JP2013250064A - Stop determination method, program performing stop determination and stop determination device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To propose a new technique for performing stop determination for a moving body.SOLUTION: A stop determination method for a moving body includes: a first step of calculating a first error variance that is an error variance of angular acceleration in movement of the moving body; a second step of calculating a second error variance that is an error variance of angular velocity in the movement of the moving body; and a third step of calculating a first mean value that is a mean value of velocity in the movement of the moving body. When a moving state of the moving body is determined to be at a stop in any of determination using the first error variance, determination using the second error variance and determination using the first mean value, the moving body is determined to be at a stop.

Description

  The present invention relates to a moving body monitoring apparatus, and more particularly, to a moving body stop determination method, a stop determination program, and a stop determination apparatus equipped with a gyro sensor and an acceleration sensor.

  Conventionally, the use of inertial sensors has attracted attention in various fields such as so-called seamless positioning, motion sensing, and attitude control. As the inertia sensor, a gyro sensor, an acceleration sensor, a pressure sensor, a geomagnetic sensor, and the like are widely known, and these inertia sensors are used by being mounted on various moving bodies.

  For example, as described in Patent Document 1, a technique for determining stop of a vehicle using dispersion of output data of a gyro sensor or an acceleration sensor is disclosed.

Japanese Patent Laid-Open No. 9-96535

  In recent years, as a small and inexpensive inertial sensor, a MEMS (Micro Electro Mechanical Systems) sensor to which a semiconductor microfabrication technology is applied has been mounted on various moving bodies. However, these MEMS sensors are more susceptible to external influences such as shock, vibration, and temperature change than conventional sensors. For this reason, the stop determination described in Patent Document 1 may not be able to make an accurate stop determination. Conceivable.

  SUMMARY An advantage of some aspects of the invention is to solve at least a part of the problems described above, and the invention can be implemented as the following forms or application examples.

[Application Example 1]
The stop determination method according to this application example includes a first step of calculating a first error variance that is an error variance of angular acceleration in the movement of the moving body, and a second step that is an error variance of the angular velocity in the movement of the moving body. And a third step of calculating a first average value that is an average of speeds of movement of the moving body, wherein the moving state of the moving body includes the first step In any of the determination using the error variance of 1, the determination using the second error variance, and the determination using the first average value, it is determined to be stopped. It is characterized by that.

  According to this method, the influence of the error component included in the output of the gyro sensor can be eliminated by using the result in the first step and the result in the second step.

  In addition, when the vehicle travels straight at a constant speed, the result in the first step, the result in the second step, and the output data of the acceleration sensor are not changed, and there is a possibility that the stop is erroneously determined. . Therefore, the result in the third step is also used to estimate whether the moving body is stopped. Thereby, stop determination can be performed appropriately.

[Application Example 2]
In the stop determination method according to the application example, the moving body includes a gyro sensor and an acceleration sensor, the angular velocity is obtained from output data of the gyro sensor, and the angular acceleration is obtained from the angular velocity, The speed is preferably obtained from output data of the acceleration sensor.

  According to this method, the mobile body has a gyro sensor and an acceleration sensor, the angular velocity is obtained from the output data of the gyro sensor, and the speed is obtained from the acceleration sensor. It can be easily performed, and the accuracy of determination can be further enhanced by using two types of sensors.

[Application Example 3]
In the stop determination method according to the application example, the stop determination of the moving body using the second error variance is performed by comparing with a predetermined threshold, and the calibration with respect to the predetermined threshold is performed by the calibration. This is preferably performed when it is determined that the moving body is stopped.

  According to this method, a predetermined threshold value is used for determining whether the moving body is stopped using the second error variance, and calibration is performed when the moving body is determined to be stopped by performing calibration of the predetermined threshold value. Since the result calculated immediately before can be used as the data for the operation, even if the output of the sensor shifts due to the influence of the outside world, the threshold value can be maintained at an appropriate value for judgment. The stop determination based on the output data and the output data of the acceleration sensor data can be made appropriate.

[Application Example 4]
The program for determining whether to stop according to this application example is a program used in a mobile body having a gyro sensor and an acceleration sensor, and calculates angular velocity from output data of the gyro sensor, and calculates angular acceleration from the angular velocity. A process for calculating, a process for calculating a velocity from output data of the acceleration sensor, a process for calculating a first error variance which is an error variance of the angular acceleration, and a second error variance which is an error variance of the angular velocity. And a process of calculating an average value of the speeds, and the stop state of the moving body is determined by the first error variance, the second error variance, and the average value of the speeds. The threshold value calibration used for determining the stoppage due to the second error variance is performed when the moving body is determined to be stopped. To.

  According to this configuration, the stop determination program in the moving body having the gyro sensor and the acceleration sensor calculates the angular velocity from the output data of the gyro sensor, the processing of calculating the angular acceleration from the angular velocity, and the output data of the acceleration sensor. , Calculating the first error variance that is the error variance of the angular acceleration, processing calculating the second error variance that is the error variance of the angular velocity, and calculating the average value of the velocity And determining a stop state of the moving object based on an average value of the first error variance, the second error variance, and the speed, and calibrating a threshold value used for determining the stop due to the second error variance. By being performed when it is determined that the moving body is stopped, it is possible to appropriately make a determination regarding the stop of the moving body.

