JP7833295B2 - Relative position detection system, relative position detection method, and position information transmission device - Google Patents

Description

This invention relates to a relative position detection system, a relative position detection method, and a position information transmission device.

In areas prone to landslides and other disasters, ground monitoring systems using GPS (Global Positioning System) or GNSS (Global Navigation Satellite System) have been proposed to monitor even slight displacements of the ground and structures. Here, GPS refers to satellites operated by the United States, while GNSS refers to satellite systems launched by countries including the United States, Japan, Europe, and China. While this invention will use the term GNSS consistently, it goes without saying that this invention can also be applied to GPS.

GNSS position detection methods can be broadly classified into two types: methods that detect code phase and methods that detect carrier phase. The method widely used in applications such as car navigation systems is the method that detects code phase. Since the code signal transmitted from GNSS satellites changes at approximately 1 MHz, the Earth's position can be determined with an error of a few meters to tens of meters. In contrast, the method that detects carrier phase achieves significantly higher accuracy (less than a few centimeters) by detecting the carrier frequency (1.5 GHz).

This invention relates to a relative position detection system and method that detects the carrier phase of radio waves transmitted from a GNSS satellite, thereby enabling the detection of slight displacements in the ground and structures, and to a position information transmission device that transmits the carrier phase of radio waves transmitted from a GNSS satellite.

Two receiving methods are known for GNSS (Guided Navigation System) for detecting carrier phase: static and kinematic. The static method achieves ultra-high precision (error of a few millimeters) and is used in surveying, but requires continuous observation of the carrier phase. Therefore, it consumes a large amount of power and cannot be used for detecting landslides or mudslides in locations without commercial power.

The kinematic method enables highly accurate relative position detection (error of a few centimeters) by transmitting the integrated value of the carrier wave phase every second. For example, Patent Document 1 describes a method in which radio waves transmitted by a GNSS satellite are received, the integrated value of the carrier wave phase is counted, and transmitted to a center at regular intervals. The center then determines the reception point position (baseline vector) based on the received integrated phase value.

To detect landslides and mudslides using this kinematic method, it is necessary to continuously transmit carrier phase integration values from multiple satellites without interruption. Therefore, a transmission rate of at least 1 kbps is required. However, in many areas prone to landslides and mudslides, the network is not well-established, posing a problem in such cases where the method cannot be applied.

Patent No. 3532267

Therefore, the objective of this invention is to provide a relative position detection system, a relative position detection method, and a position information transmission device that can be installed on cliffs or similar terrain, operate for extended periods using batteries, and achieve positional accuracy of several centimeters.

[1] The relative position detection system 1 of the present invention includes a calculation means 6, a receiving base station 5 connected to the calculation means 6 by a communication line, a plurality of position information transmitting devices 2A, 2B each acquiring instantaneous values of carrier phase of radio waves transmitted from a plurality of GNSS (Global Navigation Satellite System) satellites at a predetermined sampling timing synchronized with GNSS time, and transmitting the acquired instantaneous values of carrier phase to the receiving base station wirelessly, thereby transmitting the instantaneous values of carrier phase to the calculation means 6, and a satellite information acquisition means 4 connected to the calculation means by a communication line, which acquires satellite orbit information of GNSS satellites and transmits the acquired satellite orbit information to the calculation means 6. The calculation means 6 is a relative position detection system 1 that detects the relative positions of the plurality of position information transmitting devices using the instantaneous values of carrier phase and the satellite orbit information.

[2] In the relative position detection system 1 of the present invention, it is preferable that the calculation means 6 detects the relative positions of the plurality of position information transmitting devices 2A and 2B using the double phase difference calculation of the carrier wave phase.

[3] In the relative position detection system 1 of the present invention, it is preferable that the transmission rate from the plurality of position information transmitting devices 2A and 2B to the receiving base station 5 is lower than the transmission rate from the satellite information acquisition means 4 to the calculation means 6.

[4] In the relative position detection system 1 of the present invention, it is preferable that the wireless transmission from the plurality of position information transmitting devices 2A and 2B to the receiving base station 5 is performed by LPWA (Low Power Wide Area) wireless communication means that is capable of long-distance communication and has low power consumption.

[5] In the relative position detection system 1 of the present invention, it is preferable that the interval of the predetermined sampling timing (SEC00) is set to one minute or more.

[6] The relative position detection method of the present invention involves acquiring instantaneous values of carrier phases of radio waves transmitted from multiple GNSS satellites at multiple measurement points at a predetermined sampling timing synchronized with GNSS time, transmitting the acquired instantaneous values of carrier phases to a calculation means by communication means including at least wireless transmission, acquiring satellite orbit information of the multiple GNSS satellites, transmitting the acquired satellite orbit information to the calculation means, and the calculation means detecting the relative position between the multiple measurement points using the instantaneous values of carrier phases and the satellite orbit information.

[7] In the relative position detection method of the present invention, it is preferable to detect the relative positions of the plurality of position information transmitting devices 2A and 2B using the double phase difference calculation of the carrier wave phase.

[8] In the relative position detection method of the present invention, it is preferable that the transmission rate of the wireless transmission is lower than the transmission rate of the satellite information.

[9] In the relative position detection method of the present invention, it is preferable to use LPWA (Low Power Wide Area) wireless communication means capable of long-distance communication and low power consumption for the wireless transmission.

[10] In the relative position detection method of the present invention, it is preferable that the interval of the predetermined sampling timing (SEC00) is 1 minute or more.

