JP6701642B2 - Position estimation system, position estimation device, position estimation method, and position estimation program - Google Patents

Description

  The present invention relates to a position estimation system, a position estimation device, a position estimation method, and a position estimation program, for example, in a position measurement method using a plurality of sensors, a position estimation system, a position estimation device, and a position estimation device that accurately estimate the position. It is applicable to the method and the position estimation program.

  For example, it is possible to estimate the position of a pedestrian having a GPS device that communicates with GPS satellites outdoors, but it is difficult to accurately estimate the position of a pedestrian indoors or in a place that cannot communicate with GPS satellites. .. In such a case, as a position measuring method for estimating the position of a pedestrian, there is a pedestrian autonomous navigation (PDR: pedestrian dead reckoning) for estimating the position of a pedestrian using a plurality of sensors (Non-Patent Document 1). reference).

  In the direction estimation, the rotation angle can be calculated from the angular velocity sensor (gyro sensor). Further, there is a method of using an electronic compass (geomagnetic sensor) as a method of obtaining an absolute direction with respect to the North Pole of the earth. An acceleration sensor is used to estimate the distance in the traveling direction. There is a pedestrian autonomous navigation technology for positioning the position of a pedestrian using these estimation results.

JP-A-5-113337 JP, 2000-346661, A

Kei Hirose, Hitoshi Toki, Akiko Kondo, "Study on posture measurement in sports using inertial sensor and geomagnetic sensor", Sports industry research, Vol. 22, No. 2,2012, pp.255-262

  However, in the direction estimation, the rotation angle can be calculated from the acceleration and angular velocity sensors, but since the calculation method accumulates the relative angle change amount, the error is always accumulated and the accuracy is not sufficient. Furthermore, there is a method of using an electronic compass (geomagnetic sensor) as a method of obtaining an absolute direction with respect to the North Pole of the earth, but the accuracy is not sufficient due to the influence of metal in the building of the reinforcing bar. Similarly, in the estimation of the distance traveled, the accuracy is not sufficient if simply calculated from the acceleration sensor. Therefore, the conventional positioning based on pedestrian dead reckoning is not sufficiently accurate.

  Therefore, in view of the above problems, the present invention requires a position estimation system, a position estimation device, a position estimation method, and a position estimation program that accurately estimate the position of a target node.

In order to solve such a problem, the position estimation system according to the first aspect of the present invention is (1) provided in the reception node in the position estimation system for estimating the position of the reception node that receives the radio wave transmitted from the transmission node. The first estimated position of the receiving node is estimated based on the state change information and the absolute azimuth information detected by the inertial sensor, and the moving average value of the first estimated position before a predetermined number in the past and the current Inertia position estimating means for obtaining first position change information indicating time-series change of the first estimated position , which is vector information based on a difference from the moving average value at the first estimated position of (2) receiving node The second estimated position of the receiving node is estimated based on the intensity of the radio wave received from the transmitting node at, and the moving average value of the second estimated positions before a predetermined number in the past and the movement at the current second estimated position. Radio wave position estimation means for obtaining second position change information indicating time series change of the second estimated position , which is vector information based on the difference from the average value, and (3) first position change information and second position change information. Position correction means for comparing the position change information and correcting the current first estimated position and the past first estimated position according to the comparison result, and the position correction means It is characterized in that the first estimated position is rotated and corrected according to the angle between the direction of the position change information and the direction of the second position change information .

A position estimation device according to a second aspect of the present invention is a position estimation device that estimates the position of a reception node that receives a radio wave transmitted from a transmission node, (1) detected by an inertial sensor provided in the reception node, The first estimated position of the receiving node is estimated based on the state change information and the absolute azimuth information, and the moving average value of the past first estimated positions and the movement at the current first estimated position are calculated. Inertial position estimation means for obtaining first position change information indicating a time series change of the first estimated position , which is vector information based on a difference from the average value; and (2) reception radio waves from the transmission node at the reception node. A second estimation position of the receiving node is estimated based on the strength, and a vector based on the difference between the moving average value of the second estimated positions in the past and the current moving estimation value of the second estimated position. Radio wave position estimating means for obtaining second position change information indicating time series change of the second estimated position , which is information, and (3) comparing the first position change information and the second position change information, A position correction unit that corrects the current first estimated position and the past first estimated position according to the comparison result is provided , and the position correction unit determines the direction of the first position change information and the second position change information. The first estimated position is rotated and corrected according to the angle of the position change information with the direction .

A position estimation method according to a third aspect of the present invention is the position estimation method of estimating the position of a reception node that receives a radio wave transmitted from a transmission node, wherein (1) an inertial sensor provided in the reception node detects: The first estimated position of the receiving node is estimated based on the state change information and the absolute azimuth information, and the moving average value of the past first estimated positions and the movement at the current first estimated position are calculated. First position change information indicating time series change of the first estimated position , which is vector information based on the difference from the average value, is obtained, and (2) reception is performed based on the intensity of the reception radio wave from the transmission node at the reception node. The second estimated position of the node is estimated, and the vector information is based on the difference between the moving average value of the past second estimated positions and the moving average value of the current second estimated position . The second position change information indicating the time-series change of the estimated position of No. 2 is obtained, and (3) the first position change information and the second position change information are compared, and the current position is changed according to the comparison result. The first estimated position and the past first estimated position are corrected , and the first estimated position is rotated and corrected by the angle between the direction of the first position change information and the direction of the second position change information. characterized in that it.