[Application Example 5]
The stop determination device according to this application example includes a gyro sensor and an acceleration sensor, calculates a angular velocity from output data of the gyro sensor, calculates an angular acceleration from the angular velocity, and outputs the acceleration sensor A process of calculating a velocity from data, a process of calculating a first error variance which is an error variance of the angular acceleration, a process of calculating a second error variance which is an error variance of the angular velocity, and an average of the velocity A stop state of the moving body is determined based on the average value of the first error variance, the second error variance, and the velocity, and a stop state due to the second error variance is calculated. The threshold value calibration used for the determination is performed when the moving body is determined to be stopped.

  According to this configuration, the stop determination device includes the gyro sensor and the acceleration sensor, and calculates the angular velocity from the output data of the gyro sensor, the processing of calculating the angular acceleration from the angular velocity, and the output data of the acceleration sensor. Processing to calculate velocity, processing to calculate first error variance that is error variance of angular acceleration, processing to calculate second error variance that is error variance of angular velocity, and processing to calculate average value of velocity And determining a stop state of the moving body based on an average value of the first error variance, the second error variance, and the speed, and calibration of a threshold value used for determining a stop due to the second error variance, By being performed when it is determined that the moving body is stopped, it is possible to appropriately perform the determination regarding the stop of the moving body. Note that the gyro sensor and the acceleration sensor may be included in the stop determination device, or may be included on the side of the moving body on which the stop determination device outside the stop determination device is mounted.

The figure for demonstrating a 1st process. The figure for demonstrating a 2nd process. The figure for demonstrating a 3rd process. The schematic block diagram of a navigation apparatus. The figure which shows the structural example of sensor data. The figure which shows the structural example of the data for detailed stop determination. The flowchart which shows the flow of a 1st navigation process. The flowchart which shows the flow of a detailed stop determination process. The figure which shows an example of the data structure of a memory | storage part. The flowchart which shows the flow of a 2nd navigation process. The flowchart which shows the flow of a stop condition calibration process.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the following drawings, the scale of each layer and each member is made different from the actual scale so that each layer and each member can be recognized. The figure used is for convenience of explanation.

(First embodiment)
FIG. 1 is a view for explaining calculation of error variance of angular acceleration (first step). In FIG. 1, the horizontal axis represents the time axis. Arrows shown downward indicate the detection timing of output data of the gyro sensor on the time axis. An interval between adjacent detection timings is called a unit time. The unit time may be 10 milliseconds, for example. Further, a predetermined number of unit times that are continuous are referred to as accumulation time. The accumulation time may be, for example, 20 times the unit time. In this case, if the unit time is 10 milliseconds, the accumulation time is 200 milliseconds. The error variance of the angular acceleration obtained in the first step is the error variance in the accumulation time.

  First, angular acceleration time-series data is acquired using detection results for a predetermined accumulation time of the gyro sensor. Specifically, time differentiation is performed on the angular velocity corresponding to the accumulation time to obtain time series data of angular acceleration.

  Here, as a method of calculating the angular acceleration, for example, a method of calculating a time difference between detection results of gyro sensors at adjacent detection times or a smoothing process on the detection results of the gyro sensor corresponding to the accumulation time are continuously performed. Various methods such as a method of calculating a typical curve and calculating a time derivative thereof can be applied. In the present embodiment, the angular velocity is represented as “ω”, the angular acceleration is represented as “α”, and the velocity is represented as “v”.

Thereafter, the error variance of the angular acceleration “α” during the accumulation time (hereinafter referred to as “angular acceleration error variance”) is calculated. The angular acceleration error variance is an index value indicating a variation in error of the angular acceleration “α”. The error variance can be defined in various ways. In this embodiment, the sample variance of the angular acceleration “α” data corresponding to the accumulation time is defined as the angular acceleration error variance. More specifically, an average value of angular acceleration data corresponding to the accumulation time is calculated. Then, the angular acceleration error variance is calculated by dividing the square sum of the difference between each angular acceleration data and the average value by the number of data. This angular acceleration error variance is expressed as “σ ² _α ”.

In the present embodiment, the angular acceleration error variance “σ ² _α ” is calculated while shifting the time by unit time. That is, the error variance of the angular acceleration “α” corresponding to the accumulation time is calculated while shifting the time by unit time. For example, it is assumed that the unit time is “10 milliseconds” and the accumulation time is “200 milliseconds”. In this case, the angular acceleration error variance “σ ² _α ” is calculated using time series data of the angular acceleration “α” for 200 milliseconds while shifting the time by 10 milliseconds. By performing threshold determination for the latest angular acceleration error variance “σ ² _α ” calculated in this way, a first stop determination is made as to whether or not the moving body is stopped.

Here, threshold determination for the angular acceleration error variance “σ ² _α ” will be described. The threshold determination is to determine whether the vehicle is moving or stopped in comparison with a preset threshold “θ _α ”. Further, the threshold value “θ _α ” may be a value determined by experiments or the like. For example, it is suitable to acquire sample data of angular acceleration error variance “σ ² _α ” using the detection results of the moving and stopped gyro sensors, and to separate the moving state and the stopped state using this sample data As a value, the threshold value “θ _α ” can be obtained.

If the angular acceleration error variance “σ ² _α ” is less than the threshold “θ _α ” (or less than the threshold) (σ ² _α <θ _α ), it is determined that the moving body is stopped. On the other hand, if the angular acceleration error variance “σ ² _α ” is equal to or greater than the threshold “θ _α ” (or exceeds the threshold) (σ ² _α ≧ θ _α ), it is determined that the moving body is moving.

