JP2013088348A - Navigation terminal and method for acquiring position of the same - Google Patents

Abstract

PROBLEM TO BE SOLVED: To allow a position of a terminal to be acquired by lowering facility costs and securing prescribed accuracy.SOLUTION: A landmark distance determination part 3 determines a distance between a terminal 1 and a landmark according to a positioning signal from the landmark LMk, and a representative position setting part 4 sets from map information a representative position as a candidate position of the terminal 1 according to a position which is on a circle equal to the determined distance around the landmark and on a passage section where the terminal 1 is movable. An autonomous navigation position estimation part 6 estimates a position of the terminal 1 which has moved from the landmark by autonomous navigation. A terminal position correction part 7 acquires the position of the terminal 1 by correction processing based on an estimated position and the representative position.

Description

  The present invention relates to a navigation terminal that acquires its own position using a positioning signal from a landmark, autonomous navigation, and map information in combination, and a method for acquiring the position.

  With the spread of location information services on mobile terminals using GPS (Global Positioning System), there is an increasing need for location acquisition methods that can be used indoors. As one of the methods suitable for measuring the current position indoors where GPS signals cannot be received, the TDOA (Time Difference of Arrival) system disclosed in Non-Patent Documents 1 to 3 is known. Yes.

  In the TDOA system, a plurality of landmark devices installed at appropriate intervals periodically transmit a distance measurement signal, that is, a positioning signal including a sound wave signal and a radio signal. At this time, each landmark device is controlled to transmit a positioning signal almost simultaneously, and the positioning signal includes information on the installation position of the landmark device that transmitted the wireless signal.

  Assuming that the terminal receives a radio signal from a certain landmark device at time t1 and a sound wave signal from this landmark device at time t2, the distance between the terminal and the landmark device is sound speed × (t2−t1). ) Can be calculated. Since the radio signal from the landmark device includes the position information of the landmark, the terminal can calculate its current position by measuring the distance to three or more landmark devices. it can.

  Further, as one of the current position measurement methods that can be used indoors, there is autonomous navigation (inertial navigation) disclosed in Non-Patent Documents 4 to 6 and the like. According to autonomous navigation, it is possible to estimate the position autonomously without the need for environmental facilities such as landmarks in the TDOA system. Autonomous navigation is a method in which a terminal equipped with a gyroscope, an acceleration sensor, and the like is used to cumulatively obtain a movement distance relative to a movement direction from a reference point.

Shinichi Minamimoto, Sae Fuji, Hirozumi Yamguchi and Teruo HIgashino, "Local Map Generation using Position and Communication History of Mobile node", IEEE percom 2010, p2-10, 2010. S. Venkatesh, and R.M.Buehrer, "NLOS Mitigation Using Linear Programming in Ultrawideband Location-aware Networks," IEEE Trans. On vehicular technology, Vol. 56, No. 5, Sep. 2007. Lui, KWK, So, HC, and Ma, W.-K. "Maximum A Posteriori Approach to Time-of-Arrival-Based Localization in Non-Line-of-Sight Environment," Vehicular Technology, IEEE Transactions on Vol.59 , Issue.3, pp. 1517-1523, 2010. Daisuke KAMISAKA, Shigeki MURAMATSU, Takeshi IWAMOTO and Hiroyuki YOKOYAMA, "Design and Implementation of Pedestrian Dead Reckoning System on a Mobile Phone", IEICE Transactions on Information and Systems, Vol.E94.No. 6, pp1137-1146, 2011. U. Steinhoff and B. Schiele, "Dead reckoning from the pocket- An experimental study," In Proceedings of PerCom 2010. Georg GARTNER & Andrew FRANK & Gunther RETSCHER, "Pedestrian Navigation System in Mixed Indoor / Outdoor Environment"

  In a TDOA system, when measuring a position while carrying a position detection device, the number of landmarks is usually 5 or more, and the current position is estimated using the 5 or more landmarks. The measured value corrected by the above is output. Therefore, there is a problem that the introduction barrier in terms of cost is large especially when trying to cover a large-scale indoor environment. That is, it is necessary to arrange the landmarks so that positioning signals can be received from five or more landmarks at all positions in the large-scale environment, and a large number of landmarks are required, which increases costs. End up.

  In addition, autonomous navigation has a problem that it is difficult to ensure accuracy when staying indoors for a long time. In other words, since autonomous navigation uses the integration of angular velocity and acceleration, it is suitable for navigation of airplanes, ships, automobiles, etc., but the accumulated error is large especially in the case of human walking with complicated movements. Cannot be ignored. Therefore, in order to apply it to human walking, it is necessary to impose severe restrictions on the method of holding the terminal and to improve the accuracy by capturing the characteristics of sensor data generated by walking, but still accumulation of errors can be avoided. Absent.

  The object of the present invention is to solve the above-mentioned problems of the prior art, and can be used indoors, has a low equipment introduction cost, and can obtain a proper position with a predetermined accuracy. And providing a method of acquiring the position of the terminal.

  In order to solve the above-described problems of the prior art, the present invention provides a navigation terminal that acquires its own position, receives a positioning signal from a landmark placed at a predetermined position, and the landmark, the navigation terminal, Map information holding for holding a landmark distance determining unit for determining the distance of the landmark, position information of the landmark, and map information including information on a passage section around the landmark where the navigation terminal can move An autonomous navigation position estimating unit that obtains an estimated position of the navigation terminal by applying autonomous navigation to the output from the sensor unit, an output from the sensor unit, the map information, and the landmark From each of the arcs obtained as an overlap of a circle with a radius equal to the determined distance and the passage section, the navigation system A representative position setting unit that sets a representative position of the terminal, and a terminal position correction unit that acquires the position of the navigation terminal based on each of the estimated position, the representative position, and the distance between the estimated position and each representative position It is characterized by providing.

