JP3968429B2 - Position information processing device - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a position information processing apparatus for acquiring the absolute position of a measurement target (a small device such as a mobile phone or a PDA, a human body, etc.) with a small, lightweight, and inexpensive sensor group that can be attached to the measurement target. Is.
[0002]
[Prior art]
Conventionally, GPS (Global Positioning System) has been used as a position information acquisition device. This is a technology that receives signals from GPS satellites and acquires the current absolute position. In addition, as a technology related to the position information acquisition device, it is transmitted from an infrared beacon or radio beacon installed in the environment. Technology to acquire the current absolute position by comparing the image obtained from the camera or the like with each image in the image group database registered in advance. It has been.
[0003]
In addition, a technology that acquires the relative movement vector from the reference position based on the speed and direction of the wheel and an inertial sensor (acceleration sensor, angular velocity sensor, tilt sensor) are combined to integrate the measurement results. A technique for acquiring information, a technique for measuring a movement amount by detecting a pedestrian motion, and a technique for acquiring position information by measuring a moving direction by a magnetic compass or the like are known.
[Patent Document 1]
US Pat. No. 5,583,795
[Patent Document 2]
Japanese Patent Laid-Open No. 11-211479
[Patent Document 3]
Japanese Patent No. 3038452
[Patent Document 4]
JP-A-8-178681
[Non-Patent Document 1]
"Robust measurement of human daily activities using ultrasonic 3D tags-Robust position estimation based on redundant sensor information-" -(4), 2002)
[Non-Patent Document 2]
"Wearable annotation presentation system using infrared beacons and step counting"
(Science Technical Report, PRMU2002-68, pp. 71-76, 2002)
[Non-Patent Document 3]
"User View-Based Position and Orientation Acquisition Method for Wearable Augmented Reality System" (The Virtual Reality Society of Japan, vol. 7, no. 2, pp. 139-150, 2002)
[Non-Patent Document 4]
“Constellation: A Wide-range wireless motion-tracking system for augmented reality and virtual set applications,” (SIGGRAPH 1998, pp. 371-378, 1998)
[Non-Patent Document 5]
"New version Kalman filter" (Asakura Shoten, 1999)
[Non-Patent Document 6]
“Introduction to Random Signals and Applied Kalman Filtering (Third Edition)”, (John Wiley & Sons.)
[Non-Patent Document 7]
“Incremental Motion-Based Location Recognition,” (Proceedings of International Symposium on Wearable Computers (ISWC2001), pp. 123-130, 2001)
[0004]
[Problems to be solved]
The conventional technology has a problem that consideration is lacking from the viewpoint of efficiently obtaining the absolute position of the measurement target by combining the device that measures the absolute position and the device that measures the relative position. This is a problem that the absolute position estimated based on the sensing device that observes the external state is not always obtainable, but lacks consideration for this purpose.
[0005]
For example, when using GPS, the absolute position cannot be obtained in an environment where signals from satellites are blocked. Further, in the position / orientation acquisition method using an image group as a database (see Non-Patent Document 3), an absolute position cannot be acquired when an image photographed at the current position is not registered in the database. In order to obtain an absolute position, such a problem often accompanies.
[0006]
In many cases, the relative position movement estimated based on the sensing device that observes the internal state can always be acquired. For example, an inertial navigation device (IMU) composed of inertial sensor groups does not require any input from the external environment, and can calculate a relative position movement vector from a reference position.
[0007]
The present invention has been made in view of the above problems, and the present invention efficiently combines an absolute position acquisition device that can be used intermittently and a relative position acquisition device that can be used constantly. Another object of the present invention is to provide a position information processing apparatus capable of obtaining position information with high accuracy.
[0008]
[Means for Solving the Problems]
The position information processing apparatus of the present invention processes position information using the following basic principle. That is, in general, the absolute position calculated based on the observation result of the sensing device or the like is accompanied by certain uncertainties, and the nature of the absolute position varies depending on the sensing device and calculation method used. Note that here, a sensing device in which observation results are observed discretely is targeted.
[0009]
Therefore, in the present invention, taking into account the probability distribution of the observation result observed by the sensing device, the probability distribution is given to the absolute position estimation result calculated based on the observation result, Based on the probability distribution, the absolute position estimation process executed at the next discrete time is changed.