As described above, in the present embodiment, the angular acceleration error variance “σ ² _α ” is calculated while shifting the unit time by unit time. Therefore, the threshold for the angular acceleration error variance “σ ² _α ” is detected at each detection timing of the gyro sensor. A determination can be made to determine a stop.

  FIG. 2 is a diagram for explaining calculation of the angular velocity error variance (second step). In FIG. 2, the horizontal axis indicates the time axis, and the downward arrow shown above the time axis indicates the detection timing of the gyro sensor. First, time-series data of the angular velocity “ω” for a predetermined accumulation time is obtained from the detection result of the gyro sensor. Then, similarly to the calculation of the angular acceleration error variance, the error variance of the data of the angular velocity “ω” for the accumulation time (hereinafter referred to as “angular velocity error variance”) is calculated.

Also in this case, for example, the sample variance of the data of the angular velocity “ω” can be defined as the angular velocity error variance. That is, the angular velocity error variance is calculated by dividing the sum of squares of the difference between each angular velocity data and the average value by the number of data. This angular velocity error variance is expressed as “σ ² _ω ”. The angular velocity error variance “σ ² _ω ” corresponds to the error variance (detection error variance) of the detection result of the gyro sensor.

Similar to the first step, also in the second step, the angular velocity error variance “σ ² _ω ” is calculated while shifting the time by unit time. For example, it is assumed that the unit time is “10 milliseconds” and the accumulation time is “200 milliseconds”. In this case, the angular velocity error variance “σ ² _ω ” is calculated using time series data of the latest angular velocity “ω” for 200 milliseconds while shifting the time by 10 milliseconds.

Next, the angular velocity error variance “σ ² _ω ” is collected for a predetermined sample value calculation time. The sample value calculation time is a time (sample time) for calculating the sample value, and is set to a time longer than the accumulation time, for example, an integral multiple of the accumulation time. Specifically, when the accumulation time is “200 milliseconds”, for example, “1000 milliseconds = 1 second”, which is five times as long, is set as the sample value calculation time. Then, the sample value is calculated using the angular velocity error variance “σ ² _ω ” included in the sample value calculation time.

The sample value can be selected as appropriate, but in the present embodiment, the average value of the angular velocity error variance “σ ² _ω ” included in the sample value calculation time is calculated and used as the sample value. In the present embodiment, this sample value (average value of angular velocity error variance) is expressed as “σ ² _AVE ”. In FIG. 2, a group of rectangles surrounded by a solid line corresponds to a combination of angular velocity error variances “σ ² _ω ” for calculating one sample value “σ ² _AVE ”.

A sample value “σ ² _AVE ” is calculated for each of a plurality of consecutive sample value calculation times, and a set of sample values “σ ² _AVE (1), σ ² _AVE (2) _,. .., Σ ² _AVE (n) ”is generated. The sample value change determination time is a time for determining the time change of the sample value, and is set to a time longer than the sample value calculation time, for example, an integer multiple of the sample value calculation time. For example, when the sample value calculation time is “1000 milliseconds = 1 second”, “5 seconds”, which is five times the time, is set as the sample value change determination time.

In this embodiment, a sample value set “σ ² _AVE (1), σ ² _AVE (2),..., Σ ² _AVE (n)” for the sample value change determination time is referred to as a “sample value set”. Called. The numbers in parentheses of the sample values that make up the sample value set indicate the sample value numbers. For example, when the sample value change determination time is “5 seconds” as described above, the sample value sets “σ ² _AVE (1), σ ² _AVE (2), σ ² _AVE (3), σ ² _AVE (4), σ ² _AVE (5) ”are generated. And based on the time change of the sample value which comprises a sample value group, it is determined whether the mobile body has stopped. This is referred to as a second stop determination.

The stop determination for the sample value will be described. The horizontal axis is the time axis, the vertical axis is the sample value, the sample values “σ ² _AVE ” constituting the sample value set are plotted in time series, and approximate lines of these sample values “σ ² _AVE ” are calculated. . The approximate straight line can be calculated using, for example, the least square method. That is, a straight line that minimizes the sum of squares of the distance from each sample value “σ ² _AVE ” to the approximate straight line is calculated. Then, threshold determination is performed for the calculated slope of the approximate line. Specifically, if the slope of the approximate straight line is less than a predetermined threshold (or less than or equal to the threshold), it is determined that the moving body has stopped. On the other hand, if the inclination of the approximate line is equal to or greater than the threshold (or exceeds the threshold), it is determined that the moving body is moving.

  FIG. 3 is a diagram for explaining average speed calculation (third step). In FIG. 3, the horizontal axis indicates the time axis, and the downward arrow shown above the time axis indicates the detection timing of the acceleration sensor. In the present embodiment, the detection time interval of the acceleration sensor will be described as “unit time”. The unit time is, for example, “10 milliseconds”.

  First, speed time-series data is acquired using detection results for a predetermined accumulation time of the acceleration sensor. The accumulation time is set to a time longer than the unit time, and for example, a time that is an integral multiple of the unit time is set. The accumulation time in the first step and the accumulation time in this step may be the same time. Specifically, if the unit time is “10 milliseconds”, for example, “200 milliseconds”, which is 20 times as long, may be set as the accumulation time.

