Vehicle position estimation device

Abstract

To provide a vehicle position estimation device that is excellent in position estimation accuracy.SOLUTION: According to a vehicle position estimation device 100, a position estimation unit 120 estimates a vehicle position of a vehicle on a map on the basis of external information, a quantity of state, and map data. The position estimation unit 120 uses, according to purposes, an autonomous navigation method for estimating the vehicle position by using autonomous navigation of sequentially updating a position on the basis of, the quantity of state, and a map data method for estimating the vehicle position on the basis of, the external information and the map data, uses the autonomous navigation method while the vehicle changes its lane, and uses the map data method while the vehicle does not change its lane. Duration of the autonomous navigation method can be included in the lane change time to reduce the duration of the autonomous navigation method, thereby reducing a cumulative error and improving accuracy of estimating a position after the lane change.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The disclosure in this specification relates to a vehicle position estimation device that estimates the traveling position of a vehicle on a road. [Background technology]

[0002] Patent Document 1 discloses a self-position estimation device that determines which lane a position within a lane corresponds to, based on the correlation between the position within the lane, the absolute position, and the error therebetween, and estimates the driving lane based on the determination result. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2018-155732 Summary of the Invention [Problem to be solved by the invention]

[0004] The technology described in the aforementioned Patent Document 1 has a problem in that if the error in the estimation result of the in-lane position by the gyro sensor is larger than the lane width, the accuracy of estimating the lane candidate deteriorates, and therefore the accuracy of estimating the position also deteriorates.

[0005] The disclosed object has been made in consideration of the above-mentioned problems, and has as its object to provide a vehicle position estimation device that has excellent position estimation accuracy. [Means for solving the problem]

[0006] The present disclosure employs the following technical means to achieve the above-mentioned objectives.

[0007] The vehicle position estimation device disclosed herein is a vehicle position estimation device (100) mounted on an automobile (200) and used, and includes an external information acquisition unit (110) that acquires external information related to objects around the automobile and surrounding road markings, a vehicle state quantity acquisition unit (110) that acquires state quantities related to the traveling of the automobile, a map data acquisition unit (110) that acquires map data having road information related to lanes, a lane change determination unit (123) that determines whether the automobile is changing lanes based on the external information, the state quantities, and the map data, and a position estimation unit (120) that estimates the vehicle's position on the map based on the external information, the state quantities, and the map data, and when it is determined by the lane change determination unit that the automobile is changing lanes, the position estimation unit estimates the vehicle's position using autonomous navigation that successively updates the position based on the state quantities without using the external information and the map data, and when it is determined by the lane change determination unit that the automobile is not changing lanes, If the lane change determination unit determines that the vehicle is not changing lanes and a lane has been detected by external information, the vehicle position is estimated based on the external information and map data, and if the lane change determination unit determines that the vehicle is not changing lanes and a lane has not been detected by external information, the vehicle position is estimated using autonomous navigation. This is a vehicle position estimation device. The vehicle position estimation device disclosed herein is a vehicle position estimation device (100) mounted on an automobile (200) and used, and includes an external information acquisition unit (110) that acquires external information related to objects around the automobile and surrounding road markings, an automobile state quantity acquisition unit (110) that acquires state quantities related to the traveling of the automobile, a map data acquisition unit (110) that acquires map data having road information related to lanes, a lane change determination unit (123) that determines whether the automobile is changing lanes based on the external information, the state quantities, and the map data, and a position estimation unit (120) that estimates the position of the automobile on the map based on the external information, the state quantities, and the map data, and the position estimation unit In this case, the vehicle position estimation device estimates the vehicle position using autonomous navigation that successively updates the position based on state quantities without using external information or map data, and if the lane change determination unit determines that the vehicle is not changing lanes, the vehicle position estimation unit estimates the vehicle position based on the external information and map data, the external information acquisition unit acquires the distance to the road edge on both sides of the vehicle, and if the lane change determination unit determines that the vehicle is not changing lanes and the distance to the road edge has been detected using the external information, the position estimation unit estimates the vehicle position based on the external information and map data, and if the lane change determination unit determines that the vehicle is not changing lanes and the distance to the road edge has not been detected using the external information, the vehicle position estimation device estimates the vehicle position using autonomous navigation. Furthermore, the subject vehicle position estimation device disclosed herein is a subject vehicle position estimation device (100) mounted on and used in a vehicle (200), and includes an external information acquisition unit (110) that acquires external information related to objects around the vehicle and surrounding road markings, a subject vehicle state quantity acquisition unit (110) that acquires state quantities related to the running of the vehicle, a map data acquisition unit (110) that acquires map data having road information related to lanes, a lane change determination unit (123) that determines whether the vehicle is changing lanes based on the external information, the state quantities, and the map data, and a position estimation unit (120) that estimates the subject vehicle's position on a map based on the external information, the state quantities, and the map data, and the position estimation unit determines whether the vehicle is changing lanes based on the external information, the state quantities, and the map data. If it is determined that the vehicle is changing lanes, the vehicle position is estimated using autonomous navigation that successively updates the position based on state quantities without using external information or map data, and if the lane change determination unit determines that the vehicle is not changing lanes, the vehicle position is estimated based on external information and map data. The position estimation unit further includes a reliability calculation unit (121) that, if the vehicle is located on a road with multiple lanes, calculates, for each lane, a reliability indicating the probability that the vehicle is located in one of the multiple lanes using the external information and map data, and the lane change determination unit determines that the lane change is not in progress and that the lane change has been completed when the lane with the highest reliability calculated by the reliability calculation unit has changed to another lane.