[0010]
In the position information processing apparatus of the present invention, an information database apparatus is provided which becomes prior knowledge about the external environment within a range in which the measurement target can move. This information database apparatus can use the absolute position as a key for information retrieval. In the position information processing, the current absolute position is based on the estimation result of the current absolute position and the probability distribution. By defining the candidate range, the information corresponding to the current state is searched and referred to, so that the absolute position estimation process can be efficiently performed.
[0011]
  Specifically, the position information processing apparatus of the present invention includes, as a first aspect, an external state sensing device that observes an external state of a measurement target, an internal state sensing device that observes the internal state of the measurement target, A position storage device that stores the estimation result of the current absolute position of the measurement target, a distribution storage device that stores a probability distribution of the estimation result of the current absolute position of the measurement target, and a range in which the measurement target can move Information on the external environment of theSearchInformation database device equipped with a mechanism for retrieving stored information by retrieving stored information as an external device and an external state sensing deviceWhen the observation result of the external state is obtained byThe current absolute position estimation result read from the position storage deviceandThe current absolute position candidate range determined by the probability distribution of the current absolute position estimation result read from the distribution storage device is detected.SearchAs,The verification retrieved from the information database device.SearchAnd information applicable to, Check the observation results obtained by the external state sensing deviceThe latest absolute position of the measurement targetofAbsolute position to obtain the estimation result and the probability distribution of the estimation resultdistributionBased on an observation result of the acquisition device, the internal state sensing device, and a probability distribution of the observation result, a relative position movement vector of the measurement targetRelative position ofAlways obtain the estimation result and the probability distribution of the estimation resultDorelative positiondistributionAn acquisition device;Acquired by absolute position distribution acquisition deviceAbsolute position estimation result and the aboveAcquired by the relative position distribution acquisition deviceThe position estimation result merge / update device that merges the relative position estimation results and updates / acquires the current absolute position estimation results; andAcquired by absolute position distribution acquisition deviceThe distribution of certainty of the absolute position estimation result and the aboveAcquired by the relative position distribution acquisition deviceProbability distribution of relative position estimation resultsAccording to the Kalman filter frameworkA merged distribution update device that updates the probability distribution of the current absolute position estimation result and the latest absolute position acquired by the position estimation result merge update deviceofA storage processing device that stores an estimation result in the position storage device, and stores a probability distribution of the latest estimation result acquired by the distribution merge update device in the distribution storage device. .
[0012]
  The position information processing apparatus of the present invention provides the absolute position as a second aspect.distributionAcquisition device,in frontThe absolute position candidate range determined by the absolute position estimation result read from the position storage device and the probability distribution of the absolute position estimation result read from the distribution storage device is detected.SearchAsFrom the information database deviceInformation to be retrievedAnd observation results obtained from the external state sensing deviceThe collation processing device acquires the absolute position estimation result and the probability distribution of the estimation result by collatingincludingConstitutionWhenIt is characterized by that.
[0013]
  According to a third aspect of the position information processing apparatus of the present invention, the external state sensing device to be measured includes an image photographing device, and the information database device identifies an image group whose photographing position is known as an external environment. And is determined based on the current absolute position estimation result and the distribution of the estimation resultBy search keyIncluding an image group database device having a mechanism for searching for an image group including a shooting position within a current absolute position candidate range and extracting the corresponding image group,Obtain absolute position distributionThe apparatus includes a current absolute position candidate determined based on a current absolute position estimation result read from the position storage device and a probability distribution of the current absolute position estimation result read from the distribution storage device. RangeSearchThe information database device as a keyTheSearchIn search resultsWhen the corresponding image group exists, the feature amount of each image of the image group or each image of the image group, and the feature amount of the image acquired from the image capturing device or the image acquired from the image capturing device Image collation device for collationIt is composed includingIt is characterized by that.
[0014]
  As a fourth aspect, the position information processing apparatus of the present invention targets a human walking as the measurement object, and as an internal state sensing apparatus for measuring relative position movement due to a walking motion that is a pedestrian moving means, Using the inertial sensor group and the magnetic sensor, the relative positiondistributionBased on the sensor output of the inertial sensor group and the magnetic sensor, and the probability distribution of the observation results obtained by the inertial sensor group and the magnetic sensor, the acquisition device detects the walking motion of the pedestrian, estimates the stride, and detects the moving direction. An arithmetic device is provided for calculating a position vector and calculating a probability distribution of the relative position vector.
