JPH01277715A - Navigation apparatus for vehicle - Google Patents

Abstract

PURPOSE:To perform accurate matching, by a method wherein the average azimuths of a definite running distance before and after one's own car passes a corner when it is detected that said car begins to turn the corner and the matching of the present position of one's own car is taken on the basis of the difference between the average azimuths. CONSTITUTION:It is detected that one's own car begins to turn a corner, on the basis of the data from an one's own car position estimation means 3 by a corner detection means 6. Whereupon, the first and second average azimuth calculation means 7, 8 calculate the average azimuths of a definite running distance before and after one's own car passes the corner. The passage of the corner is detected on the basis of the difference between the average azimuths calculated by the first and second average azimuth calculation means 7, 8 by a corner specifying means 9 to take the matching of the present position of one's own car with the road on the map stored in a map data memory means 4. By this method, it can be accurately detected that one's own car passes the corner or a crossing.

Description

DETAILED DESCRIPTION OF THE INVENTION (Industrial Application Field) The present invention is a vehicle navigation device, in particular, a method for storing in advance the current position of the own vehicle estimated by the own vehicle position estimating means using a direction detecting means and a distance detecting means. The present invention relates to a vehicle navigation device that guides the driving of a vehicle by matching roads on a map.

(Prior art) In recent years, the current position of the own vehicle relative to a reference point is estimated using a direction sensor and a distance sensor that utilize geomagnetism, and the estimated position and a map around the estimated position are displayed on the display screen. A vehicle navigation device has been developed that uses so-called dead reckoning navigation to display and guide the travel of the own vehicle. By the way, in the dead reckoning navigation described above, the current position of the own vehicle relative to the reference position is estimated based on the direction obtained by the azimuth sensor and the travel distance obtained by the distance sensor. As the vehicle moves away from the vehicle's location, measurement errors in travel distance and direction accumulate, resulting in a decline in the accuracy of recognition of the vehicle's current location. In order to deal with such a situation, for example, the navigation device described in Japanese Patent Application Laid-open No. 61-209316 uses dead reckoning when the vehicle passes a position with a distinctive road shape, such as an intersection. A method called map matching is adopted in which the current position of the own vehicle estimated by the above is corrected to the above position.

(Problem to be Solved by the Invention) By the way, as in the above conventional example, when the own vehicle approaches a position with a characteristic road shape, such as an intersection, the estimated current position of the own vehicle is corrected to that position. In some cases, the current position of the vehicle is estimated with measurement errors as described above, and therefore, for example, the vehicle is driving in an urban area where there are many intersections within a relatively short distance. At times, it was difficult to identify the intersection that the vehicle actually passed through. If the current position of the vehicle is matched to an intersection different from the intersection that the vehicle actually passed through, it is impossible to perform subsequent matching accurately.

SUMMARY OF THE INVENTION The present invention provides a navigation system that guides the travel of a vehicle by matching the current position of the vehicle obtained by vehicle position estimating means with roads on a pre-stored map. An object of the present invention is to provide a vehicle navigation device that can accurately match the position of a vehicle to roads on a map by reliably detecting corners.

(Means for Solving the Problems) In order to solve the above problems, a navigation device according to the present invention is characterized by using the following means.

That is, as shown in FIG. 1, the own vehicle position estimating means estimates the current position 1 of the own vehicle with respect to the reference position based on the azimuth data obtained by the azimuth detecting means 1 and the distance data obtained by the distance detecting means 2. 3, a map information storage means 4 in which map information is stored in advance, and a comparison between the current position of the own vehicle estimated by the own vehicle position estimation means 3 and the map information stored in the map information storage means 4. In a vehicle navigation device having a map matching means 5 for matching the current position of the own vehicle to a road on a map, the own vehicle starts turning a corner based on data obtained by the above-mentioned own vehicle position estimation means 3. corner detection means 6 for detecting;
A first average azimuth calculation means 7 that calculates an average azimuth over a certain distance traveled before the corner detected by the corner detection means 6; and a second average azimuth calculation means 8 that calculates an average azimuth over a certain distance traveled after passing the corner. , on the map stored in the map information storage means 4 by detecting the passage of a corner based on the difference between the average orientations calculated by the first average orientation calculation means 7 and the second average orientation calculation means 8. Corner identification means 9 for matching the current position of the own vehicle to the corner
It is characterized by having the following.

(Function) According to the above configuration, since the corner detection means 6 is provided for detecting that the own vehicle has started to turn a corner based on the data obtained by the own vehicle position estimating means 3, the own vehicle position estimating means 3 It will be reliably detected that the car has started to turn a corner.