Thereafter, the average of the speed “v” in the accumulation time (hereinafter referred to as “speed average”) is calculated. The speed average is an index value indicating an average value of the speed “v”. The average can be defined in various ways. In this embodiment, the average is defined as the average of the data of the speed “v” for the accumulation time. More specifically, the average value of the speed data for the accumulation time is calculated. This average speed is expressed as “Ave _v ”.

Similar to the first step, also in the third step, the speed average “Ave _v ” is calculated while shifting the time by unit time. That is, the speed average of the speed “v” for the latest accumulation time is calculated while shifting the time by unit time. For example, it is assumed that the unit time is “10 milliseconds” and the accumulation time is “200 milliseconds”. In this case, the speed average “Ave _v ” is calculated using time series data of the speed “v” for the latest 200 milliseconds while shifting the time by 10 milliseconds. In the third stop determination method, it is determined whether or not the moving body is stopped by performing a threshold determination on the speed average “Ave _v ” calculated in this way. This is referred to as third stop determination.

The threshold judgment for the speed average “Ave _v ” will be described. Previously sets a threshold value "v _v" for the speed average "Ave _v". The threshold value “v _v ” is desirably determined empirically as a value capable of separating the stopped state and the moving state.

If the speed average “Ave _v ” is less than the threshold “v _v ” (or less than the threshold) (Ave _v <v _v ), it is determined that the moving body is stopped. On the other hand, if the speed average “Ave _v ” is equal to or greater than the threshold “v _v ” (or exceeds the threshold) (Ave _v ≧ v _v ), it is determined that the moving body is moving.

As described above, in this embodiment, the speed average “Ave _v ” is calculated while shifting the time by unit time. Therefore, the threshold is determined for the speed average “Ave _v ” at each detection timing of the acceleration sensor, and the stop is stopped. Can be determined.

  Of course, the moving body can be determined to be stopped by using each of the first stop determination, the second stop determination, and the third stop determination. In addition, it is also possible to perform the stop determination of the moving body by using the first stop determination, the second stop determination, and the third stop determination method in combination.

  The problem is that when the first stop determination is used alone, there is an increased possibility that an erroneous determination will occur, particularly at the transition stage where the moving body moves from the moving state to the stopped state. When moving from a moving state to a stop state, the angular acceleration error variance gradually decreases, and stops when the angular acceleration error variance is less than (or below) a predetermined threshold. It will be determined. However, when the moving body is almost stopped but still moving slightly, the angular acceleration error variance level becomes the same as the level of angular acceleration noise observed at the time of stopping. It may be judged.

  Further, when the first stop determination and the second stop determination are used, the angular acceleration error variance value and the angular velocity error variance value are less than a predetermined threshold (or threshold value) when the moving body is traveling straight at a constant speed. Or less) and may be erroneously determined to be stopped by the threshold determination.

  In order to cope with such a problem, it is more preferable to perform stop determination as follows. That is, it is determined that the moving body is stopped by the first stop determination, the moving body is determined to be stopped by the second stop determination method, and the third stop determination method is also used. When it is determined that the moving body is stopped, it is determined that the moving body is stopped. That is, when all the determination results are “stop” using the first, second, and third stop determination methods as AND conditions, it is finally determined that the moving body is stopped.

  The present embodiment is an example of a case where a stop determination device that uses the first stop determination, the second stop determination, and the third stop determination is used for the navigation device 100. FIG. 4 shows a schematic configuration diagram of the navigation device 100. The navigation device 100 includes a processing unit 10, an operation unit 20, a display unit 30, a sound output unit 40, a communication unit 50, a clock unit 60, an IMU (Inertial Measurement Unit) 70, a storage unit 80, and the like.

  The processing unit 10 is a control device that comprehensively controls each unit of the navigation device 100 according to various programs such as a system program stored in the storage unit 80, and includes a processor such as a CPU (Central Processing Unit). Is done. The processing unit 10 performs inertial navigation calculation processing using the detection result of the IMU 70 to calculate the position (positional coordinates) of the moving body. And the process which displays the map which pointed the calculation position on the display part 30 is performed.

  Further, the processing unit 10 includes an angular acceleration calculation unit that obtains angular acceleration from the detection result of the gyro sensor 71, an angular acceleration error variance calculation unit that calculates an angular variance of angular acceleration for a predetermined accumulation time, and a predetermined accumulation time. It functions as an angular velocity error variance calculator that calculates angular velocity error variance, a velocity calculator that calculates velocity from the detection result of the acceleration sensor 73, and a velocity variance calculator that calculates velocity variance for a predetermined accumulation time. Further, the processing unit 10 functions as a stop determination unit that determines whether or not the moving body is stopped using the angular acceleration error variance, the angular velocity error variance, and the velocity variance.

  The operation unit 20 is an input device configured by, for example, a touch panel or a button switch, and outputs a pressed key or button signal to the processing unit 10. By operating the operation unit 20, various instructions such as destination input are input.

  The display unit 30 is configured by an LCD (Liquid Crystal Display) or the like, and is a display device that performs various displays based on display signals input from the processing unit 10. A navigation screen or the like is displayed on the display unit 30.

  The sound output unit 40 is a sound output device that includes a speaker or the like and outputs various sounds based on a sound output signal input from the processing unit 10. From the sound output unit 40, sound guidance and the like for navigation are output.