[0015]
  Moreover, the position information processing apparatus of the present invention has, as a fifth aspect,As the external state sensing device,A signal receiving device that receives a signal from a signal generating device that transmits a unique signal corresponding to an absolute position installed on the external environment side, and the information database device includes a unique signal and an absolute position A database device that stores correspondence as data,Absolute positiondistributionWhen the signal is received by the signal receiving device, the acquiring device extracts the absolute position corresponding to the received signal from the database device and acquires the estimation result of the current absolute position, such as the strength of the received signal Obtain the probability distribution of the estimation result based on the propertyRukoIt is characterized by.
[0016]
DETAILED DESCRIPTION OF THE INVENTION
FIG. 1 is a block diagram showing a first embodiment of the present invention. In the first embodiment, the measurement target is, for example, a person, a bicycle, a car, an aircraft, or the like. In addition, as an example of measurement data of the external state of the measurement target, an image acquired by a CCD camera or the like, an image sequence, a device that receives a signal transmitted from an infrared beacon or a radio beacon installed on the external environment side The acquired signal, audio data acquired by a microphone or the like, temperature acquired by a temperature sensor, magnetic vector acquired by a magnetic sensor, light amount acquired by a light amount sensor, or a combination thereof.
[0017]
The external state sensing device 102 is a device that acquires the external state of the measurement target as described above. Specifically, the external state sensing device 102 is an image capturing device such as a CCD camera, a sound acquisition device such as a microphone, and an infrared ray installed in an external environment. Receivers for signals transmitted from beacons or radio beacons (see Non-Patent Document 2), signal transponders installed in the external environment (devices that send back signals representing their own unique numbers when receiving standard signals), and their groups Signal transmitter / receiver (see Non-Patent Document 4), mobile phone or PHS that receives base station information in the vicinity of the external environment, a device that transmits ultrasonic signals, and a device that receives radio signals that are received by the external environment A device that includes a group of devices and transmits a three-dimensional position measurement result by these receiver groups by radio signal (see Non-Patent Document 1), a light amount sensor, and the like Used.
[0018]
  The information database device 103 is a database device that stores information about the external environment to which the measurement object can move,SearchAs a key, it is a device provided with a mechanism for extracting corresponding information from stored information. This is configured by a data processing device that stores necessary information in a magnetic storage device (such as a hard disk), an optical storage device (such as a CD), or a magneto-optical storage device, and reads out the information. Combination of microprocessor (CPU) and general-purpose computer programsStepsConsists of. As information stored here, for example, an image captured in advance at each point in the environment and a correspondence table of absolute positions where the image was captured, an infrared beacon or a radio beacon disposed in the external environment is transmitted. Information corresponding to the absolute position or its range as a key, with the basic unit of signal data and absolute position correspondence table where the signal can be received, base station information received by mobile phone or PHS and absolute position correspondence table, etc. Is stored in a format that can be retrieved.
[0019]
The current absolute position storage device 105 stores the estimation result of the current absolute position, and the absolute position probability distribution storage device 106 stores the probability distribution of the estimation result. It is. As an example of the probability distribution of the estimation result, there are an absolute position vector and a variance / covariance matrix when a single normal distribution is assumed. Alternatively, there is an expression method in which a non-negative number representing the probability is stored in each element in a two-dimensional or three-dimensional table in which absolute positions are discretized. Use such data.
[0020]
  Absolute position estimation result and estimation result probability distribution acquisition device(Absolute position distribution acquisition device)101 is an observation result acquired by the external state sensing deviceFromThe estimation result of the current absolute position read from the storage device 105andBased on the probability distribution of the estimation result read from the storage device 106, the latest absolute position estimation result (output 1) and the probability distribution of the estimation result (output 2) are calculated, acquired, and output. Device. In the calculation process, information stored in the information database apparatus 103 is referred to through the following process.
[0021]
  The absolute position candidate range can be determined based on the current absolute position estimation result read from the storage device 105 and the probability distribution of the estimation result read from the storage device 106. The candidate range is detected.SearchThe data corresponding to the search key is searched from the information database device 103, and if it exists, the corresponding data group is extracted. These corresponding data groups are data corresponding to the current absolute position candidate range, and by considering the corresponding data group in the calculation process of the latest absolute position estimation result and the estimation result, more appropriate and accurate The estimation result and its distribution can be acquired. When there is no corresponding data, the distribution acquisition apparatus 101 outputs data outputs indicating that a valid output cannot be obtained to outputs 1 and 2.