A first average azimuth calculation means 7 for calculating an average azimuth for a certain distance traveled before the corner, and a second average azimuth calculation means 8 for calculating an average azimuth for a certain distance traveled after passing the corner are provided. As a result, the average running direction of the vehicle before the corner and after passing the corner is calculated, and the difference between the respective average headings calculated by the first average heading calculating means 7 and the second average heading calculating means 8 is calculated. Corner specifying means 9 is provided which detects the passage of a corner based on the information and matches the current position of the vehicle with the road on the map stored in the map information storage means 4. It is possible to accurately detect when a vehicle has passed a corner (even intersections can be considered corners). This allows the subsequent matching by the map matching means to be performed more accurately.

(Example) Examples of the present invention will be described below.

FIG. 2 is a control system diagram of the vehicle navigation device according to the present embodiment. A distance sensor 12 as a distance detecting means that outputs a pulse signal every time the vehicle travels to detect the distance traveled by the vehicle, and map information such as a map showing information necessary for travel guidance of the vehicle such as roads and buildings. a map information storage bag W13 in which is stored a map information storage bag W13; a display device 14 composed of a CRT or the like to display a road map as map information stored in the map information storage device 13 on a display screen;
It has a control circuit 15.

The control circuit 15 includes an arithmetic circuit 16, a memory circuit 17 connected thereto, and an input/output interface 1.
8, and the geomagnetic sensor 11, distance sensor 12, map information storage device 13, and display 14 are connected to the arithmetic circuit 16 via the input/output interface 18.

Therefore, a signal indicating the traveling direction of the own vehicle from the geomagnetic sensor 11 and a signal indicating the traveling distance of the own vehicle from the distance sensor 12 are input to the arithmetic circuit 16 via the input/output interface 18. Accordingly, the arithmetic circuit 1
In step 6, the estimated position of the own vehicle is calculated, and the calculation result is temporarily stored in the storage circuit 17. Also,
The control circuit 16 extracts a map around the estimated position of the own vehicle detected as described above from the map information storage device 13, matches the estimated position of the own vehicle on the map, and displays it on the display. It is configured to be displayed on 14 screens. The map information storage devices 1 and 3 are composed of, for example, a CD-ROM and a CD drive (player) for reproducing the CD-ROM.

Note that in this embodiment, the map information storage means 13
As the map information stored in , a large number of nodes are set on the pre-stored road, and data indicating the connection relationship of each of these nodes is input. It is assumed that they are provided at appropriate predetermined intervals even on a straight road.

Next, the specific operation of this embodiment will be explained according to the flowcharts shown in FIG. 3 and subsequent figures.

The flowchart in FIG. 3 shows the main program of the map matching operation by the control circuit 15.
Flag F in step S1.

, Fl, and F2 are each set to 1. Then, in step S2, a pulse signal indicating the traveling distance is input from the distance sensor 12, and based on this pulse signal, step S2 is inputted.
3 to determine whether the own vehicle has traveled a predetermined distance or more,
If it is determined that the vehicle has traveled a predetermined distance or more, a signal indicating the running direction of the vehicle from the geomagnetic sensor 11 is inputted in step s4, and an estimated position of the vehicle is calculated based on smell in step S. Ru. Based on this calculation result, it is determined in step S6 whether the flag Fo=1 or not. If the determination is YES, node search processing A is executed in step S7. In this node search process A, a process is performed to search for a specific node that the own vehicle is heading from among a large number of nodes set on the memorized road.Next, in step S8, a corner Detection processing will be executed. In this corner detection process, a process is performed to determine whether or not the own vehicle has turned a corner.

Further, if it is determined as Fo-0 in step S6, node search processing B is executed in step S9. In this node search processing B, when the own vehicle deviates from the memorized road, a process is performed in which dead reckoning is newly started from the point where the own vehicle deviated and the self vehicle returns to the memorized road again. Then, in step SIO, the estimated position of the own vehicle is determined based on the results of the node search process A and the corner detection process executed in steps S7 and S8, or the result of the node search process B executed in step S9. A new matching process is executed on the memory road, and step Sll is executed.
Then, the result of the matching process executed in step S10 is output to the display 14, and the current position of the own vehicle and a map of its surroundings are displayed on the screen of the display 14.

Next, the node search process A executed in step S7 will be explained based on the flowchart shown in FIG. 4. First, in step S12, it is determined whether the flag F1=1, and if YES, Assuming that the node has just arrived at the node, in step 313, the number N of nodes that can be moved from the latest arriving node is detected. In this case, for example, as shown in FIG.

