Map generation device

Abstract

The present application provides a kind of map generation device (50), with: vehicle-mounted detector (1a), it detects the situation around the vehicle (101) in running;Feature point extraction unit (141), it extracts feature point from the detection data obtained by vehicle-mounted detector (1a);Generation unit (171), it generates map using the feature point extracted by feature point extraction unit (141) in the vehicle (101) running;Position recognition unit (131), when the vehicle (101) is in the region corresponding to the map generated by generation unit (171) running, the feature point extracted by feature point extraction unit (141) and map are compared, so as to recognize the position of the vehicle on the map;And delete unit (172), it deletes the feature point in the map from the map, the performance of the feature point in the map is lower than the feature point of the prescribed degree in the comparison of position recognition unit (131).

Description

TECHNICAL FIELD

[0001] The present application relates to a map generation device that generates a map for acquiring a position of a host vehicle. BACKGROUND

[0002] As such a device, a device has been known that extracts a feature point from an image captured by an in-vehicle camera, generates a map including the feature point, and acquires a position of the host vehicle based on an amount of displacement of the feature point obtained by tracking the extracted feature point (see, for example, Patent Literature 1). In the device described in Patent Literature 1, when a feature point under tracking cannot be extracted from a current captured image, the feature point is excluded from the tracking target.

[0003] However, among the feature points included in the map, there are sometimes unnecessary feature points that do not match the current road environment due to a change in road structure or the like. Therefore, as described in Patent Literature 1 above, merely excluding the feature points that cannot be tracked from the tracking target can not be sufficient to reduce the unnecessary feature points.

[0004] PRIOR ART DOCUMENTS

[0005] PATENT LITERATURE

[0006] Patent Literature 1: International Publication No. 2015 / 049717 (WO 2015 / 049717 Al). SUMMARY

[0007] The map generation device of the technical solution of the present application has: an in-vehicle detector that detects a situation around a host vehicle under travel; a feature point extraction section that extracts a feature point from detection data acquired by the in-vehicle detector; a generation section that generates a map using the feature point extracted by the feature point extraction section while the host vehicle is under travel; a position recognition section that, when the host vehicle is under travel in a region corresponding to the map generated by the generation section, collates the feature point extracted by the feature point extraction section with the map, thereby recognizing a position of the host vehicle on the map; and a deletion section that deletes, from the map, a feature point among the feature points included in the map, for which the collation by the position recognition section is below a prescribed level. BRIEF DESCRIPTION OF DRAWINGS

[0008] The objects, features, and advantages of the present application will be further clarified by the following description of embodiments with reference to the attached drawings.

[0009] Figure 1 is a block diagram that schematically shows an overall structure of a vehicle control system according to an embodiment of the present application.

[0010] Figure 2 is a block diagram that shows a main part structure of a map generation device according to an embodiment of the present application. is a block diagram that schematically shows an overall structure of a vehicle control system according to an embodiment of the present application.

[0011] Figure 3A This shows the environment generated by this vehicle. Figure 1 A diagram showing the state of a vehicle traveling on a road.

[0012] Figure 3B This indicates that the vehicle is in autonomous driving mode. Figure 3A A diagram showing the driving status at the location.

[0013] Figure 4 It is shown by Figure 2 A flowchart illustrating an example of the processing performed by the CPU of the controller. Detailed Implementation

[0014] The following is for reference Figures 1-4 Embodiments of the present invention will be described. The map generation apparatus of the present invention can be applied to vehicles with autonomous driving capabilities, i.e., autonomous vehicles. It should be noted that, sometimes, the vehicle using the map generation apparatus of this embodiment is referred to as "this vehicle" to distinguish it from other vehicles. This vehicle can be any of the following: an engine vehicle with an internal combustion engine as the driving source, an electric vehicle with a drive motor as the driving source, or a hybrid vehicle with both an engine and a drive motor as driving sources. This vehicle can operate not only in an autonomous driving mode that does not require driver operation, but also in a manual driving mode that requires driver operation.

[0015] First, let's explain the general structure related to autonomous driving. Figure 1 This is a block diagram schematically illustrating the overall structure of a vehicle control system 100 having a map generation apparatus according to an embodiment of the present invention. Figure 1 As shown, the vehicle control system 100 mainly includes a controller 10 and external sensor group 1, internal sensor group 2, input / output device 3, positioning unit 4, map database 5, navigation device 6, communication unit 7, and driving actuator AC, which are communicatively connected to the controller 10.