[11] In the relative position detection method of the present invention, the relative positions of the plurality of measurement points are represented by a three-dimensional relative distance vector P calculated using the GNSS antenna of a reference measurement point among the two measurement points as the starting point and the GNSS antenna of the other measurement point among the two measurement points as the ending point. Preferably, a three-dimensional estimated relative distance vector Re is estimated as an estimated value of the relative distance vector P, an error evaluation value E, which serves as an index for evaluating position error, is calculated from the estimated estimated relative distance vector Re, an estimated relative distance vector that minimizes the error evaluation value E is determined, and the determined estimated relative distance vector is output as the relative distance vector P.

[12] In the relative position detection method of the present invention, it is preferable to obtain a double phase difference φ(n) from the instantaneous value of the carrier wave phase, obtain an estimated double phase difference φ''(n) based on the estimated relative distance vector Re, and calculate the error evaluation value E according to the difference between the double phase difference φ(n) and the estimated double phase difference φ''(n).

[13] In the relative position detection method of the present invention, the estimated double phase difference φ''(n) is calculated by using the satellite orbit information of the plurality of GNSS satellites to calculate the position information of the plurality of GNSS satellites at the predetermined sampling timing, and when one of the plurality of GNSS satellites is designated as the reference satellite, the reference satellite position coordinates (X(0), Y(0), Z(0)) of the reference satellite are extracted from the position information of the plurality of GNSS satellites, and the position coordinates (X(n), Y(n), Z(n)) of each satellite other than the reference satellite are extracted from the position information of the GNSS satellites, and the position coordinates (Xr, Yr, Zr) of the reference measurement point are determined, and the reference satellite position coordinates (X(0), Y(0), Z(0)), the position coordinates (X(n), Y(n), Z(n)) of each satellite n are obtained, and the position coordinates (Xr, Yr, It is preferable to calculate the wave vector k(n) using Zr, obtain the estimated double phase difference φ'(n) by performing an inner product operation between the wave vector k(n) and the estimated relative distance vector Re, and then apply a modulo operation modulo 2π to the estimated double phase difference φ'(n).

[14] The position information transmission device 2 of the present invention includes a satellite signal receiving unit 221 that receives radio waves transmitted from a plurality of GNSS satellites and outputs carrier wave phase and satellite orbit information contained in the radio waves; a calculation unit 23 that detects GNSS time information from the satellite orbit information; a sampling means 25 that samples the carrier wave phase output from the satellite signal receiving unit 221 at a predetermined sampling timing to acquire an instantaneous value of the carrier wave phase; and a wireless communication unit 26 that wirelessly transmits the instantaneous value of the carrier wave phase acquired by the sampling means 25, wherein the predetermined sampling timing is synchronized with the GNSS time information.

[15] In the position information transmission device 2 of the present invention, it is preferable that the predetermined sampling timing is a predetermined time determined by the GNSS time information.

[16] In the position information transmission device 2 of the present invention, it is preferable that the wireless communication unit 26 wirelessly transmits the instantaneous value of the carrier phase using a long-distance communication-capable and low-power LPWA (Low Power Wide Area) wireless communication means.

[17] In the position information transmission device 2 of the present invention, it is preferable that the interval of the predetermined sampling timing (SEC00) is 1 minute or more.

According to the relative position detection system, relative position detection method, and position information transmission device of the present invention, instantaneous values of carrier phase acquired at a sampling timing synchronized with GNSS time are obtained from the position information transmission device, and satellite information is obtained from a separately provided GNSS receiver (satellite information acquisition means). The relative positions of multiple position information transmission devices can be detected using these instantaneous values of carrier phase and satellite information. Therefore, the position information transmission device only needs to acquire and transmit instantaneous values of carrier phase at an extremely low sampling rate (for example, once per minute), making it possible to perform high-precision position detection using carrier phase with extremely low power consumption, and even operate on batteries, for example. Furthermore, since the amount of information transmitted wirelessly can be made extremely small, by using a low-power, long-range wireless transmission system such as LPWA, it becomes possible to install the system in mountainous areas where conventional mobile phones have no signal and detect landslides in the mountains. As a result, the present invention makes it possible to provide a relative position detection system, a relative position detection method, and a position information transmission device that can be installed on cliffs and other similar terrain, operate for extended periods using batteries, and achieve positional accuracy of several centimeters.

This is a diagram showing the configuration of a GNSS position information transmission device 40 according to the conventional method 1. This is a diagram showing the configuration of a GNSS position information transmission device 41 according to the conventional method 2. This is a schematic diagram illustrating the difference between the instantaneous value and the integrated phase of the carrier wave. This is an experimental configuration diagram of the integrated phase detection method. These are the experimental results of the integrated phase detection method. This is a configuration diagram of the location information transmission device 2 according to Embodiment 1. This figure shows the payload configuration of the LPWA wireless signal transmitted by the location information transmission device 2 according to Embodiment 1. This is a configuration diagram of the relative position detection system 1 according to Embodiment 1. This figure shows a relative position detection method using the relative position detection system 1 according to Embodiment 1. This figure illustrates the relative position detection method according to Embodiment 2 using the relative position detection system 1 according to Embodiment 1. This is an explanatory diagram illustrating the calculation method of the calculation server 6 in the relative position detection system according to Embodiment 2. This figure shows an example of two GNSS antennas, A1 and B1, being installed at a predetermined distance on the roof of a two-story building for the purpose of conducting an experiment. This figure shows the measurement results of the relative distance R obtained through the experiment. This figure shows the measurement results of the horizontal azimuth angle α with respect to the north-south line L1 and the vertical angle β with respect to the horizontal plane L2, obtained through experiments.

The present invention will be explained below in comparison to conventional methods using the embodiments shown in the figures. First, the conventional methods will be described.

