JP2007155363A - Correction factor arithmetic unit and self-position recognition system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a correction factor arithmetic unit capable of speedily acquiring highly accurate distance correction factors with high reliability by sufficiently utilizing sequentially captured GPS information. <P>SOLUTION: A pair of pieces of GPS information, two pieces of the GPS information to be used to compute distance correction factors, are evaluated on the basis of distances traveled. Information having a greater distance traveled is more frequency used for computations. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention corrects a moving distance obtained from the pulse number information based on GPS information obtained from a GPS satellite and pulse number information obtained from a distance sensor that outputs a pulse signal corresponding to the amount of movement. The present invention relates to a correction coefficient calculation apparatus that calculates a distance correction coefficient, and also relates to a self-position recognition apparatus that calculates a movement locus of the self-position using the obtained distance correction coefficient and recognizes the self-position.

  As this type of correction coefficient calculation device, the applicant has proposed the technique disclosed in Patent Document 1. In the technique described in this document, when obtaining the distance correction coefficient using the latest GPS information and the previous GPS information, an acquisition unit that acquires pulse number information simultaneously with reception of the GPS information is provided. As a result, the distance correction coefficient can be obtained without causing a decrease in accuracy. Also, with this technique, since the sampling interval is short and the number of samplings can be increased, it is possible to obtain a distance correction coefficient with good accuracy without actually using GPS information with poor conditions.

Now, an outline of the flow of the movement distance calculation process, the distance correction coefficient determination process, and the distance correction coefficient calculation process in this type of correction coefficient calculation device will be described based on the description in Patent Document 1.
1 Travel Distance Calculation Processing As shown in FIG. 5 attached to the specification, first, a speed, time, direction, and status signal are acquired as GPS information from a GPS receiver in step S1, and simultaneously with GPS reception in step S2. A pulse signal (which is a pulse number signal) is acquired from the distance sensor, it is determined whether or not a distance correction coefficient is calculated in step S3, a distance correction coefficient is calculated in step S4, and the distance sensor is determined in step S5. The movement distance L is calculated by correcting the distance obtained by H.26 based on the distance correction coefficient.

  This movement distance L is calculated as L = k × Δpls × (distance per pulse). Here, k is a distance correction coefficient, Δpls is a pulse difference between the number of pulses from the distance sensor received this time (pls) and the number of pulses from the distance sensor received last time (pls1), and the distance per pulse is determined in advance. It is set as a default value for each vehicle. The distance per pulse is a movement distance corresponding to the unit pulse generation in a situation where the moving body is in a standard state.

  The movement distance calculation process is performed every time a GPS signal is received. That is, the reception interval of GPS information is a predetermined time interval, and generally a 1 second interval is used. Therefore, this movement distance calculation process is also executed at a 1 second interval.

Distance correction coefficient determination process This process is a distance correction coefficient determination process in step S3 of FIG. 5 attached to the above-described specification, and calculates GPS information that can be used to calculate the distance correction coefficient. It is processing. This determination has the same property as the determination called the first condition determination in the present invention. The determination condition is whether the speed between the two points is equal to or higher than a predetermined speed or not, and the angle change between the two points is the same. Whether or not the angle is within a predetermined angle, whether the speed difference between the two points is within the predetermined speed difference, whether the time difference between the two points is within the predetermined time difference, whether the two received points are the same dimension positioning information Or not.

Distance Correction Coefficient Calculation Process FIG. 7 attached to this specification is a diagram for explaining the flow of the distance correction coefficient calculation process in step S4 of FIG. 5. First, in step S1, the current GPS information is received. The time difference Δt between the two data is calculated from the time (t) and the time (t1) when the previous GPS information was received. Next, in step S2, the pulse difference Δpls between the two data is calculated from the number of pulses (pls) of the distance sensor when the current GPS information is received and the number of pulses (pls1) when the previous GPS information is received. To do.

Next, in step S3, the GPS movement distance LGPS between the two points is calculated by the following equation.
Lgps = (spd + spd1) / 2 × Δt
spd: GPS speed obtained this time spd1: GPS speed obtained last time That is, the GPS moving distance Lgps is calculated by multiplying the average value of speed obtained from GPS reception at two points by the time difference.

  Next, in step S4, the number of pulses p required to advance the unit distance D (for example, 100 m) from the calculated pulse difference (Δpls) and the GPS moving distance is calculated, and the value is incremented in the histogram. In step S5, when a certain amount of data is set in the histogram, the data is averaged, and the distance correction coefficient k is calculated from the following equation.

  k = (D / p) / (distance per pulse)

  Therefore, the method disclosed in this document is based on using the current GPS information received this time and the previous GPS information to calculate the distance correction coefficient.

The learning for obtaining the movement distance correction coefficient using such GPS information is conventionally performed after traveling a predetermined number of information, for example, 150 m.
That is, the travel distance correction coefficient obtained as a default value is used at the start of movement, and after that, a significant travel distance correction coefficient is obtained with GPS information accumulated to some extent, and the correction coefficient is used. The structure to be adopted is adopted.
Patent 2891403

When the above method is adopted, only the current and previous GPS information is used for the calculation of the distance correction coefficient. I can't say.
For such a problem, from the viewpoint of effectively using continuously captured GPS information, for example, the same processing as described above is performed using the current GPS information and the past GPS information a certain number of times ago. It is also conceivable to go to obtain the distance correction coefficient.

  In addition, when using two points, if a process similar to traveling on a horizontal surface is performed in the movement between the two points even though the region has a large difference in elevation, a relatively long gradient is obtained. In spite of moving, there is a problem that the moving distance is caught as a short distance and an incorrect distance correction coefficient is obtained.

  Furthermore, in the method disclosed in Patent Document 1, although the GPS information that is sequentially acquired is information having different error radii, it is handled as substantially equivalent information, and thus the accuracy of the distance correction coefficient is sufficient. There was a problem of not going up.

Therefore, in order to improve the accuracy (reliability) of the movement distance correction coefficient, it is possible to make the calculation employment conditions (first condition referred to in the present application) required for the GPS information used for the correction coefficient calculation stricter. When such a condition is set, the accumulation of the number of information that should be the basis of the movement distance correction coefficient calculation is naturally delayed.
For example, regarding GPS information, if only the information with a small error radius is to be used, the accumulation of the number of information is delayed. In such a situation, when GPS information that satisfies various judgments such as the number of satellites, speed judgment, straight ahead judgment, constant speed judgment, gradient judgment, etc. is used for the calculation of the movement distance correction coefficient. That's more.

  The present invention has been made in view of the above problems, and its purpose is to make full use of successively acquired GPS information, to obtain a highly accurate and reliable distance correction coefficient, and relatively An object of the present invention is to obtain a correction coefficient calculation device capable of quickly obtaining a highly reliable distance correction coefficient.