[0016] External sensor group 1 is a collective term for multiple sensors (external sensors) that detect information about the vehicle's surroundings, i.e., external conditions. For example, external sensor group 1 includes: a lidar that measures the distance from the vehicle to surrounding obstacles by measuring the scattered light relative to the omnidirectional illumination light of the vehicle; a radar that detects other vehicles or obstacles around the vehicle by irradiating electromagnetic waves and detecting the reflected waves; and a camera mounted on the vehicle and equipped with imaging elements such as CCD and CMOS to capture images of the vehicle's surroundings (front, rear, and sides).

[0017] The internal sensor group 2 is a collective term for a plurality of sensors (internal sensors) that detect the running state of the host vehicle. For example, the internal sensor group 2 includes a vehicle speed sensor that detects the vehicle speed of the host vehicle, an acceleration sensor that detects the acceleration in the front-rear direction and the acceleration in the left-right direction (lateral acceleration) of the host vehicle, respectively, a rotation speed sensor that detects the rotation speed of the running drive source, a yaw rate sensor that detects the rotation angular speed of the center of gravity of the host vehicle about the vertical axis, and the like. Sensors that detect the driver's driving operation in the manual driving mode, such as the operation of the accelerator pedal, the operation of the brake pedal, the operation of the steering wheel, and the like, are also included in the internal sensor group 2.

[0018] The input-output device 3 is a collective term for devices that input instructions from the driver and output information to the driver. For example, the input-output device 3 includes various switches through the operation of which the driver inputs various instructions, a microphone through which the driver inputs instructions by voice, a display that provides information to the driver by displaying an image, a speaker that provides information to the driver by voice, and the like.

[0019] The positioning unit (GNSS unit) 4 has a positioning sensor that receives positioning signals transmitted from positioning satellites. The positioning satellites are GPS satellites, quasi-zenith satellites, or the like. The positioning unit 4 measures the current position (latitude, longitude, altitude) of the host vehicle using the positioning information received by the positioning sensor.

[0020] The map database 5 is a device that stores general map information used in the navigation device 6, and is configured by, for example, a hard disk, a semiconductor element. The map information includes the position information of roads, the information of road shapes (curvatures and the like), the position information of intersections and branch points, the information of limit speeds set on roads. Note that the map information stored in the map database 5 is different from the high-precision map information stored in the storage section 12 of the controller 10.

[0021] The navigation device 6 is a device that searches for a target route on a road to a destination input by the driver and performs guidance along the target route. The input of the destination and the guidance along the target route are performed by the input-output device 3. The target route is calculated on the basis of the current position of the host vehicle measured by the positioning unit 4 and the map information stored in the map database 5. It is also possible to measure the current position of the host vehicle using the detection values of the external sensor group 1, and to calculate the target route on the basis of the current position and the high-precision map information stored in the storage section 12.

[0022] The communication unit 7 communicates with various servers not shown using a network including a wireless communication network typified by the Internet, a mobile phone network, and the like, and acquires map information, travel record information, traffic information, and the like from the servers periodically or at an arbitrary timing. The network includes not only a public wireless communication network but also a closed communication network such as a wireless LAN, Wi-Fi (registered trademark), Bluetooth (registered trademark), and the like, which is provided for each prescribed management area. The acquired map information is output to the map database 5, the storage 12, and the map information is updated.

[0023] The actuator AC is a travel actuator for controlling travel of the host vehicle. In the case where the travel drive source is an engine, the actuator AC includes a throttle actuator that adjusts an opening degree of a throttle valve (throttle opening degree) of the engine. In the case where the travel drive source is a travel motor, the actuator AC includes the travel motor. A brake actuator that operates a brake device of the host vehicle and a steering actuator that drives a steering device are also included in the actuator AC.