A position estimation program according to a fourth aspect of the present invention is a position estimation program for estimating the position of a reception node that receives a radio wave transmitted from a transmission node, wherein (1) an inertial sensor provided in the reception node detects a computer. The first estimated position of the receiving node is estimated based on the state change information and the absolute azimuth information, and the moving average value of the past first estimated positions and the current first estimation are calculated. Inertial position estimating means for obtaining first position change information indicating time series change of the first estimated position , which is vector information based on the difference from the moving average value at the position; and (2) from the transmitting node in the receiving node. The second estimated position of the receiving node is estimated based on the intensity of the received radio wave, and the difference between the moving average value of the second estimated position in the past and the moving average value of the current second estimated position is calculated. Radio wave position estimating means for obtaining second position change information indicating time-series change of the second estimated position , which is vector information based on (3) first position change information and second position change information. The position of the first position change information is compared with the position of the first position change information by making the position correction unit correct the current first estimated position and the past first estimated position according to the comparison result. It is characterized in that the first estimated position is rotated and corrected by the angle between the direction of the second position change information and the direction of the second position change information .

  According to the present invention, the position of a target node can be accurately estimated.

It is an internal block diagram which shows the internal structure of the target node which concerns on 1st Embodiment. 1 is an overall configuration diagram showing an overall configuration of a communication system according to a first exemplary embodiment. It is an internal block diagram which shows the internal structure of the anchor node which concerns on 1st Embodiment. It is explanatory drawing explaining the reception of the beacon signal which a target node receives. It is explanatory drawing explaining the outline|summary of the position estimation process in the communication system which concerns on 1st Embodiment. 6 is a flowchart showing an operation of position estimation processing according to the first embodiment. It is an explanatory view explaining calculation of a moving average value and vector R. It is explanatory drawing explaining the correction method of the angle of the estimated position by the integrated position estimation part which concerns on 1st Embodiment. It is explanatory drawing explaining the outline|summary of the position estimation process in the communication system which concerns on 2nd Embodiment. 9 is a flowchart showing an operation of position estimation processing according to the second embodiment. It is explanatory drawing explaining the correction method of the scale of an estimated position by the integrated position estimation part which concerns on 2nd Embodiment. It is explanatory drawing explaining the outline of the position estimation process which concerns on modified embodiment. It is a flowchart which shows operation|movement of the position estimation process which concerns on modified embodiment (the 1). It is a flowchart which shows the operation|movement of the position estimation process which concerns on modified embodiment (the 2).

(A) First Embodiment Hereinafter, a first embodiment of a position estimation system, a position estimation device, a position estimation method, and a position estimation program according to the present invention will be described in detail with reference to the drawings.

(A-1) Configuration of First Embodiment FIG. 2 is an overall configuration diagram showing an overall configuration of a communication system according to the first embodiment. The communication system 10 according to the first embodiment includes a plurality (four in FIG. 2) of anchor nodes 2-1 to 2-4 and a plurality (three in FIG. 2) of target nodes 1-1 to 1-3. Have.

  It should be noted that, in the following, the structure and processing common to all anchor nodes will be referred to as anchor node 2. Also, when describing the configuration and processing common to all target nodes, it will be referred to as target node 1.

  The communication system 10 exemplifies a system built indoors, for example. The communication system 10 may be constructed outdoors, but can be applied in an environment where it is difficult to communicate with GPS satellites, an environment in which the position of a pedestrian cannot be estimated by the GPS function, or the like.

  The anchor node 2 is, for example, a node installed indoors in a building, and, for example, an access point of a public or dedicated wireless LAN can be applied. The anchor node 2 may be fixed indoors or movable if the position of the anchor node 2 is known.

  The target node 1 is a node whose position is to be estimated, and for example, a movable node possessed by a pedestrian can be applied. The target node 1 can be widely applied as long as it has a device having a wireless communication function. More specifically, the target node 1 includes a node mounted on a smartphone, a mobile phone, a mobile terminal, a wearable terminal, a vehicle (for example, a concept including a car, a bicycle, a wheelchair, a small car, etc.) and the like.

  In the first embodiment, for convenience of explanation, a method in which the target node 1 collects information by itself and measures its own position will be described. However, the anchor node 2 has a function of positioning the position of the target node 1, estimates the position of the target node 1 of interest, and communicates a packet including the estimated position information to the target node 1 of interest. You can That is, in the first embodiment, the case where the position estimation system is installed in the target node 1 is illustrated, but the anchor node 2 may be installed with the position estimation system, or the processing function of the position estimation system may be provided. The target nodes 1 and the anchor nodes may be arranged in a distributed manner.

  FIG. 1 is an internal configuration diagram showing an internal configuration of the target node 1 according to the first embodiment. Although the hardware configuration of the target node 1 is not shown, for example, a circuit having a CPU, a ROM, a RAM, an EEPROM, an input/output interface, a communication device, or the like can be applied.