GPS information acquisition means for periodically acquiring GPS information including reception time information and speed information based on a signal from a GPS satellite according to the present invention for achieving the above object;
Pulse number information acquisition means for acquiring pulse number information from a distance sensor that outputs a pulse signal corresponding to the amount of movement at least simultaneously with acquisition of the GPS information;
A characteristic configuration of a correction coefficient calculation device that calculates a distance correction coefficient for correcting a moving distance obtained based on the pulse number information based on the GPS information and the pulse number information is as follows:
First condition determining means for determining whether or not a predetermined first condition is satisfied with respect to any two points of GPS information;
When the GPS information of the two points satisfies the first condition, a moving distance calculating means for calculating a moving distance between the two points based on the GPS information between the two points;
A relation amount calculation means for calculating a relation amount between the movement distance between the two points calculated by the movement distance calculation means and the pulse number difference indicated in the pulse number information between the two points;
Storage means for storing information on the relation amount calculated by the relation amount calculation means;
Information number adjusting means for adjusting the number of information when storing the relation amount in the storage means according to the moving distance between the two points;
A representative relation quantity computing means for computing a representative relation quantity based on a plurality of the relation quantities stored in the storage means after the number of pieces of information of the relation quantity stored in the storage means reaches a certain number or more;
Distance correction coefficient calculating means for calculating a distance correction coefficient for correcting the output of the distance sensor using the representative relation amount;
It is in the point provided with.

By providing this feature configuration, in the present application, with respect to GPS information, a distance correction coefficient is calculated for a pair of GPS information, which is arbitrary two points of GPS information.
Here, in the calculation of the distance correction coefficient, for the one GPS information pair, the first condition determination unit determines and extracts a pair of GPS information that is preferable in the calculation of the distance correction coefficient. The distance between the two points is obtained based on the GPS information, and the relation between the number of pulses and the movement distance is obtained by the relation amount calculation means.
These series of processes are performed individually for each GPS information pair to be processed. In the present application, the number of information is stored by the number-of-information adjusting means in the storage of the relation amount in the storing means. Adjusted. That is, while different GPS information pairs have been treated as equivalents in the past, they have different numbers of information (informational value) depending on the capture status of GPS information pairs. For example, as described later, the longer the moving distance, the higher the number of information (informational value). In this way, the storage / accumulation of the number of information is accelerated.

  Then, after the number of pieces of information on the relationship quantity stored in the storage means becomes a certain number or more, the representative quantity calculation means obtains the representative relation quantity that is the statistical value from the plurality of relation quantity information stored in the storage means, A distance correction coefficient is obtained from the representative relation amount.

  As a result, in the invention according to the present application, the GPS information can be stored and accumulated while evaluating the value of the GPS information by the information number adjusting means. The distance correction coefficient can be obtained quickly compared with the case where the information is handled as simple information.

A correction coefficient calculation program realized by incorporating such a correction coefficient calculation apparatus into an arithmetic processing apparatus that is hardware is configured as follows.
That is, a GPS information acquisition step for periodically acquiring GPS information including reception time information and speed information based on a signal from a GPS satellite;
A pulse number information acquisition step of acquiring pulse number information from a distance sensor that outputs a pulse signal according to the amount of movement, at least simultaneously with the acquisition of the GPS information,
A correction coefficient calculation program for causing a computer to execute a process for calculating a distance correction coefficient for correcting a moving distance obtained based on the pulse number information based on the GPS information and the pulse number information,
A first condition determination step for determining whether or not a predetermined first condition is satisfied with respect to arbitrary two points of GPS information;
A moving distance calculating step of calculating a moving distance between the two points based on the GPS information between the two points when the GPS information of the two points satisfies the first condition;
A relation amount calculation step for calculating a relation amount between the movement distance between the two points calculated in the movement distance calculation step and the difference in the number of pulses indicated in the pulse number information between the two points;
An information number adjusting step of adjusting the number of information when storing the relation amount in the storage unit according to the moving distance between the two points;
A storage step of storing, in the storage means, information on the relation amount of the number of information after adjustment by the information number adjustment step;
A representative relation amount calculation step for calculating a representative relation amount based on the plurality of relation amounts stored in the storage means after the number of pieces of information of the relation amount stored in the storage means reaches a certain number or more;
A distance correction coefficient calculating step of calculating a distance correction coefficient for correcting the output of the distance sensor using the representative relation amount;
Is a correction coefficient calculation program that causes a computer to execute the above.

  As the relational quantity, the number of pulses per reference distance or the distance per number of unit pulses can be employed. The distance correction coefficient according to the present application should be obtained to satisfactorily follow this change when the distance per unit pulse changes depending on the state of the installation target of the distance sensor, and the number of pulses per reference distance is This is because the distance per unit pulse can be obtained by the reciprocal thereof. Of course, the distance per unit pulse number may be adopted as it is.

  As the first condition, the accuracy condition for determining whether or not the information is within a predetermined error radius, the satellite for determining whether or not the number of satellites that have received the signal is greater than or equal to the predetermined number Several conditions, speed conditions for determining whether or not the vehicle is traveling at a certain speed or more, linear conditions for determining whether or not a straight traveling state is indicated, and whether or not a constant speed moving state is indicated Any one or more of the constant speed conditions for

By including the accuracy condition in the first condition, the distance correction coefficient can be obtained using highly reliable GPS information with a small error radius. The same applies to the satellite number condition.
As for speed conditions, straight travel conditions, and constant speed conditions, in the same way as has been done in the past, in order to calculate the distance correction coefficient, the movement is performed at a certain speed state, and the vehicle is traveling straight at almost constant speed. This is because it is most preferable to use GPS information.

Further, the information number adjusting means uses a plurality of classifications according to an increase in the movement distance between the two points, and when storing the relation amount in the storage means, It is preferable that the number of pieces of information regarding the relation amount be increased as the number of steps regarding the movement distance between the two points increases.
As the movement distance increases, the reliability of the relationship quantity obtained between the two points increases (the longer the distance between the two points, the more the average relation value can be represented and the abnormal value is eliminated). This is because processing is performed with a high number of pieces of information (informational value), and a suitable distance transfer coefficient can be obtained quickly.

A correction coefficient calculation program for realizing such an operation is configured as follows.
When the information quantity adjustment step uses a plurality of classifications according to an increase in the movement distance between the two points, and the relational quantity is stored in the storage means, the movement between the two points from which the relational quantity is calculated is stored. What is necessary is just to set it as the structure which increases the number of information about the said related quantity as the step regarding distance increases.

Furthermore, when any of the selected GPS information between the two points no longer satisfies the first condition, it is preferable to end the processing for the GPS information of the two points.
If this configuration is adopted, unless all the GPS information existing between the two points satisfies the first condition, the GPS information pair is not used for the calculation of the distance correction coefficient, and the most preferable GPS information is calculated by the coefficient calculation. This is because it can be used.

The correction coefficient calculation program that realizes this function is
When any of the selected GPS information between the two points no longer satisfies the first condition, the process for the current GPS information in the GPS information selection step is terminated.

In addition, it further comprises GPS information number determination means for determining whether or not a GPS information number condition that the number of GPS information between the two points is a predetermined number or more,
In the case where the GPS information count condition is not satisfied, it is preferable that the relation amount calculation means exclude the relation amount calculation target.
By doing so, only GPS information in a state in which GPS information of a certain degree or more is included can be used for calculation of the movement distance correction coefficient, and reliability is improved.

