JP2017009554A - Vehicle location determination device and vehicle location determination method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a vehicle location determination device that can accurately correct a current location of a vehicle in various travelling environments.SOLUTION: A vehicle location determination device comprises: a storage device 5 that stores map information 51 including locations of features existing around roads; a vehicle location estimation unit 61 that estimates a current location of a vehicle in the map information 51; a camera 1 that captures images surrounding the vehicle; an area specification unit 64 that, in the image, specifies an area Q in which a feature being a recognition object exists on the basis of a prescribed condition; a feature recognition unit 65 that recognizes the feature from the area Q; a relative location calculation unit 66 that calculates a relative location of the vehicle with respect to a location of the recognized feature in the map information 51; and a location correction unit 67 that corrects the current location of the vehicle P on the basis of the relative location calculated by the relative location calculation unit 66.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a vehicle position determination device and a vehicle position determination method for determining the position of a vehicle.

  As a technique for correcting the current position of the vehicle detected by the positioning device, the distance from the vehicle to the intersection is obtained using a camera mounted on the vehicle, and the position of the vehicle with respect to the intersection in the map information is specified. Has been proposed (see Patent Document 1).

JP-A-9-243389

  However, the technique described in Patent Document 1 may not be able to correct the current position of the vehicle in situations where it is difficult to detect an intersection such as when traveling on an expressway.

  In view of the above problems, an object of the present invention is to provide a vehicle position determination device and a vehicle position determination method that can accurately correct the current position of a vehicle in various traveling environments.

  The vehicle position determination device identifies a region where a feature to be recognized exists in a captured image based on a predetermined condition, and recognizes the feature from the region identified in the image, thereby detecting the position of the feature. The relative position of the vehicle with respect to is calculated, and the current position of the vehicle is corrected.

  According to the present invention, a vehicle position determination that can accurately correct the current position of a vehicle in various driving environments by specifying an image area including a feature to be recognized according to a predetermined condition. An apparatus and a vehicle position determination method can be provided.

FIG. 1 is a block diagram illustrating an example of a configuration of a vehicle position determination device according to an embodiment of the present invention. FIG. 2 is an example illustrating a region specified in an image when the vehicle position determination device according to the embodiment of the present invention is raining or snowing. FIG. 3 is an example illustrating an area specified in an image when the vehicle position determination device according to the embodiment of the present invention is clear or cloudy. FIG. 4 is an example illustrating an area specified in an image by the vehicle position determination device according to the embodiment of the present invention in a traffic jam. FIG. 5 is a flowchart illustrating a processing flow of the vehicle position determination device according to the embodiment of the present invention.

  Embodiments of the present invention will be described with reference to the drawings. In the description of the drawings, the same or similar parts are denoted by the same or similar reference numerals, and redundant description is omitted.

(Vehicle position determination device)
As shown in FIG. 1, the vehicle position determination device according to the present embodiment includes a camera 1, a positioning device 2, a communication interface (I / F) 3, a sensor group 4, a storage device 5, and a processing unit. 6, a display 7, an input I / F 8, and a speaker 9. A vehicle position determination device according to an embodiment of the present invention is mounted on a vehicle and determines a current position of the vehicle.

  The camera 1 captures an image around the vehicle. The camera 1 generates a digital image that can be processed by a solid-state imaging device such as a CCD or CMOS. The camera 1 sequentially outputs captured images to the processing unit 6. The camera 1 may be a fixed camera that captures an image in front of the vehicle, or may be an omnidirectional camera that captures images in all directions around the vehicle. Alternatively, the camera 1 may perform rotation in at least one of pan, tilt, and roll, zoom in, and out in accordance with control by the processing unit 6.

  The positioning device 2 measures the current position by a positioning system such as a global positioning system (GPS). The positioning device 2 is composed of a GPS receiver, for example. The positioning device 2 sequentially outputs the measured current position to the processing unit 6.

  The communication I / F 3 is, for example, a communication device that transmits and receives signals to and from the outside wirelessly. The communication I / F 3 receives various information from the outside by an intelligent road traffic system (ITS) that transmits traffic information such as traffic jam information and traffic regulations, weather information, and the like in real time. ITS includes VICS (Vehicle Information and Communication System: registered trademark), telematics, and the like. The communication I / F 3 sequentially outputs the received information to the processing unit 6.

