JP2017129458A - Position specification system and position specification method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To specify a position accurately even under circumstances in which a GNSS satellite is hardly captured.SOLUTION: With a position specification system, a plurality of master nodes 1 whose position is specified and a plurality of slave nodes 2 whose position is not specified can communicate with each other. Each slave node 2 obtains distance information from each master node 1 based on a transmission/receipt time of a time synchronization packet of a PTP exchanged with each master node 1 to transit to the master node 1 by specifying own positional information from the obtained distance information with reference to positional information of the master node 1 that is a transmission/reception partner of the time synchronization packet.SELECTED DRAWING: Figure 3

Description

  The present invention relates to a technology for a position specifying system and a position specifying method.

In order to specify the position of the own apparatus by car navigation or the like, GNSS (Global Navigation Satellite System) such as GPS (Global Positioning System) is widely used.
FIG. 10 shows each node that receives radio waves from the GNSS satellite.
Each node A, B, C specifies its own position based on the information received from the GNSS. When the clock time of the GNSS satellite and the time of the node are synchronized (matched), the node may acquire three or more GNSS satellites. However, if the three satellites cannot be captured simultaneously due to obstacles, bad weather, etc., as in node Z, or if GNSS radio waves cannot be received indoors, the positional relationship between the node and the GNSS satellite is changed. It cannot be determined accurately.

  FIG. 11 identifies the position of the node T from the arrival times (ToA: Time of Arrival) of radio waves received from three points A, B, and C whose positions are known, such as three GNSS satellites and three base stations. The calculation method is shown. Based on the coordinates (X, Y, Z) of each three points and the distance L between each base station and node T calculated based on the arrival time of the radio wave received by each node T from each base station, The position (x, y, z) of the node T can be obtained by solving the three simultaneous equations shown in 1).

Here, the accuracy of the position of the node T depends on the accuracy of the distance L, and the distance L is obtained from the arrival time of the radio wave. Therefore, the greater the difference between the clock time of the base station and the clock time of the node T, the more errors are included in the arrival time of the radio wave, and the position accuracy of the node T also deteriorates.
Therefore, before communicating with the base station, the node T needs to be notified of the correct time from a server that provides the correct time and synchronize (correct) its own clock time. As this time synchronization protocol, for example, PTP (Precision Time Protocol) described in Non-Patent Document 1 has been proposed.

IEEE (The Institute of Electrical and Electronics Engineers, Inc), "IEEE Standard for a Precision Clock Synchronization Protocol for Networked Measurement and Control Systems", IEEE Std 1588-2008 (Revision of IEEE Std 1588-2002)

  There are also nodes located in regions where the weather is unsatisfactory as shown in FIG. 10 and nodes located in regions where it is difficult to capture GNSS satellites, such as high-rise buildings and underground malls. As shown in FIG. 11, unless three GNSS satellites are acquired, simultaneous equations cannot be solved and the position of the node T cannot be specified. However, there is a need to specify the position of these nodes with high accuracy. Further, there is a need to specify the position with high accuracy even for an inexpensive terminal that does not include a GNSS antenna.

  Therefore, the main object of the present invention is to specify a position with high accuracy even under a situation where it is difficult to acquire a GNSS satellite.

In order to solve the above-mentioned problem, the positioning system of the present invention is:
A plurality of master nodes whose own positions are specified can communicate with a plurality of slave nodes whose own positions are not specified,
Each slave node obtains distance information from each master node based on the transmission / reception time of the time synchronization packet exchanged with each master node, and the transmission / reception partner of the time synchronization packet Based on the position information of the master node that is, the transition to the master node by specifying its position information from the obtained distance information,
Each master node notifies each slave node of its clock time and its position information in the time synchronization packet transmitted to each of the slave nodes.

  As a result, the slave node is gradually propagated to the master node thereafter (propagated), so that the position can be specified in a wide range even in an area where it is difficult to capture a GNSS satellite such as a high-rise building group or an underground shopping center.