[Conventional Method 1]
Figure 1 is a diagram of the configuration of a GNSS position information transmission device 40 according to Conventional Method 1. The GNSS position information transmission device 40 is a GNSS position information transmission device that detects code signals (code phase information), and is a typical GNSS position information transmission device installed in car navigation systems and the like. Since it is publicly known technology, it will be explained in a simplified manner. The antenna 20 receives radio waves from each GNSS satellite, and the satellite signal receiving unit 221 receives the code phase information and satellite orbit information (ephemeris) of each satellite, which are then provided to the calculation unit 23. The front end 21 filters out the weak received signals, amplifies them, converts them into low-frequency signals (IQ signals), and supplies them to multiple satellite receiving units 221. The satellite receiving unit 221 consists of a synchronization unit, a random number generation unit, a despreader unit, a decoding unit, etc. (not shown), and outputs the code phase information and satellite orbit information of each satellite.

The calculation unit 23 determines the satellite position (X(n), Y(n), Z(n)) of each satellite from the satellite orbit information, and determines the distance (pseudo-distance) between each satellite and the receiving point from the code phase information. The calculation unit 23 solves a system of equations formed from the positions of four or more satellites and the pseudo-distances to output the receiving point position (Xr, Yr, Zr) and the precise GNSS time. It outputs a "1PPS" signal, for example, once per second, as a pulse precisely matching the GNSS time.

Incidentally, since the modulation frequency of the code signal is approximately 1 MHz, the position (Xr, Yr, Zr) detected by the device shown in Figure 1 contains a measurement error of several meters to tens of meters. This lacks the accuracy necessary to detect slight displacements of the ground or structures, which is the objective of this invention.

[Conventional Method 2]
Figure 2 is a diagram of the configuration of the GNSS position information transmission device 41 according to the conventional method 2. The GNSS position information transmission device 41 is a position information transmission device that detects the carrier wave phase. Since the carrier wave frequency (1.5 GHz) is significantly higher than the modulation frequency (approximately 1 MHz), high accuracy with an error of several centimeters can be achieved. In Figure 2, the satellite signal receiving unit 221 outputs the carrier wave phase (φ1, φ2, ..., φM) of each satellite in addition to the code phase information and satellite orbit information. The carrier wave phase is a signal that changes in the range from -π to π, and a "phase jump" occurs each time a phase boundary is crossed. Therefore, the phase integration unit 222 integrates the carrier wave phases and outputs them as integrated phases (φs(1), φs(2), ..., φs(M)). By using integrated phases, the effect of "phase jumps" can be reduced.

Figure 3 is a schematic diagram illustrating the difference between the instantaneous and integrated carrier phases. In the example shown in Figure 3, assuming that carrier phase measurement begins at time t=0, the instantaneous value φ(n) and integrated phase φs(n) of the nth satellite's carrier phase at time t=T are schematically displayed. In Figure 3, the instantaneous value φ(n) of the carrier phase folds between +π and -π, so at time t=T, φ(n) = 0, making it impossible to determine how much the phase has rotated since the start of observation. In contrast, the integrated phase φs(n) at time t=T is 7π, indicating that the phase has rotated 3 times since the start of observation, and there has been a phase shift of π.

The integrated phases (φs(1), φs(2), ..., φs(M)) are held by a latch 25 every second. Since each integrated phase is an integrated value over one second, it contains more than 30 bits of information. These integrated phases are transmitted from the antenna 27 by the wireless communication unit 26. As this is a known technique, details are omitted, but in high-precision relative position detection using carrier phase, the integrated phases (φs(1), φs(2), ..., φs(M)) are received by a receiving device (not shown), the phase difference is calculated between different receiving devices, and then the phase difference between different satellites is calculated. This calculation, which involves calculating the phase difference twice, is called double-difference phase calculation, and it removes the effects of the ionosphere and frequency shifts of the receiving devices. Finally, the phase uncertainty that is an integer multiple of 2π is removed, and high-precision position information is output. The phase uncertainty that is an integer multiple of 2π is generally called the "integer value bias N". In the conventional method 2, there is a problem in that it is difficult to correctly determine the value of this integer bias N for each received satellite.

Figure 4 shows the experimental configuration diagram of the integrated phase detection method. Figure 5 shows the experimental results of the integrated phase detection method. In Figure 4, the GNSS position information transmitter 41B was initially installed at location A. Here, location A is 1 m away from the location where the GNSS position information transmitter 41A was placed (reference position). In this experiment, the installation location of 41B was shifted by 10 cm every 20 seconds, moving from A→B→C→D→E→F→G.

Using the GNSS position information transmitter 41A as the reference position and the GNSS satellite 3B as the reference satellite, the integrated phase output from the GNSS position information transmitter 41B was calculated, and the result of performing a double-difference phase calculation is shown in Figure 5. As can be seen from Figure 5, a movement of 10 cm at 20-second intervals is detected with high accuracy. Furthermore, it can be seen that by using this integrated phase, even if the phase difference exceeds the range of ±π, it can be detected as a continuous waveform.

To achieve high-precision position detection using carrier phase, however, the wireless transmission unit 26 must transmit integrated phases (φs(1), φs(2), ..., φs(M)) every second. Assuming each integrated phase is 30 bits and the number of satellites M is 6, the wireless communication unit 26 requires a transmission rate of 180 bps. In reality, redundancy such as headers and error correction necessitates several times that transmission rate (e.g., 1 kbps). While this transmission rate could be easily achieved using cellular networks, wireless devices using cellular networks have the problem of high power consumption, and furthermore, there are many places, such as mountainous areas, where cellular networks are unavailable.

