KR100216535B1 - Positioning Method of Vehicles Using Position Matching Value for Car Navigation System - Google Patents

Abstract

The present invention relates to a method for measuring a position of a driving vehicle, wherein the current position of a vehicle is obtained by using a dead reckoning technique using data collected from a direction and distance sensor attached to the vehicle, and then the position is set as a first moving position. A second step of correcting the first moving position obtained in the first step and the first moving position obtained in the first step to the road on the numerical road by a map matching method using a position matching degree, and then setting the position as the second moving position; A third step of analyzing a signal received from GPS to obtain a current position of the vehicle, and then setting the position as a third moving position; and using the third matched position obtained in the third step as a map matching method. A fourth step of correcting the road on the numerical road map and setting the position as the fourth moving position, and calculating a short-distance driving pattern of the first to fourth moving positions, respectively. Is a comparison between the fifth step and the similarity indicative of the degree of coincidence between each of the short-distance driving patterns of the first to fourth moving positions obtained in the fifth step and the road pattern of the numerical road map. In the sixth step of selecting the final current position, the positional error is not accumulated, and the position of the vehicle can be continuously measured even under a roadside, in a tunnel, in a building forest, and the like where a GPS signal is not received.

Description

Positioning Method of Vehicles Using Position Matching Value for Car Navigation System

The present invention relates to a method for measuring the position of a driving vehicle for vehicle navigation using a position matching degree corrected on a road on a numerical road map and a position measurement of a driving vehicle. The present invention relates to a method for measuring a position of a traveling vehicle to conveniently provide a driver with a safest and fastest optimal driving route to a destination in consideration of a situation.

The vehicle navigation system necessarily needs to determine the position of the driving vehicle, and the position information of the vehicle is used for displaying the current position of the vehicle, guiding the route, and recognizing the surrounding environment, and thus must be continuously provided in real time within the allowable error range. do. In addition, the provision of incorrect or discontinuous location information can seriously affect safe driving. Therefore, positioning of the vehicle occupies a very important position in the vehicle navigation system.

Such a vehicle positioning method includes a self-standing position measuring method for locating position using data collected from a sensor mounted in a vehicle, a radio navigation position measuring method for finding a position by receiving an external signal, and self-standing. There is a mixed position measuring method using an equation and a radio navigation formula together.

The vehicle location measuring methods are described in detail as follows.

Freestanding position measuring method

The self-supporting position measuring method mainly uses dead reckoning and map matching techniques, which were theoretically introduced with the advent of the vehicle navigation system in the mid-1980s.

The dead reconning method is a method of obtaining a relative position from a starting point by using travel distance and traveling direction data collected from a sensor mounted inside a vehicle without the help of an external signal. An illustration of the location tracking method is shown.

Referring to FIG. 1, the dead reconning technique refers to an initial position of the vehicle as ( x ₀ , y ₀ ), and measures the direction (θ _i ) and the moving distance ( d _i ) in which the vehicle moves from the position. How to find out where you have moved. At this time, the moving direction θ _i and the moving distance d _i of the vehicle are obtained from a sensor inside the vehicle, and a calculation equation for finding a current position is shown in Equation 1 below.

However, the direction sensor such as a gyro, electronic compass, etc. used at this time may have a different error depending on the rotation direction, and when moving, the direction of travel may be deflected or changed in one direction. The measured traveling direction may include an error, and the measurement of the moving distance by a wheel sensor or the like also includes an error due to external factors such as air pressure of a wheel and a road surface state of a road.

Therefore, the dead reconning method using the sensor data including the error may accumulate these errors to calculate the position of the wrong vehicle. In particular, the digital map data produced for displaying the current position is mainly used for the road. Since the centerline is input, the vehicle does not travel along the centerline at intersections or general roads, and may contain an input error at the time of input, and thus the position of the vehicle may be displayed at the wrong position.

The map matching method is a method of correcting an error included in a location and a numerical map measured by a dead reckoning method. The map matching method is based on the assumption that a vehicle is always on a road. How to calibrate up the road. Therefore, such a map matching method requires a digital map not only for accurate vehicle position representation, optimal route guidance, destination search, and feature information providing but also for vehicle position correction.