  The sensor group 4 can be composed of a speed sensor, an acceleration sensor, an angular velocity sensor, a steering angle sensor, and the like. The sensor group 4 detects the vehicle speed, three-axis acceleration, angular velocity, and the like in the three-dimensional orthogonal coordinate system by each sensor, and sequentially outputs the detection results to the processing unit 6. The sensor group 4 may include a distance sensor such as a laser range finder (LRF) that detects the distance and direction from the vehicle to the target, an orientation sensor that detects the direction of the vehicle, and the like.

  The storage device 5 stores map information 51 including the positions and types of features existing around the road. The storage device 5 can be composed of a semiconductor memory, a magnetic disk, or the like. In addition, the storage device 5 may store a program necessary for processing performed in the processing unit 6. The storage device 5 may be composed of a single piece of hardware or a plurality of pieces of hardware.

  The map information 51 records road information, facility information, and the like. The map information 51 records feature information including the position, size, type, and the like of various features existing around the road. Features around roads include road signs indicating speed limits, temporary stops, one-way streets, pedestrian crossings, parking prohibitions, road markings indicating stop lines, pedestrian crossings, pedestrian crossings, lane markings, directions, services Information signs indicating areas, parking areas, branches, various facilities, and the like. In addition, the features include road structures and facilities such as distance markers, traffic lights, telephone poles, tunnels, bridges, piers, emergency parking zones, emergency telephones, toll gates, gantry cranes, and railways.

  The processing unit 6 includes a vehicle position estimation unit 61, an environment information acquisition unit 62, a condition determination unit 63, a region specifying unit 64, a feature recognition unit 65, a relative position calculation unit 66, and a position correction unit 67. And a route setting unit 68. The route setting unit 68 sets the travel route of the vehicle from the current position of the vehicle to the destination in the map information 51 as the set route.

  The processing unit 6 can be configured by, for example, a microcontroller that is an integrated circuit including a central processing unit (CPU), a memory, an input / output I / F, and the like. In this case, a plurality of information processing units (61 to 68) constituting the processing unit 6 are realized by the CPU executing a computer program preinstalled in the microcontroller. Each unit constituting the processing unit 6 may be configured by integral hardware or may be configured by separate hardware. The microcontroller may also be used as an electronic control unit (ECU) used for other control related to the vehicle such as automatic driving control.

  The vehicle position estimation unit 61 estimates the current position of the vehicle in the map information 51 based on the current position measured by the positioning device 2 and the momentum of the vehicle calculated from the detection result of the sensor group 4.

  The environment information acquisition unit 62 acquires traffic information, weather information, and the like of a region where the vehicle is currently traveling from the communication I / F 3.

  The condition determination unit 63 includes an image captured by the camera 1, a detection result of the sensor group 4, a current position of the vehicle estimated by the vehicle position estimation unit 61, information acquired by the environment information acquisition unit 62, and a route setting unit 68. Based on at least one of the travel routes set by (1), the travel condition of the vehicle is determined. The travel condition determined by the condition determination unit 63 is at least one of the surrounding environment such as weather, time zone, road surface condition, and travel location, the travel state, the travel state such as the travel lane, and the travel route of the vehicle. Including.

  The condition determination unit 63 determines, for example, sunny, rain, snow, fog, etc., regarding the weather based on the weather information acquired by the environment information acquisition unit 62. The condition determination unit 63 may determine the weather based on the detection result of the raindrop sensor included in the sensor group 4, the driving state of the wiper included in the vehicle, and the like. The condition determination unit 63 determines whether the time zone is daytime or nighttime based on the current time with respect to the sunrise and sunset times or the detection result of the illuminance sensor included in the sensor group 4.

  The condition determination unit 63 is wet, snowy, frozen, or appropriately paved as the road surface state on which the vehicle travels based on the weather information acquired by the environment information acquisition unit 62, the image taken by the camera 1, or the like. A bad road or the like which is a road whose surface is not exposed in the state is determined. The condition determination unit 63 may determine the road surface state based on a detection result of a slip sensor that is included in the sensor group 4 and detects a slip of a vehicle wheel. In addition, the condition determination unit 63 determines an urban area, a suburb, an expressway, a general road, a mountain road, and the like as the travel location of the vehicle based on the current position estimated by the vehicle position estimation unit 61.