  The present invention is characterized in that the time synchronization packet is a PTP (Precision Time Protocol) packet.

  As a result, the position can be specified over a wide range using the PTP mechanism.

  In the present invention, each slave node calculates a deviation of its own clock time with respect to the clock time of the master node based on the transmission / reception time of the time synchronization packet exchanged with the master node, The present invention is characterized in that its own clock time is corrected so as to be adjusted to the clock time of the master node without the deviation.

  Thereby, even if a server that provides the correct time is not separately prepared, the correct clock time is provided from the master node, so that the time and cost for preparing the time synchronization server can be reduced.

  The present invention is characterized in that each slave node omits the process of correcting its own clock time again during a predetermined period after correcting its own clock time.

  As a result, in a period in which it is not necessary to consider the clock time difference between the master node and the slave node, the distance between the nodes can be easily reduced by only one-way packet communication without reciprocating the time synchronization packet. Can be sought.

In the present invention, the location system further includes a management device capable of communicating with the master node and the slave node,
The management device receives notification of the transmission / reception time of the time synchronization packet from the master node and the slave node, calculates the distance information between the master node and the slave node, and the position of the slave node. The calculation of information is performed instead of the slave node.

  As a result, each node does not have to perform calculation of distance information and position information by itself, so that a low-performance terminal such as a portable terminal can also be used as a node.

  In the present invention, the slave node tries to send / receive the time synchronization packet to / from the specific master node a plurality of times, and based on the trial with the shortest transmission / reception time among the trial results, The distance information with respect to the master node is obtained.

  Thereby, even when the communication situation fluctuates due to radio noise or the like, the position information can be calculated with high accuracy.

In the present invention, the slave node is:
In the process of selecting three master nodes from the set of communicable master nodes to form a node group, a plurality of node groups with different combinations of the master nodes belonging to the node group are formed,
Determining its own position information from the overlapping range of its own position information range for each of the node groups determined based on the position information of the master node belonging to each of the node groups;
When the distance information with respect to the master node belonging to each node group is closer to its own slave node, the reliability of each node group is calculated to be higher, and its position information is determined from the overlapping range. Priority is given to the range of position information obtained from the node group with high reliability.

  As a result, a slave node that can communicate with four or more master nodes can determine its own location information with higher accuracy by effectively utilizing the combination of the master nodes of the communication counterparts. The position information calculation accuracy can be improved by preferentially determining a master node close to and determining its own position information.

  According to the present invention, it is possible to specify the position with high accuracy even under a situation where it is difficult to acquire the GNSS satellite.

FIG. 1A shows a stage in which nodes A to C that have received radio waves directly from a GNSS satellite become base stations. FIG. 1B shows a stage where nodes D and E that have received radio waves from nodes A to C of the base station become new base stations. FIG. 2A shows a stage where the node F that receives radio waves from the nodes A, B, and D of the base station becomes a new base station. FIG. 2B shows a stage where the node G that has received radio waves from the nodes B, D, and F of the base station becomes a new base station. It is a lineblock diagram of the position specific system concerning this embodiment. It is a block diagram which shows a different form from FIG. 3 of the position specification system concerning this embodiment. It is a flowchart which shows the position specific process which paid its attention to the slave node concerning this embodiment. It is explanatory drawing which shows the detail of the calculation content of the time synchronization part and coordinate calculation part concerning this embodiment. FIG. 7A shows an example of obtaining the slave position range R1. FIG. 7B shows an example of obtaining the slave position range R2. FIG. 8A shows an example in which the slave position range R3 is obtained. FIG. 8B shows an example of obtaining the position F of the slave. FIG. 9A shows that the position range R3 in FIG. 8A is narrowed down by three trials (R3a to R3c). FIG. 9B shows an example in which the reliability of the node group is referred to when the position of the slave node F is specified from the overlapping region Rz in the range R1 to R3 in FIG. 8B. Each node which receives an electromagnetic wave from the GNSS satellite concerning this embodiment is shown. A calculation method for specifying the position of the node T from the arrival times (ToA: Time of Arrival) of radio waves received from the three GNSS satellites A, B, and C according to this embodiment will be described.

  Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings.

FIG. 1A shows a stage in which nodes A to C that have received radio waves directly from a GNSS satellite become base stations.
First, two types of nodes are defined as follows for a node that is a computer having a communication function.
Nodes A, B, and C are indicated by solid circles. These nodes have already been identified, and by notifying other nodes of their location information, they serve as base stations (reference points) for assisting the location of other nodes. Master node 1 (details are shown in FIG. 3).
Nodes D, E, F, G, and H are indicated by broken-line circles. These nodes are slave nodes 2 (details are shown in FIG. 3) that have not been identified yet, and that identify their positions with the assistance of the master node 1.

  Nodes A, B, and C specify their positions by receiving radio waves directly from GNSS satellites using their GNSS antennas 23 (FIG. 3) (in other words, by becoming primary nodes). If one node can receive radio waves from three or more GNSS satellites, the three-dimensional position (x, y, z) of the node can be specified. On the other hand, the other nodes D, E, F, G, H do not have the GNSS antenna 23, or even if they have the GNSS antenna 23, the radio waves from the GNSS satellite are shielded. At present, the position of the user is not specified.

FIG. 1B shows a stage where nodes D and E that have received radio waves from nodes A to C of the base station become new base stations.
The node D receives its radio waves from the nodes A to C of the base station, thereby specifying its own position. Similarly, the node E also specifies its own position by receiving radio waves from the nodes A to C of the base station. That is, nodes D and E are secondary nodes for primary nodes A to C.

Here, two types of expressions indicating node position information will be described.
“Global coordinates” refers to coordinates that can uniquely identify a position on the earth by a satellite. For example, the global coordinates of Tokyo Tower are expressed as “longitude = 139 degrees 44 minutes 43.558 seconds, north latitude = 35 degrees 39 minutes 30.89 seconds”.
“Local coordinates” is a coordinate system arbitrarily set by the measurer, and indicates relative coordinates viewed from the origin. For example, the local coordinates of Tokyo Station are expressed as “1800m to the east and 2500m to the north as seen from Tokyo Tower”.

When the global coordinate of the origin A with respect to the local coordinate of the point B is known, the global coordinate of the point B can be obtained by calculating the difference from the global coordinate of the origin A.
For example, in the vicinity of north latitude 35 degrees, longitude is 25 m per second and latitude is 31 m per second, so Tokyo Tower's "longitude = 139 degrees 44 minutes 43.558 seconds, north latitude = 35 degrees 39 minutes 30.89 seconds" and the difference "longitude = By adding "about 82 seconds, north latitude = + 74 seconds", the global coordinates of Tokyo Station "longitude = 139 degrees 45 minutes 57.902 seconds, north latitude = 35 degrees 40 minutes 52.975 seconds" are obtained.

  In FIG. 1B, since the global coordinates of the nodes A to C of the base station are known, the position of the node D is relatively determined as the local coordinates from the global coordinates, and then the local coordinates of the node D are obtained. The global coordinate of the node D is obtained by performing coordinate conversion to the global coordinate. As a result, the node D moves from the slave node 2 to the master node 1 and operates as a base station in the same manner as the other nodes A to C. Similarly, the node E shifts to the master node 1.

  FIG. 2A shows a stage where the node F that receives radio waves from the nodes A, B, and D of the base station becomes a new base station. Unlike node D, node F is located far from node C, so it is difficult to receive radio waves from node C. Therefore, it is possible to move from the slave node 2 to the master node 1 with assistance from a total of three nodes including the primary nodes A and B and the secondary node D located near the node F. That is, the node F moves to the tertiary node.