  Further, in order to solve the above-described problems of the prior art, the present invention is a method for acquiring a position of a terminal, which receives a positioning signal transmitted by a landmark placed at a predetermined position, and the landmark and the terminal The terminal comprises: a step of determining a distance from the terminal; a step of acquiring map information including position information of the landmark and information of a passage section around the landmark that the terminal can move; and Applying autonomous navigation to the output from the sensor to obtain an estimated position of the terminal; referring to the map information; a circle having a radius equal to the determined distance centered on the landmark and the passage section Setting the representative position of the terminal from each of the arcs obtained as overlapping portions with the estimated position, each of the estimated position, the representative position, and the estimated position, Based on the distance between each of the table position, characterized in that it comprises the steps of obtaining a position of the terminal.

  According to the present invention, as a region determined from a positioning signal, a circle centering on a landmark that is determined to be located on the terminal with high accuracy, and a region where the terminal can move is determined from map information. Obtain multiple representative positions as candidate positions where the terminal exists from the overlap with the passage section, and obtain the position of the terminal by comparing the estimated position of the terminal by autonomous navigation where accuracy is a problem with the multiple representative positions To do.

  Therefore, the estimated position of autonomous navigation where accuracy is a problem is corrected by the representative position, and an appropriate terminal position can be acquired. In the correction, in consideration of the distance between the estimated position and the representative position, for example, a weight can be given by the representative position having the smaller distance. In addition, at the time of the acquisition, it is only necessary that the distance from the landmark can be used using the positioning signal of one landmark. Therefore, since the terminal does not need to use positioning signals from a plurality of landmarks at a time, the number of landmarks is reduced.

It is a functional block diagram of the navigation terminal which concerns on embodiment of this invention. It is a figure for demonstrating the schematic structure of the navigation system which concerns on embodiment of this invention. 5 is a flowchart of a terminal location acquisition method according to an embodiment of the present invention. It is a figure for demonstrating the data which each functional block of a terminal acquires. It is a flowchart of position correction etc. concerning the embodiment of the present invention. It is a figure for demonstrating the representative position which a position setting part sets. It is a figure for showing the example of calculation of the temporary position in case (1) in the flow of FIG. It is a figure for demonstrating the example which adds correction to the temporary position in the case of (1) in the flow of FIG. It is a figure for demonstrating one Example of determination of the temporary position in the case (2) in the flow of FIG. It is a figure for demonstrating the example which adds correction to the temporary position in the case of (2) in the flow of FIG. It is a figure for showing the example of arrangement of the landmark in the present invention, and the passage section around it.

  FIG. 1 is a functional block diagram of a navigation terminal that acquires its own position according to an embodiment of the present invention. The navigation terminal 1 (abbreviated as terminal 1) includes a landmark signal receiving unit 2, a landmark distance determining unit 3, a representative position setting unit 4, a sensor unit 5, an autonomous navigation position estimating unit 6, a terminal position correcting unit 7, and map information. A holding unit 8 is provided.

The landmark signal receiving unit 2 is a positioning signal (radio wave signal and sound wave signal) from any one of the landmarks LM _k (k = 1, 2, 3,...) Fixedly arranged at a predetermined reference position. Receive. The landmark distance determination unit 3 identifies the ID of the landmark that transmitted the signal from the received positioning signal, analyzes the received signal according to the TDOA method, and obtains the time required for the sound wave to reach. By multiplying the sound speed, the distance between the terminal 1 and the landmark LM _k is calculated.

The representative position setting unit 4 uses the map information held in the map information holding unit 8 so that the distance from the landmark LM _k satisfies the distance determined by the landmark distance determining unit 3. A plurality of representative positions are set as candidate positions where the terminal 1 can exist. This process will be described later.

  The sensor unit 5 includes various autonomous navigation sensors, for example, a triaxial acceleration sensor, a geomagnetic sensor, and a gyro sensor. The autonomous navigation position estimation unit 6 estimates the position of the terminal 1 by applying autonomous navigation to the output of the sensor unit 5 and obtains an estimated distance described later. The estimation is started at a predetermined timing, which will be described later, and the position where the terminal 1 has moved with the position at the start as a reference position is estimated by integration processing for the sensor output.

  As will be described later, the terminal position correction unit 7 performs correction using both the representative position set by the representative position setting unit 4 and the position estimated by the autonomous navigation position estimation unit 6, so that the terminal 1 as a final result is obtained. The position information is obtained and presented to the user by displaying the position information on the map together with the map information as necessary. For example, the correction is performed by assigning a weight to the closer representative position based on the distance between the estimated position and the representative position.

  At the time of presentation to the user, a destination input from the user is received in advance from an input unit (not shown) in the terminal 1, and a known route guidance technique is applied with the obtained position information of the terminal 1 as the current position. It may be presented to the user together with the route to the destination. The map information holding unit 8 holds map information so that it can be referred to as necessary.

FIG. 2 is a diagram schematically illustrating the configuration of the navigation system for the position of the terminal 1 according to the embodiment of the present invention. In the navigation system S1, a plurality of stores or the like represented by each landmark LM _k (here, k = 1, 2,..., 8) and rectangular hatched areas cannot use GPS or are difficult to use. It is placed in the target area such as indoors. The information such as the predetermined arrangement is described in the map information M1 held by the map information holding unit 8. The appropriate position of the terminal 1 that moves while drawing the trajectory T1 (the trajectory of the actual position of the terminal 1) within the target area thus configured is acquired by the present invention.

In the map information M1, the position information of each landmark LM _k , information on the passage as a region where the user and the terminal 1 can move, and the user and the terminal 1 move directly from the passage due to the presence of a wall or the like. And information on stores and the like as areas that cannot be used. That is, the map information M1 includes the shape and position information of each area in the target area and the information on whether or not the user and terminal 1 in each area can pass and move.