Incidentally, LPWA (Low Power Wide Area) wireless communication has the advantage of long communication range, making it usable even in mountainous areas, and also low power consumption. However, the amount of data that can be transmitted with LPWA is small, for example, about 128 bits per minute. Converted to a transmission rate, this is 2 bps, which is two orders of magnitude lower than the 180 bps required for the transmission of integrated phase. A GNSS position information transmission device that can detect carrier phase and bridge this two-order-of-magnitude difference is desired.

Next, embodiments of the present invention will be described.
[Embodiment 1]
Figure 6 is a configuration diagram of the position information transmission device 2 according to Embodiment 1. The position information transmission device 2, which is installed at multiple measurement points, receives radio waves from GNSS satellites using the receiving antenna 20 as shown in Figure 6, and the satellite signal processing unit 22 performs GNSS reception processing. The front end 21 extracts the weak received signal with a filter, amplifies it, converts it into a low-frequency signal, and supplies it to multiple satellite signal receiving units 221. The satellite signal receiving units 221 output code phase information, satellite orbit information, and carrier phase (φ1, φ2, , φM). The calculation unit 23 calculates the position (Xr, Yr, Zr) where the receiving antenna 20 is placed and the accurate GNSS time. In the implementation of the present invention, it is assumed that the carrier frequency transmitted from the GNSS satellite is approximately 1.5 GHz.

The latch (sampling means) 25 samples (acquires) the carrier phase (φ1, φ2, ..., φM) at the timing of 00 seconds every minute of the GNSS time (sampling timing SEC00). The sampling timing SEC00 can be created, for example, by generating one pulse per second (1 PPS pulse) and dividing it by 60 in synchronization with the GNSS time. Here, the value sampled by the latch 25 is the instantaneous value of the carrier phase, not the integrated value.

In other words, the latch 25 obtains the instantaneous value of the carrier phase by sampling the carrier phase (φ1, φ2, ..., φM) output from the satellite signal receiving unit 221 at a predetermined sampling timing (sampling timing SEC00). Note that the "instantaneous value of the carrier phase" obtained by sampling the carrier phase (φ1, φ2, ..., φM) is represented as "φ(1), φ(2), ..., φ(M)" with parentheses around the subscripts 1, 2, ..., M, to distinguish it from the actual carrier phase (φ1, φ2, ..., φM).

The instantaneous values of the carrier phase (φ(1), φ(2), ..., φ(M)) sampled by latch 25 are then wirelessly transmitted by the wireless communication unit 26. Since the timing of latching the instantaneous values of the carrier phase is precisely determined by GNSS time, the calculation server (calculation means) 6, described later, can achieve high-precision relative position detection.

Figure 7 shows the payload configuration of the LPWA wireless signal transmitted by the location information transmission device 2 according to Embodiment 1. The first two bytes are the header, which contains information such as the ID number of the location information transmission device 2 and the battery voltage. The remaining 14 bytes contain two bytes of information obtained from up to seven satellites (satellite number (6 bits), CNR (2 bits), carrier phase (8 bits)). This CNR is an index representing carrier strength and corresponds to the Carrier to Noise Ratio in English. Here, in order to compress the amount of information during transmission, the CNR is simplified and only two bits are transmitted. That is, "11" is transmitted when the CNR is good, and "10" or "01" is transmitted when it is moderate. If no satellite is received, the CNR is set to "00".

Figure 8 is a configuration diagram of the relative position detection system 1 according to Embodiment 1. As shown in Figure 8, the relative position detection system 1 is configured using two of the above-mentioned position information transmitting devices 2. Note that when describing the two position information transmitting devices 2 separately, they may be referred to as position information transmitting device 2A and position information transmitting device 2B. These position information transmitting devices 2A and 2B are installed at two measurement points. Specifically, position information transmitting device 2A is installed in a location with bedrock where there is no risk of landslides. On the other hand, position information transmitting device 2B is installed in a location where positional displacement may occur due to precursory phenomena of landslides. By accurately measuring the relative distance from position information transmitting device 2A to position information transmitting device 2B, precursory phenomena such as landslides can be detected.

As described above, position information transmission devices 2A and 2B receive radio waves from GNSS satellites 3, set the instantaneous carrier phase values of each satellite as payloads, and transmit them as LPWA radio to the receiving base station 5 installed at the foot of the mountain. The receiving base station 5 transmits the payload sent from position information transmission device 2 to the computing server (computation means) 6 located in the cloud. The satellite orbit information acquisition means (GNSS receiver) 4 is composed of a commercially available GNSS receiver, etc., and receives satellite orbit information (ephemeris) from the radio waves of the GNSS satellites and transmits it to the computing server 6. The computing server 6 can accurately determine the positions of all satellites using the satellite orbit information provided by the GNSS receiver 4.

The calculation server 6 calculates the relative distance R using a calculation described later. It is known that when a landslide occurs, the relative distance changes slowly, and eventually, when it exceeds a certain value, the landslide occurs. Therefore, the calculation server 6 determines that the risk of a landslide is increasing when the relative distance exceeds a predetermined value. When the risk is high, the calculation server 6 prompts residents to evacuate by displaying a warning on the smartphone 7.

Figure 9 is a diagram illustrating a relative position detection method using the relative position detection system 1 according to Embodiment 1. The method for calculating relative distance will be explained using Figure 9. In this calculation, one of the received GNSS satellites is designated as the reference satellite and its satellite number is set to "0". The difference calculation shown in [Equation 1] and [Equation 2] below is performed on the instantaneous values of the carrier phase obtained from the reference satellite and the satellite number n. Note that "instantaneous value of carrier phase" may also be written as "instantaneous phase".