[0024] The controller 10 is constituted by an electronic control unit (ECU). More specifically, the controller 10 is constituted by a computer including an arithmetic unit 11 such as a CPU (microprocessor), a storage 12 such as a ROM (read only memory), a RAM (random access memory), and other peripheral circuits such as an I / O (input / output) interface, and the like not shown. Note that a plurality of ECUs having different functions such as an engine control ECU, a travel motor control ECU, a brake device ECU, and the like can be provided separately, but for the sake of convenience, the controller 10 is shown as a collection of these ECUs in the following description. Figure 1

[0025] ​High-precision detailed map information (referred to as high-precision map information) is stored in the storage section 12. The high-precision map information includes: position information of a road, information of a road shape (curvature, etc.), information of a slope of a road, position information of an intersection, a branch, information of a number of lanes, a width of a lane, and position information of each lane (a central position of a lane, information of a boundary line of a lane position), position information of a landmark (a signal, a sign, a building, etc.) as a marker on a map, information of a road surface profile such as a concave-convex of a road surface, etc. The high-precision map information stored in the storage section 12 includes: map information acquired from outside of the host vehicle by the communication unit 7, for example, information of a map (referred to as a cloud map) acquired by a cloud server, and information of a map (referred to as an environmental map) composed of point cloud data generated by mapping using a technology such as SLAM (Simultaneous Localization and Mapping) and the like, which is made by the host vehicle itself using detection values of the external sensor group 1. Various control programs, information on threshold values used in the programs, and the like are also stored in the storage section 12.

[0026] The arithmetic section 11 has a host vehicle position recognition section 13, an outside recognition section 14, a behavior plan generation section 15, a travel control section 16, and a map generation section 17 as functional structures.

[0027] The host vehicle position recognition section 13 recognizes a position of the host vehicle on a map (host vehicle position) based on position information of the host vehicle acquired by the positioning unit 4 and map information of the map database 5. The host vehicle position can also be recognized using map information stored in the storage section 12 and surrounding information of the host vehicle detected by the external sensor group 1, whereby the host vehicle position can be recognized with high precision. Note that when the host vehicle position can be measured with a sensor provided outside on or beside a road, the host vehicle position can also be recognized by communicating with the sensor via the communication unit 7.

[0028] The outside recognition section 14 recognizes an outside situation around the host vehicle based on signals from the external sensor group 1 such as a laser radar, a radar, a camera, etc. For example, the outside recognition section 14 recognizes a position, a travel speed, an acceleration of a surrounding vehicle (a front vehicle, a rear vehicle) traveling in the vicinity of the host vehicle, a position of a surrounding vehicle parked or standing in the vicinity of the host vehicle, and a position, a state of another object, etc. The another object includes: a sign, a signal, a marker (road surface marker) such as a division line or a stop line of a road, a building, a guardrail, a utility pole, a billboard, a pedestrian, a bicycle, etc. The state of the another object includes: a color (red, green, yellow) of a signal, a moving speed, an orientation of a pedestrian, a bicycle, etc. A part of the stationary object among the another object constitutes a landmark as a marker on a map, and the outside recognition section 14 also recognizes a position and a category of the landmark.

[0029] The action plan generating section 15 generates a travel trajectory (target trajectory) of the host vehicle from the current time point to a prescribed time based on, for example, the target path calculated by the navigation device 6, the host vehicle position recognized by the host vehicle position recognizing section 13, and the external situation recognized by the external recognizing section 14. When there are a plurality of trajectories as candidates for the target trajectory on the target path, the action plan generating section 15 selects the best trajectory that complies with the law and satisfies the criteria for efficient and safe travel, etc. from among them, and sets the selected trajectory as the target trajectory. Then, the action plan generating section 15 generates an action plan corresponding to the generated target trajectory. The action plan generating section 15 generates various action plans corresponding to the travel modes such as the overtaking travel for overtaking a preceding vehicle, the lane changing travel for changing the travel lane, the following travel for following a preceding vehicle, the lane keeping travel for keeping the lane without deviating from the travel lane, the deceleration travel, or the acceleration travel, etc. The action plan generating section 15 first determines the travel mode when generating the target trajectory, and generates the target trajectory based on the travel mode.

[0030] In the automatic driving mode, the travel control section 16 controls each actuator AC so that the host vehicle travels along the target trajectory generated by the action plan generating section 15. More specifically, the travel control section 16 calculates the required driving force for obtaining the target acceleration per unit time calculated by the action plan generating section 15, taking into account the travel resistance determined by the road slope, etc. in the automatic driving mode. Then, for example, feedback control is performed on the actuator AC so that the actual acceleration detected by the internal sensor group 2 becomes the target acceleration. That is, the actuator AC is controlled so that the host vehicle travels at the target vehicle speed and the target acceleration. Note that in the manual driving mode, the travel control section 16 controls each actuator AC in accordance with the travel instruction (steering operation, etc.) from the driver taken by the internal sensor group 2.