Each landmark LM _k has a predetermined range Ci in which the terminal 1 can receive the transmitted TDOA positioning signal (here, i = 10, as an example corresponding to k = 1, 2,..., 8 in LM _k , respectively). 20, ..., 80). In the present invention, signal reflection and interference are not considered, and each range Ci is a circular region having a predetermined radius determined by each landmark LM _k and the signal transmission / reception performance of the terminal 1.

The navigation system S1 is configured such that the above ranges Ci do not overlap each other, and the configuration is recorded in the map information M1. That is, in the navigation system S1, the terminal 1 can receive or cannot receive a positioning signal according to its position, and if it can, the terminal 1 always receives only from one of the landmarks LM _k . No more than two are received.

As described above, in the present invention, the positioning signal is compared to the case where the terminal needs to receive positioning signals from at least three landmarks in principle and five landmarks in consideration of accuracy in the conventional TDOA system. Only one can be used. Therefore, in the present invention, the number of landmarks LM _k when constructing the navigation system S1 in the area of the same area is reduced as compared with the prior art, and there is an effect of cost reduction.

  FIG. 3 shows a flowchart of a method for acquiring the position of the terminal 1 in the navigation system S1 according to the embodiment of the present invention. In order to describe FIG. 3, first, data acquired by each functional block constituting the terminal 1 will be described with reference to FIG. 4.

In FIG. 4, a certain landmark LM _k is arranged at the center in the figure. R _TDOA (t) shown in the figure is a distance between the LM _k determined by the landmark distance determination unit 3 and the terminal 1, is determined at each time t, and the value changes appropriately according to the movement of the terminal 1. .

The circle Ck1 is a circle whose center is at the position of the landmark LMk and whose length of the radius R1 is r _TDOA (t). Here, the following matters are assumed as the premise of the present invention. The accuracy of the distance determined by the landmark distance determination unit 3 is higher than the accuracy of the position estimated by the autonomous navigation position estimation unit 5 because no error accumulation occurs. The terminal 1 is positioned with sufficient accuracy at any point on the circle Ck1 determined by the landmark distance determination unit 3 determining the distance. In the description of the present invention, the circle Ck1 is particularly referred to as a “determined circle”.

The representative position setting unit 4 sets a predetermined representative position as described later using map information from the circumference of the determined circle Ck1 where the terminal 1 is determined to be located. In FIG. 4, four points P1, P2, P3 and P4 on the decision circle Ck1 are shown as examples of representative positions. The representative positions P1 to P4 are located on the determined circle Ck1 as the determined circle Ck1 changes in size according to the movement of the terminal 1 as time elapses with the radius value r _TDOA (t). Will move. The representative position is determined from the map information including the number of the representative positions.

The circle Ck10 is a circle whose center is at the position of the landmark LM _k and whose length of the radius R2 is r _LMk shown in the drawing. Here, the length r _LMk is the maximum length at which the terminal 1 can receive a positioning signal from the landmark LM _k . That is, the circle Ck10 is the same as the circles C10 to C80 shown for the landmarks LM1 to LM8 in FIG. Therefore, when it exists outside the circle Ck10, the terminal 1 cannot determine the distance r _TDOA (t) from the landmark LM1.

The trajectory T3 is a trajectory of the position of the terminal 1 estimated by the autonomous navigation position estimation unit 6. Further, the position on the movement trajectory at the time t, that is, the estimated position is P30. The distance d _est (t) is equal to the length of the line segment R3 connecting the estimated position P30 and the position of the landmark LM _k , and the autonomous navigation position estimation unit 6 obtains the distance as the estimated distance d _est (t).

  If the accuracy of the autonomous navigation position estimation unit 6 is sufficiently high as much as the landmark distance determination unit 3, the estimated position P30 represents the actual position of the terminal 1 with sufficient accuracy, and therefore, on the determination circle Ck1 Will exist. However, since the accuracy of autonomous navigation is limited, generally the estimated position P30 does not exist on the decision circle Ck1, but is also inside the decision circle Ck1 as shown in (a) (R3 <R1). , (B), when it is outside the decision circle Ck1 and inside the circle Ck10 (R1 <R3 <R2; this case is illustrated in FIG. 4), and also as shown in (c), the circle Ck10 It may be external (R3> R2). In the case of (c), the decision circle Ck1 is no longer obtained. In the present invention, using the estimated position P30 and the representative positions on the determination circle Ck1 (P1 to P4 in this example), a position with high validity is determined as the position of the terminal 1 according to the flow of FIG.

In FIG. 3, when the flow is started in step S0, the landmark distance determination unit 3 is activated in step S1, and reception of a positioning signal from any of the landmarks LM _k is started.

In step S2, whether the positioning signal from any of the landmarks LM _k can be received and the distance r _TDOA (t) between the terminal 1 and the landmark LMk determined by the landmark distance determining unit 3 is sufficiently small. That is, it is determined whether the terminal 1 is sufficiently close to the landmark LM _k . Note that whether or not they are sufficiently close may be determined by providing a predetermined threshold value for the distance r _TDOA (t).

  If it is determined in step S2 that the positioning signal cannot be received in the first place, it is determined that the signals are not sufficiently close, and step S2 is repeated again at the next time step. In this way, step S2 is continued until it is determined that the terminal 1 moves with the user and is sufficiently close.

If it is determined to be sufficiently close in Step S2, the position information and the map information of the landmark LM _k as an initial position information in step S3 terminal 1 acquires. Note that the map information is stored in the map information holding unit 8 by the terminal 1 acquiring predetermined information as shown as the map information M1 in FIG. 2 via the Internet or the like, and the landmark LM _k on the map information. Position information is given. At this time, the position information is given to the map information by referring to the ID information of the landmark LM _k included in the positioning signal. For the sake of explanation, the position information of the landmark LM _k is represented by coordinates (x _k , y _k ) (where the landmark identifier k = 1, 2, 3,...).