  The communication unit 50 is a communication device for performing wireless communication with an external device such as a management server of the navigation device 100. This function is realized by using a known wireless communication technique such as Bluetooth (registered trademark).

  The timepiece unit 60 is an internal timepiece of the navigation device 100 and includes an oscillation circuit having a crystal oscillator or the like. The time measured by the clock unit 60 is output to the processing unit 10 as needed.

  The IMU 70 is a sensor unit that includes an inertial sensor, and includes a gyro sensor 71 and an acceleration sensor 73, for example. The IMU 70 functions as a motion sensor unit that detects the operation of the apparatus main body, and is configured to detect the acceleration of each of the three orthogonal axes of the local coordinates previously associated with the sensor and the angular velocity around the axis of each detection axis. Is done. Note that the gyro sensor 71 and the acceleration sensor 73 may be independent sensors or may be integrated sensors.

  The storage unit 80 is configured by a storage device such as a ROM (Read Only Memory), a flash ROM, or a RAM (Random Access Memory), and various programs for realizing various functions such as a system program of the navigation device 100 and a navigation function. , Data, etc. are stored. In addition, it has a work area for temporarily storing data being processed and results of various processes.

  As shown in FIG. 4, the storage unit 80 stores a first navigation program 81 that is read as a program by the processing unit 10 and executed as a first navigation process (see FIG. 7). The first navigation program 81 stores a detailed stop determination program 811 executed as a detailed stop determination process (see FIG. 8) as a subroutine. The detailed stop determination program 811 includes a first stop determination program 8111, a second stop determination program 8112, and a third stop determination program 8113.

  The first navigation process is a process in which the processing unit 10 performs a detailed stop determination process to determine the stop of the moving object, performs a positioning process according to the determination result, and calculates the position of the moving object. . And the process part 10 produces | generates the navigation screen which plotted the calculation position, and displays it on the display part 30. FIG.

  The detailed stop determination process is a process in which the processing unit 10 performs stop determination based on the first stop determination method, the second stop determination method, and the third stop determination method described in the principle, and the moving body stops. This is a process for finely determining whether or not it is in progress. These processes will be described later in detail using a flowchart.

  The storage unit 80 also stores sensor data 83, detailed stop determination data 85, and navigation data 87 as data.

  The sensor data 83 is data in which the detection results of the IMU 70 are stored in time series, and an example of the data configuration is shown in FIG. In the sensor data 83, a detection time 831 of the IMU 70, an angular velocity detection value 833 detected by the gyro sensor 71, and an acceleration detection value 835 detected by the acceleration sensor 73 are stored in association with each other. One accumulation time is formed by k detection times 831.

  The detailed stop determination data 85 is processing data for the detailed stop determination process, and a data configuration example thereof is shown in FIG. The detailed stop determination data 85 is associated with a determination time 851 for determining stop, first stop determination data 853, second stop determination data 855, and third stop determination data 857. Remembered. One sample value calculation time is formed by m determination times 851.

  The first stop determination data 853 is data for first stop determination processing, and for each determination time 851, the angular acceleration error variance and the first stop determination result are stored in association with each other. . “Stop” or “Move” is stored in the first stop determination result.

  The second stop determination data 855 is data for the second stop determination process, and for each determination time 851, the angular velocity error variance, the sample value set, and the second stop determination result are associated with each other. Remembered. In the sample value set, n sample values corresponding to the sample value change determination time are stored. Further, “stop” or “movement” is stored in the second stop determination result.

  The third stop determination data 857 is data for third stop determination processing, and for each determination time 851, a speed average and a third stop determination result are stored in association with each other. “Stop” or “movement” is stored in the third stop determination result.

  The navigation data 87 is processing data for the first navigation processing. The navigation data 87 stores data such as the velocity vector of the moving object obtained by integrating the acceleration detected by the acceleration sensor 73 and the position of the moving object calculated using the velocity vector.

  FIG. 7 is a flowchart showing the flow of the first navigation process executed by the processing unit 10 in accordance with the first navigation program 81 stored in the storage unit 80.

  First, the processing unit 10 waits until the sensor data 83 is input from the IMU 70 (step A1; No). When the sensor data 83 is input (step A1; Yes), the detailed stop determination program stored in the storage unit 80 Detailed stop determination processing is performed according to 811 (step A3).

FIG. 8 is a flowchart showing the flow of the detailed stop determination process.
First, the processing unit 10 performs a first stop determination process (step B1). Specifically, time differentiation is performed on the time-series data of angular velocities for the past accumulated time (for example, 200 milliseconds) to obtain time-series data of angular acceleration. Then, using the acquired time series data of the angular acceleration, the angular acceleration error variance during the accumulation time is calculated and stored in the first stop determination data 853. Then, threshold determination is performed on the calculated angular acceleration error variance. If the angular acceleration error variance is less than a predetermined threshold (or less than the threshold), the result of the first stop determination (first stop determination data 853) 1 stop determination result) is set to “stop”.