  In FIG. 1, the target node 1 includes a position estimating unit 11, an acceleration sensor 12, a gyro sensor 13, a geomagnetic sensor 14, a wireless unit 15, and a battery 16. The target node 1 may also have a position display unit 17 that displays the estimated position information. For example, when the target node 1 is a smartphone, the display unit of the smartphone may be the position display unit 17.

  The position estimation unit 11 is a functional unit or device that estimates the position of the target node 1 itself. The position estimation unit 11 uses the position information acquired from the anchor node 2 through the wireless unit 15 and the sensor information of various sensors (acceleration sensor 12, gyro sensor 13, geomagnetic sensor 14) mounted on the target node 1 itself, It estimates the position of itself. The position estimation unit 11 may be software (for example, an application program) activated on the OS (operating system) of the target node 1 or may be realized by a hardware configuration.

  The position estimating unit 11 has a sensor position estimating unit 111, a radio wave position estimating unit 112, and an integrated position estimating unit 113 as a position correcting means.

  The sensor position estimation unit 111 acquires sensor information from each of the acceleration sensor 12, the gyro sensor 13, and the geomagnetic sensor 14, and estimates the position based on these sensor information. The sensor position estimation unit 111 gives the estimation result to the integrated position estimation unit 113. In addition, the sensor position estimation unit 111 acquires the correction amount of the estimated position from the integrated position estimation unit 113 in order to avoid the accumulation of errors in the position estimation process using the sensor information, and the next position estimation process is performed. At that time, the position is estimated using the correction amount.

  The radio wave position estimation unit 112 estimates the position of the target node 1 itself based on the information on the received power value of the beacon signal acquired from the wireless unit 15. The radio wave position estimation unit 112 gives the estimation result to the integrated position estimation unit 113. Various methods can be widely applied to the position information estimation processing by the radio wave position estimation unit 112.

  For example, the radio wave position estimation unit 112 has a conversion table showing the relationship between the received power value and the distance (distance between the anchor node 2 and the target node 1), and the radio unit 15 receives from the anchor node 2. The distance between the anchor node 2 and the target node 1 is estimated based on the received power value (RSSI) of the beacon signal. Further, when beacon signals are received from a plurality of anchor nodes 2, the positions of the target nodes 1 are stochastically calculated by the maximum likelihood estimation method using the respective distances between the anchor nodes 2 and the target nodes 1. A method of estimating information can be applied.

  The integrated position estimation unit 113 corrects the position information by integrating the position information based on the sensor information from the sensor position estimation unit 111 and the position information based on the received power value of the radio wave from the radio wave position estimation unit 112. The process in which the integrated position estimation unit 113 integrates the position information based on the sensor information and the position information based on the received power value and corrects the position information will be described in detail in the section of operation.

  Further, the integrated position estimation unit 113 adds the calculated correction amount of the position information to the position information based on the sensor information, and displays the corrected position information on the position display unit 17.

  Further, the integrated position estimation unit 113 feeds back the calculated correction amount of the position information to the sensor position estimation unit 111.

  The wireless unit 15 has a function of an existing wireless device. The wireless unit 15 has a power measuring unit 151, measures the received power value of a received radio wave (beacon signal), and gives the measurement result to the radio wave position estimation unit 112. The wireless unit 15 may include the transmission unit 152. This transmits information including the position information estimated by the transmitter 152 when transmitting the position information estimated by the target node 1 to the anchor node 2.

  The acceleration sensor 12 is an inertial sensor having a function of acquiring acceleration. The acceleration sensor 12 measures the acceleration of the target node 1 and gives information on the acceleration to the sensor position estimation unit 111. The type and method of the acceleration sensor 12 are not particularly limited, and, for example, those to which a MEMS (Micro Electro Mechanical System) technology is applied can be applied.

  The gyro sensor 13 is an inertial sensor having a function of acquiring angular velocity. The gyro sensor 13 measures the angular velocity of the target node 1 and gives information on the angular velocity to the sensor position estimation unit 111. Note that the type and method of the gyro sensor 13 are not particularly limited, and for example, those to which the MEMS technology is applied can be applied.

  Here, the “state change information” described in the claims includes the acceleration obtained by the acceleration sensor 12 and the angular velocity obtained by the gyro sensor 13. That is, the state change information is intended to be information about a change in the state of the target node 1 obtained by inertia, such as displacement of the moving speed and displacement of the rotation angle.

  The geomagnetic sensor 14 is a so-called electronic compass, and has a function of acquiring the azimuth information of the absolute coordinate system based on the north direction from the geomagnetism, and the azimuth information of the target node 1 is used as the sensor position estimation unit 111. Give to.

  The battery 16 is a power supply unit that supplies power to the devices mounted on the target node 1, and for example, a general-purpose secondary battery or the like can be applied.

  The position display unit 17 displays position information. The position display unit 17 may be included in the main body of the target node 1 or may be an external display unit connected via an interface. The position display unit 17 may not be included in the target node 1. For example, the position display unit 22 (see FIG. 3) may be included in the anchor node 2, and the position information communicated by the target node 1 by packet communication is displayed on the position display unit 22 of the anchor node 2. You may do so.