FIG. 2B shows a stage where the node G that has received radio waves from the nodes B, D, and F of the base station becomes a new base station. Similarly to the node F, the node G does not distinguish between the primary node and the secondary node, and receives the assistance from the nodes B, D, and F that are base stations in the vicinity of the node G, so that the slave node 2 To the master node 1.
If the distance from the node G is substantially the same, it is desirable to receive assistance from a node having a small n-th order n. This is because errors when converting local coordinates to global coordinates are accumulated as the order n increases.
As described above, by performing the transition from the slave node 2 to the master node 1 in a propagation manner from the primary nodes A to C, the position identification range is expanded even for nodes that cannot receive radio waves from the GNSS satellite. can do. That is, the primary node, the secondary node,..., The (n−1) -order node is the master node 1, and the slave node 2 of the n-order node is the master node 1.

FIG. 3 is a configuration diagram of the position specifying system.
As described with reference to FIGS. 1 and 2, the location specifying system is configured such that each node is configured to be able to communicate with other nodes in the vicinity in a wired or wireless manner, and its own location is known. Is classified into one of the slave nodes 2 whose position is unknown. Even if the same node is a slave node 2 at a predetermined time, it may be transferred (promoted) to the master node 1 with assistance from another master node 1.

The master node 1 and the slave node 2 are each configured as a computer having a CPU (Central Processing Unit), a memory, storage means (storage unit) such as a hard disk, and a network interface.
In this computer, the CPU executes a program (also referred to as an application or its abbreviated application) read on the memory, thereby operating a control unit (control means) configured by each processing unit described later.

The master node 1 includes a communication interface 10a, a transmitter 11a, a receiver 12a, a time synchronization unit 21a, a coordinate calculation unit 22a, and a storage unit 30a. Further, in the case of a primary node, the master node 1 also includes a GNSS antenna 23. doing. The storage unit 30a stores its own clock time 31a, its own local coordinates 32a, and its own global coordinates 33a.
The slave node 2 has basically the same configuration as the master node 1, but does not have the GNSS antenna 23. In order to distinguish between the components of the master node 1 and the slave node 2 even if they are the same component, the suffix of the component of the master node 1 is “a”, and the suffix of the component of the slave node 2 is Let it be “b”.

The communication interface 10a is connected to the transmitter 11a and the receiver 12a, and the communication interface 10b is connected to the transmitter 11b and the receiver 12b. A packet from the transmitter 11a reaches the receiver 12b, and a packet from the transmitter 11b reaches the receiver 12a.
The communication standard between the communication interfaces 10a and 10b is arbitrary, and may be wired communication or wireless communication. Furthermore, unicast communication between one master node 1 and one slave node 2 may be performed, or multicast transmission may be performed from one master node 1 to N slave nodes 2, or M Multicast transmission from one slave node 2 to the master node 1 may be performed.

  The time synchronization unit 21a of the master node 1 corrects (synchronizes) the clock time 31b of the slave node 2 by notifying the time synchronization unit 21b of the time stamp information of the clock time 31a. For this time synchronization processing, for example, a time synchronization protocol such as PTP is used (details are shown in FIG. 6). Furthermore, the time synchronizers 21a and 21b also use the time stamp information of the clock times 31a and 31b used for the time synchronization process to estimate the distance between the master node 1 and the slave node 2.

The coordinate calculation unit 22b calculates the coordinates of its own slave node 2 based on the local coordinates 32a and global coordinates 33a notified from the master node 1 and the time stamp information acquired by the time synchronization unit 21b. Therefore, the coordinate calculation unit 22b first calculates the local coordinates 32b of the slave node 2, and then calculates the difference between the local coordinates 32b with the global coordinates 33a as the origin, as described in the example of the Tokyo Tower. The global coordinate 33b of the node 2 is calculated. Note that the coordinate conversion calculation from the local coordinates 32b to the global coordinates 33b may be realized by any known calculation method such as using a coordinate conversion rule such as matrix calculation.
Thereby, the local coordinates 32b and the global coordinates 33b are associated with each other and stored in the storage unit 30b. Thereafter, since the slave node 2 behaves as the master node 1, the position determination processing of the other slave nodes 2 is performed with respect to the other slave nodes 2 based on the local coordinates 32a and the global coordinates 33a in the storage unit 30a. To assist.