This self-supporting position measuring method guarantees the continuity of the position data. In particular, the dead reconning technique using a sensor mounted in a vehicle is relatively accurate when traveling a relatively short distance, while initial positioning by a user input, a sensor Accumulation of errors due to inaccuracies, etc. is disadvantageous, and once the wrong position is determined, positioning of the vehicle is very difficult thereafter.

Radionavigation Position Measurement Method

A representative system of the radio navigation positioning method is a GPS (Global Positioning System), which is a high-precision satellite that receives radio signals from satellite groups orbiting the earth and automatically measures their 3D position and time in real time. It is a navigation system.

GPS is a system capable of 24 hours absolute positioning anywhere on earth regardless of weather. It is officially called NAVSTAR GPS (Navigation Satellite Timing And Ranging Global Positioning System), and the positioning method used in GPS is based on radio waves from different satellites. The delay time is measured and the current position is obtained from the distance from each satellite orbit.

Since the error is not accumulated in the vehicle position measurement using this system, if the radio wave signal is normally received, the error may be relatively constant regardless of the driving time. In addition, there is an advantage that a distance, a direction sensor, and the like are not required, are easy to use, and relatively inexpensive.

On the other hand, because of the phenomenon that the electric wave splits into several branches due to roadsides, buildings, roads, etc., data may be inaccurate, and there may be discontinuity of data in which signals are not received in tunnels or underground spaces. In addition, there is a disadvantage that a position error from 25m to 100m may occur depending on whether S / A (Selective Availability) is implemented. Therefore, a map matching technique is required to accurately display the position measured using the radio navigation method on the road on the numerical road map.

Therefore, in the present invention, when only a signal is received, a GPS error including a limited error does not occur according to an absolute position measuring method and only a limited error, and a dead reckoning technique indicating an accurate result in a short-term movement is simultaneously received from a GPS receiver. GPS location data and sensor data collected from the direction sensor and distance sensor installed in the car are synthesized and compared with the roads of the numerical map prepared in advance to determine the location of the reliable driving vehicle for determining the location of the driving vehicle for the vehicle navigation system. An object of the present invention is to provide a method for measuring a position of a traveling vehicle for vehicle navigation using position matching. That is, the method for measuring the position of the driving vehicle provided by the present invention is a process of estimating the position of the vehicle using a dead recording technique and a GPS receiver, and obtaining a corrected position for the estimated position to select the position which is determined to be the most accurate. to be.

According to an aspect of the present invention, there is provided a method for measuring a position of a driving vehicle by using a data collected from a direction and distance sensor attached to a vehicle to obtain a current position of a vehicle by using a dead reckoning technique, and then setting the position as a first moving position. A first step, a second step of correcting the first moving position obtained in the first step on the road on the numerical road map by a map matching method using a position matching degree, and setting the position as the second moving position; Analyzing the signal received from the third step of determining the current position of the vehicle and setting the position as the third moving position; and using the map matching method using the position matching degree for the third moving position obtained in the third step. A fourth step of obtaining a fourth moving position after correcting the road on the numerical road map and a short-distance driving pattern of the first to fourth moving positions, respectively; Step 5 and the similarity indicative of the degree of coincidence between each of the short-distance driving patterns of the first to fourth moving positions obtained in the fifth step and the road pattern of the numerical road map are respectively compared to the final current. It consists of a sixth step of selecting by position.

In this case, the degree of similarity is a value obtained by calculating a position matching degree for all the trajectories of each of the first to fourth moving positions, and then adding up the position matching degree of the respective trajectories, and the position matching degree of the estimated vehicle It is calculated | required by the sum of the distance matching degree which shows the distance of the distance from a road with a position and a numerical road map, and the separation angle matching degree which shows the degree of separation of the angle with which the vehicle traveled, and the road angle.

1 is an explanatory diagram of a location tracking method using a dead reckoning technique;

2 is a conceptual diagram for the type and generation process of candidate positions according to an embodiment of the present invention;

3 is an explanatory diagram of the separation distance and the separation angle,

4 is a graph of the distance matching degree,

5 is a graph of the separation angle matching,

6 is a graph of candidate positions and trajectories of each type of the present invention;

7 is a flowchart illustrating a method for measuring a position of a driving vehicle for vehicle navigation using a position matching diagram according to an embodiment of the present invention;

8 is a block diagram of an apparatus for measuring a position of a traveling vehicle according to an exemplary embodiment of the present invention.

