JPH04213019A - Position detection accuracy determination method and vehicle guidance device using the method - Google Patents

Abstract

PURPOSE:To make it possible to perform integrated judgment of the position of a vehicle and to guide the vehicle adequately by distinguishing the state wherein there is a candidate road and the road having the largest number of similarities can be specified and the state wherein the road is not specified or there is no candidata road at all, and performing the judgment corresponding to the number of registered candidate roads. CONSTITUTION:All roads, which are located in a vehicle-presence probability region with an estimated position as the center, are selected based on the estimated position of own vehicle which is determined by a bearing sensor 3, distance sensors 4 and a road-map memory 5. The change in estimated position and the similarity with each road pattern are computed. The road whose similarity is a specified value or more is registered as a candidate road. In a position detecting device 8, the state wherein there is the candidate road and the road having the largest number of the similarities can be specified and the state wherein the road is not specified and there is no candidate road at all are distinguished. How many candidate roads remain and the like are judged in response to the number of the registered candidata roads. Thus, the integrated reliability of detection is judged, and the adequate guidance can be performed.

Description

[Detailed description of the invention]

0001

[Industrial Application Field] The present invention relates to a position detection accuracy determination method for comprehensively determining the accuracy of the current position of a vehicle detected by a position detection device included in a vehicle navigation system, the certainty of the road on which it is traveling, etc. The present invention also relates to a vehicle guidance device for effectively realizing route guidance and route guidance according to the determination results.

0002

[Background Art] Conventionally, as a method for detecting the position of a vehicle traveling at any point on a road transportation network, a distance sensor, a direction sensor, and a processing device that performs the necessary processing on the output signals from both sensors are used. A dead reckoning navigation system has been proposed which obtains the vehicle's current position data by integrating the distance change and azimuth change that occur as the vehicle travels.
This method has a problem in that errors that the distance sensor and direction sensor inevitably have accumulate as the vehicle continues to travel, and errors included in the obtained current position data also accumulate.

[0003] Considering these problems and assuming that the vehicle runs on the road, the estimated position of the vehicle is determined based on data accumulated by obtaining the distance traveled and the amount of change in direction at each predetermined timing. , and obtain the limit error amount of the estimated position determined based on the error between the integrated data and the road map, and correspond to all roads located within the limit error amount range (vehicle existence probability region) centered on the estimated position. The estimated position is registered as the self-position, the correlation coefficient for each road of the registered estimated position is calculated, the correlation coefficient showing the least error for the road is selected, and the selected correlation coefficient is calculated. A map matching method is known that outputs the estimated position corresponding to the number as the current position (Japanese Patent Laid-Open No. 6
3-148115, Japanese Patent Application Laid-Open No. 64-53112, Doi et al., "Development of Automobile Position Detection Method," Sumitomo Electric No. 137, pp. 105-112, September 1990).

[0004] With this method, the distance traveled and the amount of change in direction are obtained and integrated at predetermined timings, and based on the integrated data, the vehicle's own position on any road on the road map stored in memory is determined. When detecting, in addition to obtaining the estimated position based on the integrated data,
Obtain the marginal error corresponding to the estimated position determined based on the error between the cumulative data and the road map, and calculate the estimated position automatically by making it correspond to all roads located within the marginal error around the estimated position. Register the estimated position as a position, calculate the correlation coefficient for each road at the registered estimated position, select the correlation coefficient that shows the least error for the road, and then perform estimation corresponding to the selected correlation coefficient. The position can be output as the current position.

[0005]

However, even if the above method is used, the current position of the detected vehicle is always 100%.
% accuracy cannot be determined. For example, in the above method, even if an attempt is made to select the correlation coefficient that shows the smallest error for a road, if the correlation coefficient for any road is extremely low, it will be determined based on the integrated data that no matching road exists. The vehicle will be driven relying on its own estimated position.

[0006] Furthermore, when the correlation coefficients corresponding to all roads located within the limit error range take similar values, the correlation coefficient that shows the least error for the road is selected and selected. Even if the estimated position corresponding to the calculated correlation coefficient is output as the current position, it cannot be said that the probability that the current position is located on another road is small. Therefore, while registering a plurality of roads, the vehicle continues to drive without being able to determine which road it is on.