[0008] According to this vehicle position estimation device, the position estimation unit estimates the vehicle's position on a map based on external information, state quantities, and map data. The position estimation unit selectively uses a first method for estimating the vehicle's position using autonomous navigation, which successively updates the position based on state quantities, and a second method for estimating the vehicle's position based on external information and map data. Specifically, the position estimation unit uses the first method when the vehicle is changing lanes and the second method when the vehicle is not changing lanes. With the first method, the cumulative error increases as the duration increases, resulting in a decrease in estimation accuracy. However, when the vehicle is changing lanes, lane recognition becomes unstable using external information, so the first method is preferable to the second method using external information. Therefore, by limiting the duration of the first method to when the vehicle is changing lanes, the duration of the first method can be shortened to suppress the cumulative error, while improving the estimation accuracy of the position after the lane change.

[0009] The symbols in parentheses for the above-mentioned means are examples showing the correspondence with the specific means described in the embodiments to be described later. [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 1 is a block diagram showing a vehicle system. [Figure 2] FIG. [Figure 3] FIG. 10 is a diagram illustrating an example of a method for calculating reliability using a lane marking; [Figure 4] FIG. 10 is a diagram illustrating an example of a method for calculating reliability using the number of lanes. [Figure 5] FIG. 10 is a diagram illustrating an example of a method for calculating reliability using road markings. [Figure 6] FIG. 10 is a diagram illustrating an example of a method for calculating reliability using lane changes. [Figure 7] FIG. 10 is a diagram illustrating an example of a method for determining whether a vehicle is changing lanes. [Figure 8] 10 is a flowchart illustrating switching of the position estimation method. [Figure 9] 10 is a flowchart illustrating another example of switching of the position estimation method. [Figure 10] 10 is a flowchart illustrating yet another example of switching between position estimation methods. DETAILED DESCRIPTION OF THE INVENTION

[0011] (First embodiment) A first embodiment of the present disclosure will be described with reference to FIGS. 1 to 7. A vehicle position estimation device 100 of this embodiment is mounted, for example, as part of a vehicle system 10 of a vehicle equipped with a navigation system or a vehicle with an automatic driving function. While a vehicle 200 is actually traveling, the vehicle position estimation device 100 estimates the vehicle position, that is, the position on a map, specifically, which road and which lane the vehicle is traveling in, based on various data described below. The vehicle position estimation device 100 outputs the estimated vehicle position including the lane to another device.

[0012] The vehicle position estimation device 100 estimates the position of the vehicle 200, thereby providing, for example, assistance to the driver for safe driving and assistance for autonomous driving. The vehicle 200 corresponds to an automobile. As shown in FIG. 1 , the vehicle system 10 includes a periphery monitoring sensor 20, a vehicle state quantity sensor unit 30, a GNSS receiver 40, a map data storage unit 50, and the vehicle position estimation device 100.

[0013] As shown in Fig. 2, the perimeter monitoring sensor 20 is a sensor that monitors the environment surrounding the host vehicle 200. The perimeter monitoring sensor 20 can detect moving objects such as pedestrians, cyclists, non-human animals, and other vehicles, as well as stationary objects such as fallen objects on the road, guardrails, curbs, road signs, road markings, and roadside structures, from within a detection range around the host vehicle. The perimeter monitoring sensor 20 outputs sensing information, which is external information obtained by detecting objects around the host vehicle 200, to the host vehicle position estimation device 100.

[0014] The perimeter monitoring sensor 20 acquires external information related to objects and road markings around the host vehicle 200. Specifically, the perimeter monitoring sensor 20 detects information on the dividing lines adjacent to both sides of the host vehicle 200, the total number of lanes on the road, and the number of lanes located on both sides of the host vehicle 200. The perimeter monitoring sensor 20 also acquires marking information such as road markings ahead of the lane in which the host vehicle 200 is traveling, and the distance to the road edges on both sides of the vehicle.