<Explanation of symbols for main parts of drawing>

10: direction sensor 20: distance sensor

30: GPS receiver 40: peripheral input unit

50: database 60: central processing unit

70: LCD screen 80: remote control

Hereinafter, with reference to the accompanying drawings will be described in detail the present invention.

2 is a conceptual diagram of the type and generation process of candidate positions according to an embodiment of the present invention, FIG. 3 is an explanatory diagram of a separation distance and a separation angle, FIG. 4 is a graph of the separation distance matching diagram, and FIG. 5 6 is a graph of a distance matching angle, FIG. 6 is a graph of candidate positions and trajectories of each type of the present invention, and FIG. 7 is a flowchart of a method of measuring a position of a driving vehicle for vehicle navigation according to an embodiment of the present invention. 8 is a configuration diagram of an apparatus for measuring a position of a driving vehicle according to an exemplary embodiment of the present invention.

2 is a conceptual diagram of a type and a generation process of candidate locations according to an embodiment of the present invention, in which the type and generation process of candidate locations generated to determine a current location of a driving vehicle are shown. At this time, the candidate position is a position of a vehicle that can be generated according to the synthesis result of the collected data, and a plurality of candidate positions can be generated depending on the type of data used and whether the map matching method using the matching degree is applied. Errors may be included according to each generation method. According to the present invention, the candidate position with the least error among these candidate positions is determined to determine the current position of the vehicle that travels.

The types of candidate positions as described above include a first moving position 15, a second moving position 18, a third moving position 22, and a fourth moving position 24. The vehicle's final position (13), GPS position (21), numerical road map (17), and traveling angle (11) of the vehicle collected from the sensor and travel distance (12) from the previous position to the current position Used as data. At this time, the potential value of the determined vehicle represents the potential value determined by the method proposed by the present invention as the position where the vehicle was located at the moment before.

GPS location refers to the current location of the vehicle indicated by the signal received from the GPS system. In this case, the GPS signal used should be a certain number of satellites tracked more than a certain number, and must be a reliable signal with a value of Geometry Dilution Of Precision (GDOP) within a certain number. In order to synthesize each data, map matching technique using dead reckoning and matching is used. There are many types of candidate positions generated by synthesizing various data by this technique, depending on the result of the combination. However, in the present invention, as shown in Fig. 2, detailed descriptions are as follows.

The first moving position 15 is a position obtained by the dead reckoning 14 using the determined total position data 13, the traveling angle 11 collected from the sensor, and the moving distance 12. to be. At this time, since the first moving position 15 is obtained from the definite position 13 by the dead reckoning technique 14, the road of the numerical map due to the inaccuracy of the sensor data, the error of the numerical map, the lane change of the vehicle, etc. May deviate from.

Therefore, it is the second moving position 18 that corrects this position error, which is the road map 17 of the numerical road map in which the first moving position 15 is input in advance by the map matching technique 16. It is the best matching position among the locations on the road of the numerical map by comparing the separation distance and the progress angle of the vehicle.

In addition, the third moving position 22 is the current position of the vehicle obtained by analyzing the GPS position 22 received from the GPS system, and the third moving position 22 is the performance of the receiver and the selective validity (S / A). The error may be included depending on whether or not: Selective Availability is implemented, multipath phenomenon, and the like may not match the road of the numerical road map. Therefore, in order to correct the position error of the GPS and to match the road on the numerical road map, the distance between the road and the GPS position entered in the numerical road map is close, and the progress angle input from the sensor and the numerical road The position on the road of the numerical map that best matches the slope of the road is set as the fourth moving position 24.

At this time, the map matching method used to obtain the second moving position 18 and the fourth moving position 24 uses a position matching degree.

At this time, the position matching degree is calculated by using the separation distance and the separation angle indicating the distance deviated from the road of the numerical map. Referring to FIG. 3, the separation distance Pd is input closest to the estimated position of the vehicle. It is the straight line distance to the road 25 on the numerical road map, which is the length of the line segment passing perpendicular to the road 25 through the candidate position 26 of the vehicle.