  FIG. 3 is an internal configuration diagram showing an internal configuration of the anchor node 2 according to the first embodiment.

  In FIG. 3, the anchor node 2 of the first embodiment has a wireless unit 21.

  The wireless unit 21 has a beacon transmission unit 211 that transmits a beacon signal.

  Beacon transmitter 211 transmits a beacon signal, whether it is regular or irregular. The beacon signal may be transmitted as a packet or may be transmitted as a continuous wave (carrier).

  Further, the wireless unit 21 may include a receiving unit 212 that receives the packet from the target node 1 when the packet including the position information is received. The reception unit 212 gives the position information included in the received packet to the position display unit 22 to display the position information. In that case, the anchor node 2 includes a position display unit 22 for external connection or a position display unit 22 mounted inside the anchor node 2.

  FIG. 4 is an explanatory diagram illustrating reception of a beacon signal received by the target node 1.

  As shown in FIG. 4, each pedestrian carries the target node 1. The anchor nodes 2-1 to 2-4 regularly or irregularly transmit beacon signals, and the target node 1 receives the beacon signals.

  Here, at least the identification information of the anchor node 2 and the position information of the anchor node 2 are included in the beacon signal. If the position information of the anchor node 2 is known when the position estimation system (file data related to the position estimation program) is installed in the target node 1, the position information of the anchor node 2 needs to be included in the beacon signal. There is no. In this case, the beacon signal may be a continuous wave instead of a packet. The target node 1 can acquire the position information from the anchor nodes 2-1 to 2-4, and the target node 1 can measure the received power value. Therefore, the position estimation unit 11 uses the position information and the received power value of the anchor node 2. Can be used to estimate the position.

  In the example of FIG. 4, the target node 1 estimates its own position using the position information and the received power values of the four anchor nodes 2-1 to 2-4. Beacon signals may be irregular, and the target node 1 also moves. Therefore, when the delay is large, if the position estimation is performed using the position information and the received power value of the anchor node 2, the accuracy of the estimated position decreases. Therefore, the position estimation unit 11 of the target node 1 estimates the position based on the received power value acquired within a predetermined unit time. When there are a plurality of beacon signals received from the same anchor node 2 per unit time, they may all be used for position estimation calculation.

(A-2) Operation of First Embodiment Next, the operation of the position estimation processing in the communication system 10 according to the first embodiment will be described in detail with reference to the drawings.

  The position estimation processing based on the received power value of the radio wave has poor positioning accuracy at each location, but by taking the moving average, a certain accuracy can be relied upon as a global traveling direction. On the other hand, the accuracy of position estimation based on sensor information is high at the start of estimation. However, when an error occurs, the error accumulates with the lapse of time, and the accuracy of position estimation deteriorates. In consideration of these performance characteristics, the first embodiment performs position estimation by hybridizing an estimated position by radio waves and an estimated position by sensor information.

  FIG. 5 is an explanatory diagram illustrating an overview of the position estimation process in the communication system 10 according to the first embodiment.

  The position estimation unit 11 according to the first embodiment estimates a global traveling direction by taking, for example, a moving average value with respect to the estimated position based on the radio wave from the anchor node 2 (step S1). In this embodiment, the case of estimating the overall traveling direction by taking the moving average is exemplified, but, for example, the moving average in which the weight is changed according to the information at the time of radio wave reception is performed. Is also good.

  The position estimating unit 11 estimates a global traveling direction by taking, for example, a moving average value with respect to the estimated position based on the sensor information (step S2).

  The position estimation unit 11 compares these estimated positions (the general traveling direction of the estimated position based on the radio wave and the general traveling direction of the estimated position based on the sensor information), and when the deviation exceeds a certain value. , The estimated position based on the radio wave is determined to be correct, and the rotation correction amount is calculated from the angle of each traveling direction vector. The estimated position based on the sensor information is corrected based on the calculated rotation amount (step S3).

  FIG. 6 is a flowchart showing the operation of the position estimation processing according to the first embodiment. In FIG. 6, the process by the radio wave position estimating unit 112, the process by the sensor position estimating unit 111, and the process by the integrated position estimating unit 113 are shown separately.

  The radio wave position estimation unit 112 calculates the estimated position of the target node 1 based on the received power value of the radio wave measured by the wireless unit 15 (step S101). For example, the maximum likelihood estimation method may be used to estimate the position by radio waves. The estimated position is expressed in a format such as (X coordinate, Y coordinate). The estimated position may be expressed in two dimensions or three dimensions.

  An example in which the radio wave position estimation unit 112 calculates an estimated position using a statistical method will be described. Here, an example using maximum likelihood estimation as a statistical method is shown.

The estimated position (x^, y^, z^) can be obtained by calculating x, y, z when maximizing l(x, y, z) as shown in Expression (1). it can.

Further, l(x,y,z) can be expressed by equations (2) and (3).

  In Expressions (2) and (3), α and β are values empirically determined by the environment of radio wave propagation. N is the number of pieces of information of the received power value acquired within the unit time.

The information can be represented as (x _n , y _n , z _n , P _n ) from the position coordinates of the anchor node 2 stored in the received packet and the received power value P _n of the radio wave received at this time. .