[Formula 1] φAr(n)=φA(n) - φA(0)
[Formula 2] φBr(n)=φB(n) - φB(0)

In [Equation 1] and [Equation 2], φA(n) is the instantaneous phase obtained by the latch 25 of the position information transmitting device 2A after receiving radio waves from satellite number n, and φB(n) is the instantaneous phase obtained by the latch 25 of the position information transmitting device 2B after receiving radio waves in the same manner.

Next, using the position information transmitting device 2A as the reference position, the double phase difference φd(n) of the carrier wave phase is determined by the following [Equation 3].

[Formula 3] φd(n)=φBr(n) - φAr(n)

The relationship between the double phase difference φd(n) obtained in [Equation 3] above and the relative distance is shown in [Equation 4] below. Since the range of change of the relative distance is known, [Equation 4] can be solved by applying the least squares method to multiple satellites n.

[Formula 4] 2πN(n)+φd(n)=R×(2π/λ)×{cosθ(n)−cosθ(0))}

In [Equation 4], N(n) is an integer and represents the phase uncertainty with respect to satellite number n. λ is the wavelength of the carrier wave, θ(0) is the angle between the movement direction vector R and the direction vector from the position information transmitter 2A to the reference satellite (0), and θ(n) is the angle between the movement direction vector R and the direction vector from the position information transmitter 2B to satellite n.

Here, the direction vector R is the vector from the coordinates (Xr, Yr, Zr) of the position information transmitter 2A to a virtual point (Xt, Yt, Zt) set in the assumed direction of movement of the position information transmitter 2B. The direction vector to satellite n is the vector from the coordinates (Xr, Yr, Zr) of the position information transmitter 2A to the position of satellite n (X(n), Y(n), Z(n)). The position of each GNSS satellite can be determined by applying Kepler's equations to the satellite orbit information (ephemeris).

To avoid complexity, the above explanation assumes that the direction of movement due to the landslide is known beforehand, and only the length R of the movement direction vector R is calculated. However, as shown in Embodiment 2 described later, by considering a three-dimensional movement direction vector P and applying the same process to correspond to the three-dimensional movement direction, the scope of application can be further expanded. Furthermore, in [Equation 4] above, the integer N(n) representing the phase uncertainty differs for each satellite, and since the integer N(n) must be calculated for each satellite, the computational complexity increases. Errors may occur if the estimation of integer N(n) is incorrect.

[Embodiment 2]
Figure 10 is a diagram illustrating a relative position detection method according to Embodiment 2 using the relative position detection system 1 according to Embodiment 1. In Embodiment 2, a three-dimensional relative distance vector P is obtained by a calculation described later. As shown in Figure 10, the relative distance vector P is a three-dimensional vector that starts from the GNSS antenna of the position information transmission device 2A and ends at the GNNS antenna of the position information transmission device 2B, and represents the relative positions of the two position information transmission devices 2A and 2B.

Figure 10 shows that the relative distance vector P can be decomposed into three components. The relative distance vector P can be expressed in polar coordinates (R, α, β), which consist of the relative distance R between A and B, the angle α on the horizontal plane with respect to the north-south line L1 (the horizontal azimuth angle with respect to the north-south line L1), and the vertical angle β with respect to the horizontal plane L2. While the relative distance vector P can also be expressed using Cartesian coordinates, the explanation will consistently use polar coordinates (R, α, β) from here on.

When a landslide occurs, not only does the relative distance R slowly increase, but the vertical angle β also changes. Furthermore, depending on the cliff structure, horizontal movement is also possible (the horizontal azimuth angle α also changes). Therefore, the calculation server 6 calculates the relative distance vector P at predetermined time intervals, and determines that the risk of a landslide has increased when the changes in the relative distance R, horizontal azimuth angle α, and vertical angle β exceed predetermined values. Similar to the previously described embodiment 1, the calculation server 6 can prompt residents to evacuate by displaying a warning on the smartphone 7.

The calculation server 6 performs the following calculations (steps S1 to S12) to obtain the relative distance vector P.
Figure 11 is an explanatory diagram illustrating the calculation method of the calculation server 6 in the relative position detection system according to Embodiment 2. When explaining steps S1 and beyond, refer to Figure 11 as needed. Note that the relative position detection system according to Embodiment 2 has the same system configuration as the relative position detection system 1 according to Embodiment 1.

[Step S1]
First, in step S1, the instantaneous value of the carrier phase of each satellite (n) transmitted by LPWA, i.e., the instantaneous phase φ(n), is subjected to a two-stage difference calculation as described in [Equation 3] to obtain the double phase difference φd(n).

[Step S2]
In step S2, using the satellite orbit information (ephemeris) provided by the GNSS receiver 4, the reference satellite position coordinates (X(0),Y(0),Z(0)) of the reference satellite (0) and the position coordinates (X(n),Y(n),Z(n)) of each satellite other than the reference satellite (n) at the time the instantaneous phase φ(n) was acquired are determined. In the following, "reference satellite position coordinates (X(0),Y(0),Z(0)) of the reference satellite (0)" may be written as "reference satellite position coordinates (X(0),Y(0),Z(0))", and "position coordinates (X(n),Y(n),Z(n)) of each satellite (n)" may be written as "each satellite position coordinates (X(n),Y(n),Z(n))".

[Step S3]
In step S3, the location where the reference position information transmitting device 2A is installed (reference measurement point) is determined as the receiving point position coordinates (Xr, Yr, Zr).

[Step S4]
In step S4, the direction vector e(n) is calculated from the receiving point coordinates (Xr, Yr, Zr) to each satellite position coordinate (X(n), Y(n), Z(n)). Similarly, the direction vector e(0) is calculated from the receiving point coordinates (Xr, Yr, Zr) to the reference satellite position coordinates (X(0), Y(0), Z(0)). Here, the direction vector is a vector that points only to the direction component, and its magnitude is normalized to "1".