[0015] The perimeter monitoring sensor 20 has, for example, a front camera and a millimeter-wave radar as detection components for detecting objects. The front camera outputs at least one of image data capturing an image of the area ahead of the vehicle 200 and an analysis result of the image data as sensing information. A plurality of millimeter-wave radars are arranged at intervals on the front and rear bumpers of the vehicle 200, for example. The millimeter-wave radar irradiates millimeter waves or quasi-millimeter waves toward the periphery of the vehicle 200. The millimeter-wave radar generates sensing information by receiving waves reflected by moving objects, stationary objects, etc.

[0016] The host vehicle state quantity sensor unit 30 detects state quantities related to the traveling of the host vehicle 200, such as the vehicle speed, acceleration, and yaw rate. The host vehicle state quantity sensor unit 30 outputs data of the detected state quantities to the host vehicle position estimation device 100.

[0017] As shown in FIG. 2, the GNSS (Global Navigation Satellite System) receiver 40 receives positioning signals transmitted from a plurality of artificial satellites. Artificial satellites are also called positioning satellites. The GNSS receiver 40 can receive positioning signals from positioning satellites of at least one of the satellite positioning systems GPS, GLONASS, Galileo, IRNSS, QZSS, and Beidou. The GNSS receiver 40 outputs the received positioning signals to the vehicle position estimation device 100 as GPS information.

[0018] The map data storage unit 50 is a component that stores map data. The map data storage unit 50 is connected to the vehicle position estimation device 100, and the map data can be read by the vehicle position estimation device 100. The map data defines a map that represents roads using, for example, links and nodes. Specifically, the map data is formed by links formed as line segments of a predetermined length along the roads, which are sequentially connected by nodes.

[0019] Map data is data that shows so-called high-precision three-dimensional maps, and is a precise 3D mapping of roads and their surrounding areas. Map data includes road information related to roads. Road information includes the number of lanes, lane positions, lane shapes, and marking information. Marking information includes information on symbols, arrows, figures, and other symbols formed on the road surface, and includes information on road markings. Marking information includes information other than road markings stipulated by laws such as the Road Traffic Act, such as figures unique to a particular region. Road marking information includes information on lane markings and road markings.

[0020] The lane markings include roadway exterior lines and lane boundary lines. Roadway exterior lines are solid lines that indicate the boundary between the roadway and the shoulder. Lane boundary lines are solid or dashed lines that indicate the boundary between lanes. Information about roadway exterior lines also includes the color of the lines, such as yellow or white. Road markings are painted on the road surface for traffic control and regulation, such as no-turn restrictions, traffic divisions for different directions of travel, and maximum speed limits.

[0021] The map data also includes information on laneless sections, which are sections where lanes are not separated. In laneless sections, no lane boundaries are shown, and only the outer lines of the roadway are shown. The information on laneless sections includes information indicating the distance of the laneless section and information indicating its location.

[0022] The map data storage unit 50 may be, for example, a cloud server, instead of being provided in the vehicle position estimation device 100. The same function can be provided by transmitting map data from the cloud server to the vehicle position estimation device 100.

[0023] The vehicle position estimation device 100 generates highly accurate position information of the vehicle 200 by performing composite positioning that combines multiple pieces of acquired information. Furthermore, the vehicle position estimation device 100 estimates one lane in which the vehicle 200 is traveling on a road that includes multiple lanes.

[0024] The vehicle position estimation device 100 is a control device that executes a program stored in a storage medium and controls each unit. The vehicle position estimation device 100 has at least one central processing unit (CPU) and a storage medium that stores programs and data. The vehicle position estimation device 100 is realized, for example, by a microcomputer equipped with a computer-readable storage medium. The storage medium is a non-transient, tangible storage medium that non-temporarily stores computer-readable programs and data. The storage medium is realized by a semiconductor memory, a magnetic disk, or the like.

[0025] The vehicle position estimation device 100 has, as functional blocks, an information acquisition unit 110 and a position estimation unit 120. The information acquisition unit 110 acquires sensing information from the periphery monitoring sensor 20, state quantity data from the vehicle state quantity sensor unit 30, GPS information from the GNSS receiver 40, and map data from the map data storage unit 50. Therefore, the information acquisition unit 110 functions as an external information acquisition unit, a vehicle state quantity acquisition unit, a satellite positioning acquisition unit, and a map data acquisition unit. The information acquisition unit 110 provides the acquired information to the position estimation unit 120.