  Based on the traffic information acquired by the environment information acquisition unit 62, the condition determination unit 63 determines, for example, when driving on an expressway (or a car road) or when there is a traffic jam. The condition determination unit 63 may determine a traveling state such as during high-speed traveling or low-speed traveling based on the detection result of the speed sensor provided in the sensor group 4. Alternatively, the condition determination unit 63 calculates the distance to the other vehicles on the front and rear sides based on the image taken by the camera 1, and determines that the traffic is congested when the short distance is maintained for a predetermined time. It may be.

  The condition determination unit 63 uses the image taken by the camera 1, the current position estimated by the vehicle position estimation unit 61, the movement of the vehicle calculated based on the detection result of the sensor group 4, and the like as a traveling lane. The right lane, the left lane, the right lane, and the center lane between the left lane and the like are determined.

  The region specifying unit 64 is a region Q (see FIG. 2-4) as a range in which the feature to be recognized exists in the image taken by the camera 1 based on each traveling condition determined by the condition determining unit 63. Is identified. For example, the area specifying unit 64 specifies the area Q in the image in the traveling direction of the vehicle.

  The feature recognizing unit 65 recognizes the feature from the region Q specified by the region specifying unit 64 among the images taken by the camera 1. The feature recognizing unit 65 applies predetermined image processing such as edge detection and pattern matching for calculating the similarity to a template stored in advance to the region Q of the image, thereby applying the predetermined feature to the feature in the map information 51. Recognize corresponding feature from region Q. The feature recognition unit 65 associates the feature in the map information 51 with the recognized feature. Further, the feature recognition unit 65 calculates the direction of the recognized feature with respect to the vehicle and the distance from the vehicle based on the image taken by the camera 1. The feature recognition unit 65 may acquire the distance and direction to the recognized feature using the LRF or the like provided in the sensor group 4.

  For example, as shown in FIG. 2, when the condition determination unit 63 determines that the weather is snow as the surrounding environment, the region specifying unit 64 becomes difficult to accurately recognize the road marking due to snow. The upper half of the image is specified as a region Q1 in order to recognize information signs, road signs, etc. The area specifying unit 64 performs image processing by specifying the area Q in a range where a feature that is expected to be recognizable is located, except for a range on the image where a feature with low likelihood of recognition is located. The range of images to be processed is limited, and the processing load on the feature recognition unit 65 can be reduced.

  In the example shown in FIG. 2, only the road sign M1 and the guide sign M2 are included in the region Q among the road sign M1, the guide sign M2, and the guide sign M3 existing on the image. The area specifying unit 64 may specify the area excluding the lower part of the image as the area Q so that the guide sign M3 that is a feature existing at a recognizable distance is also included in the area Q. When the weather is raining, the area specifying unit 64 makes it difficult to accurately recognize a feature such as a painted road sign due to reflection of a wet road surface. May be.

  In addition, as shown in FIG. 3, the region specifying unit 64 specifies the region Q2 in the entire range of the image when the feature on the image is expected to be generally recognizable, such as when the weather is sunny or cloudy. It may be. Alternatively, for example, when the weather is fine, the region specifying unit 64 may display a lower part or an upper part of an image where a road sign, a distance sign, etc., which are located at a relatively short distance from the vehicle, in order to improve recognition accuracy. The region Q may be specified in the excluded range. In addition, when the weather is foggy, it becomes difficult for the area specifying unit 64 to recognize a feature away from the vehicle. Similarly, the area specifying unit 64 specifies the area Q in the range excluding the lower part or the upper part of the image. Also good. Further, the area specifying unit 64 may similarly specify the area Q based on the road surface state determined by the condition determining unit 63.

  As shown in FIG. 4, in the case of a general road that is congested, the road marking is not easily recognized because it is hidden behind other vehicles. In this case, the area specifying unit 64 specifies the area Q3 in the upper half of the image in which the feature existing above the road is considered to be located. The region Q3 may be specified in the range excluding the lower portion or the upper portion of the image. In the example shown in FIG. 4, the area Q3 includes a traffic light M4, a road sign M5, and a telephone pole M6. As described above, the region specifying unit 64 sets the region Q in a range where a feature that is expected to be recognizable is located.