[Step S5]
In step S5, the calculation server 6 obtains the wave vector k(n) = e(n) - e(0) by subtracting the direction vector e(0) toward the reference satellite (0) from the direction vector e(n). Here, the equiphase surfaces of the double phase difference φd(n) are formed in a direction perpendicular to the wave vector k(n). Also, the spacing of the double phase difference φd(n) is inversely proportional to the magnitude of the wave vector k(n).

[Step S6]
In step S6, the computing server 6 determines a three-dimensional estimated relative distance vector Re as an estimated value of the relative distance vector P, as shown in [Equation 5] below.

[Formula 5] Re=(R'+ΔR, α'+Δα, β'+Δβ)

In the above [Equation 5], (R', α', β') is the initial vector used as the starting point for the calculation. Since the displacement due to the landslide is a slow movement, the relative distance vector P is expected to be not significantly different from the initial vector (R', α', β'). Therefore, the initial vector (R', α', β') can be determined by pre-measuring the position of the receiving point B. Alternatively, the relative distance vector P measured a few minutes earlier can be used as the initial vector (R', α', β').

The values we want to find in the following calculations are the combinations of (ΔR, Δα, Δβ) that minimize the distance between the estimated relative distance vector Re and the relative distance vector P. Since the positional change due to a landslide begins slowly, it is expected that these (ΔR, Δα, Δβ) values will be small when precursory phenomena for a landslide occur.

[Step S7]
In step S7, the calculation server 6 scans the values of (ΔR, Δα, Δβ). For example, for ΔR, it scans from ΔR = -0.5m to ΔR = +0.5m in 1mm increments (1000 combinations), for Δα, it scans from Δα = -30 degrees to +30 degrees in 0.5 degree increments (120 combinations), and for Δβ, it scans from Δβ = -15 degrees to +15 degrees in 0.5 degree increments (60 combinations). The total number of scan combinations is 1000 × 120 × 60 = 7,200,000.

By applying various optimal value search techniques, the total number of scan combinations can be reduced, thereby shortening the computation time. In Embodiment 2, the simplification of the calculation is not described, and the following explanation assumes that all combinations are scanned.

[Step S8]
In step S8, the computing server 6 calculates the estimated double phase difference φ'd(n) shown in [Equation 6] below by calculating the dot product of the estimated relative distance vector Re and the wave vector k(n) that have been estimated by the scan.

[Formula 6] φ'd(n)=(2π/λ)×{Re・k(n)}

In [Equation 6], the "•" in Re・k(n) within {} represents the dot product of the two vectors Re and k(n).

The calculations performed by the computation server 6 described above are schematically illustrated in Figure 11 using vectors. Specifically, at receiving point A, the direction vectors e(0) and e(n) of the reference satellite (0) and satellite (n) are calculated, and the wave vector k(n) is obtained by subtracting these two direction vectors. An equiphase surface with a double phase difference φd(n) is formed in a direction orthogonal to the wave vector k(n). The relative distance vector P is a three-dimensional vector extending from receiving point A to receiving point B, which is located at an unknown location. Near receiving point B, the estimated relative distance vector Re extends to the estimated receiving point B'.

If we consider the case where receiving point B is at the position of estimated receiving point B', the estimated double phase difference observed at that time is φ'd(n) determined by [Equation 6] above.

Therefore, by searching for the estimated receiving point B' where the difference between the double phase difference φd(n) observed from the instantaneous phase transmitted via LPWA and the estimated double phase difference φ’d(n) is minimized, we can determine that this point is closest to the coordinates of the desired receiving point B.

However, as mentioned earlier, the instantaneous phase transmitted in LPWA is folded into the range from +π to -π, so the value of the double phase difference φd(n) is similarly folded. In contrast, the estimated double phase difference φ'd(n) is obtained by performing the operation [Equation 6] on the estimated relative distance vector Re, so it is not folded and may, for example, be a value significantly exceeding π. Therefore, step S9 below is performed.

[Step S9]
In step S9, the calculation server 6 performs the calculation shown in [Equation 7] below, and applies a modulo operation (mod) modulo 2π to the estimated double phase difference φ'd(n) to fold it into a range from +π to -π.

[Formula 7] φ“d(n)=mod{φ'd(n), 2π}

When the distance between the estimated receiving point B' and receiving point B is closest, the double phase difference φ“d(n)” calculated by [Equation] 7 will be close to the value of the double phase difference φd(n).

Here, when comparing the double phase difference φ“d(n) shown in [Equation 7] and the double phase difference φd(n) shown in [Equation 3], it is necessary to consider a phase jump of 2π. For example, if we simply subtract φ“d(n) = 0.9π and φd(n) = -0.9π, we get φ“d(n) - φd(n) = 1.8π. However, in this example, the correct phase difference is 0.2π. Since simple subtraction is not possible, some kind of ingenuity is required to determine the phase difference. Therefore, we perform step S10 below.

[Step S10]
As step S10, for example, the degree to which the phases of φ“d(n) and φd(n) differ is evaluated by performing the following calculation [Equation 8].

[Formula 8]
F(n)={cosφ“d(n)−cosφd(n)} ² +{sinφ”d(n)−sinφd(n)} ²

[Step S11]
Finally, in step S11, the calculation shown in [Equation 9] below is performed to sum up all evaluation results from satellite number 1 to satellite number M, and an error evaluation value E, which serves as an index for position error evaluation, is obtained.
For all combinations (7.2 million) of (ΔR, Δα, Δβ) given in step S7, the calculations from steps S8 to S11 are performed to obtain the error evaluation value E. Then, in step S12, the combination of (ΔR, Δα, Δβ) that minimizes the error evaluation value E is extracted, and the estimated relative distance vector Re is obtained using [Equation 5], thereby obtaining the optimal estimated value of the relative distance vector P.