[0031] The map generation section 17 generates an environmental map composed of three-dimensional point cloud data using the detection values detected by the external sensor group 1 while traveling in the manual driving mode. Specifically, from the captured image taken by the camera, an edge showing the outline of an object is extracted based on the information of the brightness and color of each pixel, and a feature point is extracted using the edge information. The feature point is, for example, an intersection of edges, and corresponds to a corner of a building, a corner of a road sign, and the like. The map generation section 17 sequentially plots the extracted feature points on the environmental map, thereby generating an environmental map of the surroundings of the road on which the host vehicle is traveling. Instead of the camera, data taken by a radar, a lidar, or the like can be used to extract feature points of objects around the host vehicle, and an environmental map can be generated. Also, the map generation section 17 determines whether a landmark such as a signal, a sign, a building, or the like as a marker on the map is included in the captured image taken by the camera by, for example, a template matching process when generating the environmental map. Then, when it is determined that the landmark is included, the position and the category of the landmark on the environmental map are recognized based on the captured image. This landmark information is included in the environmental map and stored in the storage section 12.

[0032] The host vehicle position recognition section 13 performs a position estimation process of the host vehicle in parallel with the map creation process of the map generation section 17. That is, based on the change in the position of the feature point over time, the position of the host vehicle is estimated and acquired. Also, the host vehicle position recognition section 13 estimates and acquires the host vehicle position based on the relative positional relationship with the landmarks around the host vehicle. The map creation process and the position estimation process are performed simultaneously in accordance with, for example, a SLAM algorithm. The map generation section 17 can generate an environmental map in the same manner not only when traveling in the manual driving mode but also when traveling in the automatic driving mode. In a case where an environmental map has already been generated and stored in the storage section 12, the map generation section 17 can also update the environmental map according to newly obtained feature points.

[0033] However, the environmental map composed of point cloud data has a large amount of data, and when creating an environmental map corresponding to a wide area, the capacity of the storage section 12 can be significantly occupied. Therefore, in the present embodiment, the map generation device is configured as follows in order to reduce the amount of data of the environmental map.

[0034] Figure 2 is a block diagram showing the main part structure of a map generation device 50 according to an embodiment of the present application. The map generation device 50 is a device that generates a point cloud map (an environmental map) based on feature points extracted from a captured image of a camera la, and constitutes a part of a vehicle control system 100 of Figure 1 As shown in Figure 2 , the map generation device 50 has a controller 10, the camera la, a radar lb, and a lidar lc.

[0035] Camera 1a is a single-lens camera with imaging elements (image sensors) such as CCD (charge-coupled device) and CMOS (complementary metal-oxide-semiconductor), constituting... Figure 1 The vehicle is part of an external sensor group 1. Camera 1a can also be a stereo camera. Camera 1a captures images of the area around the vehicle. Camera 1a is mounted at a predetermined position, for example, at the front of the vehicle, and continuously captures images of the space in front of the vehicle to obtain image data of objects (hereinafter referred to as captured image data or simply captured images). Camera 1a outputs the captured images to controller 10. Radar 1b is mounted on the vehicle and detects other vehicles, obstacles, etc., around the vehicle by irradiating electromagnetic waves and detecting reflected waves. Radar 1b outputs the detection values ​​(detection data) to controller 10. LiDAR 1c is mounted on the vehicle and measures the scattered light from the vehicle relative to the omnidirectional illumination light to detect the distance from the vehicle to surrounding obstacles. LiDAR 1c outputs the detection values ​​(detection data) to controller 10.

[0036] The controller 10 includes a position recognition unit 131, a counting unit 132, a path information generation unit 133, a feature point extraction unit 141, a generation unit 171, and a deletion unit 172, and serves as an arithmetic unit 11. Figure 1 The functional structure undertaken by the feature point extraction unit 141, for example, is... Figure 1 It consists of an external identification unit 14.

[0037] The feature point extraction unit 141 extracts feature points from the captured image obtained by the camera 1a. The generation unit 171 uses the feature points extracted by the feature point extraction unit 141 to generate an environment map while the vehicle 101 is in motion.