Next, in step S4, the autonomous navigation position estimation unit 6 is activated, and the position information (x _k , y _k ) acquired in step S3 is used as an initial value to start position estimation by autonomous navigation. For the sake of explanation, the estimated position at time t (position P30 in FIG. 4) is expressed as coordinates (x _est (t), y _est (t)). It is assumed that step S4 is executed immediately (substantially simultaneously) immediately after step S3. That is, when the time at which the steps S3 and S4 are executed is the initial time t = 0,
(x _est (0), y _est (0)) = (x _k , y _k )… [Formula 1]
It is.

In this way, with the position of the terminal 1 at the initial time t = 0, that is, the position of the landmark LM _k as a reference, the estimated values (x _est (t), y _est (t)) of the autonomous navigation at the subsequent time t are given. It is done. Further, as apparent from the setting, the estimated distance d _est (t) from the landmark LMk described in FIG. 4 is given as follows.
d _est (t) = sqrt {(x _est (t) −x _k ) ² + (y _est (t) −y _k ) ² } [Expression 2] (where sqrt {} represents a square root operation)

  After the autonomous navigation position estimation unit 6 starts position estimation in steps S3 and S4, the user moves with the terminal 1 in steps not shown. In step S5, it is determined whether there is an instruction to acquire the position of the terminal 1 from the user. If there is no instruction, step S5 is repeated until there is an instruction, whereby the terminal 1 enters a state of waiting for the instruction.

If there is a position acquisition instruction, the process proceeds to step S6, and the terminal position correction unit 7 determines whether or not the landmark signal receiving unit 2 can receive a positioning signal from the landmark when the instruction is received. Note that the landmark is the same landmark LM _k as determined in step S2.

If it is determined in step S6 that reception is possible, the process proceeds to step S7, where the terminal position correction unit 7 receives the instruction at the time t (x _est (t), y _est (t)) and the time t. As the position corrected based on the representative position set by the representative position setting unit 4, the position of the terminal 1 at the time t is obtained and presented to the user. Details will be described later.

If it is determined in step S6 that reception is not possible, the process proceeds to step S8, and the terminal position correction unit 7 determines the estimated position (x _est (t), y _est (t)) at the time point t when the instruction is received. The position is determined and presented to the user. Incidentally, if it is determined not to be received, also includes if the landmark LM _k determined in step S2 that receives positioning signals from another landmark LMl (k ≠ l).

Note that when the process proceeds to step S8, the determination circle at the center of the landmark LM _k cannot be obtained and the representative position cannot be used, and the correction (step S7) according to the present invention cannot be applied, so the estimated position (x _est (t ), y _est (t)) only. Therefore, in step S8, as another example, instead of the estimated position, information indicating that the position cannot be obtained may be presented to the user, or there is a high possibility that the error is large when presenting the estimated position. This information may also be provided to the user.

  As described above, when the process proceeds from step S6 to either step S7 or S8, the process proceeds to step S9, and the terminal 1 confirms whether there is an instruction to continue position acquisition from the user. If there is such an instruction, the process returns to step S5, and thereafter each step is repeated in the terminal 1 that moves with the user. If there is no such instruction, the flow ends in step S10. In step S9, instead of confirming the continuation instruction, the presence / absence of a position acquisition end instruction may be confirmed.

The supplementary items in the flow of FIG. 3 are as follows. That is, after it is determined in step S2 that a certain landmark LM _k is sufficiently close and the flow proceeds, the radius r _TDOA (t in the positioning signal of any landmark LM _j (including the case of j = k) ), At the moment when the same sufficiently close condition as in step S2 is satisfied, the process jumps back to step S2 by a flow (not shown). Then, it is assumed that the flow is continued from step S3 onward for the landmark LM _j .

By the jump processing, even when the terminal 1 moves from one landmark LM _k to another landmark LM _j , the estimated position (x _est (t), y _est (t) ) Is reset to a position based on the other landmark LM _j , and the flow is continued. Further, the estimated position with respect to the same landmarks _{LM k (x est (t)} , y est (t)) also when sufficiently close to the landmark LM _k, by being reset back to the initial value, also the accumulation of errors Reset.

  Further, steps S5 (confirmation of position acquisition instruction) and S9 (confirmation of continuation instruction) in the flow of FIG. 3 are merely an example when the user uses the terminal 1. Steps S5 and S9 may be skipped (the flow is always advanced to the side indicated by Y in the figure), and the terminal 1 may continue to record or output terminal position information.

FIG. 5 is a detailed flowchart of step S7 in FIG. 3, which is a characteristic step of the present invention, that is, correction by the terminal position correction unit 7 and the like. When the flow is started in step S20, in step S21, the terminal position correction unit 7 checks the magnitude of the estimated distance d _est (t) obtained by the autonomous navigation position estimation unit 6 according to the above [Expression 2]. A case-by-case determination for switching the subsequent processing according to the size is performed.

Specifically, as shown in the figure, the case determination is made to the following (1) to (3).
(1) d _est (t) ≦ r _LMk
(2) r _LMk <d _est (t) ≤ R
(3) d _est (t)> R
Here, R is a predetermined value larger than the usable radius r _LMk (radius R2 in FIG. 4) of the positioning signal from the landmark LM _k , and is set in advance to a predetermined value that satisfies the condition of R ≧ 2r _LMk . Note that the value of the usable radius r _LMk as the threshold and the value of R as the threshold are acquired by the terminal 1 in advance in step S3 and the like as basic data used in the present invention together with the map information and the like. Shall.

In FIG. 5, a provisional position to be described later is calculated for each of the cases (1) to (3), and the provisional position appropriately corrected is finally given to the user as the location of the terminal 1 in step S22. It is determined that it should be presented and presented to the user. In particular, in the cases (2) and (3), the estimated distance d _est (t) is within a range where the terminal 1 can exist when the flow of FIG. 5 is executed (a positioning signal of the landmark LM _k can be used). Since the value indicates the outside of the range), the estimated position surely includes an error, and a corresponding calculation process is performed.