  In addition, when the traveling state is high-speed traveling at a speed equal to or higher than a predetermined threshold, it is difficult to recognize the feature located on the side, so that the region specifying unit 64 is in a range excluding both ends in the horizontal direction of the image. The region Q may be specified. In addition, the region specifying unit 64 may specify the region Q in a range excluding the left portion of the image when there are a plurality of lanes (for example, three lanes) and the vehicle is driving in the right lane. Similarly, the area specifying unit 64 may specify the area Q in the range excluding the right part of the image when traveling in the left lane, and specify the area Q in the range excluding both ends in the horizontal direction of the image when driving in the center lane. You may make it do.

  Alternatively, the area specifying unit 64 may select an image range to be specified as the area Q in an integrated manner from the combination of the traveling conditions determined by the condition determining unit 63. As described above, the region specifying unit 64 has features that can be recognized effectively even in environments where there are almost no intersections such as expressways or in environments where features that are difficult to recognize are expected. The area Q to be obtained can be specified, and the processing load on the feature recognition unit 65 can be reduced.

  The relative position calculation unit 66 calculates the relative position of the vehicle with respect to the position of the feature recognized by the feature recognition unit 65. The relative position calculation unit 66 specifies the position of the feature in the map information 51 associated with the feature recognized by the feature recognition unit 65. Further, the relative position calculation unit 66 calculates the relative position of the vehicle in the map information 51 from the direction of the feature calculated by the feature recognition unit 65 and the distance to the feature.

  The position correction unit 67 corrects the current position of the vehicle estimated by the vehicle position estimation unit 61 based on the relative position of the vehicle calculated by the relative position calculation unit 66. That is, the position correction unit 67 determines the current position of the vehicle using the relative position of the vehicle calculated by the relative position calculation unit 66 as the current position of the vehicle estimated by the vehicle position estimation unit 61.

  The display 7 presents various information to the vehicle occupant by displaying images and characters in accordance with control by the processing unit 6. The display 7 includes a display device such as a liquid crystal display. The display 7 displays a map including roads and facilities recorded in the map information 51 according to control by the processing unit 6, and displays the current position of the vehicle corrected by the position correction unit 67 on the displayed map. Can be displayed.

  For example, the input I / F 8 accepts an occupant's operation and outputs a signal corresponding to the operation to the processing unit 6. The display 7 and the input I / F 8 may be integrally configured as a touch panel display. The speaker 9 presents various kinds of information to the vehicle occupant by playing back sound according to control by the processing unit 6.

(Vehicle position determination method)
With reference to the flowchart of FIG. 15, an example of operation | movement of the vehicle position determination apparatus which concerns on this Embodiment is demonstrated.

  In step S <b> 1, the vehicle position estimation unit 61 corrects the current position measured by the positioning device 2 based on the vehicle momentum calculated from the detection result of the sensor group 4. Estimate the current position.

  In step S <b> 2, the condition determination unit 63 determines the surrounding environment of at least one of weather, time zone, road surface condition, and travel location. In step S <b> 3, the condition determination unit 63 determines a traveling state including at least one of a traveling state and a traveling lane. In step S4, the condition determination unit 63 determines the travel route of the vehicle.

  In step S5, the area specifying unit 64 specifies the area Q in the image photographed by the camera 1 as a range where the recognition target exists based on the running conditions determined in steps S2 to S4.

  In step S <b> 6, the feature recognition unit 65 recognizes from the image in the region Q specified by the region specifying unit 64. Further, the feature recognition unit 65 calculates the direction of the recognized feature with respect to the vehicle and the distance from the vehicle based on the image taken by the camera 1.

  In step S <b> 7, the relative position calculation unit 66 calculates the relative position of the vehicle with respect to the position in the map information 51 of the feature recognized by the feature recognition unit 65. In step S <b> 8, the position correction unit 67 corrects the current position of the vehicle estimated by the vehicle position estimation unit 61 based on the relative position of the vehicle calculated by the relative position calculation unit 66.

  As described above, according to the vehicle position determination device according to the present embodiment, the vehicle is identified by identifying the region Q in the image based on the traveling condition of the vehicle and recognizing various features existing in the region Q. Correct the current position of. As a result, the vehicle position determination apparatus according to the present embodiment increases the frequency of correction, and can maintain high accuracy of the current position of the vehicle in various traveling environments. In addition, the vehicle position determination device according to the present embodiment can efficiently limit the range of the image to be subjected to image processing by specifying the region Q, thereby reducing the processing load for recognition. Can do.