[0038] When the vehicle 101 travels in an area where the environment map is generated by the generation unit 171, the position recognition unit 131 compares the feature points extracted by the feature point extraction unit 141 with the environment map to identify the position of the vehicle 101 on the environment map. The counting unit 132 counts the number of times each feature point in the environment map is compared by the position recognition unit 131 (hereinafter referred to as the comparison count). The deletion unit 172 deletes feature points from the environment map whose comparison count is below a predetermined level. Based on the position of the vehicle 101 obtained by the position recognition unit 131 during the travel of the vehicle 101, the path information generation unit 133 generates information showing the path traveled by the vehicle 101 on the environment map (hereinafter referred to as path information) and stores it in the storage unit 12. When the vehicle 101 travels multiple times in the same area, information showing the area and the number of times it travels is recorded in the path information.

[0039] Here, the counting unit 132 and the deletion unit 172 will be explained. Figure 3A This shows the environment generated by the vehicle 101. Figure 1Fig. 1 is a diagram showing a state in which the host vehicle 101 is traveling on a one-lane road RD on which vehicles travel on the left side toward the intersection IS. In Figure 3A the example shown, the host vehicle 101 is traveling on a one-lane road RD on which vehicles travel on the left side toward the intersection IS. In Figure 3A the example shown, the host vehicle 101 is traveling on a one-lane road RD on which vehicles travel on the left side toward the intersection IS. In Figure 3A the example shown, the host vehicle 101 is traveling on a one-lane road RD on which vehicles travel on the left side toward the intersection IS. In Figure 3A the example shown, the host vehicle 101 is traveling on a one-lane road RD on which vehicles travel on the left side toward the intersection IS. In Figure 3B the example shown, the host vehicle 101 is traveling on a one-lane road RD on which vehicles travel on the left side toward the intersection IS. In

[0040] Figure 3B Fig. 1 is a diagram showing a state in which the host vehicle 101 is traveling on a one-lane road RD on which vehicles travel on the left side toward the intersection IS. In Figure 3A the example shown, the host vehicle 101 is traveling on a one-lane road RD on which vehicles travel on the left side toward the intersection IS. In Figure 3B the example shown, the host vehicle 101 is traveling on a one-lane road RD on which vehicles travel on the left side toward the intersection IS. In Figure 3A the example shown, the host vehicle 101 is traveling on a one-lane road RD on which vehicles travel on the left side toward the intersection IS. In Figure 3A the example shown, the host vehicle 101 is traveling on a one-lane road RD on which vehicles travel on the left side toward the intersection IS. In

[0041] Therefore, when the comparison of the feature points performed by the position recognition section 131 is performed, the counting section 132 recognizes the feature points in the environmental map that are compared with the feature points extracted by the feature point extraction section 141. Then, the counting section 132 increments the number of times of comparison of the recognized feature points by one. The counting section 132 updates the information (counting number information) showing the number of times of comparison of each feature point constituting the environmental map stored in the storage section 12 each time the comparison of the feature points performed by the position recognition section 131 is performed.

[0042] The deletion section 172 determines a region (hereinafter referred to as an object region) from which the feature points are to be deleted from the environmental map. The deletion section 172 deletes the feature points included in the object region, the comparison performance of which by the position recognition section 131 is lower than a prescribed degree. Specifically, the deletion section 172 calculates the average of the number of times of comparison of each feature point included in the object region based on the counting number information stored in the storage section 12, and judges that the feature points whose number of times of comparison is lower than the average by a prescribed degree or more are unnecessary data and deletes them. The object region is a region including a path on which the host vehicle 101 has traveled a prescribed number of times or more, and is a region of a prescribed size. The deletion section 172 determines the object region based on the path information stored in the storage section 12. Specifically, the deletion section 172 extracts a path on which the host vehicle 101 has traveled a prescribed number of times or more from the environmental map, judges the size of the region including the extracted path, and determines the region as the object region when the size is a prescribed size or more. Note that the object region can also be determined based on an instruction by the user through the input and output device 3. That is, the object region can also be determined by the user. In this way, the amount of data of the environmental map can be reduced. Further, by thus deleting the unnecessary data, the comparison error of the feature points can be suppressed.