FIG. 4 is a configuration diagram showing a form different from that of FIG. 3 of the position specifying system.
The management device 3 cuts out some functions of the master node 1 and the slave node 2 in FIG.
The coordinate calculator 22z obtains local coordinates 32a and 32b and global coordinates 33a and 33b of each node on behalf of the coordinate calculators 22a and 22b. Therefore, the coordinate calculation requesting units 22x and 22y of each node notify the coordinate calculation unit 22z of the time stamp information acquired by its own time synchronization units 21a and 21b, and requests the coordinate calculation of each node from the management apparatus 3. .
The coordinate calculation unit 22z receives the request, and based on the notified time stamp information, the local coordinates 32a of each node stored by itself, and the global coordinates 33a of each node stored by itself, the master node 1- The distance between the slave nodes 2 and the local coordinates 32b and global coordinates 33b of the slave nodes 2 are obtained.
The coordinate calculation unit 22z stores the global coordinates 33a and 33b as global coordinates 33z in its own storage unit, and stores the local coordinates 32a and 32b as local coordinates 32z in its own storage unit, so that the position of another node thereafter Also use for identification.

As a result, the management apparatus 3 can centrally manage the position information of each distributed node. For example, the position of each node is displayed on a map so that the administrator can grasp the distribution of the nodes. , Location information can be managed centrally. Furthermore, since it is not necessary for each node to perform coordinate calculation, a low-performance portable terminal can be used as a node.
Alternatively, each node of the position specifying system in FIG. 3 may perform unified management of position information by the management apparatus 3 by notifying the management apparatus 3 of its own position information calculated by itself.

FIG. 5 is a flowchart showing the position specifying process focusing on the slave node.
S11 to S14 are loops that sequentially select the master nodes 1 that can communicate with the slave node 2 one by one.
As S12, the time synchronization unit 21b performs time synchronization by transmitting and receiving a time synchronization packet such as PTP to and from the time synchronization unit 21a of the master node 1 selected in the loop. That is, the clock time 31b of the slave node 2 is set to the clock time 31a of the master node 1. In this time synchronization process, the time synchronization unit 21b acquires the following four time stamp information from the time synchronization packet.
(Time t1) Transmission time of a packet transmitted from the master node 1 to the slave node 2 (time on the master node 1 side)
(Time t2) Reception time of a packet transmitted from the master node 1 to the slave node 2 (time on the slave node 2 side)
(Time t3) Transmission time of a packet transmitted from the slave node 2 to the master node 1 (time on the slave node 2 side)
(Time t4) Reception time of packet transmitted from slave node 2 to master node 1 (time on master node 1 side)
In addition, the slave node 2 can acquire the position information of each master node 1 from the global coordinates 33a included in the time synchronization packet transmitted by each master node 1.

  Note that the time synchronization process of S12 may be repeatedly executed (trial) on the same master node 1-slave node 2 (see FIG. 9A). Even if the transmission / reception nodes are the same, the arrival time (delay time) of each packet may fluctuate due to radio wave noise or the like in a plurality of time synchronization processes. Therefore, the minimum delay time with the least influence of noise may be adopted among the arrival times of the packets measured by repeated execution.

Moreover, since the time synchronization process of S12 is in the loop of S11 to S14, it is executed many times in a short time. On the other hand, when the clock function in the slave node 2 has high performance, once the time is updated, the accurate time can be maintained for a while. Therefore, the update of the time may be omitted as a period in which the accurate time can be maintained in the predetermined period after the time is updated.
When time update is omitted, if there is one-way time information (a combination of t1 and t2 or a combination of t3 and t4) of the times t1 to t4, the master node 1 is connected to the slave node 2. The arrival time (delay time) of the packet can be obtained by a simple expression “reception time−transmission time”. That is, the clock-synchronized measurement packet can be shortened from the round-trip packet to the one-way packet.