Therefore, the distance matching degree Zd with respect to the position of the vehicle spaced apart by the distance Pd in a straight line from the registration position 27 on the road 25 input to the numerical road map is as follows.

Iii. In the case of separation distance Pd ≤ search distance Hd

Distance Match (Zd) = 1-Distance (Pd) / Search Distance (Hd)

Ii. In the case of separation distance (Pd) search distance (Hd)

Distance match (Zd) = 0

At this time, the search distance Hd is an allowable range of an error in which the position of the vehicle may deviate from the road, and is a range for searching the road position of the numerical road map at the position where the vehicle is presently determined to exist. If the separation distance Pd is larger than the search distance Hd, the matching degree is '0', and the graph of the distance matching degree is shown in FIG. 4.

Referring to FIG. 4, the distance matching degree Zd is inversely proportional to the separation distance Pd.

The position matching degree of the vehicle with respect to the numerical road map should consider not only the distance matching degree but also the matching degree with respect to the separation angle, and the separation direction Pa of the traveling direction of the vehicle and the straight road 25 on the numerical map is As shown in FIG. 3, the difference is between the traveling angle θ ₁ and the road angle θ ₂ . The matching degree with respect to the separation angle Pa with this straight road 25 is as follows.

Iii. When separation angle (Pa) ≤ allowable angle (Ha)

Separation Angle Matching Degree (Za) = 1-Separation Angle (Pa) / Permissible Angle (Ha)

Ii. In the case of a separation angle (Pa) and an allowable angle (Ha)

Separation Angle Match (Za) = 0

In this case, the degree of distance matching is obtained from a straight road having a distance matching degree of '0' or more, and the allowable angle may be related to the traveling angle θ ₁ , which is the traveling direction of the vehicle, and the road angle θ ₂ . The maximum separation angle judged to be. A graph of the matching degree of the separation angle is shown in FIG. 5. Referring to FIG. 5, the spacing angle matching degree Za is inversely proportional to the spacing angle Pa.

Accordingly, as shown in FIG. 3, the position matching degree of one point having a straight line and directionality is obtained by the sum of the separation distance matching degree Zd and the separation angle matching degree Za.

The map matching is performed by correcting the position of the tracked vehicle on the road input in the numerical map using the obtained position matching degree.

In this case, the numerical map used represents the actual road structure, and includes nodes and arcs indicating the geometry of the road. The nodes represent the starting point, the end point, and the intersection of each road. Is a road connecting each node and the node includes shape coordinates indicating the shape of the road. In addition, all roads input to the numerical road map may be configured by a combination of straight lines connecting the respective shape coordinates.

Therefore, in order to perform map matching, first, each position matching degree is obtained for all the straight roads composed of the geometric coordinates of the numerical map, and the matching position on the straight line having the highest position matching degree is the corrected position by map matching. Decide

The second moving position 18 and the fourth moving position 24 of FIG. 2 may be configured such that the first moving position 15 and the third moving position 22 are matched with each other by the above-described position matching. This is the location of each road on the numerical road map.

The trajectories for each of the moving positions are obtained using the moving distance and the traveling direction collected from the distance sensor and the direction sensor, and the trajectories of the moving positions are shown in FIG. 6. In this case, '□' represents the first movement position, '■' represents the second movement position, '○' represents the third movement position, and ● represents the fourth movement position, and each movement position proceeds in the direction of the arrow shown in the drawing. do. That is, if the current time to measure the position of the vehicle is t, each movement position has a trajectory at t-1, t-2, and t-3 times.

The first moving position is a candidate position calculated by the dead drone corning method, which calculates the candidate position by the moving distance and the traveling direction data collected from the in-vehicle sensor, so that an error accumulates as time passes, but is generally short. The advantage is that when you move very accurate. That is, since the first moving position calculates the candidate position from the determined previous position by the dead drone corning method, all the positions and the candidate position are the driving trajectories of the first moving position.