The correction coefficient calculation program that realizes this function is
A GPS information number determination step for determining whether or not a GPS information number condition that the number of GPS information between the two points is a predetermined number or more is satisfied,
When the GPS information count condition is not satisfied, the correction coefficient calculation program is excluded from the processing in the steps after the relation amount calculation step.

In addition, it further comprises a gradient determination means for determining whether or not the gradient of a predetermined angle or more is moving when the current GPS information is acquired,
If it is determined that the gradient is moving, it is also preferable to end the processing for the current GPS information by the GPS information selection means.
Originally, when moving on a gradient, an error is likely to occur between the movement distance obtained only from the GPS information side and the true movement distance due to the altitude difference. Although not preferable, such low-reliability GPS information can be well excluded from the calculation.

The correction coefficient calculation program that realizes this function is
A gradient determination step of determining whether or not a gradient of a predetermined angle or more is moving when the current GPS information is acquired;
When it is determined that the gradient is moving, the correction coefficient calculation program ends the processing for the current GPS information in the GPS information selection step.

Now, the self-recognition device is equipped with the correction coefficient calculation device described so far,
A direction information acquisition means for acquiring direction information from the direction sensor is provided.
Based on the pulse number information, the azimuth information, and the distance correction coefficient obtained by the correction coefficient calculation device, a movement trajectory of the own position is calculated, and a self-position calculation means for recognizing the own position is provided. Recognize your location quickly and reliably.

The self-position recognition program configured to include the correction coefficient calculation program described so far, which can realize this function,
Having an orientation information acquisition step of acquiring orientation information from the orientation sensor;
Based on the pulse number information, the azimuth information, and the distance correction coefficient obtained by the correction coefficient calculation program, the self-position calculation step for calculating the movement locus of the own position and recognizing the own position is executed by the computer. It becomes a position recognition program.

Hereinafter, an embodiment of a self-position recognition device 2 including a correction coefficient calculation device 1 according to the present application will be described with reference to the drawings. This type of self-position recognition device 2 is installed in a navigation device (not shown) such as an automobile and is used for recognition of the position of the own vehicle.
FIG. 1 is a functional block diagram showing an outline of the configuration of the self-position recognition device 2.

In the figure, the functional unit 50 related to the self-position recognition is shown on the left side, and the functional unit 10 related to the distance correction coefficient calculation is shown on the right side. The portion related to the correction coefficient calculation device 1 is a portion surrounded by a one-dot chain line.
Hereinafter, the functional unit 50 related to the self-position recognition process and the functional unit 10 related to the distance correction coefficient calculation process will be described in this order.

1 Functional unit 50 related to own position recognition processing
The self-position recognition process follows the self-position recognition processing method employed in the prior art, and is not greatly different from that described in the technique disclosed in Patent Document 1 above.
That is, based on the GPS information obtained from the GPS information obtaining unit 52, the pulse number information obtained from the pulse number information obtaining unit 55, and the gyro information obtained from the gyro information obtaining unit 58, the distance correction obtained from the correction coefficient computing device 1 The moving distance is calculated using the coefficient k, and the own position is calculated and recognized using the moving distance.

  Since the movement distance used on the self-position recognition apparatus side is a movement distance obtained by the pulse number increment Δpls, the distance Ld per unit pulse, and the distance correction coefficient k, this movement distance is hereinafter referred to as Lpls.

Hereinafter, the configuration of each unit and its function will be described.
The GPS receiver 51 receives a signal from the GPS satellite S. A signal from the GPS satellite S is received every second and sent to the GPS information acquisition unit 52. The GPS information acquisition unit 52 analyzes the received signal from the GPS satellite S, and obtains information such as own position information, traveling direction information, moving speed information, positioning satellite number information, error radius information, and height direction speed information. Deriving and generating GPS information including these pieces of information. The own position information is information including the latitude coordinate and the longitude coordinate of the own position.
The GPS information is periodically generated every time a signal from the GPS satellite S is received, that is, every second, and stored in the GPS information storage unit 53.

  The distance sensor 54 is a sensor for detecting the moving distance of the vehicle CAR on which the position recognition device is mounted. As the distance sensor 54, a so-called vehicle speed pulse sensor is adopted. This type of distance sensor 54 outputs a pulse signal each time a rotating member related to traveling such as a drive shaft of a vehicle CAR rotates a certain amount. The output from the distance sensor 54 is sent to the pulse number information acquisition unit 55. The pulse number information acquisition unit 55 integrates the number of pulses of the pulse signal output from the distance sensor 54 to generate pulse number information. This number-of-pulses information is periodically generated at least every time GPS information is acquired by the GPS information acquisition unit 52, here every second. The generated pulse number information is stored in the pulse number information storage unit 56.

  FIG. 2 shows how GPS information is captured. FIG. 2A shows a state in which the vehicle CAR in the traveling state receives signals from the GPS satellite S sequentially. FIG. 2B shows the time on the horizontal axis, the GPS information capture timing on the upper side of the time axis, and the pulse signal generation timing on the lower side. In this traveling state, it can be seen that pulse signals are frequently generated because the moving speed is high.

  The gyro sensor 57 is a direction sensor that detects a change in the direction of the vehicle. As the gyro sensor 57, various gyroscopes such as a vibration gyroscope and an optical fiber gyroscope can be used. In this example, a vibration gyroscope is used. The gyro sensor 57 generates an output such as a voltage proportional to the rotational angular velocity around a predetermined detection axis. The output from the gyro sensor 57 is sent to the gyro information acquisition unit 58. The gyro information acquisition unit 58 performs analog / digital conversion that converts an output of an analog signal such as a voltage from the gyro sensor 57 into a digital signal, and generates a gyro output value including a digital signal at a predetermined stage. This gyro output value is periodically generated at least every second when GPS information is acquired by the GPS information acquisition unit 52. The generated gyro output value is stored in the gyro information storage unit 59 as gyro information.

  The self-position storage unit 60 is a storage unit that stores self-positions that have been sequentially recognized up to the previous time, and the self-position is stored as a combination of latitude and longitude coordinates. The previous own position information is sent to the own position calculation unit 61 and used for calculation / recognition of the current own vehicle position.

  The own position calculation unit 61 is the base point for calculating the pulse number information, which is information related to the movement distance, gyro information, which is information related to the direction of the movement locus, and the movement locus, which is sent to the calculation unit 61. The current position is calculated using the previous position information. In the calculation of the own position, the distance correction coefficient k sent from the correction coefficient calculation device 1 is used for the calculation of the movement distance Lpls.

  The local position is periodically calculated at least every second when GPS information is acquired by the GPS information acquisition unit 52, and the local position obtained by the calculation is stored in the local position storage unit 60 described above. The

  In the calculation of the movement trajectory, the movement trajectory is obtained on the assumption that it has moved from the previous position in the current azimuth direction obtained as gyro information from that position by a movement distance Lpls equivalent to the increment of the number of pulses. Here, as described above, the movement distance Lpls is calculated by multiplying the increment Δpls of the number of pulses between the present time and the previous time, the distance Ld per pulse, and the distance correction coefficient k, Lpls = Δpls. × Ld × k.

  In this way, the own position is obtained based on the movement locus calculated by the own position calculation unit 61, sent back to the own position storage unit 60, and stored as the current own position. On the other hand, it is also output to the navigation device (not shown) side.