  The radio wave position estimation unit 112 moves the past A point with respect to the time-series estimated position (x, y) calculated as described above (for convenience of explanation, it is expressed in two dimensions and the Z coordinate is omitted). An average value is calculated (step S102).

  Next, the radio wave position estimation unit 112 obtains the vector R using each moving average value B-1 points before and the moving average value at the present time (step S103). That is, the radio wave position estimation unit 112 obtains the vector R using the moving average value of the past B point including the current time. The value of A and the value of B for calculating the moving average value and the vector R can be set arbitrarily.

  FIG. 7 is an explanatory diagram illustrating the calculation of the moving average value and the vector R. In the example of FIG. 7, the moving average value is obtained by using the data of the past 3 points (in the case of A=3), and the moving average value before B-1 point (in the case of B=5), each moving average value, and the current movement A case where the vector R is calculated using the average value will be exemplified.

  The radio wave position estimation unit 112 averages the X-axis data of the estimated position by the radio wave for the past 3 points, and also takes the average of the Y axis data for the past 3 points. That is, the radio wave position estimation unit 112 estimates the estimated position (x(t), y(t)), (x(t-1), y(t-1)), (x(t-2)) at the current time t. , Y(t−2)), the average value of the X-axis data and the average value of the Y-axis data are obtained.

  The radio wave position estimation unit 112 obtains the vector R using the moving average value at the present time point and each moving average value up to 4 (=B-1) points before.

  Next, the sensor position estimation unit 111 calculates the estimated position using the sensor information from the acceleration sensor 12 (step S104).

The estimated position [x^(t)y^(t)] ^{T at the} current time t is calculated by the equation (4) using the previous position of the target node 1, the traveling direction of the target node 1, and the traveling distance of the target node 1. Can be expressed as

  In Expression (4), D(t) is the distance traveled by the target node 1 in one step. Ψ(t) is the traveling direction of the target node 1 and is an angle with the north pole (north direction) as a reference.

D(t) is calculated from the equation (5) using the output from the acceleration sensor 12.

In Expression (5), α is a proportional constant. In this embodiment, the proportional constant α is called a speed factor. Expression (5) means that the distance increases as the vibration of the acceleration sensor 12 increases in proportion to the proportional constant α. That is, the output at the current time t from the acceleration sensor 12 _{(a x,} a _{y, a} z) (here, is a 3-axis output.) From the more vibrations of the acceleration a is large, the travel distance D of the target node 1 ( It means that t) is large.

  Next, Ψ(t) is calculated. It is assumed that the target node 1 is attached to the body of a pedestrian in any posture. Therefore, it is necessary to rotate the posture of the target node 1 according to the Euler angle to convert the posture in the world coordinate system.

The posture [Ψ _t θ _t φ _t ] ^T of the terminal (target node 1) is calculated by the following equations (6) and (7) using the output [ω _t ω _t ω _t ] ^T of the gyro sensor 13. Can be done (Euler angle ZYX system rotation matrix).

[Ψ ₀ θ ₀ φ ₀ ] ^T , which is the initial value of the equation (7), is calculated using the output of the geomagnetic sensor 14 and the output of the acceleration sensor 12. The method of calculating the initial value [Ψ ₀ θ ₀ φ ₀ ] ^T will be described below.

When the target node 1 is completely stationary, the acceleration sensor 12 detects only the gravitational acceleration g as the acceleration. At this time, the acceleration can be represented by (8) using the roll angle, the pitch angle, and the yaw angle.

By solving the equation (8), the roll angle and the pitch angle can be expressed as the equations (9) and (10).

Next, the yaw angle is calculated. The output of the geomagnetic sensor _{14 (m x, m y,} m z) Converting the equation (11).

The yaw angle can be calculated from the equation (11).

  The position estimated by the sensor position estimation unit 111 is, for example, two-dimensional (X coordinate, Y coordinate) or three-dimensional (X coordinate, Y coordinate, Z coordinate) as in the case of position estimation by radio waves. It is expressed in the form.

  The sensor position estimation unit 111 calculates the moving average value of the past Z points in the same manner as described with reference to FIG. 7 (step S105). Further, the sensor position estimation unit 111 obtains the vector P by using each moving average value before W points (that is, the moving average value at the current time point and each moving average value up to W-1 points before) (step S106). ..

  Next, the integrated position estimation unit 113 outputs the estimated position corrected using the information of the estimated position of the radio wave and the estimated position of the sensor information. More specifically, the integrated position estimation unit 113 calculates the internal angle θ of the vector from the vector R and the vector P (step S107).

  The integrated position estimation unit 113 compares the angle θ with the threshold λ (step S108), and when the angle θ exceeds the threshold λ, the integrated position estimation unit 113 uses the angle θ to estimate the estimated position based on the past sensor information. Is corrected (step S109). Then, the integrated position estimation unit 113 outputs the estimated position based on the corrected sensor information.

  On the other hand, if the angle θ does not exceed the threshold λ (step S108), the integrated position estimation unit 113 outputs the estimated position of the sensor without correction.

  FIG. 8 is an explanatory diagram illustrating a method of correcting the angle of the estimated position by the integrated position estimating unit 113 according to the first embodiment.