In order to describe FIG. 5, first, a representative position set by the representative position setting unit 4 will be described with reference to FIG. 6. _6A is a decision circle having a radius r _TDOA (t) at the time t, centered on the landmark LM _k (here, the position is indicated by a point), as described in FIG. Ck1 and the surrounding area are shown. As described above, the terminal 1 is considered to exist at any position on the determined circle Ck1.

Taking (A) as an example, the setting of representative positions P1 to P4 by the representative position setting unit 4 will be described. (B) in the figure shows the same area as (A) with its map information highlighted, and a passage through which the user can pass is arranged around the landmark LM _k. The map information is divided into passage sections W1 to W4. Each of the passage sections W1 to W4 has a rectangular or polygonal region having an end formed as a side in the vicinity of the landmark LM _k (provided in the vicinity of the landmark LM _k) , as shown in FIG. The area information is given in the map information as only a part thereof is shown).

Further, the intersection points CR12 to CR14 of each passage section are also included in the map information. For example, the intersection point CR12 is defined as an intersection point between the passage sections W1 and W2 and given a position. Note that the rectangular area around the four intersections CR12 to CR14 around the landmark LM _{k is} also movable as an area connected to the passage sections W1 to W4, but includes the landmark LM _k inside. Therefore, it is described in the map information as a partitioned area distinguished from the passage section.

  Conversely, areas other than the passage sections and section areas such as W1 to W4 that are shaded on the boundary in (A) are the stores described in FIG. 2 and are blocked by walls or the like. The area where the user of the terminal 1 cannot freely move from the passage sections W1 to W4 and the like is also included in the map information M1.

(A) As shown in the representative position setting unit 4 includes a determination circle Ck1 at the time t, a plurality of arcs defined as overlap between the passage section W1~W4 near the landmark LM _k (here In this example, the middle points are obtained as representative positions P1 to P4. As shown in the figure, the midpoint is the midpoint of the length on the arc, and is the midpoint where the angle from the landmark LM _k is the center on the arc.

  That is, by referring to the map information, for example, arcs Q11 to Q12 that are overlapping portions of the decision circle Ck1 and the passage section W1 are obtained, and the representative position P1 corresponding to the passage section W1 is obtained as a midpoint on the arc. It is done. Similarly, arcs Q21 to Q22, Q31 to Q32 and Q41 to Q42, which are overlapping portions of the passage sections W2, W3, and W4 and the determined circle Ck1, are obtained, and the passage sections W2, W3, and Representative positions P2, P3 and P4 corresponding to W4 are obtained.

For explanation, when the positioning signal of the landmark LM _k is received at time t, the decision circle Ck1 is determined as the midpoint of the arc that overlaps the passage section Wj (j = 1 to 4 in the above example). The coordinates of the representative position are (x _{[j, k]} (t), y _{[j, k]} (t)) ([Definition 1]). For example, the coordinates of the representative position P1 in FIG. 6 are (x _{[1, k]} (t), y _{[1, k]} (t)).

Returning to FIG. 5, in the case of (1) in step S <b> 21, the process proceeds to step S <b> 31, and the passage position history that the terminal 1 has moved after the initial time, that is, after steps S <b> 3 and S <b> 4 in FIG. Get. The passage section history is a predetermined time step from the initial time to the current time t = t ₁ , t ₂ , t ₃ , ..., t _i , t _{i + 1} , ..., t _N (t _N is the current time ) _Are estimated positions (x _est (t _i ), y _est (t _i )) {i = 1, 2, 3,..., N}, for example, in any of the passage sections W1 to W4 in FIG. It is expressed as history as to whether it corresponds, that is, in which passage section.

Here, by referring to the map information,
Estimated position (x _est (t _i ), y _est (t _i )) ∈ passage section Wj (in the example of FIG. 6, any of j = 1 to 4)... [Formula 3]
If so, the passage section history is configured as if the terminal 1 existed in the passage Wj at the time step t _i .

In addition, [Equation 3] may not be established for any passage section. For example, this may be the case when the estimated position becomes a position that cannot be originally moved except for the passage section due to problems such as accuracy. In such a case, by referring to the map information in the same manner, the passage section history is configured as being present in the passage section closest to the estimated position. That is, the distance Lj between the estimated position (x _est (t _i ), y _est (t _i )) and each passage section Wj is obtained, and the terminal 1 exists in the passage section Wj that minimizes the distance Lj. The passage section history is constructed.

Here, the distance Lj may be obtained as the minimum value of the line segment extending from the estimated position (x _est (t _i ), y _est (t _i )) to the boundary of the passage Wj. As another example preferable or, on the distance Lj, the estimated position _{(x est (t i),} y est (t i)) and the representative position on the passage section Wj in the time step t _i (x _{[j, k]} (t _i ), y _{[j, k]} (t _i )) or the like may be used.

That is, as a general rule, at each time step t _i , the estimated position (x _est (t _i ), y _est (t _i )) of the terminal 1 belongs to the passage section to which the estimated position belongs, or if there is no assigned passage, the estimated position is the highest. A nearby passage section is defined as a passage section Wj (t _i ) where the terminal was present, and a passage section history {Wj (t _i ) | i = 1, 2, 3,..., N} is obtained. Here, since the index j for specifying the passage section depends on each time step t _i , it is expressed as j (t _i ) in Wj (t _i ).

Even when the estimated position exists within a predetermined section area (which is distinguished from the passage section) including the landmark LM _k itself, any of the surrounding passage sections Wj is determined by performing the determination based on the distance Lj. The passage history is configured on the assumption that the terminal 1 exists in the closest one of them. For example, as shown in FIG. 6, if there is an estimated position in a predetermined section area (a quadrangle having intersections CR12 to CR14 as vertices) including the landmark LM _k, the closest one of the surrounding passage sections W1 to W4 is the path history. Chosen.

