JP4247715B2 - Vehicle navigation device - Google Patents

Description

  The present invention relates to a vehicle navigation device that improves the current position detection accuracy.

  The vehicle travel distance is obtained by counting pulses (vehicle speed pulses) generated according to the rotation of the tire and multiplying this counted value by a coefficient (distance coefficient) for converting the number of pulses into a distance. The current position is being detected. However, since the vehicle navigation system does not know what type of tire is installed, it calculates and learns the distance count for each vehicle using the map data and past travel data, and the detection accuracy of the current position is increased. I try to raise it. The learning of the distance coefficient is usually performed by detecting that the vehicle has traveled a certain distance (straight road) on the map data using GPS positioning data and obtaining the number of vehicle speed pulses during this period. In order to further improve the detection accuracy, the altitude of the vehicle is also detected using an inclination angle sensor.

By the way, if the tire diameter or pressure changes during high-speed running, an error occurs only with the learned distance coefficient, and the current position detection accuracy deteriorates. For this reason, there has been proposed a method for correcting the travel distance by measuring a change in the diameter of the tire (Patent Document 1), but it is not necessarily sufficient for improving the current position accuracy. In addition, even when an error occurs due to traveling in an area where a difference in height occurs, an error occurs only with the learned distance coefficient, and the current position accuracy deteriorates. For example, on a slope with a steep slope, the error between the distance on the two-dimensional map and the actual distance increases, and the vehicle position detection accuracy deteriorates. On the other hand, when the tilt angle sensor is used, the sensor cost is increased accordingly.
Japanese Patent Laid-Open No. 10-239092

The present invention is intended to solve the above-described problems, and an object thereof is to improve the current position detection accuracy of a vehicle.
For this purpose, the present invention comprises distance coefficient learning means for learning a distance coefficient of a vehicle speed pulse with respect to the travel distance of the vehicle, and current position detection means for calculating the current position of the vehicle using the learned distance coefficient, The detecting means registers the positional deviation data in units of districts when traveling in an area where a positional deviation has occurred between the current position calculated using the distance coefficient and the current position calculated based on the GPS positioning data. Features.

  According to the present invention, it is possible to reduce the cause of detection error when detecting the current position using the learned distance coefficient as much as possible, and to greatly improve the current position detection accuracy.

Hereinafter, embodiments will be described.
FIG. 1 is a conceptual diagram illustrating an example of an embodiment of a vehicle navigation device.
The navigation control unit 1 is a control unit that executes each process of navigation. In this embodiment, the navigation control unit 1 includes a distance coefficient learning unit 2, a current position detection unit 3, and a search unit 4, and includes a GPS (Global Positioning System) position detection device ( Hereinafter, the GPS (10), the measurement data of the vehicle speed sensor 11 that generates a pulse according to the rotation of the tire is taken in, and the data retrieved from the map data 20 stored in the information storage means 12 by the retrieval means 4 is referred to. The distance coefficient is learned and the current position is detected. In addition to the map data, the information storage means 12 stores elevation polygon data 21, travel data 22, distance coefficient 23, and other various data 24 necessary for navigation.

  The distance coefficient learning means 2 travels 100 m using, for example, GPS 10 data on a straight road searched from the map data 20 stored in the information storage means 12, and calculates the number of pulses acquired from the vehicle speed sensor 5 during this distance. The coefficient (distance coefficient) to be converted into is calculated. The distance coefficient is learned by performing such processing as needed when the vehicle travels, and the accuracy is improved.

  The current position detection means 3 detects the current position from the learned distance coefficient and the vehicle speed pulse obtained from the vehicle speed sensor 11. In the present embodiment, for example, when traveling in an area that can be recognized as having a height difference from the elevation polygon data 21, or when using a distance coefficient even when there is no elevation polygon data, and when using GPS positioning data. If position errors occur frequently, it is judged that there is a difference in elevation, and the effectiveness of the distance coefficient is lowered.In the case of high speed driving, a distance coefficient corresponding to the traveling speed is used, and after distance coefficient learning When a position error is detected on a road that has traveled in the past, the current position cost that reflects the error is registered on the road, and as a result, the factors that cause the position detection error are minimized to improve the current position detection accuracy. I try to let them. Note that these processes are performed after the vehicle has traveled a predetermined distance or more and the distance coefficient has been accurately learned.

Next, an embodiment for improving the position detection accuracy using the elevation polygon data will be described.
FIG. 2 is a plan view for explaining an example of elevation polygon data. In this example, each small section having coordinates (latitude and longitude) at the four corners is a section of several tens of meters (for example, 30 m), and FIG. 2 (a) shows an altitude difference (in the area) for each small section ( Difference between maximum altitude and minimum altitude) [Delta] H1, [Delta] H2... FIG. 2 (b) shows the average altitude H1, H2... An example is shown.

  When traveling in the area where the elevation polygon data shown in FIG. 2A is set, there is a difference in height, so the error between the distance on the two-dimensional map and the actual distance becomes large, and the two-dimensional map The vehicle position detection accuracy detected above will deteriorate. Therefore, when traveling in an area where such elevation polygon data is set, it is assumed that the distance coefficient does not apply, and the effectiveness of the distance coefficient is reduced, for example, GPS positioning without using the distance coefficient Current position detection is performed using data to prevent a decrease in position detection accuracy. In the area where the elevation polygon data shown in FIG. 2 (b) is set, the distance coefficient is considered to be a case where the distance coefficient is not applicable when it can be recognized that there is an elevation difference of a predetermined value or more from the difference in average elevation of each area. It is sufficient to lower the effectiveness of.