[0022]
  As a means for reflecting the current absolute position estimation result distribution in the latest absolute position estimation result and the estimation result distribution, a Kalman filter framework (Non-Patent Document 5 or Non-Patent Document 5) regards the probability distribution as a normal distribution. (Ref. 6)The
[0023]
  When a plurality of external state sensing devices 102 are provided here, as will be described later, the relationship between the distribution acquisition device 101, the external state sensing device 102, and the information database device 103 is a plurality of configurations as shown in FIG. ofDistribution acquisition device (Absolute positiondistributionAcquisition device)101 (N in the case of FIG. 12) output results are merged, and a final output absolute position estimation result (output 1) and a probability distribution (output 2) of the estimation result are generated, An absolute position estimation result to be output and a probability merger / update processor 111 are provided.
[0024]
  The internal state sensing device 109 is a device that observes the internal state of the measurement target. Examples of the internal state here include a velocity vector, an acceleration vector, an angular velocity vector, an angular acceleration vector, a gravitational acceleration vector (tilt angle), a magnetic vector, and the like to be measured. When the measurement target is a human, the presence / absence of the motion taken by the human and the magnitude thereof (for example, walking motion and stride) are listed, and the internal state sensing device 109 observes any one or a combination thereof. It is a sensor group, and it is possible to simultaneously obtain a probability distribution with respect to the observation result. Relative position estimation result and estimation result probability acquisition device(Relative position distribution acquisition device)104 calculates the relative movement vector of the measurement target based on the observation result acquired by the internal state sensing device 109 and the probability distribution of the observation result, and outputs it as the relative position estimation result (output 1). At the same time, it is a device that calculates and outputs a probability distribution (output 2) of the estimation result.
[0025]
As a method of calculating the relative movement vector, the acceleration vector at each discrete time is integrated by double integration, the speed vector is integrated and integrated, and when the measurement target is a human, the walking motion is human Therefore, a relative movement vector due to human walking can be calculated by a walking motion detection process, a stride estimation process, and a movement direction estimation process (see Patent Document 1). The human moving direction is estimated using a method of acquiring a geomagnetic vector using a magnetic sensor, or a method of measuring a relative change in posture angle from a reference posture angle (see Patent Document 2 or Patent Document 3). The direction of movement is estimated.
[0026]
When a plurality of internal state sensing devices are provided here, as will be described later, the relationship between the internal state sensing device 109 and the relative position estimation result and the estimation result probability distribution acquisition device 104 is shown in FIG. As shown in the configuration, the apparatus 107 that acquires the latest current position by combining the absolute position estimation result and the relative position estimation result is the output 1 from the distribution acquisition apparatus 101 (the latest absolute position estimation result). The output 1 (relative position estimation result) from the distribution acquisition device 104 is merged, and the latest absolute position estimation result is updated and output.
[0027]
The apparatus 108 that acquires the likelihood distribution of the latest absolute position estimation result by combining the probability distribution of the absolute position estimation result and the reliability distribution of the relative position estimation result is an output 2 from the distribution acquisition apparatus 101. (The probability distribution of the latest absolute position estimation result) and the output 2 from the distribution acquisition device 104 (the probability distribution of the relative position estimation result) are merged, and the reliability of the latest absolute position estimation result is calculated. Calculate and output the distribution. The storage processing device 110 is a device that performs processing for storing the output of the device 107 and the output of the device 108 in the storage device 105 and the storage device 106, respectively.
[0028]
Signal processing by each device in the above embodiment is also realized by program processing of a general purpose computer (CPU). Next, the flow of processing by such program processing will be described.
[0029]
The flowchart shown in FIG. 2 is a flowchart showing the flow of processing corresponding to the first embodiment. Here, in the process, the observation result is acquired from the external state sensing device 102 in step S101, the estimation result of the current absolute position is acquired from the current absolute position storage device 105 in step S102, and the accuracy of the absolute position is determined in step S103. A probability distribution of the estimation result is acquired from the distribution storage device 106. In step S104, the current absolute position candidate range is determined based on the absolute position estimation result acquired in step S102 and the probability distribution of the estimation result acquired in step S103.
[0030]
In step S105, corresponding information is extracted from the information database apparatus 103 using the absolute position candidate range determined in step S104 as a key. In the next step S106, the observation result obtained in step S101, the current position estimation result obtained in step S102, the probability distribution of the estimation result obtained in step S103, and the step S105 are extracted. Based on the obtained information, the latest absolute position estimation result and the probability distribution of the estimation result are calculated.