, five movable nodes N1, N2, N3, N4 are selected based on the data indicating the location of a large number of nodes stored as map information in the map information storage device 13 and the connection relationships between these nodes. .
is set to 0 and then returned to the main program, and when the determination is No, i is set to 0 in step S17.
=0. Then, in step 818, the angular difference θi between the candidate node Ni and the running direction of the host vehicle detected by the geomagnetic sensor 11 is calculated. For example, as shown in FIG. 5, node N2 is
And the angular differences θ1 and θ2 with the node N3 are calculated.

(The same thing is done for the other nodes N1, N4, and N.) Then, in step S19, the angular difference θi between the running direction of the own vehicle and the candidate node Ni is multiplied by the evaluation function W, and , the distance traveled by the own vehicle from the latest arrival node to calculate the evaluation reference value θni. Note that the evaluation function W is preset according to the value of the angular difference θi, and the larger the angular difference θi, the larger the value of the evaluation function W is set. This is to exclude candidate nodes Ni that have a large angle difference θi from the running direction of the own vehicle from nodes that the own vehicle can reach.9Next, in step S20, the above θni is set to a predetermined threshold. It is determined whether the value is greater than or equal to the value, and if YES,
In step S21, the candidate node Ni is excluded from the reachable candidate nodes, and in step S2□, N=N-1 is set. Then, i=i+1 is set in step S23, and steps S18 to S23 are repeatedly executed for all candidate nodes Ni, until it is determined that i=N in step S24. Subsequently, in step S25, the distance from the estimated position of the host vehicle to each remaining candidate node Ni that is determined to be reachable is calculated, and in step S26, it is determined whether there is a candidate node Ni with a remaining distance of O. If the determination is YES, the node Ni is newly added to the own vehicle's sealed node list in step S27, and then the flag F1 is reset to 1 in step 52111.

Here, flag F is set in step S6 of the main program shown in FIG. The node search process B executed when it is determined that is not 1 will be explained according to the flowchart shown in FIG. 6. First, in step S29, as shown in FIG. 7, for example, a point p off the stored road is found. A new dead reckoning is started from , and in step SSO, point P is off the memorized road. Travel distance Lx of own vehicle from
(corresponding to the distance from Po to Pl or from Po to P2 in FIG. 7) is calculated, and in step S3□, a predetermined reference value β preset for the moving distance Lx is calculated.
By multiplying by , the radius r (corresponding to rl and r2 in FIG. 7) that is the reference of the node search range is set, and then in step S32, the area with the radius r based on the estimated position of the own vehicle is set. Node Ni located within
Explore. Then, in step S33, it is determined whether there is a node Nf within the search area, and if the determination is YES, in step S34, the distance Lx between the node Ni within the search area and the estimated position of the own vehicle is determined. '
is calculated, and if the above-mentioned Lx' is equal to or less than a predetermined value H in step S35, the node Ni is added to the arrival node list of the host vehicle in step S36, and the flag F1 is set to 1 in step S37. After that, return.

Next, the corner detection process executed in step S8 of the main flowchart will be explained according to the flowchart shown in FIG. First, in step 838, as shown in FIG. 9, the average direction D1 for the latest constant mileage l1 traveled by the own vehicle is calculated, and in step S39, the predetermined direction D1 after traveling the constant mileage I11 is calculated. It is determined whether there is a difference between the direction of the travel distance ΔJ" (see FIG. 9) and Dl, and if the determination is No, that is, if it is determined that the constant travel route Mar is a straight line, step S40 The process returns after setting the flag F2 to 1.If the determination is YES, that is, if it is determined that the own vehicle has started to turn a corner, in step S41, as shown in FIG. 9(I), the process returns. Determine whether you have traveled the distance 11 that is considered necessary to complete the turn, and confirm YE.
If S, it is determined in step S4□ whether the vehicle has traveled a certain distance (2) after turning the corner.

Then, in step S43, it is determined whether the travel distance 1□ is a straight line. In this case, for example, if the constant travel path Mj2 is divided into unit vectors for each pulse output from the distance sensor 12 and the difference between each unit vector is within a predetermined value, it is determined that )2 is a straight line. It can be determined whether or not. Then, when the determination is YES, in step S44, a certain mileage! ;I2
In step S45, the average azimuth D2 for a certain traveling distance p1 before turning the corner is calculated.