As an exception process of the process constituting the passage history as described above, there is the following process. That is, small decision circle Ck1, if there is a corresponding passage section that does not overlap with the determined circle Ck1 even one in the passage section of the immediate surrounding landmarks LM _k is defined a history in the time step t _i Without deleting the term of the time step t _i from the passage section history. Here, the nearest surrounding passage section is a passage section connected to a predetermined section area to which the landmark LM _k belongs. In the example of FIG. 6, the passage sections W1 to W4 connected to the predetermined sections of the quadrilateral CR14 to CR14 are the nearest passage sections around the landmark LM _j .

For example, if the corresponding passage sections are {W1, W1, (decision circle Ck1 is small), W4, W4} at time steps {t ₁ , t ₂ , t ₃ , t ₄ , t ₅ }, respectively, the time steps to erase the portion of t _3, the time step _{{t 1, t 2, t} 4, t 5} passage section history defined by adopting {W1, W1, W4, W4 }.

Next, in step S32, the terminal position correction unit 7 uses the obtained passage section history {Wj (t _i ) | i = 1, 2, 3,..., N} and the estimated position at the current time t _N. Thus, the provisional position (x _out , y _out ) of the terminal at the current time t _N is obtained as follows.

In [Expression 4] and [Expression 5], k is an identifier of the landmark LM _k that has received the positioning signal at the time, and x _{[j (ti), k]} (t _N ) etc. as per the definition 1], a representative position determined relative to the determined circle at the current time t _N, the passage sections Wj appearing on history and (t _i).

The provisional position by [Formula 4] and [Formula 5] That is, the N representative position on the determination circle at the current time t _N (including the case where counting duplicate), the estimated position at the current time t _N It is the average position of the total (N + 1) positions with (1). In particular, when using the representative position, the number of times that appears in the passage section history Wj (t _i ) [i = 1, 2,... Note that the method of assigning the weight is not limited to the method in [Expression 4] and [Expression 5], and other given weight assignments according to the number of times of appearing in the history, its rank, and the like may be adopted. .

FIG. 7 is a diagram for explaining calculation examples of the above [Expression 4] and [Expression 5]. Here, passage sections W1 to W4 are provided around the landmark LM _k , and N = 3 and time The step consists of {t ₁ , t ₂ , t ₃ }. FIG. 7 shows the determination circle of the radius r _TDOA (t _i ) at each time step and the locus of the estimated position indicated by the arrows as examples (A) and (B). The current time is t _3. Example (A) is an example in which the passage history of each time step {t ₁ , t ₂ , t ₃ } is {W1, W1, W1}.

Therefore, in (A), for example, Equation [4] is as follows.
x _out = {x _est (t ₃ ) + 3x _{[1, k]} (t ₃ )} / 4
That is, the representative position of W1 that appears three times is added three times, and a total of four average positions (x coordinates) of one estimated position and three representative positions are obtained. In addition, the example (B) is an example in which the passage history of each time step {t ₁ , t ₂ , t ₃ } is {W1, W4, W4}. Therefore, in (B), for example, Equation [4] is as follows.
x _out = {x _est (t ₃ ) + x _{[1, k]} (t ₃ ) + 2x _{[4, k]} (t ₃ )} / 4
That is, the representative position of W1 that appears once is added once, the representative position of W4 that appears twice is added twice, and the average of four estimated positions and three representative positions in total The position (x coordinate) is determined.

Returning to FIG. 5, in step S33, the provisional position (x _out , y _out ) obtained in step S32 is a position on the map information that can be reached, that is, any part of the predetermined section area closest to the passage section or landmark. The terminal position correction unit 7 confirms whether or not the terminal exists. If the temporary position exists at a passable position, the process proceeds to step S22, and the terminal position correction unit 7 finally sets the temporary position to be presented to the user.

If the temporary position does not exist at the passable position, the process proceeds to step S34 and step S35 to make correction. Terminal position correcting unit 7 in step S34 is the temporary position (x _out, y _out) and the distance between the representative position in the present time t _N is calculated and sorted in the order of magnitude of the distance. In step S35, the terminal position correcting unit 7 further selects the upper two passage sections Wi (path having the smallest distance) and passage section Wj (second passage from the smallest distance) on the side where the sorted distance is smaller. Then, the value of the provisional position (x _out , y _out ) is overwritten and corrected by the value of the intersection (x _{cross, ij} y _{cross, ij} ) of the upper two passage sections. Then, the corrected provisional position is determined as a position to be presented to the user in step S22.

FIG. 8 is an example for explaining the processing of steps S34 and S35 taking the area of FIG. 6 as an example. In FIG. 8, as a point not drawn in FIG. 6, the provisional position (x _out , y _out ) to be corrected is indicated by × as a point E1. As illustrated by the hatching, the point E1 is at a position where the terminal 1 cannot exist.

In general, when the distance between the point X and the point Y is expressed as dist (X, Y) ([Definition 2]), in the example of FIG. 8, the distances from the representative positions P1 to P4 are calculated in step S34. And sorted as follows:
dist (E1, P2) <dist (E1, P1) <dist (E1, P3) <dist (E1, P4)

From the sorting result, the upper two passages are a passage section W2 corresponding to the first representative position P2 and a passage section W1 corresponding to the second representative position P1. Therefore, in step S35, the point CR12 that is the intersection of the passage sections W2 and W1 is determined as the position where the provisional position is to be corrected. That is, the provisional position (x _out , y _out ) is corrected from the point E1 to the point CR12.

As is clear from the above description, in the case (1) in FIG. 5, the estimated distance d _est (t) itself is small, and the terminal 1 is in the vicinity of the landmark LM _{k in} the same way with respect to the actual position. The present invention corresponds to a case where the passage section W1 to W4 is slightly entered, or the terminal 1 enters the passage section W1 to W4 and then enters another passage section. Gives a reasonable position. Further, even if a place where movement is impossible is determined as a temporary position by the determination in step S33, the temporary position can be corrected to the intersection of the two nearest passage sections as a reasonable position.

