of approximately the first half of the positioning pulse portion A. The calculation ranges ΔT (ΔT0, ΔT1, ΔT2) for each channel 0, 1, and 2 are set to the same time length, although their time ranges are different.

[0039] The arrival time difference calculation unit 2033 calculates the arrival time difference Δd of positioning pulses (positioning pulse portion A) between channels using the iterative gradient method. The time difference d between channels (between hydrophones) detected by the gradient method is given by equation (1) below, where d0 is the initial delay difference.

[0040]

number

[0041] Here, the arrival time difference Δd is calculated using the following equation (2), where Δti is the slope of the signal at each time point i and Δci is the difference between channels displaced by d0.

[0042]

number

[0043] In the iterative gradient method, the arrival time difference Δd can be calculated by starting with an initial delay difference of 0 and repeating equation (1) several times. The iterative gradient calculation may be performed while decimating a portion of the signal. For details of the iterative gradient method, the contents of Non-Patent Documents 3 and 4 by the inventors are referenced here.

[0044] In this embodiment, the arrival time difference Δd1 between channels 0 and 1 is obtained by applying the iterative gradient method to the signals between the calculation range ΔT0 of channel 0 and the calculation range ΔT1 of channel 1. Similarly, the arrival time difference Δd2 between channels 0 and 2 is obtained by applying the iterative gradient method to the signals between the calculation range ΔT0 of channel 0 and the calculation range ΔT2 of channel 2.

[0045] Here, the arrival time difference Δd1 is the time (delay) difference in the x direction, and the arrival time difference Δd2 is the time (delay) difference in the y direction. Therefore, the direction calculation unit 203a can calculate the azimuth angle and depression angle of the object based on the respective arrival time differences Δd1 and Δd2.

[0046] Figure 9 is a functional block diagram showing the configuration of the main parts of the direction calculation unit 203a according to the second embodiment of the present invention, where the same reference numerals represent the same or equivalent parts. This embodiment is characterized in that it includes a reflected wave pattern detection unit 2034 that detects the peak value of the reflected wave portion B, and the calculation range setting unit 2032 adaptively sets the calculation range ΔT based on the peak value of the reflected wave portion B.

[0047] Figure 10 is a schematic diagram showing the method for setting the calculation range ΔT in this embodiment. When the reflected wave pattern detection unit 2034 detects the peak value of the reflected wave portion B, the calculation range setting unit 2032 moves the termination position of the calculation range ΔT further forward as the peak value increases, thereby limiting the calculation range ΔT to a narrower forward portion of the positioning pulse portion A.

[0048] In other words, the starting point of the calculation range ΔT is the same regardless of the peak value of the reflected wave portion B, but the ending point of the calculation range ΔT is the same regardless of the relatively high peak value P. H In this case, the peak value P is low. L This can be accelerated compared to the previous case. Therefore, the peak value P is relatively high. H The corresponding calculation range ΔT(H) is the low peak value P L This becomes shorter than the corresponding calculation range ΔT(L).

[0049] Generally, the later portion of the preceding positioning pulse A is more strongly affected by the larger the peak value of the following reflected wave portion B. According to this embodiment, the larger the peak value of the reflected wave portion B, the longer the latter half of the positioning pulse portion A is excluded from the calculation range ΔT, making it possible to calculate the arrival time difference while suppressing the influence of the reflected wave.

[0050] Figure 11 is a functional block diagram showing the configuration of the main parts of the direction calculation unit 203a according to the third embodiment of the present invention, where the same reference numerals represent the same or equivalent parts. This embodiment is characterized in that it includes a reflected wave pattern detection unit 2035 that detects the interval Δt between the positioning pulse portion A and the reflected wave portion B, and the calculation range setting unit 2032 adaptively sets the calculation range ΔT according to the length of the interval Δt of each portion.

[0051] Figure 12 schematically shows the method for setting the calculation range ΔT in this embodiment. When the reflected wave pattern detection unit 2035 detects the interval Δt of each part, the calculation range setting unit 2032 moves the termination position of the calculation range ΔT further forward as the interval Δt narrows, thereby limiting the calculation range ΔT to the narrower forward portion of the positioning pulse portion A.

[0052] In other words, the starting point of the calculation range ΔT is the same regardless of the interval Δt between each part, but the ending point of the calculation range ΔT is earlier when the interval Δt is relatively narrow (Δt(H)) than when it is wide (Δt(L)). Therefore, the calculation range ΔT(H) corresponding to a relatively narrow interval is shorter than the calculation range ΔT(L) corresponding to a wide interval.

[0053] Generally, the later portion of the preceding positioning pulse A is more strongly affected by the following reflected wave portion B as the interval Δt between them narrows. According to this embodiment, the narrower the interval Δt between the reflected wave portion B and the positioning pulse A, the longer the latter half of the positioning pulse A is excluded from the calculation range ΔT, making it possible to calculate the arrival time difference while suppressing the influence of the reflected wave.