[0007] As described above, even though the certainty of the detected current position of the vehicle is unknown, guiding the driver without indicating the degree of uncertainty may be rather dangerous. Therefore, the present invention provides a method for finding and outputting the position of your own vehicle on a road using a map matching method.
To provide a position detection accuracy determination method that comprehensively determines the detection accuracy of a vehicle position, and to provide a vehicle guidance device that can provide appropriate vehicle guidance to a driver based on the results determined by this method. The purpose is to provide.

[0008]

[Means for Solving the Problem] In order to achieve the above object, the position detection accuracy determination method of claim 1 obtains and integrates the traveling distance and the amount of change in direction at each predetermined timing, and based on this integrated data, Obtain the determined estimated position of the vehicle, select all roads located in the vehicle existence probability region centered on the estimated position from the road map stored in memory,
Calculate the degree of similarity between the change in the estimated position and each road pattern, register roads with a degree of similarity above a certain value as candidate roads, identify the road with the highest degree of similarity among the registered candidate roads, In the vehicle position detection method that sets the current position of the vehicle on the road, there are two situations in which there are candidate roads and the road with the highest degree of similarity has been identified, and another where there are no candidate roads or the road with the highest degree of similarity has been identified. A method of determining vehicle position detection accuracy according to a first criterion for distinguishing between a state in which a road with a high degree of similarity cannot be identified and a second criterion according to the number of registered candidate roads. be.

Further, the vehicle guidance device according to claim 4 has a road map memory storing a road map, and obtains and integrates travel distance and azimuth change amount at each predetermined timing, and calculates the estimated position of the vehicle based on this integrated data. The estimated position detection means detects the estimated position, and selects all roads located in the vehicle presence probability region centered on the estimated position from the road map stored in the road map memory, and detects changes in the estimated position and each road pattern. The system calculates the degree of similarity between roads, registers roads with a degree of similarity above a certain value as candidate roads, identifies the road with the highest degree of similarity among the registered candidate roads, and locates the current position of the vehicle on that road. The vehicle guidance device has a position specifying means to set, and there are two states: a state where there is a candidate road and the road with the highest degree of similarity can be specified by the position specifying means, and a state where there is no candidate road or there is no candidate road. The vehicle position detection accuracy is determined according to a first criterion that distinguishes between a state where the road with the highest degree of similarity cannot be identified, and a second criterion that corresponds to the number of registered candidate roads. The position detection accuracy determining means includes a position detection accuracy determining means, and an information providing means for changing the manner of providing information to the driver according to the determination result of the position detection accuracy determining means.

[0010] The vehicle guidance device has means for calculating an optimal route from the departure point to the destination of the vehicle, and the information providing means indicates the identified vehicle position and the calculated optimal route to the driver along with a road map. The method of guidance along the optimum route may be changed depending on the determination result of the position detection accuracy determination means (Claim 7).

[0011]

[Operation] According to the position detection accuracy determining method of claim 1, it is possible to determine whether a road has been identified by the first criterion, and how many candidate roads remain by the second criterion. can be determined. That is,
If there is one candidate road and the road with the highest degree of similarity has been identified, it can be assumed that the probability that the driver is driving on this road is quite high. If a road with a high degree of similarity has not been identified, the road has been lost, and the reliability of the detected position will be quite low. In addition, even if the road you are driving on can be identified, if it is selected from among many candidate roads, the reliability will be relatively low; If so, the probability that you are driving on that road increases.

[0012] The reliability and accuracy of the detected vehicle position can be determined using the above-mentioned criteria. Note that the longer the road is identified or the road is lost, the higher the probability of being in that state increases, and if it does not continue for a long time, the probability of being in that state decreases, so in the above method, the first criterion, In addition to the second criterion, if a third criterion is adopted that takes into consideration the period of time during which any of the judgment states according to the first criterion continues, more accurate judgments can be made (Claim Item 2).