  As S13, the coordinate calculation unit 22b measures the distance from its own slave node 2 to the master node 1 from the time stamp information at times t1 to t4 (details are shown in FIG. 6). As a result, the slave node 2 can determine the distance to each master node 1 that can communicate with itself, and this distance information is used for the position specifying process of the slave node 2 as described later in S21 to S24. .

  S21 to S24 is a loop that selects each combination of node groups in which three master nodes 1 are extracted from the set of master nodes 1 selected in S11 to S14. For example, there are 10 combinations of <A, B, C>, <A, B, D>, etc. in total for selecting 3 nodes from 5 nodes (A, B, C, D, E). , S21 to S24 are executed 10 times.

  As S22, the coordinate calculation unit 22b calculates the position range of the slave from the distance to each node belonging to the node group (the distance obtained in S13) (details are shown in FIGS. 7 to 9). Here, since the time stamp information of the times t1 to t4 and the global coordinates 33a of the master node 1 may include some errors, there is a possibility that the slave position is not a point coordinate but a slave. It becomes a position range.

As S23, the coordinate calculation unit 22b calculates the reliability of the node group selected in the current loop. Therefore, first, the individual reliability of each of the three master nodes 1 belonging to the node group is obtained. Basically, as the distance from the master node 1 determined in S13 is closer to the slave node 2, the individual reliability is calculated to be higher. The following is an example of correspondence between distance and individual reliability.
If 0m ≦ distance <100m, individual reliability = 3
If 100m ≦ distance <500m, individual reliability = 2
If 500m ≦ distance, individual reliability = 1
Next, the group reliability is calculated by integrating the individual reliability. For example, the example which makes the total score of individual reliability the group reliability is mentioned.

As S31, the coordinate calculation unit 22b specifies the local coordinate 32b of the slave position from the overlapping area of the slave position range in each node group obtained in S22. Alternatively, the slave positions may be narrowed down from an area where more (majority) slave position ranges overlap, instead of overlapping areas of all slave position ranges. Here, the coordinate calculation unit 22b may narrow down the slave position from the overlapping region of the slave position range with reference to the group reliability obtained in S23 (see FIG. 9B).
As described above, by combining a plurality of node groups and narrowing down the slave position, the accuracy of the slave position is improved.

  In S32, the coordinate calculation unit 22b converts the local coordinate 32b of the specified slave position into a global position (global coordinate 33b). The local coordinates 32b of S31 and the global coordinates 33b of S32 are respectively stored in the storage unit 30b as position information of the slave node 2 itself.

FIG. 6 is an explanatory diagram showing details of calculation contents of the time synchronization unit 21b and the coordinate calculation unit 22b.
The node T that is the slave node 2 synchronizes the time with the nodes A, B, and C that are the master nodes 1 by the PTP protocol. Details of PTP are described in Non-Patent Document 1.
The slave node 2 obtains four pieces of time stamp information (time t1 to t4) by reciprocating the PTP time synchronization packet between the master node 1 and the slave node 2 as described in S12. Based on the time stamp information, the time synchronization unit 21b obtains the accurate clock time 31b and the distance from the master node 1 according to the equation (2) in FIG.

Hereinafter, exemplified packet P _AT sent to the node A → node T, in the case of packet P _TA sent to the node T → node A.
First, the transmission time t1 on the node A side of the packet P _AT and the reception time t2 on the node T side of the packet have the following relationship.
(Time t1) + (Delay time taken for transmission) + (Offset that is a time lag on the slave node 2 side) = (Time t2)
Similarly, the transmission time t3 at node T side for the packet P _TA, the reception time t4 at the node A side of the packet is the next relationship.
(Time t3) + (Delay time taken for transmission) − (Offset that is a time lag on the slave node 2 side) = (Time t4)
Therefore, as in the first line of Equation (2), the delay time ΔT _A required for transmission and the offset β _A which is a time lag are obtained (S12). Here, T _AT is the time difference (t2-t1) of the packet P _AT , and T _TA is the time difference (t4-t3) of the packet P _TA . Note that the slave node 2 may correct its own clock time so as to match the clock time of the master node 1 by eliminating the offset β _A (S12).