[0054] Here, the calculation range of the iterative gradient method is based on channel 0, as in this embodiment, when calculating the arrival time difference Δd1 between channels 0 and 1 and the arrival time difference Δd2 between channels 0 and 2. That is, the calculation range ΔT = ΔT0 (= ΔT1 = ΔT2), and the starting points of ΔT1 and ΔT2 depend on the initial displacement vector. If the initial displacement is 0, it will be the same as ΔT0.

[0055] Furthermore, although the above embodiments have described the calculation of each arrival time difference Δd using an iterative gradient method, the present invention is not limited to this, and the calculation may also be performed using a well-known gradient method that does not involve iterative calculation.

[0056] Furthermore, although the above embodiments have been described using the example of a case where the interfering wave following the positioning pulse in the received acoustic signal is a reflected wave, the present invention is not limited to this and can be similarly applied to a wide variety of interfering waves other than reflected waves.

[0057] Furthermore, according to each of the embodiments described above, it becomes possible to perform iterative gradient calculations that suppress the effects of interference from reflected waves, and even when reflected waves are received, SSBL calculations can be performed accurately and efficiently. Therefore, it becomes possible to contribute to United Nations-led Sustainable Development Goals (SDGs) Goal 9, "Build resilient infrastructure and promote inclusive and sustainable industrialization," and Goal 11, "Make cities inclusive, safe, resilient and sustainable." [Explanation of Symbols]

[0058] 1…Acoustic positioning system, 10…Pinger, 20…Receiver, 101…Period setting unit, 101…Pressure sensor, 102…Depth measurement unit, 103…Pulse generation unit, 104…Pulse transmission unit, 201…Hydrophone, 202…Pulse receiving unit, 203…Positioning unit, 203a…Direction calculation unit, 203b…Depth calculation unit, 2031…Envelope detection unit, 2032…Calculation range setting unit, 2033…Arrival time difference calculation unit, 2034,2035…Reflected wave pattern detection unit

Claims

1. In an acoustic positioning device that receives acoustic signals containing positioning pulses transmitted by an object underwater using multiple hydrophones and determines the object's position based on the difference in arrival times of each positioning pulse, Each acoustic signal is enveloped by pulse compression, squaring, and low-pass filtering using the positioning pulses, Means for detecting the waveform pattern of the interference wave portion that follows the positioning pulse portion of each envelope, Means for setting the forward portion of each envelope's positioning pulse portion, corresponding to the detection result of the waveform pattern, within the calculation range of the arrival time difference, An acoustic positioning device characterized by comprising means for calculating the difference in arrival times between the calculation ranges of each envelope using the gradient method.

2. The means for detecting the waveform pattern detects the peak value of the interfering wave portion, The acoustic positioning device according to claim 1, characterized in that the means for setting the calculation range is set to move further forward as the peak value increases.

3. The means for detecting the waveform pattern detects the interval between the positioning pulse portion and the interference wave portion, The acoustic positioning device according to claim 1, characterized in that the means for setting the calculation range is set to move further forward the shorter the interval.

4. The acoustic positioning device according to any one of claims 1 to 3, characterized in that the means for calculation calculates the arrival time difference using an iterative gradient method.

5. The acoustic positioning device according to claim 4, characterized in that the means for calculation performs calculations using the iterative gradient method by downsampling the data.

6. The acoustic positioning device according to any one of claims 1 to 3, characterized in that the interfering wave portion is a reflected wave portion corresponding to the reflected wave of the positioning pulse.

7. In an acoustic positioning method in which an object underwater transmits an acoustic signal including a positioning pulse, which is received by multiple hydrophones, and a computer determines the position of the object based on the difference in arrival times of each positioning pulse, Each acoustic signal is processed by pulse compression, squaring, and low-pass filtering using the positioning pulses to obtain an envelope. The waveform pattern of the interference wave portion that follows the positioning pulse portion of each envelope is detected. The portion ahead of the positioning pulse portion of each envelope, corresponding to the detection result of the waveform pattern, is set as the calculation range for the arrival time difference. An acoustic positioning method characterized by calculating the difference in arrival times between the calculation ranges of each envelope using the gradient method.

8. In an acoustic positioning program that receives acoustic signals containing positioning pulses transmitted by an object underwater using multiple hydrophones and determines the object's position based on the difference in arrival times of each positioning pulse, A procedure for processing each acoustic signal into an envelope by pulse compression, squaring, and low-pass filtering using the positioning pulse, A procedure for detecting the waveform pattern of the interfering wave portion that follows the positioning pulse portion of each envelope, A procedure for setting the forward portion of the positioning pulse portion of each envelope, corresponding to the detection result of the waveform pattern, within the calculation range for the arrival time difference, An acoustic positioning program characterized by causing a computer to perform a procedure for calculating the difference in arrival times between the calculation ranges of each envelope using the gradient method.

Images