[0043] Figure 4 is an example of the processing performed by the controller 10 according to a predetermined program, and specifically, an example of the flow of the processing related to the map generation. Figure 2 The processing shown in the flowchart of FIG. 10 is started, for example, when the controller 10 is powered on, and is repeatedly performed at a prescribed period. Figure 4 The processing shown in the flowchart of FIG. 10 is started, for example, when the controller 10 is powered on, and is repeatedly performed at a prescribed period.

[0044] First, in Sll (S: processing step), a region from which the feature points are to be deleted, that is, an object region is determined. In S12, the number of times of comparison of the feature points in the object region determined in Sll is acquired. In S13, the average of the number of times of comparison of each feature point acquired in S12 is calculated. In S14, it is judged whether there is a feature point in the object region whose number of times of comparison is lower than the average by a prescribed degree or more.

[0045] When S14 is negative (S14: No), the processing ends. When S14 is affirmative (S14: Yes), in S15, the deletion processing of the feature points is performed. Specifically, among the feature points within the object region, the feature points whose number of collations is lower than the average value of the number of collations calculated in S13 by a prescribed degree or more are deleted. For example, the feature points whose number of collations is lower than the average value by a prescribed percentage (x%) or less are deleted. The value of x can be fixed, can be determined based on the accuracy required for the environmental map, or can be determined based on the number of times of travel shown by the route information, specifically, based on the number of times of travel of the host vehicle 101 in the object region recorded in the route information. When the deletion processing of the feature points is performed, the route information is reset. More specifically, the number of times of travel of the route within the object region is reset to 0. Also, the number of times of counting information is reset. More specifically, the number of collations of the feature points within the object region is reset to 0.

[0046] The embodiment of the present application can achieve the following effects.

[0047] (1) The map generation device 50 includes the camera 1a that detects the situation around the host vehicle 101 that is traveling, the feature point extraction section 141 that extracts the feature points from the detection data acquired by the camera 1a, the generation section 171 that generates a map (environmental map) using the feature points extracted by the feature point extraction section 141 while the host vehicle 101 is traveling, the position recognition section 131 that collates the feature points extracted by the feature point extraction section 141 with the environmental map to thereby recognize the position of the host vehicle 101 on the environmental map when the host vehicle 101 is traveling in the region corresponding to the environmental map generated by the generation section 171, and the deletion section 172 that deletes, from the environmental map, the feature points included in the environmental map whose performance of collation by the position recognition section 131 is lower than a prescribed degree. Thus, it is possible to delete unnecessary feature points that cannot be referred to when recognizing the position of the host vehicle from the map, and it is possible to reduce the data amount of the map.

[0048] (2) The deletion section 172 deletes, from the environmental map, the feature points extracted by the feature point extraction section 141 while the host vehicle 101 is traveling in the object range including the route in which the host vehicle 101 has traveled a prescribed number of times or more and the route of a prescribed length or more, among the feature points, whose performance of collation by the position recognition section 131 is lower than a prescribed degree. Thus, it is possible to delete unnecessary data from the map without degrading the accuracy of the map.

[0049] (3) The counting unit 132 counts the number of times each feature point included in the environmental map is compared by the location identification unit 131, and uses this count as the comparison result of the location identification unit 131. The deletion unit 172 calculates the average number of comparisons of each feature point counted by the counting unit 132, and deletes feature points from the environmental map whose comparison count is lower than a predetermined level of the average. As a result, it is possible to more accurately determine whether each feature point included in the target area is unwanted data.

[0050] The above-described embodiments can be modified in various ways. Several modifications are described below. In the above-described embodiments, the camera 1a detects the situation around the vehicle 101, but as long as the situation around the vehicle 101 is detected, the configuration of the vehicle-mounted detector can be any form. For example, the vehicle-mounted detector can also be radar 1b or lidar 1c. Furthermore, in the above-described embodiments, the process begins when the controller 10 is powered on. Figure 4 The process shown can be performed, but it can also be executed according to instructions from the user. Figure 4 The processing shown.