That is, in case (1), the estimated distance d _est (t) itself is small compared to cases (2) and (3), so the error in the estimated position (x _est (t), y _est (t)) is also small. Conceivable. In the present invention, an appropriate position is obtained by reducing or adjusting the error to some extent by using a passage section history reflecting that a restriction is imposed on a movable portion of the terminal 1 for map information. .

Note that cases (2) and (3) described below are cases where the estimated distance d _est (t) exceeds the range that can be calculated from the actual position of the terminal 1, and is more error than case (1). The cumulative effect of is increasing. Therefore, in cases (2) and (3), an appropriate position is obtained by placing more emphasis on the information of an appropriate passage section determined by the passage section history.

In addition, as is clear from the consideration in the above case (1), the passage section history {Wj (t _i ) | i = 1, 2, 3,..., N} represents the temporary position (x _out , y _out ). using only latest predetermined number of time t _N to determine, it may be applied to expression 4] and [formula 5]. The same effect as the application can also be obtained by applying a larger weight to the representative position as the side closer to the current time t _N in the history.

For example, in the example as shown in FIG. 7, N = 6 and the passage section history is {W1, W1, W1, for the time steps {t ₁ , t ₂ , t ₃ , t ₄ , t ₅ , t ₆ }. W1, W4, W4} if it like, the terminal 1 has been moved to W4 from W1, the actual position at the time t ₆ is likely to be on W4. Thus, the last three time steps _{{t 4, t 5, t} 6} passage sections histories limited to {W1, W4, W4} provisional position (x _out, y _out) may be obtained.

Specifically, in [Expression 4] and [Expression 5], instead of obtaining the sum of the right side by i = 1 to N, it is obtained by i = M to N (1 <M <N), and the right side denominator ( By changing (N + 1) to (N−M + 2), the temporary position can be obtained from the passage section history at the most recent time steps {t _M , t _{M + 1} ,..., T _N }.

Returning to the description of step S21 in FIG. 5, when the determination in the case (2) is made in step S21, the process proceeds to step S41, where the terminal position correcting unit 7 determines the estimated position (x _est (t), The distance between y _est (t)) and each representative position at the current time t is calculated, and the distances are sorted.

In step S42, the terminal position correction unit 7 assigns a larger weight to the representative position on the side with the smaller distance sorted in step S41, and determines a predetermined radius from the determined circle of the radius r _TDOA (t) at the current time t. Select a position and make it a provisional position. That is, the provisional position is a point on the decision circle and is a position determined by a predetermined method so as to be closer to the representative position with the smaller sorted distance.

An embodiment of steps S41 and S42 will be described with reference to FIG. In FIG. 9, the representative position as the midpoint of the arc on the passage sections W1 to W4 (only a part thereof is drawn) on the determination circle Ck1 of the radius r _TDOA (t) around the landmark LM _k . P1 to P4 are given. A point E2 is a point at an estimated position (x _est (t), y _est (t)). Here, when the symbol of [Definition 2] is used, the sorting result in step S41 is as follows.
dist (E2, P1) <dist (E2, P2) <dist (E2, P4) <dist (E2, P3)

  In one embodiment of step S42, first, the midpoint P12 on the arc is obtained in the arc (P1 to P2) having the representative position P1 that minimizes the distance and the representative position P2 that has the second smallest distance as both ends. . Further, in the arc (P1 to P4) having the representative position P1 that minimizes the distance and the representative position P4 having the third smallest distance as both ends, a midpoint P14 on the arc is obtained. Then, as a temporary position, a midpoint P124 on an arc of arcs (P12 to P14) having the midpoints P12 and P14 as both ends is set.

  That is, when an angle from a predetermined position of the point P on the decision circle is expressed as θ (P), P124 is a point where the angle is {2θ (P1) + θ (P2) + θ (P4)} / 4. . In addition to the embodiment, a provisional position can be obtained by applying a predetermined weight according to the sorted distance.

  In step S43, the terminal position correction unit 7 confirms from the map information M1 whether or not the provisional position set in step S42 exists at the passable position. If it exists, the process proceeds to step S22 and the provisional position is directly used by the user. Is set as the position to be presented. If it does not exist, the terminal position correcting unit 7 corrects the temporary position in step S43, and the corrected temporary position is set as the user presentation position in step S22.

  The correction in step S43 may use the same method as in steps S34 and S35. As shown in FIG. 10, the closest passage section on the circumference of the decision circle Ck1 from the provisional position obtained in step S42. You may modify it to a location that crosses. In FIG. 10, the provisional position before the correction that existed at the inaccessible position is corrected to a location that crosses the closest passage section W1.

  As described above, in the case (2), the actual position of the terminal 1 is almost the same as the assumption in the case (1), but the error in the estimated position is larger than the case (1). A reasonable position is determined by using the information of the passage section more actively.

  Returning to the explanation of step S21 in FIG. 5, when the determination in the case of (3) is made in step S21, the process proceeds to step S51. Wj) is estimated and acquired. The acquisition may be performed in the same manner as the method of acquiring the passage section history described in step S31. However, at this time, it is only necessary to acquire only the corresponding passage section at the current time t, not as a history.

In step S52, the terminal position correcting unit 7 determines an arbitrary point (x _random , y _random ) on the arc of the overlapping portion of the estimated passage section Wj and the determined circle at the time as a provisional position. In step S22, the provisional position is determined as a position to be presented to the user. In another embodiment, a representative point (x _{[j, k]} (t), y _{[j, k]} (t)) on the same arc may be used as a temporary position and a position to be presented to the user instead of an arbitrary point. .

As described above, since the error in (3) is larger than that in (2), an appropriate position is determined with more emphasis on information on the passage section than in (2). In the case of (3), there is a possibility that the actual position of the terminal 1 is further away from the landmark LM _k than the position assumed in (1) and (2) and enters some of the passage sections Wj to some extent. Is expensive. Since the estimated position is considered to retain some information regarding the direction from the landmark LMk, the estimated position is determined to be an appropriate position by obtaining the most appropriate belonging passage section.