[0026] The position estimation unit 120 estimates the position of the host vehicle 200 on a map based on sensing information, state quantity data, GPS information, map data, etc. For example, the position estimation unit 120 estimates the latitude and longitude indicating the current position of the host vehicle 200 from GPS information acquired by the GNSS receiver 40. The position estimation unit 120 also estimates, from the state quantity data detected by the host vehicle state quantity sensor unit 30, whether the host vehicle 200 is traveling on a straight road, whether the host vehicle 200 is traveling on a curved road with a certain degree of curvature, whether the host vehicle 200 is traveling so as to deviate from its lane, etc.

[0027] The position estimation unit 120 estimates the position using an autonomous navigation method and a map data method. The autonomous navigation method is a method for estimating the vehicle position using autonomous navigation, which successively updates the position based on state quantity data. Specifically, the autonomous navigation method calculates the relative position with respect to the previous position based on the speed and traveling direction included in the state quantity, and successively determines the position by updating the position based on the relative position.

[0028] The map data method is a method for estimating the vehicle's position based on external information and map data. Specifically, the map data method compares external information, such as image data from an on-board camera, with map data to calculate an estimated position on the map data.

[0029] The position estimation unit 120 has, as sub-function blocks, a reliability calculation unit 121, a lane estimation unit 122, and a lane change determination unit 123. When the vehicle 200 is located on a road with multiple lanes, the reliability calculation unit 121 calculates, for each lane, a reliability indicating the probability that the vehicle 200 is located in one of the multiple lanes.

[0030] The lane estimation unit 122 estimates the lane in which the host vehicle 200 is located using the reliability calculated by the reliability calculation unit 121. For example, the lane estimation unit 122 estimates the lane with the highest reliability as the lane in which the host vehicle 200 is located. Furthermore, for example, if there are multiple lanes with the highest reliability, the lane estimation unit 122 does not estimate the lane in which the host vehicle 200 is located to be one lane.

[0031] The lane change determination unit 123 determines whether the vehicle 200 is changing lanes based on the sensing information, state quantity data, GPS information, and map data. A lane change refers to a change in the lane in which the vehicle is traveling by moving from the lane in which the vehicle is traveling to an adjacent lane on the right or left. The lane change determination unit 123 also determines whether the vehicle is changing lanes.

[0032] Next, the calculation of reliability will be described. In the example shown in Fig. 3, the position estimation unit 120 estimates that the vehicle is currently traveling in the first lane L1 of a road with four lanes on each side. However, the vehicle is actually traveling in the second lane L2. In the sensing information, the information on the adjacent demarcation lines on both sides of the vehicle is dashed. The map data stores information on the currently traveling demarcation lines, with five demarcation lines, the left and right sides being solid lines and the central three dashed lines. Therefore, the lanes with dashed demarcation lines on both sides are the second lane L2 and the third lane L3. When traveling in the first lane L1, the left demarcation line should be detected as a solid line and the right demarcation line as a dashed line.

[0033] When the sensing information and the map data's lane marking information are compared, the information for the second lane L2 and the third lane L3 matches, but the information for the first lane L1 and the fourth lane L4 does not match. Therefore, in this case, the probability that the vehicle is in the second lane L2 and the third lane L3 is higher than the probability that the vehicle is in the first lane L1 and the fourth lane L4. Therefore, as shown in FIG. 3, the reliability indicating the probability is set higher for the second lane L2 and the third lane L3 than for the first lane L1 and the fourth lane L4. Using the reliability, the position estimation unit 120 determines that the vehicle is not in the first lane L1 and re-estimates that the vehicle is in the second lane L2 or the third lane L3.

[0034] In this way, the reliability calculation unit 121 compares the information on the adjacent lane lines on both sides obtained from the sensing information with the information on the lane lines included in the map data, and sets a higher reliability for lanes where the acquired adjacent lane lines on both sides match the lane lines in the map data than for lanes where they do not match. For example, the reliability calculation unit 121 can also use the color of the lane lines to make a determination. The reliability is set based on whether the information included in the map data matches, for example, based on whether the adjacent lane lines on both sides are white or yellow.

[0035] Next, a method for calculating reliability using the number of lanes will be described. As shown in Fig. 4, the host vehicle 200 is traveling on the second lane L2 of a road with three lanes on each side. The sensing information includes the total number of lanes on the road the host vehicle 200 is traveling on, the number of lanes located to the left of the host vehicle 200, and the number of lanes located to the right of the host vehicle 200. Since the sensing information includes lane marking information, the number of lanes is obtained using the number of lane marks. In the example shown in Fig. 4, if the sensing information is appropriate, the number of lanes on the left side will be 1 and the number of lanes on the right side will be 1.