[0031]
Next, in step S107, an observation result and a probability distribution of the observation result are acquired from the internal state sensing device 109, and in the next step S108, the observation result and the probability distribution obtained in the previous step S107 are acquired. Based on the above, the estimation result of the relative position (that is, the movement vector of the measurement target) and its probability distribution are calculated. Next, in step S109, the latest absolute position estimation result obtained in step S106 and the relative position estimation result obtained in step S108 are merged, and the latest absolute position estimation result is updated. In step S110, the latest absolute position estimation result obtained in step S106 is merged with the probability distribution of the latest absolute position estimation result and the relative position estimation result obtained in step S108. Update the probability distribution of the estimation result. In step S111, the latest estimation result of the current position updated in step S109 is stored in the storage device 105 using the storage processing device 110, and the likelihood of the estimation result of the latest absolute position updated in step S110 is confirmed. The distribution is stored in the storage device 106 using the storage processing device 110. In the next step S112, the latest absolute position estimation result updated in step S109 is output as the output 1 of this apparatus. In step S113, the probability distribution of the latest absolute position estimation result updated in step S110 is output as output 2 of the present apparatus.
[0032]
FIG. 3 is a block diagram showing a second embodiment of the present invention. In the configuration of the first embodiment (FIG. 1), the second embodiment replaces the device for absolute position acquisition (the absolute position estimation result and the estimation result probability distribution acquisition device 101), and performs the matching process. The apparatus 201 is provided. As an example of the collation process that is a process performed by the collation processing device 201, collation processing between single data at discrete times observed by the external state sensing device 102 and model pattern data acquired in advance, the external state sensing device 102 Is a collation process (for example, a matching process based on dynamic programming) between time series data at discrete times observed by the model time series pattern data registered in advance.
[0033]
Signal processing by each device in the above embodiment is also realized by program processing of a general purpose computer (CPU). Next, the flow of processing by such program processing will be described.
[0034]
  The flowchart shown in FIG. 4 is a flowchart showing the flow of processing corresponding to the second embodiment. In this process, the observation result is acquired from the external state sensing device 102 in step S201, the current absolute position is acquired from the current absolute position storage device 105 in step S202, and the probability distribution of the absolute position is determined in step S203. The probability distribution of the absolute position is acquired from the storage device 106. Next, in step S204, the current absolute position candidate range is determined based on the absolute position estimation result acquired in step S202 and the probability distribution of the absolute position acquired in step S203. In step S205, the current absolute position candidate range determined in step S204 is detected.SearchAs a result, the corresponding data group is retrieved from the database apparatus 103 and extracted. In step S206, the observation result obtained in step S201 and the corresponding data group obtained in step S205 are collated, and the collation result and the current position estimation result obtained in step S202 are obtained in step S203. Based on the probability distribution of the estimation result of the current position, the latest absolute position estimation result and the probability distribution of the estimation result are calculated using the apparatus 201. In step S207, an observation result and its probability distribution are acquired from 109, and in step S208, the relative position estimation result and the probability of the estimation result are obtained based on the observation result acquired in step S207 and its probability distribution. Is calculated using the device 104. In step S209, the latest absolute position estimation result calculated in step S206 and the relative position estimation result calculated in step S208 are merged, and the latest absolute position estimation result is calculated using the apparatus 107. Update. In step S210, the probability distribution of the latest absolute position estimation result calculated in step S206 and the probability distribution of the relative position estimation result calculated in step S208 are merged to estimate the latest absolute position. The probability distribution of the result is calculated using the device 108. In step S211, the latest absolute position estimation result calculated in step S209 is stored in the storage device 105 using the storage processing device 110, and the latest absolute position estimation result distribution calculated in step S210 is stored. The data is stored in the storage device 106 using the device 110.
[0035]
In step S212, the latest absolute position estimation result calculated in step S209 is output as output 1, and in step S213, the latest absolute position estimation result distribution calculated in step S210 is output as output 2. To do.
[0036]
FIG. 5 is a block diagram showing a third embodiment of the present invention. The third embodiment includes an image capturing device 301 instead of the external state sensing device 102 having the configuration of the first embodiment (FIG. 1), and an apparatus for obtaining an absolute position (absolute position estimation results and estimation results). Instead of the probability distribution acquisition device 101), an image collating device 302 is provided. Therefore, an image group database device 303 is provided instead of the information database device 103 having the configuration of the first embodiment (FIG. 1).