Further, the propagation velocity (speed of light) c, by using the refractive index n of the medium, the node A, the distance L _A between T can also be determined (S13). Further, δ _A is a measurement error such as a time synchronization error, and α is a measurement error of the receiver (slave node 2).
The third to fifth lines of Equation (2) are simultaneous equations that are expanded when the right side of Equation (1) in FIG. 11 includes an error. By obtaining these simultaneous equations, the position (x, y, z) of the node T is obtained as the local coordinates 32b (S22).

Therefore, if the measurement error α of the receiver (slave node 2) (for example, the measurement error peculiar to the device) is known, the slave node 2 sets the offset β _A to the equation (2) in FIG. 6 every time PTP is executed. By substituting, the distance L _A can be obtained with high accuracy.
When performing time synchronization and delay (distance) measurement at the same time in one PTP execution,
Time synchronization processing: Offset β _A that is a time lag is corrected.
Delay measurement processing: _A more accurate delay time is obtained by correcting a measurement error (adding or subtracting offset β _A ) to time stamp information (time t1 to t4).
By performing these two processes in parallel, since there is no fear of time lag compared to when delay measurement is performed after a predetermined period of time has elapsed since the time synchronization process, more accurate measurement can be performed.

The details of the processing for each node group (S21 to S24) will be described below with reference to FIGS.
First, it is assumed that the first node group “B, C, E” is selected in the first loop. The coordinate calculation unit 22b solves the simultaneous equations described in FIG. 6 based on the distance information from each node in addition to the global coordinates 33a of each node B, C, E belonging to the group, thereby obtaining the position of the slave. A range R1 is obtained (FIG. 7A).
Next, when the second node group “A, C, E” is selected in the second loop, the slave position range R2 is also obtained in the same manner as the first node group (FIG. 7B). Here, since both the slave position ranges R1 and R2 include some errors, the two ranges do not completely coincide but partially overlap.

Further, when the third node group “A, B, D” is selected in the third loop, the position range R3 of the slave is also obtained in the same way as the first and second node groups (FIG. 8A).
Since the overlapping ranges of these slave position ranges R1 to R3 have narrowed considerably, the overlapping area of each of the position ranges R1 to R3 is specified as the position of the slave node F as shown in S31 (FIG. 8B). ).

FIG. 9A shows that the position range R3 in FIG. 8A is narrowed down by three trials (R3a to R3c). As described in S12, the PTP time synchronization process may be repeatedly executed between the same master node 1 and slave node 2.
Here, the slave position range R3a calculated based on the result of the first time synchronization process (for example, the distance from the node B), and the slave calculated based on the result of the second time synchronization process. The position range R3b is narrow. Therefore, since the second time synchronization process is measured with a shorter distance (higher accuracy), the range R3b is narrower than the range R3a and the distance from the node B is shorter.
Similarly, since the third range R3c is more accurate than the second range R3b, the range R3c is adopted as the position range R3 of the third node group “A, B, D”.

FIG. 9B shows an example in which the reliability of the node group is referred to when the position of the slave node F is specified from the overlapping region Rz in the range R1 to R3 in FIG. 8B.
As shown in S31, the group reliability for each node group calculated in S23 can be used to narrow down the point positions of the slave nodes F in the overlapping region Rz.
For example, the point position F1 before reflecting the group reliability is the center of gravity of the overlapping region Rz. On the other hand, the point position F2 after reflecting the group reliability is corrected so as to be closer to the center C3 of the region R3 and the center C1 side of the region R1 than the point position F1. This is because the reliability of the region R3 = 8 and the reliability of the region R1 = 6 are greater (more reliable) than the reliability of the region R2 = 4. That is, the method illustrated in FIG. 9B is a method of determining the point position of the slave node F so that the region with higher reliability is closer to the region.