2 Function unit 10 related to distance correction coefficient calculation processing
Hereinafter, the configuration of the correction coefficient calculation apparatus 1 will be described.
As shown in FIG. 1, in the relationship with the own position calculation device 2, the correction coefficient calculation device 1 includes a GPS receiver 51, a GPS information acquisition unit 52, a GPS information storage unit 53, a distance sensor 54, a pulse number information income. The unit 55 and the pulse number information storage unit 56 are shared. Accordingly, the GPS information obtained from the GPS information acquisition unit 52 and the pulse number information obtained from the pulse number information acquisition unit 55 can be used for the calculation process in the correction coefficient calculation device 1.
The distance correction coefficient k obtained by the correction coefficient calculation device 1 is sent to the own position calculation unit 61 and used for calculation of the own position in the unit 61.

  The calculation processing of the distance correction coefficient k is roughly composed of three main parts.

  In the first part, the GPS information selection unit 110 mainly works, and sequentially selects a pair of GPS information that is a determination target of the first condition determination, the second condition determination, and the third condition determination, and specifies the determination target. When the determination is satisfied, the obtained reference pulse number NSpls is used to obtain a histogram of the reference pulse number NSpls, and finally a macro for calculating the distance correction coefficient k from the average value of the reference pulse numbers NSpls. This part controls the flow.

  In the present application, the pair of GPS information is established as a pair of newly acquired current GPS information and past GPS information before the current GPS information. Further, as will be described later, there is a limit on the number of return to the past regarding the past GPS information to be paired, and the processing is repeated within this number.

The second part is a part that is used as the condition determination unit 120, and includes a first condition determination unit 121, a second condition determination unit 122, and a third condition determination unit 123 as shown in the figure.
The first condition determination unit 121 determines whether or not the information can be directly used for the calculation of the distance correction coefficient k with respect to the GPS information pair generated in the first part.

  The second condition determination unit 122 is configured such that the information obtained between the two points regarding the distance traveled between the two points at which the two GPS information pairs forming the GPS information pair are received and the number of GPS information included therebetween is the distance. It is determined whether or not it can be used for calculating the correction coefficient k. Since the movement distance obtained here is a movement distance obtained based on GPS information as will be described later, it is described as Lgps.

  Whether the third condition determination unit 123 can be used to calculate the distance correction coefficient k for the number of pulses included between the two points where the two GPS information pairs forming the GPS information pair are received, etc. Make a decision. The number of pulses per predetermined distance obtained by this processing is the reference pulse number NSpls, which is a kind of relational quantity referred to in the present application.

The third part uses the GPS information pair that is determined to be usable satisfying the first condition determination, the second condition determination, and the third condition determination, and sequentially stores the reference pulse number NSpls in the reference pulse number storage unit 124. Accumulate and remember. This accumulation is performed during the execution of the process relating to the current GPS information, and the results obtained sequentially with respect to the past GPS information are also accumulated. That is, in the third part, the processing related to past GPS information is utilized.
Then, the average distance calculation unit 125 obtains a representative value from the accumulated number of reference pulses NSpls, and the distance correction coefficient calculation unit 126 calculates a preferred distance correction coefficient k at the present time.

Hereinafter, the configuration of each unit and its function will be described.
Each time new GPS information is sent from the GPS information acquisition unit 52, the GPS information selection unit 110 uses the new GPS information as the current GPS information for the subsequent processing in relation to the current GPS information. A pair of current GPS information and past GPS information should be selected and specified.

  That is, processing control is performed so that the current GPS information is sequentially learned in combination with GPS information up to a specific number (for example, 32) before returning to the past, and learning about the current GPS information is terminated. To do.

Specifically, when the current GPS information is captured, the selection unit 110 determines that the GPS information is a combination of the current GPS information and the previous GPS information from the current, and a combination of the previous GPS information. The GPS information pairs are sequentially sent to the subsequent processing while sequentially tracing in the past. When the repetitive operation reaches a preset number, the learning for the current GPS information is finished. In this example, the return count to the past GPS information is 32.
The GPS information selection unit 110 serves as GPS information selection means.

The first condition determination unit 121 performs the first condition determination on the GPS information pairs sent sequentially. As shown in the figure, the first condition determination unit 121 is provided with six types of determination means, and only GPS information that satisfies all the determinations in each determination means is sent to the subsequent processing.
The first condition determination unit 121 serves as a first condition determination unit.

  The accuracy determination unit 121a determines whether the current GPS information and the past GPS information are information satisfying a preset error radius with respect to the error radius of the GPS information sent as a pair. In this example, the upper limit is 300 m as the error radius.

  The number-of-satellite number determination means 121b includes, as GPS information, information based on three satellites (2D information) and information based on four satellites (3D information) regarding the number-of-satellite information included in the GPS information. In this case, a combination of only 3D information with high information accuracy is to be used for the subsequent processing. Also in this means 121b, it is determined that both pieces of GPS information forming a GPS information pair are 3D information.

  The speed determination unit 121c determines whether or not the GPS speed is equal to or higher than a preset speed with respect to the GPS speed information included in the GPS information. This speed is a lower limit speed, and only GPS information in a situation where the GPS speed is equal to or higher than a predetermined speed is used for the following processing. In this example, 10 m / s is the lower limit. This is because the distance correction coefficient k is obtained as a highly reliable value when the movement is a movement at a predetermined speed or higher.

  The straight traveling determination unit 121d moves by comparing the GPS azimuth information for two points between the time when the current GPS information is captured and the time when the past GPS information paired with this GPS information is captured. It is determined whether or not the vehicle is moving straight, and the GPS information acquired between the two points where the movement is performed in a straight line is used for the subsequent processing. In this example, when the GPS azimuth information is less than 2 ° between two points, the following processing is performed assuming that the vehicle is moving straight. This is because the distance correction coefficient k is obtained as a highly reliable value when the movement is a straight movement.

  The constant speed determination means 121e has a problem of the difference between the GPS speed information regarding the current GPS information and the past GPS information paired with the GPS information, and is running at a constant speed. This is intended to use the GPS information of the two points for the subsequent processing. In this example, when the GPS speed information is less than 1 m / s, the following processing is performed assuming that constant speed movement is performed.

  In the situation where the gradient determination means 121f takes the current GPS information and the past GPS information that is paired with this GPS information, the difference in the GPS height information is a problem, and the vehicle is running on a gradient. This is to prevent the two points of GPS information from being used for subsequent processing. In this example, when there is a difference larger than 0.2 m / s in the GPS height direction speed, it is assumed that the vehicle is traveling in a gradient and is excluded from the following processing.

  In the first condition determination unit 121, the determination by the six determination means 121a to 121f is performed, and a combination of GPS information satisfying all these conditions is sent to the process thereafter. On the other hand, when any of the conditions is not satisfied, the learning regarding the current GPS information is terminated.