  Next, the processing unit 10 performs a second stop determination process (step B3). Specifically, the average value of the angular velocity error variance is calculated by performing an averaging process on the angular velocity error variance for the latest sample value calculation time (for example, 1 second), and the sample value at the sample value calculation time is calculated. To do. Then, a sample value set for a predetermined sample value change determination time (for example, 5 seconds) is generated, and an approximate straight line is calculated using, for example, the least square method using the sample value included in the sample value set. . Then, threshold determination is performed on the calculated inclination of the approximate straight line. If the inclination is less than a predetermined threshold (or less than the threshold), the determination result (second stop determination result) by the second stop determination is “stop”. And

  Next, the processing unit 10 performs a third stop determination process (step B5). Specifically, the time series data of the velocity is obtained by integrating the time series data of the acceleration for the past accumulation time (for example, 200 milliseconds). Then, using the time series data of the acquired speed, a speed average in the accumulation time is calculated and stored in the third stop determination data 857. Then, a threshold determination is performed on the calculated speed average, and if the speed average is less than a predetermined threshold (or less than the threshold), the result of the third stop determination of the third stop determination data 857 (third stop determination) Result) is “stop”.

  Thereafter, the processing unit 10 determines whether the first stop determination result, the second stop determination result, and the third stop determination result are each stopped (steps B7, B9, and B11). If it is “stop” (step B7; Yes, B9; Yes, B11; Yes), it is finally determined that the moving body is stopped (step B13). Further, if the stop determination result is not “stop” in any one of the cases (step B7; No or B9; No or B11; No), the processing unit 10 finally determines that the moving body is moving (step B15). ).

  For example, in the detailed stop determination data 85 of FIG. 6, at the determination times “t1” and “t2”, the first stop determination result, the second stop determination result, and the third stop determination result are all “stop”. Therefore, the processing unit 10 finally determines that the moving body is stopped. However, at the determination time “t3”, the second stop determination result and the third stop determination result are “stop”, but the first stop determination result is “move”. Final decision that the body is moving. At the determination time “t4”, the first stop determination result and the second stop determination result are “stop”, but the third stop determination result is “move”. Final decision that the body is moving. After making the final determination, the processing unit 10 ends the detailed stop determination process.

  Returning to the first navigation process of FIG. 7, the processing unit 10 determines whether the determination result of the detailed stop determination process is stop (step A5), and if it is a stop determination (step A5; Yes), the position Is held (step A7). That is, during the period determined to be stopped, the latest position calculated by the inertial navigation calculation process is held (fixed) at that position so as not to change.

  On the other hand, when it determines with the mobile body moving (step A5; No), the process part 10 performs an inertial navigation calculation process (step A9). Specifically, the velocity vector of the moving body is calculated by integrating the acceleration detected by the acceleration sensor 73. Then, the moving direction and moving distance per unit time are obtained from the calculated velocity vector, and added to the previous calculated position, thereby newly calculating / updating the position of the moving body.

  Next, the processing unit 10 updates the display of the navigation screen of the display unit 30 at the latest calculated position (step A11). Then, the processing unit 10 determines whether or not to end the process (step A13). If it is determined that the process is not yet ended (step A13; No), the processing unit 10 returns to step A1. When it is determined that the process is to be ended (step A13; Yes), the processing unit 10 ends the first navigation process.

  As described above, the angular acceleration is calculated by performing time differentiation on the angular velocity detected per unit time by the gyro sensor 71 mounted on the moving body. Then, an error variance of angular acceleration for a predetermined accumulation time is calculated, and a threshold determination for the error variance is performed, thereby performing a first stop determination for determining whether or not the moving body is stopped.

  The angular velocity detected by the gyro sensor 71 may include an error component (noise component) due to the influence of the outside world such as temperature and vibration. Therefore, in the present embodiment, the time series angular velocity detected by the gyro sensor 71 is time-differentiated to calculate the angular acceleration, thereby eliminating the influence of the error component included in the angular velocity. Then, by determining whether or not the moving body is stopped using the error variance of the angular acceleration, it is possible to appropriately determine whether the moving body is stopped.

  Further, the angular velocity error variance corresponding to the accumulation time is calculated from the detection result of the gyro sensor 71. Then, a sample value set for a predetermined sample value change determination time is generated using an average value of error variances of angular velocities for a predetermined sample value calculation time as a sample value. Then, based on the temporal change of the sample values included in the generated sample value set, a second stop determination is performed to determine whether or not the moving body is stopped.

  As described above, since various error components can be included in the detection result of the gyro sensor 71, the error component is also superimposed on the calculated error distribution of the angular velocity. However, instead of using the error variance of the angular velocity when viewed in one time section, the time variation of the error distribution of the angular velocity viewed in a plurality of time sections is used to appropriately determine the stop of the moving body. be able to.

  Further, the acceleration is detected by the acceleration sensor 73 mounted on the moving body, and integration is performed on the acceleration detected every unit time, thereby calculating the speed. And the average of the speed | rate for predetermined | prescribed accumulation | storage time is calculated, and the 3rd stop determination which determines whether the moving body has stopped by performing the threshold value determination with respect to the said speed average is performed.

  As described above, in a state where the moving body is moving straight at a constant speed, there is no change in the sample set of angular acceleration and angular velocity, and there is a high possibility that it will be less than the stop determination threshold (or less than the threshold). Since it may be erroneously determined as a stop, the stop determination of the moving body can be appropriately performed by using the average speed.