[0037]
As the image photographing device 301, for example, a CCD camera or a CMOS camera is used. The image matching device 302 calculates a geometric transformation parameter between images based on the alignment processing between the images, and when the luminance difference between the images becomes equal to or less than a certain threshold using the transformation parameter, the two images are displayed. The images are collated using a method that is regarded as collated (see Non-Patent Document 3). It is implemented by a microprocessor (CPU) and a processing program of the method. Each image is associated with a corresponding absolute position, and the absolute position is obtained by this image collation. Therefore, the image group database device 303 here is a storage device (or storage medium) that stores each image of the image group in association with the absolute position where each image was taken. When the absolute position and its range are given as search keys, the database apparatus includes a mechanism for searching for and retrieving an image group corresponding to the shooting position. A storage medium (such as a magnetic disk, an optical disk, or a magneto-optical disk) that stores the contents of a database of images to be stored, and a general-purpose computer program that retrieves the contents and retrieves the information.
[0038]
FIG. 6 is a block diagram showing a fourth embodiment of the present invention. In the configuration of this embodiment, a response to dead-recking based on human walking motion is performed in the relative position vector estimation process. For this reason, it replaces with the internal state sensing apparatus 109 of the structure of 1st Example (FIG. 1), and is provided with the inertial sensor group and the magnetic sensor 401, and also acquires the distribution of the estimation result of a relative position, and the probability of an estimation result. Instead of the device 104, a device 402 is provided that acquires a relative position estimation result and a probability distribution based on walking motion detection.
[0039]
As a sensor for detecting a human walking motion, the human walking motion is detected with high accuracy by detecting the acceleration component in the vertical direction and the acceleration component in the traveling direction, and collating the time series pattern. For this reason, a magnetic sensor for detecting a geomagnetic vector is used as means for detecting the direction of human walking movement.
[0040]
In this embodiment, as the internal state sensing device 109, that is, instead of the internal state sensing device 109, an inertial sensor group (for example, an acceleration sensor (three axes) and a gyro sensor (three axes)) and a magnetic sensor that measures a geomagnetic vector. (3-axis) is used. All of these sensors have known error distribution characteristics (can be approximated using a normal distribution), and therefore, a probability distribution of observation results can be obtained. In addition, as described above, instead of the relative position estimation result and the estimation result probability distribution acquisition device 104, detection of walking motion based on the observation results of the inertial sensor group and the magnetic sensor and the probability distribution thereof. And a device 402 for processing the estimation of the stride and the detection of the moving direction. Based on the observation results of the inertial sensor group and the magnetic sensor and the distribution of the certainty thereof, a method for detecting the walking motion, estimating the stride, and detecting the moving direction is known and is not directly related to the present invention. Detailed description is omitted (see Patent Document 1 and Non-Patent Document 7).
[0041]
FIG. 7 is a block diagram showing a fifth embodiment of the present invention, and FIG. 8 is a flowchart showing a flow of processing corresponding to the fifth embodiment. The fifth embodiment shown in FIG. 7 is a combination of the elements of the third embodiment shown in FIG. 5 and the elements of the fourth embodiment shown in FIG.
[0042]
The signal processing by each device in the above embodiment is also realized by program processing by a general purpose computer (CPU) as shown in the flowchart of FIG.
[0043]
  FIG. 9 is a block diagram showing a sixth embodiment of the present invention. In this embodiment, in order to use the beacon signal generator 503 that transmits a beacon signal installed on the external environment side (a signal unique to the installed absolute position).ofIt has a configuration. Therefore, instead of the external state sensing device 102, a beacon signal receiving device 502 that receives a beacon signal is provided, and instead of the absolute position estimation result and the estimation result probability distribution acquisition device 101, a beacon signal matching device and a beacon The signal strength processing device 501 is provided. Moreover, it replaces with the information database apparatus 103, and is set as the structure provided with the beacon signal information database apparatus 504. FIG. The beacon signal information database device 504 stores the correspondence between the beacon signal and the absolute position to which the signal corresponds, and stores the beacon signal.SearchIt has a mechanism for searching for and extracting the absolute position corresponding to the key.
[0044]
The beacon signal verification processing and beacon signal strength processing device 502 searches the beacon signal information database device 504 using the beacon signal transmitted from the beacon signal generating device 503 and received by the beacon signal receiving device 502 as a search key. Then, the current absolute position estimation result is acquired by taking out the corresponding absolute position, and the probability distribution of the estimation result is calculated from the intensity and direction of the received beacon signal.