[0013] Furthermore, in the above method, the first criterion,
In addition to the second criterion, a fourth criterion may be adopted that takes into account that the vehicle position signal is obtained by means of obtaining vehicle position detection information from outside the vehicle (claim 3). This is because information obtained from outside the vehicle can also be useful vehicle position information. According to the vehicle guidance system of claim 4, when the position detection accuracy can be determined by the method of claim 1, the method of providing information by the information providing means can be changed according to the determination result, so that the driver can: It is possible to continue driving while always knowing with what degree of certainty the currently indicated vehicle position has been calculated.

[0014] Also in this vehicle guidance system, the third
Alternatively, if the fourth criterion is adopted, the position detection accuracy can be determined more appropriately (claims 5 and 6). In the case of claim 7, the probability that the vehicle is running on the optimal route can be determined according to the determination result of the position detection accuracy determining means, so if the probability that the vehicle is running on the optimal route is high, If the probability that the vehicle is traveling along the optimal route is low, it is possible to refrain from guiding the vehicle or change the guidance method to ensure that the accuracy of vehicle position detection is always commensurate with the driver. It is possible to carry out further guidance.

[0015]

Embodiments Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. FIG. 1 is a block diagram showing a vehicle guidance system for implementing the position detection accuracy determination method of the present invention. As shown in FIG. 1, the vehicle guidance system includes a display 2, a console 1, a direction sensor 3, a distance sensor 4, a road map memory 5 storing road map data, and a road map memory 5. A memory drive 6 that reads stored data, and a travel distance and direction sensor 3 that is detected by a distance sensor 4.
a position detection device that integrates the amount of change in the driving direction detected by each, and detects the vehicle position based on a comparison between this integrated data and the road map data read out by the memory drive 6, and calculates the overall reliability of the detected position. 8 (estimated position detection means, position specification means, position detection accuracy determination means)
and a navigation system that performs various controls such as reading out a road map of a predetermined range, generating display data for guiding the vehicle, controlling the audio output device 10, controlling the display device 2, and controlling the position detection device 8. A controller 9 (display control means) and a G receiving radio waves from the orbiting satellites 7a to 7c.
PS receiver 7.

To explain in more detail, the console 1 is as follows:
It has a key input board (not shown) for starting and stopping the device, moving a cursor on the screen, scrolling the road map displayed on the screen, etc. On display 2,
A transparent touch panel is attached to a screen such as a CRT or liquid crystal panel. The display device 2 displays an initial setting menu supplied from the navigation controller 9,
The driver touches the display position on the screen to set initial data such as the destination.

The road map memory 5 is composed of large-capacity storage media such as a CD-ROM, an IC memory card, and a magnetic tape. The road map memory 5 is
Road maps (highway national highways, motorways, general national roads, major local roads, general prefectural roads, general city roads of designated cities,
Road map data consisting of a combination of nodes and links is divided into meshes (including other community roads), and road map data consisting of a combination of nodes and links is generated for each mesh. ), and a lower layer that also includes data on non-arterial roads (hereinafter referred to as "general roads") such as general prefectural roads, general city roads in designated cities, and other community roads in addition to main roads. I remember it separately from the map.

[0018] Here, a node generally refers to a coordinate position for specifying a branching point or a turning point on a road, and a node representing a branching point is called a turning point node, and a turning point on a road (a turning point is a turning point). ) is sometimes called an interpolation point node. The node data includes a node number, an address of an adjacent mesh node corresponding to the node, an address of a link connected to the node, and the like.

[0019] A link is a link between each branch point node. Link data includes link numbers, addresses of the start and end nodes of the link, link distance, direction to pass through the link, time required in that direction, road type, road width, and traffic regulation data such as one-way streets and toll roads. Become. The orientation sensor 3 detects changes in orientation as the vehicle travels, and can use a geomagnetic sensor, a gyro, or the like.

The distance sensor 4 detects the speed of the vehicle or
The distance traveled is detected based on the number of rotations of the wheels, and wheel speed sensors, vehicle speed sensors, etc. can be used. The navigation controller 9 has a function of obtaining the vehicle position from the position detection device 8 and displaying the current position and destination overlappingly on the map. Specifically, it consists of a microcomputer, graphics processing processor, image processing memory, etc., and is used to display menu screens, search maps, change scales,
It scrolls, displays the current position and direction of the vehicle, displays destinations and landmarks, displays direction and distance to the destination, etc.