The second condition determination unit 122 includes a movement distance calculation unit 122a, a movement distance determination unit 122b, a GPS information number calculation unit 122c, and a GPS information number determination unit 122d.
The second condition determination unit 122 makes the movement distance itself a problem with respect to the movement distance between the two points at which the respective GPS information forming the GPS information pair is taken in, and determines the number of GPS information included therein. Make it a problem.
If the second condition determination unit 122 does not satisfy the second determination condition, the paired GPS information is not subjected to subsequent processing, and the paired past GPS information is converted to the previous one. Go to and execute processing for further pairs.

  The movement distance calculation means 122a, for the GPS information pair that satisfies the first condition in the first condition determination unit 121, between the time point when the current GPS information is captured and the time point when the past GPS information is captured. Calculate the distance traveled. The movement distance obtained based on only the GPS information in this way is referred to as a movement distance Lgps in the present application.

  For the calculation process in the calculation means 122a, the moving speed information spd included in the GPS information and the time information t from which each GPS information is acquired are used. That is, the movement speed information related to the current GPS information is spd0, the movement speed information related to the paired GPS information is spdi (where i is a natural number from 1 to 32), and the movement distance Lgps based on the GPS information. Is obtained by the following formula.

  Lgps = (spd0 + spd1) / 2 × Δt1 + (spd1 + spd2) / 2 × Δt2 +... + (Spd (i−1) + spdi) / 2 × Δti

  Here, spd1 and spd2 are the movement speed information one before and two before spd0, respectively, and spd (i-1) is one time after the time when the movement speed information spdi was taken in (at the current time). This is the movement speed information on the near side. Δt1, Δt2,..., Δti is the difference (elapsed time) between the current GPS information capture time and the previous capture time, the previous capture time and the previous capture time, respectively. Difference from the time (elapsed time) ..... Difference from the time when the GPS information was captured immediately after the time when the moving speed information spdi was captured (the time closer to the current time) (elapsed time) It is.

  Therefore, the moving distance Lgps calculated by the calculating means 122a is obtained by using the average moving speed × elapsed time using all the GPS information captured between the two pairs of the GPS information pair currently being processed. The travel distance Lgps is obtained as an integrated value of time (Σ (spd (i−1) + spdi) / 2 × Δti). Here, in this example, the elapsed time Δt is obtained by using a timer count that is provided in the navigation device on the side that uses the output of its own position and that can recognize the time in ms units. High accuracy is assumed.

  The movement distance determination unit 122b determines whether or not the distance Lgps between the two points related to the GPS information pair calculated by the movement distance calculation unit 122a is equal to or greater than a predetermined distance. In this example, 100 m is adopted as the predetermined distance. If the distance is less than the predetermined distance, the GPS information pair currently being processed is excluded from the subsequent processing, and if it is equal to or longer than the predetermined distance, it is subjected to the subsequent processing. As described above, when the GPS information pair that is the target of processing at the present time is excluded from the processing target, the processing is newly advanced with respect to the pair with the previous past GPS information.

The GPS information number calculation means 122c calculates the degree of influence e with reference to the number of GPS information between two points where each GPS information forming a GPS information pair is taken.
The influence e is obtained as e = LS / Lgps / N, where LS is the reference moving distance, N is the number of GPS information, and Lgps is the moving distance between the two points. In the case of this example, 100 m is adopted as the reference movement distance LS.
This degree of influence e means that 20 GPS information is captured within the reference movement distance LS = 100 m when the threshold is 0.05 and LS is 100 m. Further, the influence degree e will be described with reference to FIG.

  In the figure, the horizontal axis indicates time (which is also the moving distance), the left end indicates the time when the current GPS information is captured, and the GPS information is traced in the past as it goes to the right. The block row shown in the upper part corresponds to the fact that GPS information is individually captured and moved. The middle row shows the situation where the GPS information of 7 is returned from the present and the degree of influence e between the current and 7 previous points is obtained, and the lower row returns the GPS information of 12 previous times from the present, A situation in which the degree of influence e between the previous two points is obtained is shown.

In the following description, it is assumed that the movement occurs 15 m at every GPS information capturing timing.
When the GPS information of 7 shown in the middle is processed, the influence degree e is 100/105/7, calculated as 0.136, and the single GPS information has an influence degree of 13.6%. It will be. Therefore, when the threshold value of the influence degree e is set to 0.05 (5%), this GPS information pair is not used. In the case of this threshold, only 100 GPS information is included in a 100 m travel distance.

  When the 12 GPS information shown in the lower row is to be processed, the impact e is 100/180/12, calculated as 0.046, and the single GPS information has an impact of 4.6%. Yes. Therefore, when the threshold value of the influence degree e is set to 0.05 (5%), this GPS information pair can be used.

  The GPS information number determination unit 122d determines whether or not the influence degree e calculated by the GPS information number calculation unit 122c is equal to or less than a predetermined value. In this example, 0.05, which was previously introduced as a threshold value, is adopted as this value. If it is less than the predetermined value, the GPS information pair that is currently the subject of processing is sent to the subsequent processing. On the other hand, if it is larger than the predetermined value, it is excluded from the target, and the process is newly advanced for the pair with the previous past GPS information.

The third condition determination unit 123 has a problem with the reference pulse number NSpls, which is the number of pulses per reference movement distance LS, regarding the movement distance Lgps between the two points obtained as described above. The reference pulse number NSpls intends to use the GPS information pair in the standard moving state for the following processing.
If the third condition determination unit 123 does not satisfy the third determination condition, the learning process regarding the current GPS information that is the target of the current process ends. This is because it is considered that the current traveling state deviates from a standard moving state suitable for obtaining the distance correction coefficient k.

  The determination unit 123 includes a reference pulse number calculation unit 123a and a reference pulse number determination unit 123b that performs a determination on the reference pulse number NSpls calculated by the calculation unit 123a.

  The reference pulse number calculation means 123a calculates the number of pulses per reference movement distance LS between two points as the reference pulse number NSpls for the GPS information pair currently being processed. In this example, 100 m is adopted as the reference movement distance LS. The number of pulses at the point at which the current GPS information is received is pls, the number of pulses at the point at which a pair of past GPS information has been received is plsI, and the reference pulse number is calculated as (pls−plsI) × LS / Lgps. Is done. This reference pulse number calculation means 123a is a kind of relational quantity calculation means in the present application.

The reference pulse number determination means 123b determines whether or not the reference pulse number NSpls obtained as described above is within a predetermined range with respect to the average value of the reference pulse numbers NSpls obtained so far. . In this example, when it is within the range of 0.7 to 1.3 times the average value as being within the predetermined range, it is determined that it is within the range.
Then, when the reference pulse number NSpls to be determined is within the range, the GPS information pair currently being processed is set as a target for subsequent processing. If it is not within the range, the learning process for the current GPS information to be processed ends.

As described above, a GPS information pair that satisfies all the determinations of the first condition determination, the second condition determination, and the third condition determination is the target of the representative relation amount calculation process.
In order to perform the representative relation amount calculation, as shown in FIG. 1, a reference pulse number storage unit 124 and an average distance calculation unit 125 that calculates an average distance that is a representative relation amount are provided. Then, the distance correction coefficient calculator 126 calculates the distance correction coefficient k from the average distance obtained by the average distance calculator 125.