  In FIG. 8, regarding the past C point data including the present time point, the angle θ is rotated with the data C-1 points before as the starting point of the rotation. For example, when C=5, the internal angle θ between the vector P and the vector R (the vector R is not shown in FIG. 8) is obtained using the data at time t-4 as the starting point of the rotation. Then, when the angle θ exceeds the threshold value λ, it is rotated by the angle θ starting from the time t-4. The vector P'obtained by this becomes the corrected estimated position.

  Further, in step S109, when the integrated position estimation unit 113 corrects the estimated position, the angle θ that is the correction amount is fed back to the sensor position estimation unit 111.

The sensor position estimation unit 111 uses the correction amount (angle θ) from the integrated position estimation unit 113 to correct the value of the estimated position based on the sensor information according to equation (13). When calculating the estimated position based on the sensor information next time, the sensor position estimation unit 111 performs the position estimation process using the corrected estimated position.

(A-3) Effect of First Embodiment As described above, according to the first embodiment, the error of the estimated position (angle) based on the sensor information accumulated with the passage of time is calculated from the estimated position based on the radio wave. It can be corrected by using the moving average value. Therefore, the final output result (estimated position) can be expected to improve the position estimation accuracy.

(B) Second Embodiment Next, a second embodiment of the position estimation system, the position estimation device, the position estimation method, and the position estimation program according to the present invention will be described in detail with reference to the drawings.

(B-1) Configuration and Operation of Second Embodiment In the above-described first embodiment, the position estimation unit 11 has described the method of correcting the error in the accumulated angle of the estimated position based on the sensor information.

  On the other hand, the second embodiment will explain a method of correcting the error of the scale of the estimated position based on the sensor information.

  The configurations of the target node 1 and the anchor node 2 according to the second embodiment can be the same as or corresponding to the configurations shown in FIGS. 1 to 3 according to the first embodiment. Therefore, the second embodiment will describe in detail a method of correcting the scale of the estimated position based on the sensor information with reference to FIGS. 1 to 3.

  FIG. 9 is an explanatory diagram illustrating an overview of position estimation processing in the communication system 10 according to the second embodiment.

  The position estimating unit 11 according to the second embodiment estimates a global moving distance by taking, for example, a moving average value with respect to the estimated position based on the radio wave from the anchor node 2 (step S5).

  The position estimation unit 11 takes a moving average value for the estimated position based on the sensor information and estimates the global moving distance (step S6).

  The position estimation unit 11 compares the global movement distance of the estimated position based on the radio wave with the global movement distance of the estimated position based on the sensor information, and when the deviation exceeds a certain value, the radio wave is detected. The estimated position based on this is judged to be correct, and the scale correction amount is calculated from the length of each traveling direction vector. The estimated position based on the sensor information is corrected based on the calculated scale (step S7).

  FIG. 10 is a flowchart showing the operation of the position estimation processing according to the second embodiment. In FIG. 10, the process by the radio wave position estimating unit 112, the process by the sensor position estimating unit 111, and the process by the integrated position estimating unit 113 are shown separately.

  In FIG. 10, the process by the radio wave position estimation unit 112 is similar to the process of steps S101 to S103 shown in FIG. That is, the radio wave position estimation unit 112 calculates the estimated position by radio waves. The radio wave position estimation unit 112 obtains a moving average value of past A points for the calculated time-series estimated position (X, Y). Further, the radio wave position estimation unit 112 obtains the vector R using each moving average value before the B point including the current time.

  Further, the processing by the sensor position estimating unit 111 is the same as the processing of steps S104 to S106 shown in FIG. That is, the sensor position estimation unit 111 calculates the estimated position based on the sensor information. The sensor position estimation unit 111 obtains a moving average value of past Z points for the calculated time-series estimated position (X, Y). Further, the sensor position estimation unit 111 obtains the vector P using each moving average value before the W point including the present time.

  Next, the integrated position estimation unit 113 outputs the corrected estimated position using the information of the estimated position by radio waves and the estimated position by sensor information. That is, the integrated position estimation unit 113 calculates the ratio of the vector lengths of the vector R and the vector P, and sets this as the scale (scale) Z (step S201).

  The integrated position estimation unit 113 compares the scale Z with the threshold value ω, and when the scale Z exceeds the threshold value ω (step S202), the estimated position based on the past sensor information is corrected using the scale Z (step S202). S203). Then, the integrated position estimation unit 113 outputs the estimated position based on the corrected sensor information.

  On the other hand, when the scale Z does not exceed the threshold value ω (step S202), the integrated position estimation unit 113 outputs the estimated position of the sensor without correction.

  FIG. 11 is an explanatory diagram illustrating a method of correcting the estimated position scale by the integrated position estimation unit 113 according to the second embodiment.

  In FIG. 11, with respect to the past C point data including the current time point, the scale (vector length) is enlarged or reduced with the data C-1 points before as the starting point of the rotation. For example, when C=5, the data at time t-4 is used as the starting point of rotation, and the ratio of the lengths of the vector P and the vector R (the vector R is not shown in FIG. 8) (vector R/vector). P) and obtain the result as a scale Z. Then, when the scale Z exceeds the threshold value ω, the length of the vector P is enlarged or reduced based on the scale Z, starting from the time t-4. That is, the length of the vector P is multiplied by the scale Z. The vector P'obtained by this becomes the corrected estimated position.