The difference ΔD between and the average direction D2 of the constant traveling distance J2 after turning the corner is calculated, and in step S46,
When it is determined that this ΔD is greater than or equal to the predetermined value α, it is determined in step S47 that the own vehicle has turned a corner. After setting the flag F2 to 0 in step 848, the process returns. Note that when it is determined in step S46 that ΔD is less than or equal to the predetermined value α, the flag F2 is set to 1 in step S49, and then the process returns.

In addition, when it is determined in step 343 that the fixed travel distance j2 is not a straight line, j is set to 0 in step SSO, and then in step S51, as shown in FIG.
As shown in n), it is confirmed that the vehicle has traveled a certain distance M42 and then further traveled A1'', and in step S52, j is set to j+1.Then, in step S53
In this case, the constant traveling path M1 after traveling the above 71″
2 (corresponding to Shimezu J2' in FIG. 9(n)) is determined to be a straight line, and if YES, step S44
, the average orientation D2 of ρ2 is calculated, and
If the determination is NO, in step S54
and a predetermined reference value j. By comparing the above, Stella 1551 to Step S54 are repeatedly executed until j = jo, and if the determination is YES, the flag F2 is set to 1 in Step S55.
After that, it will be returned.

Next, the matching process executed in step StO in the main program will be explained based on the flowchart shown in FIG. First, in step S56, it is determined whether flag F=1, and if NO, step S57
In this step, the estimated position of the own vehicle calculated in step S5 of the main program is set as the current position, and when the determination is YES, the current position is corrected to the node Ni detected by the node search process A in step SSa. Then, in step S59, flag F2 is set to 1.
If the result is YES, the process returns; if the result is No, that is, if the corner detection process detects that the own vehicle has turned a corner, then in step S68 the result is within a predetermined range, i.e. , it is determined whether there is a node Ni provided near the corner, and if there is a node Ni within a predetermined range, the current position is corrected to that node Nt in step S6+.

(Effects of the Invention) As described above, according to the vehicle navigation device according to the present invention, when the corner detection means detects that the own vehicle has started to turn a corner, the first and second average azimuth calculation means The average azimuth for a certain distance traveled before the corner and the average azimuth for a certain distance traveled after passing the corner are calculated, and passage of the corner is detected based on the difference between these average azimuths and stored in the map information storage means. Since corner identification means is provided to match the current position of the vehicle to the road on the map, it is possible to detect corners or intersections that the vehicle has passed with high accuracy, thereby improving map matching. It can be done more accurately.

[Brief explanation of the drawing]

Fig. 1 is a block diagram showing the overall configuration of a vehicle navigation device according to the present invention, Figs. 2 to 10 show embodiments of the invention, and Fig. 2 is a control system diagram of the navigation device according to the present embodiment. , FIG. 3 is a flowchart showing the main program of the control operation by the control circuit, FIG. 4 is a flowchart showing the node search process A while driving on a memory road, and FIG. 5 is a schematic explanation of the control operation of the node search process A. 6 is a flowchart showing the node search process B, FIG. 7 is a schematic explanatory diagram of the control operation of the node search process B, FIG. 8 is a flowchart showing the corner detection process, and FIG. 9 is the corner detection process. A schematic explanatory diagram of the control operation of
FIG. 10 is a flowchart showing the matching process. 11... Orientation detection means (geomagnetic sensor), 12...
Distance detection means (distance sensor), 13...Map information storage means (map information storage device), 15... Own vehicle position detection means, corner detection means, first and second average direction detection means,
Corner identification means (M control circuit). Figure 6 Figure 7

Claims (1)

[Claims] (1) Vehicle position estimation means for estimating the current position of the vehicle relative to a reference position based on the direction data obtained by the direction detection means and the distance data obtained by the distance detection means, and a map in which map information is stored in advance. an information storage means, and a map for matching the current position of the own vehicle with a road on the map by comparing the current position of the own vehicle estimated by the above-mentioned own vehicle position estimating means with the map information stored in the above-mentioned map information storage means; a corner detection means for detecting that the own vehicle has begun to turn a corner based on data obtained by the own vehicle position estimation means; and a corner detection means for detecting when the own vehicle has started turning a corner, a first average azimuth calculation means for calculating an average azimuth for a certain distance traveled before the corner; a second average azimuth calculation means for calculating an average azimuth for a certain distance traveled after passing the corner; and the first average azimuth calculation means. Corner identification is performed by detecting the passing of a corner based on the difference between the average orientations calculated by the second average orientation calculation means and matching the current position of the own vehicle with the corner on the map stored in the map information storage means. A navigation device for a vehicle, characterized in that it is equipped with means.