[0036] Since the map data reveals that the number of lanes the vehicle is traveling on is three, the lane the vehicle is traveling in can be estimated using the numbers of lanes on both sides indicated by the sensing information. Specifically, when the vehicle is traveling in the first lane L1 of a road with three lanes on each side, the number of lanes on the left side is 0 and the number of lanes on the right side is 2. When the vehicle is traveling in the second lane L2 of a road with three lanes on each side, the number of lanes on the left side is 1 and the number of lanes on the right side is 1. When the vehicle is traveling in the third lane L3 of a road with three lanes on each side, the number of lanes on the left side is 2 and the number of lanes on the right side is 0. Therefore, the reliability calculation unit 121 uses the detected number of lanes to increase the reliability of the traveling lane and decrease the reliability of the other lanes.

[0037] However, there are cases where the perimeter monitoring sensor 20 cannot detect the number of lanes. For example, if another surrounding vehicle 201 is traveling in the adjacent lane on the right, the right lane cannot be recognized, and the number of lanes on the left side may be 1 and the number of lanes on the right side may be 0 based on the sensing information. Therefore, since the number of lanes differs from the number of lanes in the map data, it is difficult for the reliability calculation unit 121 to assign a high or low reliability to each lane.

[0038] Therefore, when the number of lanes in the map data matches the total number of lanes in the sensing information, the reliability of the lane location according to the number of lanes on both sides is calculated to be higher than for lanes that do not match, as described above. Conversely, when the number of lanes in the map data does not match the total number of lanes in the sensing information, there is a possibility that the sensing information contains an error, so the reliability of all lanes is calculated to be the same.

[0039] In this way, the reliability calculation unit 121 compares the total number of lanes in the sensing information with the number of lanes included in the map data, and if the acquired total number of lanes matches the number of lanes in the map data, it sets the reliability of the lane in which the vehicle is located, identified using the number of lanes on both sides, to a higher level than the reliability of other lanes.

[0040] Next, a method for calculating reliability using road markings will be described. In the example shown in FIG. 5, the position estimation unit 120 estimates that the host vehicle 200 is currently in the second lane L2, the center of a three-lane road. However, the host vehicle 200 is actually traveling in the first lane L1. The sensing information indicates that the road marking ahead of the host vehicle 200 is a straight-ahead left-turn arrow. The map data stores that the lane with the road marking with the straight-ahead left-turn arrow is located in the first lane L1. Therefore, the road marking ahead of the sensing information does not match the road marking included in the map data. If the host vehicle 200 is traveling in the second lane L2, the road marking ahead of the host vehicle 200 is a straight-ahead arrow. Therefore, in this case, as shown in FIG. 5, the reliability of the first lane L1, which matches the straight-ahead left-turn arrow, is set higher than the reliability of the second lane L2 and the third lane L3, which do not match.

[0041] In this way, the reliability calculation unit 121 compares the road markings in the sensing information with the road markings included in the map data, and if the acquired road markings match the road markings in the map data, the reliability calculation unit 121 sets the reliability of the matching lane higher than the reliability of the non-matching lane. As a result, the position estimation unit 120 uses the reliability to re-estimate that the vehicle is located in the first lane L1.

[0042] Next, a method for calculating reliability using lane changes will be described. As shown in Fig. 6, the position estimation unit 120 estimates that the vehicle 200 is currently traveling in the center of a road with three lanes on each side. The vehicle is actually traveling in the second lane L2. The case where the lane change determination unit 123 determines that the vehicle has changed its traveling lane to the right lane will be described.

[0043] If it is determined that the lane has changed to the right, the probability of being in the first lane L1 is lower than the probability of being in the second lane L2 or the third lane L3. Conversely, if it is determined that the lane has changed to the left, the probability of being in the third lane L3 is lower than the probability of being in the first lane L1 or the second lane L2. Therefore, in the example shown in Figure 6, since the lane has changed to the right, the probability of being in the second lane L2 or the third lane L3 is calculated to be higher than that of the first lane L1.

[0044] In the case of a road with two lanes on each side, if it is determined that the lane has changed to the right, the probability that the vehicle is in the first lane L1 is lower than the probability that the vehicle is in the second lane L2. Conversely, if it is determined that the lane has changed to the left, the probability that the vehicle is in the second lane L2 is lower than the probability that the vehicle is in the first lane L1.

[0045] In this way, when the lane change determination unit 123 determines that a lane change has occurred, the reliability calculation unit 121 lowers the reliability of the lane located at the end opposite to the direction of the lane change compared to the other lanes.

[0046] In this way, the reliability calculation unit 121 calculates the reliability of each lane using the number of lanes, road markings, lane changes, etc. Then, the reliability calculation unit 121 calculates a reliability by integrating the reliability calculated using different features using a weighting coefficient. For example, the reliability using road markings is weighted more heavily than the reliability using lane changes. This allows the position estimation unit 120 to estimate the lane in which the vehicle is traveling using the integrated reliability.