  Further, in step S203, when the integrated position estimation unit 113 corrects the estimated position, the scale Z that is the correction amount is fed back to the sensor position estimation unit 111.

The sensor position estimation unit 111 corrects the value of the speed factor α by the equation (14) using the correction amount (scale Z) from the integrated position estimation unit 113. When calculating the estimated position by the next sensor information, the sensor position estimation unit 111 performs the position estimation process using the corrected speed factor α.

(B-2) Effect of Second Embodiment As described above, according to the second embodiment, the error of the scale of the estimated position based on the sensor information accumulated with the passage of time is used to calculate the movement of the estimated position by radio waves. It can be corrected by using the average information. Therefore, the final output result can be expected to improve the accuracy of position estimation.

(C) Other Embodiments (C-1) In the above-described first and second embodiments, the radio wave position estimation unit always performs the position estimation process by the radio wave, and the integrated position estimation unit calculates the radio wave estimated position and the sensor. The case where the estimated position based on the sensor information is corrected using the moving average value of the estimated position based on the information is illustrated.

  However, the position estimation process using radio waves has a characteristic that the accuracy of the estimated position becomes higher as the received power value becomes higher. Therefore, it is also effective to correct the position estimated by the sensor information only when the pedestrian passes near the anchor node. This method also has the advantage of simplifying the position estimation process using radio waves.

  FIG. 12 is an explanatory diagram illustrating an overview of the position estimation process according to the modified embodiment.

  13 and 14 are flowcharts showing the operation of the position estimation processing according to the modified embodiment. FIG. 13 is a flowchart showing the processing operation when correcting the angle of the estimated position based on the sensor information. FIG. 14 is a flowchart showing a processing operation when the scale of the estimated position based on the sensor information is modified.

  In FIG. 12, it is assumed that the pedestrian passes near the anchor node 2-1 and further passes near the anchor node 2-3.

  At this time, in the position estimation unit 11 of the target node 1, the radio wave position estimation unit 112 determines that the radio wave received from the anchor node 2 has a high received power value, that is, if the received power value of the received radio wave exceeds the threshold value. This is the reset point for resetting the estimated position by. That is, when the received power value exceeds the threshold, the radio wave position estimation unit 112 resets a reset function for resetting the estimated position by radio waves and a timer for checking whether or not a beacon signal is received within a predetermined unit time. A timer reset function may be provided.

  For example, when a pedestrian passes near the anchor node 2-1 and the received power value exceeds the threshold value, the radio wave position estimation unit 112 resets the estimated position and also resets the timer for measuring the unit time. Here, the time when the target node 1 passes the anchor node 2-1 (that is, the time when the estimated position is reset) is set as the previous reset point.

  Next, when the pedestrian passes near the anchor node 2-3 and the received power value exceeds the threshold value, the radio wave position estimation unit 112 resets the estimated position and the timer (step S301). The time when the current target node 1 passes through the anchor node 2-1 (that is, the time when the estimated position is reset) is set as the current reset point.

  Here, the positions of the anchor node 2-1 and the anchor node 2-3 are known. Therefore, the radio wave position estimation unit 112 calculates the vector R based on the coordinates (position information) of the previous reset point and the coordinates (position information) of the current reset point (current position) (step S302). That is, the radio wave position estimation unit 112 has a relative position change information calculation unit that calculates position change information from the relative position between the coordinates of the previous reset point and the coordinates of the current reset point.

  In this way, when the received power value is sufficiently high, the radio wave position estimation unit resets the estimated position, and the vector R is calculated based on the relative coordinates between the position information of the previous reset point and the position information of the current reset point. The calculation can reduce the load of position estimation processing by radio waves.

  (C-2) In the above-described first embodiment, the case where the error in the angle of the estimated position based on the sensor information is corrected, and in the second embodiment, the case where the error in the scale of the estimated position based on the sensor information is corrected, respectively. explained. You may make it combine 1st Embodiment and 2nd Embodiment. That is, both the angle of the estimated position and the error of the scale based on the sensor information may be corrected at the same time.

  (C-3) In the above-described first embodiment, the geomagnetic sensor is used, but the geomagnetic sensor may be omitted. In that case, if arbitrary position information is given as a value obtained from the output of the geomagnetic sensor and the position is updated by the estimated position by radio waves at the above-mentioned reset reset timing, the same operation can be performed thereafter. The “absolute azimuth information” described in the claims is not limited to the absolute azimuth information obtained by the geomagnetic sensor, but may be set as information indicating the absolute azimuth of the receiving node.

  (C-4) The "first position change information" described in the claims includes the vector P described in the first and second embodiments, and the "second position change information" is , The vector R described in the first and second embodiments is included.

  10... Communication system, 1... Target node, 11... Position estimation part, 111... Sensor position estimation part, 112... Radio wave position estimation part, 113... Integrated position estimation part, 12... Acceleration sensor, 13... Gyro sensor, 14... Geomagnetism Sensor, 15... Wireless unit, 151... Power measuring unit, 2... Anchor node, 21... Wireless unit, 211... Beacon transmitting unit, 17 and 22... Position display unit.