[0045]
Signal processing by each device in the above embodiment is also realized by program processing of a general purpose computer (CPU). Next, the flow of processing by such program processing will be described.
[0046]
The flowchart shown in FIG. 10 is a flowchart showing the flow of processing corresponding to the sixth embodiment. In this process, first, in step S501, a beacon signal is acquired from the beacon signal receiving apparatus 502 (limited to the case where it exists). In step S502, the beacon signal information database device 504 is searched using the received beacon signal as a search key, and the absolute position corresponding to the beacon signal is extracted to obtain the absolute position estimation result.
[0047]
In step S503, the probability distribution of the absolute position estimation result obtained in step S502 is calculated based on the properties such as the strength and direction of the beacon signal received in step S501. Next, in step S504, an observation result and its probability distribution are acquired from the internal state sensing device. In step S505, the estimation result of the relative position and the probability distribution of the estimation result are calculated based on the observation result acquired in step S504 and the probability distribution of the observation result. In step S506, the absolute position estimation result obtained in step S502 and the relative position estimation result obtained in step S505 are merged to calculate the latest absolute position estimation result.
[0048]
In the next step S507, the latest absolute position is obtained by combining the probability distribution of the absolute position estimation result obtained in step S503 and the probability distribution of the relative position estimation result obtained in step S505. The probability distribution of the estimation result is calculated. In step S508, the absolute position estimation result calculated in step S506 is stored in the storage device 105 using the storage processing device 110, and the probability distribution of the absolute position estimation result acquired in step S507 is stored. The data is stored in the storage device 106 using the processing device 110.
[0049]
In step S509, the absolute position estimation result calculated in step S506 is output as the output 1 of the apparatus, and in step S510, the probability distribution of the absolute position estimation result calculated in step S507 is output from the apparatus. 2 is output.
[0050]
FIG. 11 is a block diagram showing a seventh embodiment of the present invention. The seventh embodiment shown in FIG. 11 is a combination of the elements of the fourth embodiment shown in FIG. 6 and the elements of the sixth embodiment shown in FIG.
[0051]
【The invention's effect】
  As described above, by using the configuration of each embodiment of the position information processing apparatus according to the present invention, an absolute position acquisition process that generally requires high calculation cost and is not always guaranteed to be executable. It can be executed efficiently. Specifically, the absolute position candidate range determined from the absolute position estimation result and the probability distribution is detected.SearchSearch the corresponding data from the information database deviceShiWhen the corresponding data does not exist, the absolute position acquisition process is not executed and the corresponding data exists.By matching the observation data obtained by the external state sensing device with the corresponding data,Since it is only necessary to execute the absolute position acquisition based only on the corresponding data, the necessary calculation cost can be kept low.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a first embodiment of the present invention.
FIG. 2 is a flowchart showing a flow of processing corresponding to the first embodiment.
FIG. 3 is a block diagram showing a second embodiment of the present invention.
FIG. 4 is a flowchart showing a flow of processing corresponding to the second embodiment.
FIG. 5 is a block diagram showing a third embodiment of the present invention.
FIG. 6 is a block diagram showing a fourth embodiment of the present invention.
FIG. 7 is a block diagram showing a fifth embodiment of the present invention.
FIG. 8 is a flowchart showing a flow of processing corresponding to the fifth embodiment.
FIG. 9 is a block diagram showing a sixth embodiment of the present invention.
FIG. 10 is a flowchart showing a flow of processing corresponding to the sixth embodiment.
FIG. 11 is a block diagram showing a seventh embodiment of the present invention.
FIG. 12 is a diagram showing a configuration of a modified example in the case where a plurality of external state sensing devices are provided.
FIG. 13 is a diagram showing a configuration of a modification in the case where a plurality of internal state sensing devices are provided.