The second moving position is the result of matching the first moving position with the road on the numerical road map and calculating it. Therefore, the position trajectory of the second movement position is dead-reconned in the reverse direction of the movement distance and the travel direction at the candidate position to obtain a past driving trajectory.

The driving trajectory for the third and fourth moving positions is also the same as the method for obtaining the second moving position trajectory. Find the position trajectory.

In FIG. 6, when the current time is t, the first moving position is a position obtained by a dead reckoning technique for the travel distance and the moving direction for each time zone from the position determined at t-3 time, and the third moving position is obtained from the GPS. The position of the vehicle calculated by analyzing the received signal, and the second and fourth movement positions are positions in which the first and third movement positions are corrected to the position on the road of the numerical road map having the largest match. to be. In this case, the matching degree is a value obtained using the set search distance and the allowable angle.

When the first to fourth moving positions, that is, four kinds of candidate positions are derived through the above process, the candidate position including the least error among them is determined as the current position of the vehicle. That is, the driving trajectory for each type of candidate position is compared with the road pattern of the numerical road map, and the candidate position of the most similar type is set as the current position.

In this case, the similarity representing the degree to which the candidate position trajectory coincides with the road structure is a sum obtained by calculating the position matching degree for all the trajectories of the candidate position, and since the type having the highest similarity is the best match with the road structure, This type of candidate position is determined as the current position.

7 is a flowchart illustrating a method for measuring a position of a driving vehicle according to an exemplary embodiment of the present invention.

Referring to FIG. 7, in the method of measuring a position of a driving vehicle of the present invention, first, a current position of a vehicle is obtained by a dead reckoning technique using data collected from a direction and distance sensor attached to a vehicle, and then the first position is determined. Set the moving position (S100), correct the first moving position on the road on the numerical road map by a map matching method, and set the position as the second moving position (S200), and analyze the signal received from the GPS. After the current position of the vehicle is obtained, the position is set as the third movement position (S300), and the third movement position is corrected on the road on the numerical road map by a map matching method, and then the position is changed to the fourth movement position. Set (S400).

When the first to fourth moving positions are obtained as described above, after obtaining the short-range driving patterns of the first to fourth moving positions, respectively (S500), the degree to which the respective short-range driving patterns coincide with the road pattern of the numerical road map By comparing the similarities each representing a to select the moving position having the highest similarity as the final current position (S600).

In this case, the similarity is a value obtained by calculating the position matching degree for all the trajectories of each of the first to fourth moving positions, and the position matching degree is spaced apart from the position on which the vehicle traveled and the position on the road on the numerical road map. It is a sum of the distance matching degree which shows the degree which was completed, and the separation angle matching degree which shows the matching degree of the angle which the vehicle traveled, and the road angle.

In addition, in the case of the second moving position, it is considered that the accuracy is higher than that of the other moving positions, and thus the weight is added to the sum of the position matching degrees when the similarity of the second moving position is obtained.

In addition, a case in which the vehicle travels off-road, that is, an alley or a new road, which is not entered in the numerical map, should be considered. For this purpose, the third moving position may be moved at the maximum similarity that varies according to the number of driving trajectories. If the value of subtraction of the similarity is larger than the similarity of other moving positions, the driving at that time is regarded as off-road driving, which is a road other than the road on the numerical road map, and the third moving position is selected as the current position.

That is, when the similarity of the third moving position is '0' and the number of trajectories is four and the maximum similarity is '8', the value obtained by subtracting the similarity of the third moving position from the maximum similarity becomes '8'. Therefore, this value is larger than the similarity value of other moving positions, and this case is regarded as off-road driving and sets the third moving position to the current position.

In FIG. 6, since the second moving position has the largest pattern matching degree, the second moving position is set to the current position of the vehicle that travels.