  The reference pulse number storage unit 124 is a part that stores the reference pulse number NSpls information regarding the GPS information pair that satisfies all the three conditions described so far. This storage and storage state is different from that performed in the prior art described above, and gives an adjusted number of information to one GPS information pair corresponding to the movement distance Lgps of the GPS information pair. The data is classified according to the reference pulse number NSpls and stored in a predetermined storage area.

In response to this request, as shown in FIG. 6, the reference pulse number storage unit 124 is provided with information number determination means 124a and accumulation storage means 124b.
The number-of-information determination means 124a uses the information represented by [integer satisfying Lgps / 50-1 (rounded down to the nearest decimal point)] with respect to the GPS information pair currently being processed as the movement distance Lgs. Determine the number. In this determination method, the number of information is 0 when the moving distance is 50 m or less, and the information number is 1 when the moving distance is greater than 50 and 100 m or less. Furthermore, the number of information increases as the moving distance Lgps increases. This information number determination means 124a is an adjustment means in the present application.
The storage unit 124b stores and stores the reference pulse number NSpls in a predetermined storage area for the number of pieces of information determined as described above for the GPS information pair currently being processed.
The reference pulse number storage unit 124 is a storage unit in the present application.

As shown in FIG. 4, the average distance calculation unit 125 includes a stored information number confirmation unit 125a, a stored information number determination unit 125b, and an average distance calculation unit 125c.
The accumulated information number confirmation unit 125a confirms the total information number accumulated in the reference pulse number storage unit 124, and the accumulated information number determination unit 125b determines whether or not the total information number has reached a predetermined value. . As shown in the figure, this means includes a first information number determination means 125b-1, a second information number determination means 125b-2, and a third information number determination means 125b-3, which are based on the total number of information. Are configured to execute different processes.
The average distance calculation means 125c calculates an average of movement distances per unit pulse (this value is referred to as a unit pulse average distance Lav) for all GPS information pairs. This average distance calculation means 125 is a representative relation amount calculation means.

  In the correction coefficient calculation device 1 according to the present application, as described above, the current GPS information is sequentially fetched and a predetermined process is executed, and then new reference pulse information NSpls is sequentially accumulated. The newly calculated reference pulse information NSpls is added (the total of the accumulated amounts can be about 4000), and old information is discarded, so that the unit pulse average distance is always maintained at a value that can represent the current state.

As shown in FIG. 5, the distance correction coefficient calculation unit 126 includes a distance correction coefficient calculation means 126a and a distance correction coefficient output means 126b.
The distance correction coefficient calculation means 126a calculates a ratio (Lav / Ld) between the unit pulse average distance Lav obtained by the average distance calculation unit 125 and the distance Ld per pulse that is set in advance as a default value. Is the distance correction coefficient k referred to in the present application.

  The distance correction coefficient k obtained in this way is configured to be output by the distance correction coefficient output means 126b.

The above is the description of the configuration of the correction coefficient calculation apparatus 1 according to the present application. Hereinafter, correction coefficient calculation processing will be described using the flowcharts shown in FIGS. 7 to 11.
FIG. 7 is a main flow of the correction coefficient calculation process, and taking this new GPS information (in this processing flow, this GPS information is called “current GPS information”) as a start condition, A learning process based on GPS information is performed. Processing according to this main flow is mainly executed by the GPS information selection unit 110.

  The learning process is started by taking in new GPS information (current GPS information) (step 1: yes). In a state where new GPS information is not captured, a waiting state for capturing new GPS information is entered (step 1: no).

  When the current GPS information is captured, a learning process related to the GPS information is started. The learning process returns to a predetermined number of past GPS information before the current GPS information, creates a pair of current GPS information and past GPS information, and creates a pair of GPS information This process repeats the process sequentially. Therefore, the counter I is set to 1 (step 2), and the counter I is sequentially incremented (step 9) under predetermined conditions.

  It is determined whether or not the counter I has reached a predetermined number (32 in this example) (step 3). Until the counter I reaches 32, the current GPS information newly acquired and the GPS information of the past counter I from the present are acquired, and the GPS information pair (current GPS) to be subjected to the following processing is executed. A pair of information and GPS information before I is generated (step 4).

  As described above, the first to third condition determination processing (steps 5, 6, and 7) is executed for each GPS information pair that is sequentially generated, and GPS information pairs that satisfy all these conditions can be used. The distance correction coefficient k is calculated and output using the information obtained so far and the current GPS information (step 8).

Here, as shown in FIG. 7, in the first condition determination (step 3) executed by the first condition determination unit 121, if a GPS information pair determined not to meet this condition appears, the counter loop (Loop around step 3 to step 9) is taken off and the learning about the current GPS information is finished.
In the second condition determination executed by the second condition determination unit 122, if a GPS information pair that is determined not to meet this condition appears, the GPS information pair remains in the counter loop. And the counter I increment processing I = I + 1 only is executed (step 9).
The third condition determination executed by the third condition determination unit 123 is the same as the first condition determination.

  Therefore, in this main flow, when new GPS information is taken in, the operation for returning to the past GPS information from the current GPS information as the GPS information is repeated, and the first condition determination and the third condition are repeated. As long as the determination is satisfied, the calculation of the distance correction coefficient is sequentially repeated. Therefore, all the GPS information existing in the GPS information pair individually satisfies both conditions with respect to the condition determination.

FIG. 8 is a flowchart showing the first condition determination process.
Table 1 shown below shows determination condition types in the first condition determination process, determination condition means for executing the determination, and actual determination conditions.

As shown in the flow, in the first condition determination, the determination is configured to be executed in three stages. Regarding the GPS information pair, first, the accuracy determination, the satellite number determination, the speed determination, and the straight traveling are performed. If the determination is made and the determination is satisfied, the subsequent processing is performed (step 11: yes). Next, constant speed determination is performed, and if the determination is satisfied, the subsequent processing is performed (step 12: yes), further gradient determination is performed, and if the determination is satisfied, the subsequent processing is performed (step 13: yes). On the other hand, when any one of the conditions is not satisfied, the processing regarding the current GPS information to be processed is finished (step 11, step 12, and step 13: no).
In this way, in the first condition determination, GPS information that cannot be appropriately used for the calculation of the distance correction coefficient is excluded from the processing target.

FIG. 9 is a flowchart showing the contents of the second condition determination process.
Table 2 shown below shows determination condition types in the second condition determination process, determination condition means for executing the determination, and actual determination conditions.

As shown in the flow, in the second condition determination, the determination is executed in two stages.
First, regarding the GPS information pair to be processed, the movement distance calculation means 122a obtains the movement distance Lgps between the two points that received these GPS information (step 21). The movement distance Lgps thus determined is determined by the movement distance determination means 122b to be equal to or greater than a predetermined movement distance (step 22). In the case of this example, as shown in Table 2, the determination condition is 100 m or more. In a situation where the total frequency is sufficiently secured, if it is 100 m or more (step 22: yes), it is assumed that it is a target of the subsequent processing, and if it is less than 100 m (step 22: no), this GPS information The pair increments the counter I by 1 without being used for subsequent processing (step 9). As a result, the processing shifts to a pair of the previous GPS information immediately before the current processing target and the current GPS information.