  In the present embodiment described above, even in the case of the node Z in FIG. 10 in which the GNSS satellite cannot be directly captured due to obstacles, bad weather, etc., the information on the inter-node distance obtained by the time synchronization calculation with the surrounding master node 1 Can be used to specify the position of the slave node 2 with high accuracy. Further, as explained in FIG. 1 and FIG. 2, by gradually shifting (propagating) the slave node 2 of the n-th node (n> 1) to the master node 1, Even in areas where it is difficult to capture GNSS satellites, it is possible to locate a wide range.

  Further, even the inexpensive terminal that does not include the GNSS antenna 23 like the slave node 2 in FIG. 3 can identify its own position with high accuracy. Furthermore, with the configuration in which the management device 3 in FIG. 4 performs the position calculation of the master node 1 and the slave node 2, a low-performance node terminal can be used for position identification.

DESCRIPTION OF SYMBOLS 1 Master node 2 Slave node 3 Management apparatus 10a, 10b Communication interface 11a, 11b Transmitter 12a, 12b Receiver 21a, 21b Time synchronization part 22a, 22b, 22z Coordinate calculation part 22x Coordinate calculation request part 22z Coordinate calculation part 23 GNSS antenna 30a, 30b Storage unit 31a, 31b Clock time 32a, 32b, 32z Local coordinates 33a, 33b, 33z Global coordinates

Claims (8)

A plurality of master nodes whose own positions are specified can communicate with a plurality of slave nodes whose own positions are not specified,
Each of the slave nodes obtains distance information from each of the master nodes based on the transmission / reception time of the time synchronization packet exchanged with each of the master nodes, and transmits / receives the time synchronization packet. Based on the position information of the master node that is, the transition to the master node by specifying its position information from the obtained distance information,
Each of the master nodes notifies each of the slave nodes of its clock time and its own position information in the time synchronization packet transmitted to each of the slave nodes.   The position identifying system according to claim 1, wherein the time synchronization packet is a PTP (Precision Time Protocol) packet. Each of the slave nodes calculates a deviation of its own clock time from the clock time of the master node based on the transmission / reception time of the time synchronization packet exchanged with the master node, and eliminates the deviation. The position specifying system according to claim 1, wherein the clock time of the master node is corrected to match the clock time of the master node. The location system further comprises a management device capable of communicating with the master node and the slave node,
The management device receives notification of the transmission / reception time of the time synchronization packet from the master node and the slave node, calculates the distance information between the master node and the slave node, and positions of the slave nodes. The position specifying system according to any one of claims 1 to 3, wherein information calculation is executed instead of the slave node. The slave node makes a plurality of attempts to send and receive the time synchronization packet to and from the specific master node, and based on the trial with the shortest transmission and reception time among the trial results, The position specifying system according to any one of claims 1 to 4, wherein the distance information between the two is obtained. The slave node is
In the process of selecting three master nodes from the set of communicable master nodes to form a node group, a plurality of node groups with different combinations of the master nodes belonging to the node group are formed,
Determining its own position information from the overlapping range of its own position information range for each of the node groups determined based on the position information of the master node belonging to each of the node groups;
When the distance information with respect to the master node belonging to each node group is closer to its own slave node, the reliability of each node group is calculated to be higher, and its position information is determined from the overlapping range. The position specifying system according to any one of claims 1 to 5, wherein priority is given to a range of position information obtained from the node group having high reliability. A plurality of master nodes whose own positions are specified and a plurality of slave nodes whose own positions are not specified are executed by a position determination system configured to be able to communicate,
Each of the slave nodes obtains distance information from each of the master nodes based on the transmission / reception time of the time synchronization packet exchanged with each of the master nodes, and transmits / receives the time synchronization packet. Based on the position information of the master node that is, the transition to the master node by specifying its position information from the obtained distance information,
Each of the master nodes notifies each of the slave nodes of its clock time and its position information in the time synchronization packet transmitted to each of the slave nodes.   The position specifying method according to claim 7, wherein the time synchronization packet is a PTP (Precision Time Protocol) packet.

Images