FIG. 8 is a configuration diagram of an apparatus for measuring a position of a driving vehicle according to an exemplary embodiment of the present invention. Referring to this, the apparatus for measuring a position of a driving vehicle provided in the present invention is mounted inside a vehicle to move and move a vehicle. A distance sensor 200 and a direction sensor 100 for detecting a direction, a GPS receiver 300 for receiving a radio wave signal from a satellite group and automatically measuring its own 3D position and time in real time, and the distance And the peripheral device input unit 400 receiving the moving distance and the direction and the current location information of the vehicle from the direction sensors 200 and 100 and the GPS receiver 300, and information on the road and the surrounding road on which the vehicle is traveling. Database 500 storing the map as a numerical value converted into a numerical value, the position measurement values of a plurality of vehicles input from the peripheral device input unit 400 and the numerical value of the database 500 Compare the road pattern of the road map to select the location of the vehicle with less error, and display to the driver the location of the vehicle selected from the central processing unit 600 and the central processing unit 600 controlled by the remote controller 800 It consists of a liquid crystal display device 700.

As described above, the present invention uses a dead-drening technique, such as GPS, so that the error is not accumulated, and the position of the vehicle can be continuously measured even under the trees, tunnels, and forests where GPS signals are not received. There is an advantage. In particular, the initial position can be automatically set using GPS, and the road can be used in alleys or off-roads where the road has not yet been entered in the numerical map.

Claims (7)

A first step of obtaining the current position of the vehicle using a dead reckoning technique using data collected from a direction and distance sensor attached to the vehicle, and setting the position as the first moving position; A second step of correcting the first moving position obtained in the first step to the road on the numerical road map by a map matching method using a position matching degree and setting the position as the second moving position; A third step of analyzing the signal received from the GPS to obtain the current position of the vehicle and setting the position as the third moving position; A fourth step of correcting the third moving position obtained in the third step to the road on the numerical road map by a map matching method using a position matching degree and setting the position as the fourth moving position; A fifth step of obtaining a short-range driving pattern of the first to fourth moving positions, respectively; Comparing the degree of similarity between each of the short-distance driving pattern of the first to fourth moving positions obtained in the fifth step and the road pattern of the numerical road map, respectively, to select the moving position having the highest similarity as the final current position. A position measuring method of a traveling vehicle for vehicle navigation using a position matching degree comprising the sixth step. The method of claim 1, The similarity of the sixth step Obtain and add a position matching degree for a predetermined trajectory of each of the first to fourth moving positions, adding weights to the position matching of the second moving position and the most recent position matching degree among the trajectories of the respective moving positions, The position matching degree is obtained by a sum of a distance matching degree indicating a degree of matching between the position at which the vehicle is driven and the numerical road map, and a separation angle matching degree indicating a degree of matching between the angle at which the vehicle is driven and the road angle. A method for measuring the position of a traveling vehicle for vehicle navigation using the degree of matching. The method of claim 2, The distance match is The separation distance divided by the search distance minus one. The separation distance is a distance that the vehicle is a straight line away from the registration position on the numerical road map, the search distance is a range of error that the position of the vehicle can deviate from the road, If the separation distance is greater than the search distance And a position matching degree of the vehicle, wherein the distance matching degree is set to '0'. The method of claim 2, The position matching degree is The separation angle match is the separation angle divided by the allowable angle minus one, The separation angle is a difference between the traveling angle of the vehicle and the linear road angle of the numerical road map, and the allowable angle represents the maximum separation angle at which it is determined that the traveling direction and the road angle of the vehicle may be related. If the separation angle is larger than the allowable angle The method of measuring the position of the driving vehicle for vehicle navigation using the position matching degree, characterized in that for setting the separation angle matching degree to '0'. The method of claim 1, When the maximum similarity that varies depending on the number of driving trajectories minus the similarity of the third moving position is greater than the similarity of the other moving positions. And the third moving position is selected as the current position by looking at a road other than the road on the numerical road map as driving at that time, and selecting the third moving position as the current position. The method of claim 1, The fifth step is The position of the traveling vehicle for vehicle navigation using position matching, wherein a short distance driving pattern of each moving position is obtained by storing a predetermined number of sample data of the first to fourth moving positions of the vehicle measured at predetermined time intervals. How to measure. The method of claim 1, The sixth step If the difference between the similarity for the movement position pattern selected as the final position of the current vehicle and the similarity for the movement position pattern to be selected as the final position of the next vehicle is within 5% of the maximum similarity, the current position without changing the vehicle's movement position pattern is changed. A position measuring method of a traveling vehicle for vehicle navigation using a position matching degree, which maintains a movement position pattern as it is.

Images