FIG. 3 is a diagram showing a processing flow for detecting the position by lowering the effectiveness of the distance coefficient due to the difference in elevation, and the elevation polygon data is shown in FIG. 2A.
When the vehicle has moved, it is determined whether or not there is elevation polygon data (step S1). (Step S2), the present location is detected. In the case of the elevation polygon data shown in FIG. 2B, it is only necessary to reduce the effectiveness of the distance coefficient by adding a process for checking whether there is a difference in height from the difference in average elevation.

  In addition, even when traveling in an area where elevation polygon data is not set but there is a height difference, a position detection error will occur in the same way. In such a case as well, the effectiveness of the distance coefficient may be reduced. Next, the form will be described.

FIG. 4 is a diagram showing another processing flow for reducing the effectiveness of the distance coefficient.
If there is an error between the current position calculated using the learned distance coefficient and the current position calculated based on the GPS positioning data, for example, there was a position shift in units of districts such as large and small characters. Is registered in advance (position shift data). Normally, such a positional shift is considered to be caused by a difference in height, and can be handled in the same manner as in the case of elevation polygon data in the processing flow of FIG. Therefore, the next time the vehicle travels, it is determined whether or not there is positional deviation data (step S11). If so, the effectiveness of the distance coefficient is lowered (step S14), and the current location is detected using GPS positioning data. . When there is no positional deviation data, it is determined whether there is a deviation between the current position calculated using the distance coefficient and the current position calculated based on the GPS positioning data. The positional deviation data is registered in units of districts such as level (step S13), the effectiveness of the distance coefficient is lowered (step S14), and the current location is detected.

  The above example is mainly an example of traveling in an area where there is a height difference, but tire expansion also occurs during high-speed traveling on an expressway or a toll road. Thus, an example of registering or learning a distance coefficient corresponding to the traveling speed for such a road will be described.

FIG. 5 is a diagram showing a current position detection processing flow during high-speed traveling.
When traveling at high speed, the air pressure rises due to frictional heat of the tire, and the tire expands. For this reason, the moving distance becomes long even at the same rotation speed, and if the position is calculated using the distance coefficient learned at the normal speed, a deviation occurs from the current position calculated based on the GPS positioning data. Therefore, for example, a distance coefficient corresponding to the speed is learned and registered at intervals of about 10 km such as 80 km to 90 km per hour, 90 km to 100 km, 100 km to 110 km, and so on. Then, during high speed traveling, it is determined whether or not a distance coefficient corresponding to the traveling speed is registered (step S21). If not registered, a distance coefficient corresponding to the traveling speed is registered (step S22). Next, it is determined whether there is a distance shift using the registered distance coefficient (step S23). If there is a distance deviation, the already registered distance coefficient is corrected and learned (step S24). Thus, the current position is detected using the distance coefficient corresponding to the traveling speed at which no distance deviation occurs.

  When there is an error between the position calculated by GPS positioning data and the position calculated by the distance coefficient on a road once traveled, it is considered that an error has occurred due to the road, and a current location cost that reflects the error is assigned to the road. And register. For example, when the position is supposed to be 100 m from the starting point but detected as a position 110 m, the current location cost is 11/10. Next, an example of registering such a current location cost for a road will be described.

FIG. 6 is a diagram showing a current location detection processing flow using the current location cost.
When traveling, it is determined whether or not the road is a registered road cost (step S31). If the road is a registered road, the current road cost is used (step S32). Next, it is determined whether or not there is an error between the position calculated from the GPS positioning data and the position calculated from the distance coefficient (step S33). At this time, if the current location cost is not a registered road and there is a distance shift, the current location cost is registered (step S34). Also, if there is a distance shift in the current location cost, the current location cost is re-registered and learned. Thus, the current location is detected in consideration of the current location cost.

  In the above description, it is assumed that the map data used for learning the distance coefficient is accurate, but the map data may be incorrect. For example, even though the distance coefficient learning level has advanced considerably such that the position error from the current position calculated based on the GPS positioning data after the distance coefficient learning ends and after traveling a predetermined distance is below a predetermined value. In some cases, a distance shift may occur. In such a case, it is considered that there is a problem with the map data, and the current location cost as described above may be assigned to the road for registration, or the map data itself may be corrected.

  According to the present invention, since the current position detection accuracy can be remarkably improved, the industrial utility value is extremely large.

It is a conceptual diagram explaining the example of this Embodiment. It is a figure explaining the example of elevation polygon data. It is a figure which shows the processing flow which lowers the effectiveness of a distance coefficient and performs present location detection. It is a figure which shows the other processing flow which lowers the effectiveness of a distance coefficient and performs present location detection. It is a figure which shows the present location detection process flow at the time of high speed driving | running | working. It is a figure which shows the present location detection process flow using present location cost.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Navigation control part, 2 ... Distance learning means, 3 ... Current position detection means, 4 ... Search means, 10 ... GPS, 11 ... Vehicle speed sensor, 12 ... Information storage means.

Claims (1)

A distance coefficient learning means for learning a distance coefficient of a vehicle speed pulse with respect to a travel distance of the vehicle, and a current position detection means for calculating the current position of the vehicle using the learned distance coefficient,
The current position detection means registers position shift data in units of districts when traveling in an area where a position shift has occurred between the current position calculated using a distance coefficient and the current position calculated based on GPS positioning data. A navigation device for a vehicle.

Images