  Next, with respect to the GPS information pair to be processed, the GPS number calculation means 122c obtains the influence degree e between the two points at which the GPS information is received (step 23). The GPS number determination means 122d determines whether or not the influence degree e thus obtained is equal to or less than a predetermined influence degree. In this example, as shown in Table 2, 0.05 or less is set as the determination condition. If it is 0.05 or less (step 24: yes), it will be the target of the subsequent processing. If it is larger than 0.05 (step 24: no), this GPS information pair shall be used for the subsequent processing. Instead, the counter I is incremented by 1 as in the case of the distance. As a result, the processing shifts to a pair of the previous GPS information immediately before the current processing target and the current GPS information.

FIG. 10 is a flowchart showing the contents of the third condition determination process.
Table 3 below shows determination condition types in the third condition determination process, determination condition means for executing the determination, and actual determination conditions.

  In this condition determination, the reference pulse number calculation means 123a uses the movement distance Lgps between the two points obtained in the previous second condition determination to determine the GPS information for the GPS information pair to be processed. The reference pulse number NSpls between the two received points is obtained (step 31). The reference pulse number NSpls is obtained by converting the pulse number measured between two points into a reference distance LS, and serves as an evaluation index for whether the movement is a standard movement or a movement in an abnormal state.

  The reference pulse number determination means 123b determines whether or not the reference pulse number NSpls thus determined is within a predetermined range with respect to the reference pulse number NSpls determined so far. In the case of this example, the determination condition is whether it is within the range of 0.7 to 1.3 of the average value of the reference pulse number NSpls obtained so far. If it is within the range (step 32: yes), it is assumed that it is an object of the subsequent processing, and if it is out of the range (step 32: no), the learning regarding this GPS information pair ends.

  In the calculation process of the distance correction coefficient k shown in FIG. 11, the selected GPS information pair is handled as information useful for the distance correction coefficient calculation by performing the condition determination as described above, and the information is accumulated. In this process, the distance correction coefficient k is calculated on the condition that accumulation of a predetermined amount of information is completed, and the calculation result is output.

  The latter calculation processing of the distance correction coefficient k is divided into two systems according to the number of information to be accumulated. When the device has a short history of use and the number of information to be accumulated is smaller than the predetermined number of information, FIG. If the number of accumulated information traced in the left processing system is sufficient, the right processing system is traced. Hereinafter, it demonstrates in order.

For the GPS information pair that satisfies the first to third determination conditions, the reference pulse number NSpls is obtained, and this processing stage is reached.
The processing in the reference pulse number storage unit 124 is the processing in step 41 shown in FIG. 11. As shown in the detailed flowchart in FIG. 12, the information number determination is performed on condition that the processing has reached this stage unit times. The number of information is determined by means 124a (step 51), and the storage unit 124b classifies the information into a predetermined storage area according to the value of the reference pulse number NSpls, and stores that number of information (step 52).

Further, returning to FIG. 11 and continuing the description, the process proceeds to the processing in the average distance calculation unit 125. The accumulated information number confirmation unit 125a confirms the total number of information accumulated in the storage area. (Step 42).
Each process shown on the left and right in the figure is executed according to the total number of confirmed information. In the processing by the first information number determination unit 125b-1, when the total number of information is larger than the first information number (64 in this example) (step 43: yes), the second information number determination unit 125b-2 In the processing according to (4), it is determined whether or not the total number is a multiple of the second information number (32 in this example) (step 44-1). And when it corresponds to the multiple of the 2nd information number (Step 44-1: yes), average distance calculation of 1 pulse by average distance calculation means 125c (Step 45-1), distance correction by distance correction coefficient calculation means 126a After the coefficient calculation (step 46-1), the calculation result obtained by the distance correction coefficient output means 126b is output as it is (step 47-1).
On the other hand, if it is not a multiple of the second information number, the counter I is incremented (step 9).

  The processing described above is a processing situation in which a predetermined period has passed since the use of the apparatus, a large amount of information equal to or greater than the first information number is secured, and a stable distance correction coefficient can be calculated and output.

  On the other hand, when the total number of information is smaller than the first information number (64 in this example) in the processing by the first information number determination unit 125b-1, the second information number determination unit 125b (step 43: no). As a process of -3, it is determined whether or not the total number is a multiple of the third information number (8 in this example) (step 44-2). And when it corresponds to the multiple of the 3rd information number (step 44-2: yes), average distance calculation of 1 pulse by average distance calculation means 125c (step 45-2), distance correction by distance correction coefficient calculation means 126a After the coefficient calculation (step 46-2), the distance correction coefficient output means 126b executes the distance correction coefficient output (step 47-2).

In the processing system shown on the right side of FIG. 11 described above, the obtained calculation result is output as it is, but the output processing form is followed also in the processing system shown on the left side. However, from the relationship between the number of second information and the number of third information, in the left flow, the distance correction coefficient k is output with high frequency from the stage where the number of information is small (step 47-2). .
On the other hand, when it does not correspond to the multiple of the second information number, the increment of the counter I is executed (step 9).

  In this way, in a situation where the amount of information that is accumulated is relatively short since the start of use of the device and the accumulated amount of information is less than the first information number, the true value is obtained while excluding abnormal values. A process of calculating and outputting a distance correction coefficient k considered to be close can be executed.

  As a result, the distance correction coefficient suitable for the current situation can be obtained using the correction coefficient calculation apparatus 1 according to the present application, and the own position can be obtained using the own position recognition apparatus, which can be suitably used for navigation. .

[Another embodiment]
(1) In the above embodiment, the reference pulse information is used as the relation quantity, but naturally the distance (movement distance) per unit pulse is directly used as the relation quantity using the movement distance Lgps obtained from the GPS information. You may ask for it.
(2) In the above embodiment, the GPS information pair that satisfies all the conditions belonging to the first condition is used for the calculation of the distance correction coefficient. However, depending on the application, any one of these condition items is adopted. It is good also as what to do.
(3) In the above embodiment, every time new GPS information is captured, a pair with the past GPS information is created and the distance correction coefficient is calculated. This can also be applied to the case where any two pieces of GPS information are selected as a pair from the GPS information group stored in the above and used for calculating the distance correction coefficient. That is, it does not need to be performed every time new GPS information is captured.
(4) In the above embodiment, the return number (predetermined number) of the past GPS information constituting the GPS information pair is set to 32. However, this numerical value is not lost in the reliability of the distance correction coefficient and is processed simultaneously. The time can be changed as long as the time required is not increased excessively.
(5) The number of first information, the number of second information, and the number of third information adopted by the average distance calculation unit 125 is a combination of (64, 32, 8) in this embodiment, but a larger number of sets A combination may be employed, and any combination may be employed. However, due to the nature of the number of information, the number is reduced as the first, second, and third proceed, and at the same time, a combination of numerical values that facilitates the calculation is preferable.
(6) In the above embodiment, the straightness of movement is determined using the traveling direction information included in the GPS information. However, from the gyro sensor provided separately, etc. The straightness of movement may be determined from the azimuth information.

  Self-recognition that can make use of GPS information that is sequentially captured to obtain a correction coefficient calculation device that can quickly obtain a highly accurate and reliable distance correction coefficient and that can perform highly reliable self-position recognition. Could get the device.