[0047] Next, a description will be given of a lane change determination method of the lane change determination unit 123. As a first determination method, when a dividing line is detected, the lane change determination unit 123 determines that a lane change has occurred when the dividing line is crossed. As a second determination method, when the host vehicle 200 is traveling near a lane center line, the lane change determination unit 123 determines that a lane change has occurred when the distance between the lane center line and the center position of the host vehicle 200 is increasing and exceeds a predetermined threshold.

[0048] As a third determination method, the lane change determination unit 123 determines whether the vehicle is changing lanes to one side by using a change in the distance from the vehicle to the road edge on one side that decreases, or a change in the distance from the vehicle to the road edge on the other side that increases. This method is effective when the perimeter monitoring sensor 20 cannot recognize marking lines but recognizes road edges. Marking lines can be difficult to detect due to deterioration or puddles, but road edges are often easy to detect due to bumps and other factors.

[0049] Specifically, as shown in Fig. 7, the distances to the road edges are detected as a left edge distance W2 from the center of the vehicle to the left edge of the road and a right edge distance W1 to the right edge of the road. When changing lanes to the left lane, as shown in Fig. 7, the left edge distance W2 decreases, and the right edge distance W1 increases along with this decrease. If the amount of decrease or increase exceeds a threshold, it is determined that the vehicle has changed lanes.

[0050] Furthermore, as a fourth determination method, the lane change determination unit 123 determines that a lane change has occurred when the lane with the highest reliability calculated by the reliability calculation unit 121 changes to another lane. Reliability is calculated for each lane, and if there is a lane with a high reliability, there is a high possibility that the vehicle is located in that lane. Therefore, if a lane with a high reliability changes to another lane, it is determined that a lane change has occurred.

[0051] In this way, the lane change determination unit 123 determines whether a lane change has occurred using the four determination methods. The lane change determination unit 123 may determine whether a lane change has occurred using only one of the four determination methods, or may determine whether a lane change has occurred by combining the determination results of two or more determination methods.

[0052] The lane change determination unit 123 also determines whether or not a lane change is in progress. In a first determination method, the lane change determination unit 123 determines that a lane change is in progress when a lane marking is being crossed. In a second determination method, the lane change determination unit 123 determines that a lane change is in progress when the distance between the lane center line and the center position of the vehicle 200 exceeds a threshold and is increasing.

[0053] The third determination method determines whether a lane change is occurring by using a decrease in the distance from the vehicle to the edge of the road on one side or an increase in the distance to the edge of the road on the other side. For example, the third determination method determines that a lane change is occurring if the distance to the edge of the road exceeds a threshold and is increasing or decreasing.

[0054] In the fourth determination method, when the lane with the highest reliability calculated by the reliability calculation unit 121 changes to another lane, it is determined that the lane change is complete rather than in progress. This is because a change in reliability can be considered synonymous with the lane change being completed. This is because the reliability cannot be updated while the lane change is in progress.

[0055] Next, a description will be given of a method for switching between the autonomous navigation method and the map data method by the position estimation unit 120. Figures 8 to 10 are flowcharts for explaining the switching method. Each flowchart is repeatedly executed by the position estimation unit 120 in a short period of time.

[0056] In step S1, it is determined whether or not a lane change is occurring. If a lane change is occurring, the process proceeds to step S2. If a lane change is not occurring, the process proceeds to step S3. In step S2, since a lane change is occurring, the position is estimated using the autonomous navigation method, and this flow ends. In step S3, since a lane change is not occurring, the position is estimated using the map data method, and this flow ends.

[0057] In this way, the position estimation unit 120 changes the estimation method depending on whether or not the vehicle is changing lanes, in order to utilize the characteristics of each position estimation method and reduce errors.

[0058] Next, the flow will be explained using the flowchart in Figure 9. In step S11, it is determined whether or not a lane change is in progress, and if so, the process proceeds to step S12, and if not, the process proceeds to step S13. In step S13, it is determined whether or not a lane is being detected, and if so, the process proceeds to step S14, and if not, the process proceeds to step S12. If a lane is not being detected, it may be that there is a lane-free section, or the markings cannot be detected due to deterioration of the markings, for example.

[0059] In step S12, since the vehicle is either changing lanes or not detecting lanes, the position is estimated using the autonomous navigation method, and the flow ends. In step S14, since the vehicle is either not changing lanes or not detecting lanes, the position is estimated using the map data method, and the flow ends.

[0060] In this way, the position estimation unit 120 changes the estimation method depending on whether or not the lane is being detected. This also makes use of the characteristics of each position estimation method and reduces errors.