[0021] The GPS receiver 7 has three or four GPs.
By decoding the pseudo-noise codes received from the S satellites, it measures the arrival time of radio waves from each satellite and calculates its estimated position on the ground. The position detection device 8 calculates traveling trajectory data by integrating the distance data detected by the distance sensor 4 and the direction change data detected by the azimuth sensor 3, and stores the traveling trajectory data and the road map memory 5. This method detects the vehicle position on the road, taking into consideration the probability of the presence of the vehicle, based on comparison with the road pattern obtained by the GPS receiver 7 (so-called map matching method). The vehicle position is appropriately corrected using data related to the estimated position, and the final position and orientation of the vehicle are output.

A functional block diagram of the position detection device 8 is shown in FIG. The input preprocessing unit 22 inputs the azimuth data sampled from the azimuth sensor 3 and the distance data sampled from the distance sensor 4, performs calibration and filtering, and obtains the optimal estimated azimuth. The current position estimating unit 23 calculates the estimated position by dead reckoning from the optimal estimated direction and distance data obtained from the input preprocessing unit 22, and also takes in the corrected current position obtained from the position resetting unit.
The above estimated position will be updated. The traveling road estimating unit 24
While determining a vehicle existence probability region centered on the determined estimated position, it iteratively calculates the degree of similarity between the road network data obtained from the road map memory 5 and the estimated position,
The vehicle position on the link included in the vehicle presence probability area is calculated. Here, the vehicle presence probability region refers to a region where the probability of vehicle presence is greater than a certain location, which is determined based on the accuracy of the vehicle position estimation method. The position reset unit 25 corrects the estimated position using the vehicle position obtained by the above method.

By the way, the position error between the estimated vehicle position obtained from the current position estimating section 23 and the actual corrected vehicle position is plotted on the horizontal axis, and the degree of similarity with the road network data is calculated based on the above estimated position. If the probability that the vehicle position matches the link is taken on the vertical axis, a correlation as shown in FIG. 3 is obtained. That is, if the position error is small, the optimal vehicle position can be reliably obtained on the link while the vehicle is traveling for a while and the similarity between the estimated position and the road pattern is repeatedly calculated. In other words, the vehicle position can be narrowed down to a specific link. This state is called a "road detection state." Conversely, as the position error increases, the probability that the vehicle can be located on a specific link decreases even if the similarity with the road pattern is calculated, and when the position error becomes large (Fig. 3), no matter how many times you calculate the similarity, you will not be able to find a match with the road. In this case, it is no longer possible to locate the vehicle on the road. This state is called a "road non-detection state." If this happens, the position reset unit 25 will not be able to correct the estimated position, so the vehicle position output from the position output unit 26 will remain the estimated position based on dead reckoning navigation.

Furthermore, the number of links included in the vehicle existence probability area is referred to as the "number of registered links." The management judgment unit 21
It knows whether the current state is a "road detection state" or a "road non-detection state" and also knows the "number of registered links." Furthermore, in order to correct the estimated position of the vehicle, the management judgment unit 21 corrects the vehicle position using the estimated position PG acquired by the GPS receiver 7 under certain conditions, and uses this corrected position as the current position estimation unit. Return to 23. When the estimated position is corrected based on the estimated position PG obtained by this GPS system, the management judgment unit 21 receives a certain signal (hereinafter referred to as
(referred to as "external position detection signal").

[0025] The position detection accuracy determination method in the management determination section 21 will be explained below. First, reduce your current points to 100.
As a perfect score, the management judgment unit 21 deducts points according to Table 1 depending on whether the current and past vehicle position detection states are "road detection state" or "road non-detection state".

[0026]

[Table 1]

Furthermore, the management judgment unit 21 deducts points according to the current number of registered links according to Table 2.