Functional block diagram of correction coefficient calculation device and self-recognition location recognition device according to the present application Explanatory drawing which shows the capture state of GPS information and pulse signal Explanatory drawing of the degree of influence on the number of GPS information and travel distance Functional block diagram showing the detailed configuration of the average distance calculator Functional block diagram showing the detailed configuration of the distance correction coefficient calculation unit The figure which shows the histogram memorize | stored in a memory | storage part Main flow flowchart of distance correction coefficient calculation First condition determination flowchart Second condition determination flowchart Third condition determination flowchart Distance correction coefficient calculation output flowchart Flow chart of information storage with information count determination

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Correction coefficient calculating device 2 Self-position recognition apparatus 52 GPS information acquisition part 55 Pulse information acquisition part 58 Gyro information acquisition part 60 Self-position storage part 61 Self-position calculation part 110 GPS information selection part 120 Determination part 121 First condition determination part 122 Second condition determination unit 123 Third condition determination unit 124 Reference pulse number storage unit 125 Average distance calculation unit 126 Distance correction coefficient calculation unit

Claims (14)

GPS information acquisition means for periodically acquiring GPS information including reception time information and speed information based on signals from GPS satellites;
Pulse number information acquisition means for acquiring pulse number information from a distance sensor that outputs a pulse signal corresponding to the amount of movement at least simultaneously with acquisition of the GPS information;
A correction coefficient calculation device for calculating a distance correction coefficient for correcting a movement distance obtained based on the pulse number information based on the GPS information and the pulse number information,
First condition determining means for determining whether or not a predetermined first condition is satisfied with respect to any two points of GPS information;
When the GPS information of the two points satisfies the first condition, a moving distance calculating means for calculating a moving distance between the two points based on the GPS information between the two points;
A relation amount calculation means for calculating a relation amount between the movement distance between the two points calculated by the movement distance calculation means and the pulse number difference indicated in the pulse number information between the two points;
Storage means for storing information on the relation amount calculated by the relation amount calculation means;
Information number adjusting means for adjusting the number of information when storing the relation amount in the storage means according to the moving distance between the two points;
Representative relation quantity computing means for computing a representative relation quantity based on a plurality of the relation quantities stored in the storage means after the number of pieces of information of the relation quantity stored in the storage means becomes a certain number or more;
Distance correction coefficient calculating means for calculating a distance correction coefficient for correcting the output of the distance sensor using the representative relation amount;
A correction coefficient computing device comprising:   The correction coefficient calculation device according to claim 1, wherein the relation amount is a pulse number per reference distance or a distance per unit pulse number.   The first condition is an accuracy condition for determining whether or not the information is within a predetermined error radius, a satellite number condition for determining whether or not the number of satellites that have received the signal is greater than or equal to a predetermined number, A speed condition for determining whether or not the vehicle is traveling at a certain speed or more, a straight line condition for determining whether or not a straight traveling state is indicated, and whether or not a constant speed moving state is indicated The correction coefficient calculation device according to claim 1, comprising any one or more of constant speed conditions.   The information number adjusting means uses a plurality of classifications according to an increase in the moving distance between the two points, and when storing the relation amount in the storage means, the movement between the two points from which the relation amount is calculated The correction coefficient calculation device according to any one of claims 1 to 3, wherein the number of pieces of information regarding the related amount is increased as the number of steps related to the distance increases.   The correction coefficient calculation according to any one of claims 1 to 4, wherein if any of the GPS information between the two points does not satisfy the first condition, the processing for the GPS information of the two points is terminated. apparatus. GPS information number determination means for determining whether or not a GPS information number condition that the number of GPS information between the two points is a predetermined number or more is satisfied,
The correction coefficient calculation device according to any one of claims 1 to 5, wherein when the GPS information count condition is not satisfied, the relation amount calculation unit excludes the relation amount calculation target. A gradient determination means for determining whether or not the gradient of a predetermined angle or more is being moved when the current GPS information is acquired;
The correction coefficient calculation device according to any one of claims 1 to 6, wherein when it is determined that the gradient is moving, the processing for the GPS information of the two points is ended. A self-position recognition device comprising the correction coefficient calculation device according to any one of claims 1 to 7,
A direction information acquisition means for acquiring direction information from the direction sensor is provided.
Self position recognition device comprising self position calculation means for calculating a movement locus of the self position and recognizing the self position based on the pulse number information, the azimuth information and the distance correction coefficient obtained by the correction coefficient calculation device. . GPS information acquisition step for periodically acquiring GPS information including reception time information and speed information based on signals from GPS satellites;
A pulse number information acquisition step of acquiring pulse number information from a distance sensor that outputs a pulse signal according to the amount of movement, at least simultaneously with the acquisition of the GPS information,
A correction coefficient calculation program for causing a computer to execute a process for calculating a distance correction coefficient for correcting a moving distance obtained based on the pulse number information based on the GPS information and the pulse number information,
A first condition determination step for determining whether or not a predetermined first condition is satisfied with respect to arbitrary two points of GPS information;
A moving distance calculating step of calculating a moving distance between the two points based on the GPS information between the two points when the GPS information of the two points satisfies the first condition;
A relation amount calculation step for calculating a relation amount between the movement distance between the two points calculated in the movement distance calculation step and the difference in the number of pulses indicated in the pulse number information between the two points;
An information number adjusting step of adjusting the number of information when storing the relation amount in the storage unit according to the moving distance between the two points;
A storage step of storing, in the storage means, information on the relation amount of the number of information after adjustment by the information number adjustment step;
A representative relation amount calculation step of calculating a representative relation amount based on the plurality of relation amounts stored in the storage means after the number of pieces of information of the relation amount stored in the storage means reaches a certain number or more;
A distance correction coefficient calculating step of calculating a distance correction coefficient for correcting the output of the distance sensor using the representative relation amount;
A correction coefficient calculation program comprising:   The information number adjusting step uses a plurality of classifications according to an increase in the moving distance between the two points, and when storing the relation amount in the storage means, the movement between the two points from which the relation amount is calculated The correction coefficient calculation program according to claim 9, wherein the number of pieces of information about the relation amount is increased as the steps related to the distance increase.   The correction coefficient calculation program according to claim 9 or 10, wherein when any GPS information between the two points does not satisfy the first condition, the processing for the GPS information of the two points is ended. A GPS information number determination step for determining whether or not a GPS information number condition that the number of GPS information between the two points is a predetermined number or more is satisfied,
The correction coefficient calculation program according to any one of claims 9 to 11, wherein when the GPS information count condition is not satisfied, the processing is not performed in the steps after the relation amount calculation step. A gradient determination step of determining whether or not a gradient of a predetermined angle or more is moving when the current GPS information is acquired;
The correction coefficient calculation program according to any one of claims 9 to 12, wherein when it is determined that the gradient is moving, the process on the GPS information of the two points is ended. A self-position recognition program comprising the correction coefficient calculation program according to any one of claims 9 to 13,
Having an orientation information acquisition step of acquiring orientation information from the orientation sensor;
Based on the pulse number information, the azimuth information, and the distance correction coefficient obtained by the correction coefficient calculation program, the self-position calculation step for calculating the movement locus of the own position and recognizing the own position is executed by the computer. Position recognition program.

Images