[0061] Next, the flow will be explained using the flowchart in Fig. 10. In step S21, it is determined whether or not a lane change is in progress, and if so, the process proceeds to step S22, and if not, the process proceeds to step S23. In step S23, it is determined whether or not a road edge is being detected, and if so, the process proceeds to step S24, and if not, the process proceeds to step S22. If a road edge is not being detected, it may be the case that the road edge cannot be detected in a large space such as a parking lot where there is no road edge, or if the road edge has deteriorated, for example.

[0062] In step S22, since the vehicle is either changing lanes or not detecting a road edge, the position is estimated using the autonomous navigation method, and the flow ends. In step S24, since the vehicle is either not changing lanes or not detecting a road edge, the position is estimated using the map data method, and the flow ends.

[0063] In this way, the position estimation unit 120 changes the estimation method depending on whether a road edge is being detected. This is to take advantage of the characteristics of each position estimation method and reduce errors. The flowcharts of Figures 9 and 10 may be executed independently, or only one of the processes of Figures 9 and 10 may be executed. Furthermore, the processes of Figures 9 and 10 may be combined to switch to the autonomous navigation method when neither a road edge nor a lane is being detected, and to switch to the map data method when either one is being detected.

[0064] As described above, according to the vehicle position estimation device 100 of this embodiment, the position estimation unit 120 estimates the vehicle position of the vehicle 200 on a map based on external information, state quantities, and map data. The position estimation unit 120 selectively uses either an autonomous navigation method, which estimates the vehicle position using autonomous navigation, which successively updates the position based on state quantities, or a map data method, which estimates the vehicle position based on external information and map data. Specifically, the position estimation unit 120 uses the autonomous navigation method when the vehicle is changing lanes and the map data method when the vehicle is not changing lanes. With the autonomous navigation method, the cumulative error increases as the duration increases, resulting in a decrease in estimation accuracy. However, when the vehicle is changing lanes, the external information makes lane recognition unstable, so the autonomous navigation method is preferable to the map data method, which uses external information. Therefore, by setting the duration of the autonomous navigation method during a lane change, the duration of the autonomous navigation method can be shortened, suppressing the cumulative error while improving the estimation accuracy of the position after the lane change.

[0065] In this embodiment, the position estimation unit 120 changes the estimation method depending on whether or not lanes are being detected. When lanes are not being detected but lanes are not being changed, the autonomous navigation method is preferable to the map data method that uses external information. Therefore, by setting the duration of the autonomous navigation method to the period when lanes are not being detected, the duration of the autonomous navigation method can be shortened to suppress cumulative errors, while improving the accuracy of estimating the position after a lane change.

[0066] Furthermore, in this embodiment, the position estimation unit 120 changes the estimation method depending on whether a road edge is being detected. When a lane change is not in progress but a road edge has not yet been detected, the autonomous navigation method is preferable to the map data method using external information. Therefore, by setting the duration of the autonomous navigation method to the period when a road edge is not being detected, the duration of the autonomous navigation method can be shortened to suppress cumulative errors, while improving the accuracy of estimating the position after a lane change.

[0067] In this embodiment, the lane change determination unit 123 determines whether or not a lane change is occurring based on a decrease in the distance from the vehicle 200 to the road edge on one side or an increase in the distance from the vehicle 200 to the road edge on the other side. Because the lane change determination unit 123 determines whether or not the vehicle is changing lanes based on the distance to the road edge, it can determine whether or not a lane change is occurring even if a dividing line cannot be detected.

[0068] Furthermore, in this embodiment, the lane change determination unit 123 determines that the lane change is complete when the lane with the highest reliability calculated by the reliability calculation unit 121 is changed to another lane. This makes it possible to determine the completion of the lane change using the reliability.

[0069] (Other embodiments) The above describes preferred embodiments of the present disclosure, but the present disclosure is not limited to the above-described embodiments and can be implemented in various modified forms within the scope of the gist of the present disclosure.

[0070] The structures of the above-described embodiments are merely examples, and the scope of the present disclosure is not limited to the scope of these descriptions. The scope of the present disclosure is defined by the claims, and further includes all modifications within the meaning and scope equivalent to the claims.

[0071] In the first embodiment described above, the information acquisition unit 110 has the functions of an external information acquisition unit, a vehicle state quantity acquisition unit, a satellite positioning acquisition unit, and a map data acquisition unit, but is not limited to a configuration in which these are integrated, and each function may be realized in a different location.

[0072] The functions realized by the vehicle localization device 100 in the first embodiment described above may be realized by hardware and software different from those described above, or a combination of these. The vehicle localization device 100 may, for example, communicate with another control device, and the other control device may perform some or all of the processing. When the vehicle localization device 100 is realized by an electronic circuit, it may be realized by a digital circuit including a large number of logic circuits, or an analog circuit.