  Further, the final determination of the moving body stop determination is performed by using the first stop determination, the second stop determination, and the third stop determination in combination. That is, when all of the determination result of the first stop determination, the determination result of the second stop determination, and the determination result of the third stop determination are all stopped, it is finally determined that the moving body is stopped, On the other hand, if the determination result is movement, it is finally determined that the moving body is moving. As described above, the moving body stop determination can be reliably performed by performing the stop determination of the moving body using the three types of stop determination methods in combination.

(Second Embodiment)
The present embodiment is an embodiment in which the second stop determination is simpler than the second stop determination in the first embodiment. In the description of the present embodiment, regarding the same functions as those described in the first embodiment, the block diagram and the flowchart in the first embodiment are used, and the description thereof is omitted.

  In the second stop determination in the present embodiment, the stop of the moving body is determined by performing a threshold determination on the error variance of the angular velocity detected from the output data of the gyro sensor 71. That is, the stop determination is performed by performing the threshold determination for the angular velocity error variance under the condition that the angular velocity error variance is less than the predetermined simple stop determination threshold (or less than the simple stop determination threshold). Hereinafter, the second stop determination in the present embodiment is referred to as simple stop determination.

  In this case, it is preferable to calibrate the simple stop determination threshold. When a four-wheeled vehicle is assumed as the moving body, the gyro sensor 71 may detect the vibration of the vehicle body due to the operation of the engine if the vehicle is idling even if the vehicle is stopped. Further, when the air conditioner is operating in the vehicle, the gyro sensor 71 may detect vibration due to the operation. As described above, even if the vehicle is simply stopped, the overall value of the angular velocity detected by the gyro sensor 71 changes each time depending on the situation of the moving body. Therefore, it is preferable not to use a fixed threshold value but to make a stop determination by setting the threshold value variably.

  Therefore, when it is determined that the moving body is stopped by any one of the first stop determination, the second stop determination, and the third stop determination, the stop determination condition (stop condition) ) Is calibrated. In the subsequent stop determination, the stop determination is performed using the calibrated stop condition. An embodiment in this case will be described below.

  FIG. 9 is a diagram illustrating an example of a program and a data configuration stored in the storage unit 80 of the navigation device 100 according to the present embodiment.

  The storage unit 80 stores a second navigation program 82 executed as a second navigation process (see FIG. 10) as a program. The second navigation program 82 includes a detailed stop determination program 811, a stop condition calibration program 813 executed as a stop condition calibration process (see FIG. 10), and a simple stop executed as a simple stop determination process. A determination program 815 is included.

  The storage unit 80 also stores sensor data 83, detailed stop determination data 85, navigation data 87, simple stop determination data 88, and a simple stop determination threshold 89 as data. The simple stop determination data 88 is processing data for the simple stop determination process, and includes angular velocity error variance used for the simple stop determination, a determination result of the simple stop determination, and the like. The simple stop determination threshold value 89 is a threshold value set in the stop condition calibration process.

  FIG. 10 is a flowchart showing the flow of the second navigation process. Note that the same steps as those in the first navigation process in FIG. 7 are denoted by the same reference numerals and description thereof is omitted, and steps different from those in the first navigation process will be mainly described.

  First, the processing unit 10 executes a stop condition calibration process according to the stop condition calibration program 813 stored in the storage unit 80 as an initial calibration (step C1). The stop condition calibration process as the initial calibration is a process to be executed before the movement of the automobile starts after the automobile engine is started when the moving body is an automobile.

  FIG. 11 is a flowchart showing the flow of the stop condition calibration process. First, the processing unit 10 waits until data is input from the IMU 70 (step D1; No), and when data is input (step D1; Yes), a detailed stop determination process according to the detailed stop determination program 811 of the storage unit 80. (Step D3). The process flow of the detailed stop determination process is as shown in FIG.

  If the determination result of the detailed stop determination process is “stop” (step D5; Yes), the processing unit 10 performs a simple stop determination threshold setting process (step D7). Specifically, when the moving body is an automobile, the angular velocity error variance (angular velocity error variance) during the suspension period is calculated using the time-series data of the angular velocity detected by the gyro sensor 71 while the automobile is stopped. calculate. Then, the processing unit 10 sets the simple stop determination threshold value 89 using the calculated angular velocity error variance. For example, an average value (angular velocity average value) of angular velocities detected during the stop period is calculated, and a value obtained by adding angular velocity error variance to the angular velocity average value is set as the simple stop determination threshold value 89. Then, the processing unit 10 ends the stop condition calibration process.

  Returning to the second navigation process of FIG. 10, after executing the stop condition calibration process, the processing unit 10 waits until data is input from the IMU 70 (step C2; No). Step C2; Yes), simple stop determination processing is performed in accordance with the simple stop determination program 815 stored in the storage unit 80 (step C3). The simple stop determination process is realized by calculating an angular velocity error variance corresponding to the accumulation time detected by the gyro sensor 71 and performing threshold determination using the simple stop determination threshold 89 for the angular velocity error variance. .

  Next, when the determination result of the simple stop determination process is “stop” (step C5; Yes), the processing unit 10 holds the position (step A7). If the determination result of the simple stop determination process is not “stop” but “movement” (step C5; No), the processing unit 10 performs an inertial navigation calculation process (step A9).

  In the second navigation process, the stop condition calibration process is executed as the initial calibration. However, the stop condition calibration process may be executed at another predetermined execution timing. For example, the stop condition calibration process may be executed with the execution timing as the timing at which the moving body changes from the moving state to the stopped state.