[0028]

[Table 2]

Furthermore, when the external position detection signal is generated, the management judgment unit 21 considers that the road detection state has continued for 5 km or more. In other words, no points will be deducted depending on the road detection state or road non-detection state. Based on the criteria described above, the management judgment unit 21 can calculate the current points. The overall accuracy of position detection can be determined based on this score. For example, when the road detection state continues for 5 km or more and the number of registered links is one, there is no deduction of points, so it can be assumed that the reliability of detecting the current position of the vehicle is perfect. Similarly, even if the road detection state continues for 5 km or more, and the number of registered links is 2, the number of points will be 90, and the reliability of position detection will decrease somewhat. If the road non-detection state continues for 10 km or more, the number of registered links is 3 or more, and no GPS signal is obtained, the number of points will be 20, and the reliability of position detection will be extremely low.

[0030] When calculating the overall reliability as described above,
The management judgment unit 21 outputs this to the navigation controller 9. The navigation controller 9 is
The manner of display on the display 2 is changed depending on the value of this overall reliability. For example, the size of the vehicle mark to be displayed may be inversely proportional to the overall reliability, or if the reliability is low (for example, 7
(less than 0 points), the vehicle position is not displayed and instead a message indicating that the reliability is low is displayed, or if the reliability is 0, the driver is prompted to input the current position. Furthermore, information may be provided by voice.

[0031] In this way, by determining the current position of the detected vehicle and the reliability of the road it is traveling on, and changing the way the vehicle position is displayed in accordance with the determination results, the vehicle position information is provided to the driver as accurately as possible. I can tell it like it is. In addition, although the vehicle guidance device of the above-mentioned Example was equipped with the GPS receiver 7, the present invention is not limited to this, and any means for obtaining position detection information from the outside may be used. For example, it may be a means for inputting the current position through an operation by the driver, or it may be a roadside beacon receiver.

Furthermore, even if the vehicle guidance system is not equipped with a means for obtaining position detection information from the outside, the present invention can be carried out as long as it is a vehicle guidance system capable of self-contained navigation using at least an azimuth sensor, a distance sensor, and a road map memory. Further, although the management judgment unit 21 outputs the instantaneous overall reliability, it may also take a moving average of the past and output it.

Furthermore, although the above-mentioned embodiment relates to a vehicle guidance device having a vehicle position detection function, the present invention may also be applied to a vehicle guidance device having a route calculation function in addition to this. good. Here, the route calculation function reads road map data in a range including the starting point and destination point from the road map memory according to the driver's settings such as the destination point, and calculates the route from the starting point based on this road map data. Refers to the function of determining the most economical route (optimal route) to a destination (Report on Development and Research of Emergency Vehicle Travel Guidance System, Japan Traffic Management Technology Association, March 1986, Dirck Von Vliet, "Imp.
roved Shortest Path Algor
ithm for Transportation N
etwork”, Transportation R
esearch, Vol. 12, 1978).

[0034] The vehicle guidance device having the above-mentioned route calculation function can calculate the optimum route and display the calculated optimum route together with the vehicle on a road map, so it is utilized as shown in Table 3.

[0035]

[Table 3]

As a premise for calculations and processing as shown in Table 3, when determining whether a vehicle is traveling on the route or deviating from the route, the determination result of the vehicle position accuracy determined by the present invention is can be used. That is, when the vehicle is traveling on the optimal route, if the overall reliability of position detection is high, it is determined that the vehicle is actually traveling on the optimal route, and the route guidance instruction and route recalculation are performed. However, if the overall reliability of position detection is low, route guidance instructions, route recalculation, and arrival guidance are not provided. This is because if the vehicle is traveling on a route other than the optimal route, it may lead to incorrect instructions being issued, which is dangerous.

In addition, when the vehicle position deviates from the optimal route, if the overall reliability of position detection is high, route deviation processing is performed, but if the overall reliability is low, even if the vehicle deviates from the optimal route, Since it cannot be determined that the vehicle is traveling on a road other than the route, processing is not performed when the vehicle deviates from the route. In this way, route guidance instructions can be given or not depending on the overall reliability of position detection, so guidance can be given to the driver according to the reliability of the vehicle position. In turn, driving safety can be achieved.

[0038]

As described above, according to the position detection accuracy determination method and vehicle guidance device of the present invention, by providing the first and second determination criteria, the probability that the vehicle is traveling on the road can be determined. Since the determination can be made comprehensively, the driver can understand the reliability and accuracy of the vehicle detection position. Therefore, driving safety can be provided.