Claims (7)

In a position estimation system that estimates the position of the receiving node that receives the radio waves transmitted from the transmitting node,
The first estimated position of the receiving node is estimated based on the state change information and the absolute azimuth information detected by the inertial sensor provided in the receiving node, and at the same time, the first estimated position before the predetermined number in the past. Inertia for obtaining the first position change information indicating the time series change of the first estimated position , which is vector information based on the difference between the moving average value of the estimated position and the current moving average value at the first estimated position. Position estimation means,
The second estimated position of the receiving node is estimated based on the intensity of the radio wave received from the transmitting node at the receiving node, and the moving average value of the second estimated position before a predetermined number of past times and the present A radio wave position estimating means for obtaining second position change information indicating time series change of the second estimated position , which is vector information based on a difference from the moving average value at the second estimated position,
Position correction means for comparing the first position change information and the second position change information and correcting the present first estimated position and the past first estimated position according to the comparison result. equipped with a door,
It is characterized in that the position correcting means rotates and corrects the first estimated position according to an angle between the direction of the first position change information and the direction of the second position change information. Position estimation system. When the difference angle between the direction of the first position change information and the direction of the second position change information exceeds a threshold value, the position correction means gives the difference angle to the inertial position estimation means,
The position estimation according to claim 1 , wherein the inertial position estimation means corrects the estimation error between the present first estimated position and the past first estimated position using the difference angle. system. When the distance difference value between the distance of the first position change information and the distance of the second position change information exceeds a threshold value, the position correction means uses the distance difference value to change the first position change. The position estimation system according to claim 1 or 2 , wherein the distance of information is corrected. The radio wave position estimation means,
An estimated position reset unit that resets the second estimated position when the intensity of the received radio wave exceeds a threshold,
From the relative position between the estimated position and the estimated position of the current reset point of the last reset point of claim 1 to 3, characterized in that it comprises a relative position change information calculation unit for calculating the second position change information The position estimation system according to any one. In a position estimation device that estimates the position of the receiving node that receives the radio waves transmitted from the transmitting node,
The first estimated position of the receiving node is estimated based on the state change information and the absolute azimuth information detected by the inertial sensor provided in the receiving node, and at the same time, the first estimated position before the predetermined number in the past. Inertia for obtaining the first position change information indicating the time series change of the first estimated position , which is vector information based on the difference between the moving average value of the estimated position and the current moving average value at the first estimated position. Position estimation means,
The second estimated position of the receiving node is estimated based on the intensity of the radio wave received from the transmitting node at the receiving node, and the moving average value of the second estimated position before a predetermined number of past times and the present A radio wave position estimating means for obtaining second position change information indicating time series change of the second estimated position , which is vector information based on a difference from the moving average value at the second estimated position,
Position correction means for comparing the first position change information and the second position change information and correcting the present first estimated position and the past first estimated position according to the comparison result. equipped with a door,
The position estimating device rotates and corrects the first estimated position according to an angle between the direction of the first position change information and the direction of the second position change information. .. In the position estimation method for estimating the position of the receiving node that receives the radio wave transmitted from the transmitting node,
The first estimated position of the receiving node is estimated based on the state change information and the absolute azimuth information detected by the inertial sensor provided in the receiving node, and at the same time, the first estimated position before the predetermined number in the past. First position change information indicating a time series change of the first estimated position , which is vector information based on a difference between the moving average value of the estimated positions and the current moving average value of the first estimated position,
The second estimated position of the receiving node is estimated based on the intensity of the radio wave received from the transmitting node at the receiving node, and the moving average value of the second estimated position before a predetermined number in the past and the current The second position change information indicating the time-series change of the second estimated position , which is vector information based on the difference from the moving average value at the second estimated position, is obtained.
The first position change information and the second position change information are compared, and the current first estimated position and the past first estimated position are corrected according to the comparison result,
A position estimation method comprising rotating and correcting the first estimated position according to an angle between the direction of the first position change information and the direction of the second position change information . In the position estimation program that estimates the position of the receiving node that receives the radio waves transmitted from the transmitting node,
Computer,
The first estimated position of the receiving node is estimated based on the state change information and the absolute azimuth information detected by the inertial sensor provided in the receiving node, and at the same time, the first estimated position before the predetermined number in the past. Inertia for obtaining the first position change information indicating the time series change of the first estimated position , which is vector information based on the difference between the moving average value of the estimated position and the current moving average value at the first estimated position. Position estimation means,
The second estimated position of the receiving node is estimated based on the intensity of the radio wave received from the transmitting node at the receiving node, and the moving average value of the second estimated position before a predetermined number of past times and the present A radio wave position estimating means for obtaining second position change information indicating time series change of the second estimated position , which is vector information based on a difference from the moving average value at the second estimated position,
Position correction means for comparing the first position change information and the second position change information and correcting the present first estimated position and the past first estimated position according to the comparison result. And let it work ,
A position estimating program, wherein the position correcting means rotates and corrects the first estimated position according to an angle between the direction of the first position change information and the direction of the second position change information. ..

Images