  In the cases (2) and (3), the obtained result may be output from the representative position setting unit 4 instead of the terminal position correction unit 7 as shown in FIG.

As described above, in the present invention, each landmark LM _k is surrounded by the passage section by arranging the end of the passage section around the landmark LM _{k, and} is arranged in advance in the corner portion of the area where the terminal 1 and the user can pass. The position of the terminal 1 can be acquired. That is, each landmark LM _k is inside a predetermined partition area where the terminal 1 can move, and the terminal 1 can move around the landmark LM _k by connecting the passage sections so as to surround the predetermined partition area. By forming a region that is not and a region that is not, a representative position can be set from the determined circle, and the present invention can be applied.

An example of the arrangement is shown in FIG. Other examples are shown in FIGS. Here, the boundary of the non-passable area is hatched. (A) In the passage section WA1 and WA2 is arranged by disposing the ends a1 and a2 around the landmarks LM _k. (B) the passage section WB1~WB3 is arranged by arranging the end b1~b3 around the landmark LM _k. (C) the passage section WC1~WC3 is being arranged by disposing end c1~c3 around the landmark LM _k (is c3 are opposite L-shaped end portion).

As shown in the example of FIG. 11, the passages may be arranged with an arbitrary number and an arbitrary angle around the landmark LM _k . The shape of the passage section is typically rectangular or polygonal, but may be any shape including a curved shape as long as the representative position can be defined on the decision circle. Moreover, the said shape should just correspond with the boundary of an actual passage etc. and a store etc. substantially.

  DESCRIPTION OF SYMBOLS 1 ... Terminal, 2 ... Landmark signal receiving part, 3 ... Landmark distance determination part, 4 ... Representative position setting part, 5 ... Sensor part, 6 ... Autonomous navigation position estimation part, 7 ... Terminal position correction part

Claims (9)

A navigation terminal that acquires its own position,
A landmark distance determining unit that receives a positioning signal from a landmark arranged at a predetermined position and determines a distance between the landmark and the navigation terminal;
A map information holding unit for holding map information including positional information of the landmarks and information of passage sections around the landmarks to which the navigation terminal can move;
A sensor unit;
An autonomous navigation position estimation unit for obtaining an estimated position of the navigation terminal by applying autonomous navigation to an output from the sensor unit;
A representative that sets a representative position of the navigation terminal from each of arcs obtained as overlapping portions of a circle having a radius equal to the determined distance around the landmark and the passage section with reference to the map information A position setting unit;
A navigation terminal comprising: a terminal position correction unit that acquires a position of the navigation terminal based on each of the estimated position and the representative position, and a distance between the estimated position and each representative position.   The autonomous navigation position estimation unit sets the position of the landmark obtained from the map information as an initial position when the distance determined from the positioning signal satisfies a predetermined condition and becomes small, and the autonomous navigation The navigation terminal according to claim 1, wherein application of is started. The terminal position correction unit acquires the history of the passage section where the navigation terminal is estimated to be located, based on the history of the estimated position and the location information of each passage section in the map information.
When acquiring the position of the navigation terminal, among the representative positions at the time of acquisition, the average position of the representative position corresponding to the passage section appearing in the history of the passage section and the estimated position at the time of acquisition As the position of the navigation terminal,
The given weight according to the number of times of appearing in the history of the passage section is given to each of the representative positions when obtaining the average position. Navigation terminal.   The terminal position correction unit obtains the average position as the position of the navigation terminal when the estimated position falls within a first threshold on the side where the distance from the landmark is small, and if not, By calculating the distance between each of the estimated position and the representative position at the time of acquisition, and given weighting such that the calculated distance is closer to the smaller representative position on the circumference of the circle, The navigation terminal according to claim 3, wherein a position of the navigation terminal is acquired.   The terminal position correction unit obtains the position of the navigation terminal by the weighting when the distance from the landmark of the estimated position is within a second threshold value that is larger than the first threshold value, and does not fit Calculating the distance between the estimated position and each representative position at the time of acquisition, and obtaining the overlapping portion of the circle and the passage section corresponding to the representative position where the calculated distance is minimum. The navigation terminal according to claim 4, wherein a position selected at random from positions on the arc to be obtained is acquired as the position of the navigation terminal.   When the terminal position correction unit cannot receive a positioning signal from the landmark at the time of acquisition, the terminal position correction unit is based on each of the estimated position and the representative position and the distance between the estimated position and each representative position. 6. The navigation terminal according to claim 1, wherein instead of acquiring the position of the navigation terminal, the estimated position at the time of acquisition is acquired as the position of the navigation terminal. A method for obtaining the location of a terminal,
Receiving a positioning signal transmitted by a landmark placed at a predetermined position, and determining a distance between the landmark and the terminal;
Obtaining map information including position information of the landmark and information of a passage section where the terminal can move around the landmark;
Applying autonomous navigation to the output from the sensor provided in the terminal to obtain the estimated position of the terminal;
Referring to the map information, setting a representative position of the terminal from each of arcs obtained as overlapping portions of a circle having a radius equal to the determined distance with the landmark as a center and the passage section; and ,
A terminal position acquisition method comprising: acquiring the terminal position based on each of the estimated position and the representative position, and a distance between the estimated position and each representative position.   8. The terminal according to claim 7, wherein each of the landmarks is arranged at a predetermined position apart from each other so that the terminal does not receive positioning signals from two or more landmarks simultaneously. How to get the position of   The landmark is disposed inside a predetermined partition area where the terminal is movable, and a passage section is connected to the predetermined partition area, so that the terminal is not movable with an area where the terminal can move around the landmark. 9. The method according to claim 7 or 8, wherein a region is formed.