[0039] In particular, by adding a third or fourth judgment criterion, the judgment accuracy can be further improved. Furthermore, if the vehicle has an optimal route calculation means, it is possible to reliably guide the driver to the destination by changing the guidance method along the optimal route.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of a vehicle guidance system.

FIG. 2 is a block diagram of a position detection device that constitutes a part of the vehicle guidance device.

FIG. 3 is a graph showing road matching probability.

[Explanation of symbols]

2 Display, 3 Direction sensor, 4 Distance sensor,
5 road map memory, 8 position detection device, 9 navigation controller

Claims (7)

[Claims] Claim 1: Obtain and integrate the distance traveled by the vehicle and the amount of changes in direction at predetermined timings, obtain the estimated position of the vehicle determined based on this integrated data, and calculate the estimated position from the road map stored in the memory. Select all roads located in a vehicle existence probability region centered on , calculate the similarity between the above estimated position change and each road pattern, and register roads with similarity above a certain value as candidate roads, In a vehicle position detection method that identifies the road that currently has the highest degree of similarity among registered candidate roads and sets the current position of the vehicle on that road, it is possible to identify a road that has a candidate road and has the highest degree of similarity. The first criteria for distinguishing between the state where there are no candidate roads and the state where there are no candidate roads or the road with the highest degree of similarity cannot be identified even if there are candidate roads, and the number of registered candidate roads. A method for determining position detection accuracy, comprising determining the position detection accuracy of a vehicle in accordance with a corresponding second determination criterion. [Claim 2] In addition to the first criterion and the second criterion, a third criterion is adopted, which takes into account the period of time during which the state determined by the first criterion continues steadily. The position detection accuracy determination method according to claim 1. [Claim 3] In addition to the first determination criterion and the second determination criterion, a fourth determination criterion is adopted that takes into account that the vehicle position signal is obtained by means of obtaining vehicle position detection information from outside the vehicle. The position detection accuracy determination method according to claim 1, characterized in that: 4. A road map memory that stores a road map, and an estimated position detection means that obtains and integrates the distance traveled by the vehicle and the amount of change in direction at predetermined timings, and detects the estimated position of the vehicle based on this integrated data. Then, from the road map stored in the road map memory, all roads located in a vehicle existence probability region centered on the estimated position are selected, and the degree of similarity between the change in the estimated position and each road pattern is calculated, and a position specifying means for registering roads with a degree of similarity equal to or higher than a certain value as candidate roads, identifying the road with the highest degree of similarity among the registered candidate roads, and specifying the current position of the vehicle on that road. In the vehicle guidance device, there is a state in which there is a candidate road and the road with the highest degree of similarity can be identified by the position specifying means, and a state in which there is no candidate road or a road with the highest degree of similarity even if there is a candidate road. position detection accuracy determination means for determining the vehicle position detection accuracy according to a first determination criterion for distinguishing between a state in which the vehicle cannot be identified and a second determination criterion according to the number of registered candidate roads; 1. A vehicle guidance device using a position detection accuracy determination method, comprising: information provision means that changes the way information is provided to a driver according to the determination result of the position detection accuracy determination means. 5. The position detection accuracy determining means includes, in addition to the first determination criterion and the second determination criterion, a third determination criterion that considers the time period during which the determination state according to the first determination criterion continues steadily. 5. A vehicle guidance system using the position detection accuracy determination method according to claim 4. 6. In addition to the first determination criterion and the second determination criterion, the position detection accuracy determination means takes into consideration that the vehicle position signal is obtained by means for obtaining vehicle position detection information from outside the vehicle. 5. A vehicle guidance system using the position detection accuracy determination method according to claim 4, wherein the fourth determination criterion is adopted. 7. The vehicle guidance device has means for calculating an optimal route from the departure point of the vehicle to the destination, and the information providing means indicates the identified vehicle position and the calculated optimal route to the driver along with a road map. Vehicle guidance using the position detection accuracy determination method according to claim 4, 5 or 6, characterized in that the method of guidance along the optimal route is changed depending on the determination result of the position detection accuracy determination means. Device.