As explained above, the optimal estimated value of the relative distance vector P can be obtained using the instantaneous values of the carrier phase (φ(1), φ(2), ..., φ(M)). Since the instantaneous values of the carrier phase contain little information, it is possible to detect precursory landslide phenomena using low-rate, low-power wireless communication technologies such as LPWA.

Next, I will explain the experiments conducted by the inventor.
Figure 12 shows an example of two GNSS antennas A1 and B1 being installed at a predetermined distance on the roof of a two-story building for the purpose of conducting an experiment. As shown in Figure 12, two GNSS antennas A1 and B1 were installed on the roof of a two-story building at a distance of 1.6 m, and these antennas A1 and B1 (see Figure 12) were connected to two self-made GNSS receivers (not shown).

The system then received signals from six GNSS satellites every five minutes. At five-minute intervals according to GNSS time, the carrier phase from each satellite was sampled to obtain the instantaneous value of the carrier phase (instantaneous phase), which was then transmitted to a personal computer (PC). On the PC, the double phase difference and difference direction vector were calculated for the instantaneous phase of each GNSS satellite. The relative position (R, α, β) that best matched the calculated double phase difference was then searched for and determined. Here, R is the relative distance between antenna A1 and antenna B1, α is the horizontal azimuth angle with respect to the north-south line L1 (see Figure 12(b)), and β is the vertical angle β with respect to the horizontal plane L2 (see Figure 12(a)). Here, R = 1.6 m (measured value using a tape measure), α = 19 degrees (angle determined from the map), and β = 0 degrees (see Figure 12). The diagrams illustrating the double phase difference for each GNSS satellite are omitted.

Figure 13 shows the measurement results of the relative distance R obtained from the experiment. Here, the results are shown from nine measurements taken at 5-minute intervals from 18:00 (18:00) to 18:40 (18:40) on October 1, 2021. For these measurements, the relative distance R was assumed to be in the range of 1m to 2m, the horizontal azimuth angle α was assumed to be in the range of ±30 degrees, and the vertical angle β was assumed to be in the range of ±15 degrees. The search was conducted within these ranges.

According to the measurement results of the relative distance R shown in Figure 13, the average value from 18:00 to 18:40, measured every 5 minutes, was 1.591 m, and the maximum error from this average value was 1.02 cm at 18:05. This result shows a high degree of agreement with the measured value of 1.6 m.

Figure 14 shows the measurement results of the horizontal azimuth angle α with respect to the north-south line L1 and the vertical angle β with respect to the horizontal plane L2, obtained through the experiment. As shown in Figure 13, the average horizontal azimuth angle α with respect to the north-south line L1 was 20 degrees, measured every 5 minutes from 18:00 to 18:40. This result shows a high degree of agreement with the measured value of 19 degrees for the horizontal azimuth angle α with respect to the north-south line L1. On the other hand, the average vertical angle β with respect to the horizontal plane L, although with some error, was 2.5 degrees, measured every 5 minutes from 18:00 to 18:40. This result shows a relatively high degree of agreement with the measured value of 0 degrees for the vertical angle β with respect to the horizontal plane L2.

The experimental results above confirm that relative position detection using instantaneous carrier phase values is feasible. By performing relative position detection using instantaneous carrier phase values, it becomes possible to transmit these values using wireless transmission systems such as LPWA. Therefore, high-precision position detection using carrier phase becomes possible with extremely low power consumption, and it may even be possible to operate it using batteries. Consequently, it becomes possible to install a position information transmission device in mountainous areas where conventional mobile phones have no signal, to detect landslides in those areas, while also enabling long-term operation on batteries and achieving positional accuracy of several centimeters.

Furthermore, the present invention is not limited to the embodiments described above, and various modifications can be implemented without departing from the spirit of the invention. For example, the following modifications are also possible.

(1) In the above-described embodiment, it was explained that the carrier frequency transmitted from the GNSS satellite is approximately 1.5 GHz. However, since there are GNSS satellites that use carrier frequencies other than 1.5 GHz, such as 1.2 GHz, applying the present invention using two types of carrier frequencies, 1.5 GHz and 1.2 GHz, will enable even higher-precision measurements.

(2) In the embodiments described above, the example of installing a GNSS receiver 4 was given as the satellite information acquisition means 4 for acquiring satellite orbit information (ephemeris). However, satellite orbit information (ephemeris) can also be acquired, for example, from a web server on the internet. Therefore, to acquire satellite orbit information (ephemeris), not only a GNSS receiver 4 but also satellite information acquisition means including a web server connected to the internet can be used. For this reason, "satellite information acquisition means 4" includes a GNSS receiver and a web server connected to the internet.

1...Relative position detection system, 2 (2A, 2B)...Position information transmission device (measurement point), 3 (3A, 3B, 3C)...GNSS satellite, 4...GNSS receiver (satellite information acquisition means), 5...Receiving base station, 6...Computation server (computation means), 7...Smartphone, 20...Receiving antenna, 21...Front end, 22...Satellite signal processing unit, 23...Computation unit, 25...Latch (sampling means), 26...Wireless transmission unit, 27...LPWA transmitting antenna, 221...Satellite signal receiving unit, φ1, φ2, ..., φM...Carrier phase, φ(1), φ(2), ..., φ(M)...Instantaneous value of carrier phase (instantaneous phase)

Claims (9)