[0051] Furthermore, in the above embodiment, path information showing the path traveled by the vehicle 101 on the environment map is generated based on the position of the vehicle 101 obtained by the position recognition unit 131 during the vehicle's travel. However, the configuration of the path information generation unit is not limited to this. The path information generation unit may also generate path information based on map information stored in the map database 5 and the vehicle position identified by the vehicle position recognition unit 13. Furthermore, in the above embodiment, feature points in the target area that have a comparison count lower than a predetermined level are deleted. However, the configuration of the deletion unit is not limited to this. The deletion unit may also delete feature points based on other criteria. For example, feature points in the target area that have a comparison count lower than a predetermined value may also be deleted. Furthermore, in the above embodiment, the size of the area including the path traveled by the vehicle 101 more than a predetermined number of times is determined, and the area is determined as the target area when the size is a predetermined size or higher. However, the target area may also be determined based on whether the area of ​​the area is a predetermined size or higher, or based on whether the length of the path included in the area that the vehicle 101 has traveled more than a predetermined number of times is a predetermined value or higher. Furthermore, in the above embodiment, the map generation device 50 is applied to an autonomous vehicle, but the map generation device 50 can also be applied to vehicles other than autonomous vehicles. It can also be applied to manually driven vehicles equipped with, for example, ADAS (Advanced Driver-Assistance Systems).

[0052] The present application can also be used as a map generation method, including: a step of extracting feature points from detection data acquired by the camera 1a that detects the situation around the vehicle 101 being driven; a step of generating a map (environment map) using the extracted feature points while the vehicle 101 is being driven; a step of, when the vehicle 101 is driven in a region corresponding to the environment map, collating the extracted feature points and the environment map to thereby recognize the position of the vehicle 101 on the environment map; and a step of deleting, from the environment map, feature points included in the environment map that have a collation performance lower than a prescribed level.

[0053] One or more of the above-described embodiments and modified examples can be arbitrarily combined, or each modified example can be combined with another modified example.

[0054] The present application can reduce the data amount of a map.

[0055] The present application has been described above in connection with a preferred embodiment thereof, but it will be understood by those skilled in the art that various modifications and changes can be made without departing from the scope of the disclosure as set forth in the following claims.

Claims

1. A map generation device, characterized in that, have: The vehicle-mounted detector (1a) detects the conditions around the vehicle (101) while it is in motion; The feature point extraction unit (141) extracts feature points from the detection data obtained by the vehicle-mounted detector (1a); The generation unit (171) generates a map using the feature points extracted by the feature point extraction unit (141) while the vehicle is in motion. The location recognition unit (131) compares the feature points extracted by the feature point extraction unit (141) with the map when the vehicle (101) is driving in an area corresponding to the map generated by the generation unit (171), thereby identifying the location of the vehicle on the map. The deletion unit (172) deletes feature points from the map whose performance in comparison by the location identification unit (131) is below a predetermined level; and The counting unit (132) counts the number of comparisons made by the location identification unit (131) against each feature point contained in the map, and uses this count as the comparison result. The deletion unit (172) calculates the average number of comparisons of each feature point obtained by the counting unit (132), and deletes feature points from the map whose number of comparisons is less than a specified level of the average value.

2. The map generation apparatus according to claim 1, characterized in that, The deletion unit (172) takes the area including the path that the vehicle (101) has traveled more than a certain number of times and the path that is more than a certain length as the target range, and deletes from the map the feature points extracted by the feature point extraction unit (141) during the vehicle (101)'s travel within the target range the feature points whose actual performance compared by the location recognition unit (131) is lower than a certain level.

3. The map generation apparatus according to claim 2, characterized in that, The deletion unit (172) deletes feature points from the map whose ratio of the number of comparisons to the average value is below a predetermined ratio.

4. The map generation apparatus according to claim 3, characterized in that, The specified scale is determined based on the required accuracy of the map.

5. The map generation apparatus according to claim 3, characterized in that, The prescribed proportion is determined based on the number of times the vehicle has traveled within the specified range.

6. A map generation method, characterized in that, Includes the following steps: Feature points are extracted from the detection data obtained by the on-board detector (1a) that detects the conditions around the moving vehicle (101); The extracted feature points are used to generate a map while the vehicle (101) is in motion; When the vehicle (101) is driving in an area corresponding to the map, the extracted feature points are compared with the map to identify the position of the vehicle (101) on the map. as well as Deleting feature points from the map whose comparison results are below a predetermined level includes: counting the number of comparisons made to each feature point in the map as the comparison results, calculating the average number of comparisons for each feature point, and deleting feature points from the map whose comparison counts are below the predetermined level of the average.

Images