[0073] In the first embodiment described above, the vehicle position estimation device 100 is used in a vehicle, but is not limited to being mounted on a vehicle, and at least a part of it may not be mounted on a vehicle. [Explanation of symbols]

[0074] 10... Vehicle system 20... Surroundings monitoring sensor 30... Vehicle state quantity sensor unit 40...GNSS receiver 50...map data storage unit 100...vehicle position estimation device 110... Information acquisition unit (external information acquisition unit, host vehicle state quantity acquisition unit, satellite positioning acquisition unit, map data acquisition unit) 120... Position estimation unit 121... Reliability calculation unit 122... Lane estimation unit 123...Lane change determination unit 200...Own vehicle (automobile) L1...First lane L2...2nd lane L3...3rd lane L4...4th lane

Claims

1. In a vehicle position estimation device (100) mounted on an automobile (200), an external information acquisition unit (110) for acquiring external information relating to objects around the vehicle and road markings around the vehicle; a vehicle state quantity acquisition unit (110) that acquires a state quantity related to the running of the vehicle; a map data acquisition unit (110) for acquiring map data having road information related to lanes; a lane change determination unit (123) that determines whether the vehicle is changing lanes based on the external information, the state quantity, and the map data; a position estimation unit (120) that estimates a position of the vehicle on a map based on the external information, the state quantity, and the map data; The position estimation unit when it is determined by the lane change determination unit that the vehicle is changing lanes, estimating the vehicle position using an autonomous navigation method that successively updates the position based on the state quantity without using the external information and the map data; When the lane change determination unit determines that the vehicle is not changing lanes, the vehicle position is estimated based on the external information and the map data; When the lane change determination unit determines that the vehicle is not changing lanes and the lane is detected by the external information, the vehicle position is estimated based on the external information and the map data; When the lane change determination unit determines that the vehicle is not changing lanes and the external information does not detect a lane, the vehicle position estimation device estimates the vehicle position using autonomous navigation.

2. In a vehicle position estimation device (100) mounted on an automobile (200), an external information acquisition unit (110) for acquiring external information relating to objects around the vehicle and road markings around the vehicle; a vehicle state quantity acquisition unit (110) that acquires a state quantity related to the running of the vehicle; a map data acquisition unit (110) for acquiring map data having road information related to lanes; a lane change determination unit (123) that determines whether the vehicle is changing lanes based on the external information, the state quantity, and the map data; a position estimation unit (120) that estimates a position of the vehicle on a map based on the external information, the state quantity, and the map data; The position estimation unit when it is determined by the lane change determination unit that the vehicle is changing lanes, estimating the vehicle position using an autonomous navigation method that successively updates the position based on the state quantity without using the external information and the map data; When the lane change determination unit determines that the vehicle is not changing lanes, the vehicle position is estimated based on the external information and the map data; the external information acquisition unit acquires distances to road edges on both sides of the vehicle, The position estimation unit When the lane change determination unit determines that the vehicle is not changing lanes and the distance to the road edge is detected based on the external information, the vehicle position is estimated based on the external information and the map data; A vehicle position estimation device that estimates the vehicle position using autonomous navigation when the lane change determination unit determines that the vehicle is not changing lanes and the distance to the road edge has not been detected using the external information.

3. In a vehicle position estimation device (100) mounted on an automobile (200), an external information acquisition unit (110) for acquiring external information relating to objects around the vehicle and road markings around the vehicle; a vehicle state quantity acquisition unit (110) that acquires a state quantity related to the running of the vehicle; a map data acquisition unit (110) for acquiring map data having road information related to lanes; a lane change determination unit (123) that determines whether the vehicle is changing lanes based on the external information, the state quantity, and the map data; a position estimation unit (120) that estimates a position of the vehicle on a map based on the external information, the state quantity, and the map data; The position estimation unit when it is determined by the lane change determination unit that the vehicle is changing lanes, estimating the vehicle position using an autonomous navigation method that successively updates the position based on the state quantity without using the external information and the map data; When the lane change determination unit determines that the vehicle is not changing lanes, the vehicle position is estimated based on the external information and the map data; the position estimation unit further includes a reliability calculation unit (121) that, when the vehicle is located on a road having a plurality of lanes, calculates, for each lane, a reliability indicating a probability of which lane out of the plurality of lanes the vehicle is located in, using the external information and the map data; The lane change determination unit determines that the lane change is not in progress and that the lane change has been completed when the lane with the highest reliability calculated by the reliability calculation unit is changed to another lane.

4. the external information acquisition unit acquires distances to road edges on both sides of the vehicle, A vehicle position estimation device as described in any one of claims 1 to 3, wherein the lane change determination unit determines whether the vehicle is changing lanes to one side using a change in the distance from the vehicle to the edge of the road on one side that decreases, or a change in the distance from the vehicle to the edge of the road on the other side that increases.

Images