[Explanation of symbols]
101 ... Distribution acquisition device,
102 ... External state sensing device,
103. Information database device,
104 ... Distribution acquisition device,
105 ... current absolute position storage device,
106 ... Absolute position probability distribution storage device,
107 ... merger,
108 ... Probability acquisition device,
109 ... Internal state sensing device,
110. Storage processing device,

Claims (5)

An external state sensing device for observing the external state of the measurement target;
An internal state sensing device for observing the internal state of the measurement object;
A position storage device for storing a current absolute position estimation result of the measurement target;
A distribution storage device for storing a probability distribution of an estimation result of a current absolute position of the measurement target;
The measurement object stores information about the external environment of the range that can be moved, and searches the information stored with the candidate range of the absolute position as an analyzer Sakuki over, information database provided with a mechanism for taking out the relevant information Equipment,
When the observation result of an external condition has been obtained by the external condition sensing apparatus, the distribution of the probability of the position storage device from the current absolute position to be read estimation result and the distribution storage device from the current absolute position to be read estimation result matching the search Sakuki over the candidates range of the current absolute position, and information corresponding to said analyzing Sakuki over retrieved from the information database device, and a observations the external condition sensing device is acquired, which is determined by and an absolute position distribution acquiring unit that acquires the likelihood of distribution of the most recent absolute position of the measurement object estimation result and the estimation result,
Based on the observation result of the internal state sensing device and the probability distribution of the observation result, the relative position estimation result based on the relative position movement vector of the measurement target and the probability distribution of the estimation result are always obtained. A relative position distribution acquisition device,
Position estimation that updates and acquires the current absolute position estimation result by merging the absolute position estimation result acquired by the absolute position distribution acquisition apparatus and the relative position estimation result acquired by the relative position distribution acquisition apparatus Results merge update device,
The probability distribution of the absolute position estimation result acquired by the absolute position distribution acquisition device and the probability distribution of the relative position estimation result acquired by the relative position distribution acquisition device are merged by a Kalman filter framework , A distribution merge update device that updates the probability distribution of the estimation result of the current absolute position;
The location estimation result merging latest acquired by the update unit of the absolute position estimation result is stored in the position storage unit, the distribution storage the likelihood of distribution of the distribution merge latest acquired by the update device the estimated result A storage processing device for storing in the device;
A position information processing apparatus comprising: The position information processing apparatus according to claim 1,
The absolute position distribution obtaining unit, an estimation result of an absolute position is read from the previous SL position storage device, the candidate range of absolute positions determined by the probability of the distribution of the estimation result of the absolute position read from the distribution storage unit and information retrieved from the information database device as an analyzer Sakuki over and collates the observation results obtained from the external condition sensing apparatus, collating for obtaining the estimated result as likelihood distribution of the estimation result of the absolute position position information processing apparatus characterized by being configured to include a processor. The position information processing apparatus according to claim 1,
The external state sensing device to be measured includes an image capturing device,
The information database device stores a group of images whose shooting positions are known as information related to the external environment, and a current absolute position by a search key determined based on the estimation result of the current absolute position and the distribution of the estimation result. Including an image group database device having a mechanism for searching for an image group including a shooting position within a position candidate range and extracting the corresponding image group;
The absolute position distribution acquisition device is determined based on a current absolute position estimation result read from the position storage device and a probability distribution of a current absolute position estimation result read from the distribution storage device. The information database device is searched using the absolute position candidate range as a search key, and when there is an image group corresponding to the search result, each image in the image group or a feature amount of each image in the image group, and the image position information processing apparatus according to claim Rukoto is configured to include an image collation apparatus for collating a feature amount of an image obtained from the image or the image capturing apparatus is acquired from the imaging device. The position information processing apparatus according to claim 1,
Using the inertial sensor group and the magnetic sensor as an internal state sensing device for measuring a relative position movement due to a walking motion that is a pedestrian movement means, targeting a human walking as the measurement target,
The relative position distribution acquisition device includes a pedestrian's walking motion detection and stride estimation based on the sensor output of the inertial sensor group and the magnetic sensor, and the probability distribution of the observation result by the inertial sensor group and the magnetic sensor. A position information processing apparatus, comprising: an arithmetic unit that detects a relative position vector, calculates a relative position vector, and calculates a probability distribution of the relative position vector. The position information processing apparatus according to claim 1, further comprising:
As the external state sensing device, comprising a signal receiving device that receives a signal from a signal generating device that sends a unique signal corresponding to the absolute position installed on the external environment side,
The information database device is a database device that stores a correspondence between a unique signal and an absolute position as data ,
The absolute position distribution acquisition device, when a signal is received by the signal reception device, extracts an absolute position corresponding to the reception signal from the database device, acquires an estimation result of a current absolute position, and receives the reception signal. Obtain a probability distribution of the estimation result based on properties such as strength
Position information processing apparatus according to claim and this.

Images