  The stop condition calibration process has been described as the process for calibrating the stop condition when performing the simple stop determination process using angular velocity error variance. However, the stop condition when performing the detailed stop determination process is calibrated. It is good also as processing to do. In this case, for example, when it is determined that the moving body is stopped by the simple stop determination process, the threshold determination threshold value (hereinafter referred to as “fine determination threshold value”) used as the stop condition in the detailed stop determination process is determined. This is performed as a process of setting “.

  More specifically, when it is determined to be stopped by the simple stop determination process, time differentiation is performed on the time-series angular velocity detected by the gyro sensor 71 to obtain sample data of angular acceleration. Then, the angular acceleration error variance is calculated using the acquired sample data of angular acceleration, and the fine determination threshold value is set using the angular acceleration error variance. The fine determination threshold value can be, for example, a value obtained by adding the angular acceleration error variance to the angular acceleration average value during the stop period.

  In the embodiment described above, the error variance of angular acceleration and angular velocity is defined as the sample variance of angular acceleration and angular velocity data, respectively, but the definition of error variance is not limited to this. For example, the unbiased variance of angular acceleration and angular velocity may be defined as error variance, and the standard deviation represented by the positive square root of sample variance or unbiased variance may be defined as error variance. Regardless of which definition is used, stop determination can be performed by the same method as in the above embodiment.

  In the embodiment described above, the second stop determination is performed using the average value of the angular velocity error variance for the sample value calculation time as the sample value. However, the sample value is not limited to this. For example, instead of an average value of angular velocity error variance, a median value may be selected as a sample value. The error variance of the angular velocity data for the sample value calculation time may be used as the sample value. That is, the time-series data of the angular velocity detected every unit time is accumulated for the sample value calculation time, the error variance of the angular velocity data for the accumulated sample value calculation time is calculated, and the sample value at the sample value calculation time is calculated. Also good.

  In the second stop determination in the first embodiment, it has been described that the approximate straight line of the sample values constituting the sample value set is obtained by using the least square method, but the approximation of the sample values is limited to this. I can't. For example, the sample value may be approximated by a curve instead of a straight line, and the threshold value may be determined for a value (differential value of the approximate curve) obtained by differentiating the approximate curve to determine stop.

  In the third stop determination in the above-described embodiment, the average speed is calculated from the speed data corresponding to the accumulation time. However, the average speed is not limited to this. For example, the stop may be determined by performing threshold determination for the median value or the mode value.

  Note that various times used for the stop determination in the above embodiment can be appropriately changed. That is, values set as various times of unit time, accumulation time, sample value calculation time, and sample value change determination time may be appropriately set according to the system to which the present invention is applied.

  The embodiment according to the present invention has been described above, but the present invention is not limited to the above-described embodiment. The present invention can be widely applied without departing from the spirit of the present invention.

  DESCRIPTION OF SYMBOLS 100 ... Navigation apparatus, 10 ... Processing part, 20 ... Operation part, 30 ... Display part, 40 ... Sound output part, 50 ... Communication part, 60 ... Clock part, 70 ... IMU, 71 ... Gyro sensor, 73 ... Acceleration sensor, 80: Storage unit.

Claims (5)

A first step of calculating a first error variance which is an error variance of angular acceleration in the movement of the moving body;
A second step of calculating a second error variance that is an error variance of angular velocity in the movement of the moving body;
A third step of calculating a first average value that is an average of speeds of movement of the moving body;
Including
The moving state of the moving body is determined to be stopped in any of the determination using the first error variance, the determination using the second error variance, and the determination using the first average value. A stop determination method, characterized in that it is determined that the vehicle has stopped when it is stopped. The moving body has a gyro sensor and an acceleration sensor,
The angular velocity is obtained from output data of the gyro sensor,
The angular acceleration is obtained from the angular velocity,
The stop determination method according to claim 1, wherein the speed is obtained from output data of the acceleration sensor. The determination of the stop of the moving body using the second error variance is performed by comparing with a predetermined threshold,
The stop determination method according to claim 1 or 2, wherein the calibration with respect to the predetermined threshold is performed when it is determined that the moving body is stopped. A program used in a moving object having a gyro sensor and an acceleration sensor,
A process of calculating an angular velocity from output data of the gyro sensor;
A process of calculating an angular acceleration from the angular velocity;
A process of calculating a speed from output data of the acceleration sensor;
A process of calculating a first error variance which is an error variance of the angular acceleration;
A process of calculating a second error variance which is an error variance of the angular velocity;
A process of calculating an average value of the speeds;
Including
The stop state of the moving body is determined by an average value of the first error variance, the second error variance, and the speed,
A program for performing stop determination, wherein calibration of a threshold value used for determination of stop due to the second error variance is performed when the moving body is determined to be stopped. It has a gyro sensor and an acceleration sensor,
A process of calculating an angular velocity from output data of the gyro sensor;
A process of calculating an angular acceleration from the angular velocity;
A process of calculating a speed from output data of the acceleration sensor;
A process of calculating a first error variance which is an error variance of the angular acceleration;
A process of calculating a second error variance which is an error variance of the angular velocity;
A process of calculating an average value of the speeds;
Including
The stop state of the moving body is determined by an average value of the first error variance, the second error variance, and the speed,
A stop determination apparatus, wherein calibration of a threshold value used for determination of stop due to the second error variance is performed when it is determined that the moving body is stopped.

Images