  In addition, according to the vehicle position determination device according to the present embodiment, by identifying the region Q suitable for the surrounding environment of the vehicle, such as weather, time zone, road surface condition, traveling location, etc., the features around the road can be efficiently used. The vehicle's current position can be corrected with high accuracy. Moreover, since the vehicle position determination apparatus according to the present embodiment can efficiently specify the region Q including the feature that is expected to be recognizable according to the surrounding environment, the processing load for recognition is reduced. can do.

  In addition, according to the vehicle position determination device according to the present embodiment, it is possible to efficiently recognize the features around the road by specifying the region Q suitable for the traveling state of the vehicle such as the traveling state and the traveling lane. The current position of the vehicle can be corrected with high accuracy. Moreover, since the vehicle position determination apparatus according to the present embodiment can efficiently specify the region Q including the features that are expected to be recognizable according to the traveling state, the processing load for recognition is reduced. can do.

(Other embodiments)
As described above, the present invention has been described according to the above-described embodiments. However, it should not be understood that the description and drawings constituting a part of this disclosure limit the present invention. From this disclosure, various alternative embodiments, examples and operational techniques will be apparent to those skilled in the art.

  For example, in the above-described embodiment, the feature recognition unit 65 may selectively recognize one feature that is most easily recognized from the region Q. In the example shown in FIG. 2, the vehicle is traveling at high speed, so that larger features are more easily recognized. In this case, the feature recognizing unit 65 can select the largest guide sign M2 as a feature to be recognized. Therefore, the feature recognition unit 65 can further reduce the processing load by reducing the number of recognition objects.

  Similarly, in the example shown in FIG. 4, the traffic light M4 that has a characteristic shape and emits light is easily recognized. In this case, the feature recognizing unit 65 selects the traffic signal M4 that is most easily recognized as a feature to be recognized. In addition, since it is difficult for the feature recognition unit 65 to recognize features other than the feature that emits light when the time zone is nighttime, an emergency call that emits light, a guidance sign, A traffic light or the like may be selected. As described above, the feature recognition unit 65 further restricts the features to be recognized from the region Q based on the position, size, type, traveling condition, and the like of the features considered to exist in the region Q. It may be.

  Further, in the above-described embodiment, the camera 1 changes the shooting direction and magnification according to the control by the processing unit 6 in order to capture the feature in the specified area Q. May be. Further, the region specifying unit 64 may specify one feature as one region Q, such as when the size of the feature to be recognized is larger than a predetermined threshold.

  In addition to the above, of course, the present invention includes various embodiments and the like that are not described here, such as configurations in which the above-described configurations are mutually applied. Therefore, the technical scope of the present invention is defined only by the invention specifying matters according to the scope of claims reasonable from the above description.

Q, Q1 to Q3 Region 1 Camera 5 Storage device 51 Map information 61 Vehicle position estimation unit 64 Region identification unit 65 Feature recognition unit 66 Relative position calculation unit 67 Position correction unit

Claims (5)

A storage device for storing map information including the positions of features existing around the road;
A vehicle position estimation unit for estimating a current position of the vehicle in the map information;
A camera that captures an image of the surroundings of the vehicle;
In the image, an area specifying unit that specifies an area where a feature to be recognized exists based on a predetermined condition;
A feature recognition unit that recognizes a feature from the region specified by the region specification unit;
A relative position calculation unit that calculates a relative position of the vehicle with respect to a position in the map information of the feature recognized by the feature recognition unit;
A vehicle position determination device comprising: a position correction unit that corrects a current position of the vehicle based on the relative position calculated by the relative position calculation unit.   The vehicle position determination apparatus according to claim 1, wherein the area specifying unit specifies the area based on a surrounding environment of the vehicle.   The vehicle area determination device according to claim 1, wherein the area specifying unit specifies the area based on a running state of the vehicle.   The feature recognition unit selectively recognizes a feature from the region based on at least one of a position, a size, and a type of the feature that is considered to exist in the region. The vehicle position determination apparatus of any one of 1-3. Estimating the current position of the vehicle in map information including the position of features present around the road;
Taking an image of the surroundings of the vehicle;
In the map information, specifying a region where a feature to be recognized exists based on a predetermined condition;
Recognizing features in the identified region from the image;
Calculating a relative position of the vehicle with respect to a position of the recognized feature in the map information;
Correcting the current position of the vehicle based on the calculated relative position. A vehicle position determination method comprising:

Images