Calculation means and
A receiving base station connected to the calculation means via a communication line,
Multiple position information transmitting devices each acquire instantaneous values of carrier phases of radio waves transmitted from multiple GNSS (Global Navigation Satellite System) satellites at a predetermined sampling timing synchronized with GNSS time, and transmit the acquired instantaneous values of carrier phases to the receiving base station wirelessly, thereby transmitting the instantaneous values of carrier phases to the calculation means.
The system includes a satellite information acquisition means connected to the calculation means via a communication line, which acquires satellite orbit information of a GNSS satellite and transmits the acquired satellite orbit information to the calculation means,
The calculation means is a relative position detection system characterized by detecting the relative positions of the plurality of position information transmitting devices using the instantaneous value of the carrier phase and the satellite orbit information,
The relative positions of multiple measurement points are,
It is represented by a three-dimensional relative distance vector P calculated using the GNSS antenna of the reference measurement point among the two measurement points in the aforementioned plurality of measurement points as the starting point and the GNSS antenna of the other measurement point among the two measurement points as the ending point.
The aforementioned calculation means
As an estimated value of the relative distance vector, a three-dimensional estimated relative distance vector Re is estimated,
The wave vector k(n) is calculated using the reference satellite position information (X(0), Y(0), Z(0)) of a reference satellite among multiple GNSS satellites, the position information (X(n), Y(n), Z(n)) of each satellite among the multiple GNSS satellites other than the reference satellite, and the position coordinates (Xr, Yr, Zr) of the reference measurement point.
Based on the dot product calculation between the wave vector k(n) and the estimated relative distance vector Re, the estimated double phase difference φ''(n) is obtained.
The double phase difference φ(n) is obtained from the instantaneous value of the carrier phase.
An error evaluation value E is calculated based on the difference between the aforementioned double phase difference φ(n) and the estimated double phase difference φ''(n).
The estimated relative distance vector that minimizes the error evaluation value E is determined, and the determined estimated relative distance vector is output as the relative distance vector P.
Relative position detection system. The relative position detection system according to claim 1, characterized in that the transmission rate from the plurality of position information transmitting devices to the receiving base station is slower than the transmission rate from the satellite information acquisition means to the calculation means. The relative position detection system according to claim 1 or 2 , characterized in that the wireless transmission from the plurality of position information transmitting devices to the receiving base station is performed by LPWA (Low Power Wide Area) wireless communication means that is capable of long-distance communication and has low power consumption. The relative position detection system according to any one of claims 1 to 3 , characterized in that the interval between the predetermined sampling timings is set to one minute or more. At multiple measurement points, instantaneous values of the carrier phase of radio waves transmitted from multiple GNSS satellites are acquired at a predetermined sampling timing synchronized with GNSS time, and the acquired instantaneous values of the carrier phase are transmitted to a calculation means by communication means including at least wireless transmission.
The system acquires satellite orbit information of multiple GNSS satellites and transmits the acquired satellite orbit information to the calculation means.
The calculation means is characterized by detecting the relative positions between the plurality of measurement points using the instantaneous value of the carrier phase and the satellite orbit information,
The relative positions of the aforementioned multiple measurement points are,
It is represented by a three-dimensional relative distance vector P calculated using the GNSS antenna of the reference measurement point among the two measurement points in the aforementioned plurality of measurement points as the starting point and the GNSS antenna of the other measurement point among the two measurement points as the ending point.
The aforementioned calculation means is
As an estimated value of the relative distance vector, a three-dimensional estimated relative distance vector Re is estimated,
The wave vector k(n) is calculated using the reference satellite position information (X(0), Y(0), Z(0)) of the reference satellite among the multiple GNSS satellites, the position information (X(n), Y(n), Z(n)) of each satellite among the multiple GNSS satellites other than the reference satellite, and the position coordinates (Xr, Yr, Zr) of the reference measurement point.
Based on the dot product calculation between the wave vector k(n) and the estimated relative distance vector Re, the estimated double phase difference φ''(n) is obtained.
The double phase difference φ(n) is obtained from the instantaneous value of the carrier phase.
An error evaluation value E is calculated based on the difference between the aforementioned double phase difference φ(n) and the estimated double phase difference φ''(n).
The estimated relative distance vector Re that minimizes the error evaluation value E is determined, and the determined estimated relative distance vector is output as the relative distance vector P.
A method for detecting relative position. The relative position detection method according to claim 5 , characterized in that the transmission rate of the wireless transmission is slower than the transmission rate of the satellite information. The relative position detection method according to claim 5 or 6 , characterized in that the wireless transmission uses LPWA (Low Power Wide Area) wireless communication means capable of long-distance communication and low power consumption. The relative position detection method according to any one of 5 to 7 , characterized in that the interval of the predetermined sampling timing is one minute or more. The estimated double phase difference φ''(n) is,
Using the satellite orbit information of the plurality of GNSS satellites, the position information of the plurality of GNSS satellites at the predetermined sampling timing is calculated.
When one of the aforementioned multiple GNSS satellites is designated as the reference satellite, the reference satellite position coordinates (X(0), Y(0), Z(0)) are extracted from the position information of the multiple GNSS satellites, and the position coordinates (X(n), Y(n), Z(n)) of each satellite other than the reference satellite are extracted from the position information of the GNSS satellites.
Determine the position coordinates (Xr, Yr, Zr) of the aforementioned reference measurement point.
The wave vector k(n) is calculated using the reference satellite position coordinates (X(0), Y(0), Z(0)), the position coordinates of each satellite n (X(n), Y(n), Z(n)), and the position coordinates of the reference measurement point (Xr, Yr, Zr).
The relative position detection method according to claim 5, characterized in that an estimated double phase difference φ'(n) is obtained by performing an inner product operation between the wave vector k(n) and the estimated relative distance vector Re , and the relative position is determined by applying a modulo operation modulo 2π to the estimated double phase difference φ'(n).