Map generating device and driving assistance system
Patent Information
- Application Number
- MYPI2025000565
- Authority / Receiving Office
- MY · MY
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2022-07-22
- Publication Date
- 2024-01-22
AI Technical Summary
Existing automatic driving support systems face high calculation loads when generating routes due to the need to determine collision possibilities with other vehicles, leading to increased processing requirements.
A map generation device that collects environmental monitoring information to identify dangerous areas and generates a map including geographic and dangerous area information, reducing the processing load by allowing vehicles to autonomously determine routes based on this data.
The solution reduces the calculation load for route generation by pre-identifying and transmitting dangerous areas, enabling vehicles to efficiently plan routes that avoid obstacles without the need for real-time collision prediction, thus improving processing efficiency.
Smart Images

Figure PI2025000565_DRAWING_001 
Figure PI2025000565_DRAWING_002 
Figure PI2025000565_DRAWING_003
Abstract
Description
Map generation device and driving assistance system
[0001] The technology disclosed herein relates to a map generating device and a driving assistance system.
[0002] For example, Patent Document 1 discloses an automated driving assistance system. In this automated driving assistance system, each vehicle has common road map data. Each vehicle calculates its relative position with respect to a reference point in the road map data and outputs the calculated position to other vehicles. Each vehicle compares the reference point in the road map data with the relative position of the other vehicle to determine the position of the other vehicle. Each vehicle automatically drives to avoid collision with the other vehicle based on the determined position of the other vehicle.
[0003] JP 2019-132988 A
[0004] In the above-mentioned system, the vehicles generate routes for autonomous driving. At that time, each vehicle determines the possibility of a collision with other vehicles based on relative position information with other vehicles. Therefore, the amount of processing, such as the calculation load, required for generating a route in each vehicle increases.
[0005] The technology disclosed herein has been made in view of the above points, and its purpose is to reduce the amount of processing, such as the calculation load, required when generating a route.
[0006] The map generating device of the present disclosure includes a collector that collects environmental monitoring information and a generator that generates a map for assisting autonomous driving of a mobile body based on the monitoring information, and the generator determines dangerous areas in the environment that should be avoided based on the monitoring information and generates the map that includes geographical information and the dangerous areas.
[0007] The driving assistance system of the present disclosure includes a mobile body that performs autonomous driving and a map generation device that generates a map to assist the autonomous driving of the mobile body, the map generation device having a collector that collects environmental monitoring information and a generator that generates the map based on the monitoring information, the generator determining dangerous areas that should be avoided in the environment based on the monitoring information, generating the map including geographical information and the dangerous areas, and transmitting the map to the mobile body, and the mobile body generating a route for autonomous driving based on the geographical information and the dangerous areas.
[0008] Here, the term "map" is not limited to an actual representation of a map, but also includes data format. In other words, "map" can be read as "map information" or "map data."
[0009] According to the map generating device, it is possible to reduce the amount of processing, such as the calculation load, when generating a route.
[0010] According to the driving assistance system, it is possible to reduce the amount of processing, such as the calculation load, when generating a route.
[0011] FIG. 1 is a schematic diagram of a driving assistance system. FIG. 2 is an explanatory diagram showing the configuration of a map. FIG. 3 is a diagram showing a schematic hardware configuration of a first monitoring device. FIG. 4 is a functional block diagram of a controller of the first monitoring device. FIG. 5 is a diagram showing a schematic hardware configuration of a second monitoring device. FIG. 6 is a functional block diagram of a controller of the second monitoring device. FIG. 7 is a diagram showing a schematic hardware configuration of a moving object. FIG. 8 is a functional block diagram of a controller of the moving object. FIG. 9 is a diagram showing a schematic hardware configuration of a map generating device. FIG. 10 is a functional block diagram of a controller of the map generating device. FIG. 11 is a schematic diagram showing an example of geographical information. FIG. 12 is a schematic diagram showing an example of first danger area information. FIG. 13 is a schematic diagram showing an example of second danger area information. FIG. 14 is a schematic diagram showing an example of obstacle information. FIG. 15 is a schematic diagram showing an example of disturbance information. FIG. 16 is a schematic diagram showing an example of search information. FIG. 17 is an explanatory diagram showing changes in the monitoring necessity level. FIG. 18 is a flowchart of monitoring processing. Fig. 19 is a flowchart of an automated driving system. Fig. 20 is a schematic diagram of a map in which first danger area information is superimposed on geographical information. Fig. 21 is a schematic diagram of a map in which first danger area information and second danger area information are superimposed on geographical information. Fig. 22 is a schematic diagram of a map in which obstacle behavior predicted based on obstacle information is superimposed on geographical information.
[0012] An exemplary embodiment will be described in detail below with reference to the drawings. FIG. 1 is a schematic diagram of a driving assistance system 1000. The driving assistance system 1000 includes a mobile object 1 that performs autonomous driving, a monitoring device 8 that acquires monitoring information of the environment, and a map generation device 100 that generates a map 2 to assist the autonomous driving. The driving assistance system 1000 may include multiple mobile objects 1. The map generation device 100 is communicably connected to the mobile object 1 and the monitoring device 8. The map generation device 100 transmits and receives various information between the mobile object 1 and the monitoring device 8. The map generation device 100 generates a map 2 to assist the driving of the mobile object 1 based on the monitoring information acquired by the monitoring device 8 and transmits the generated map 2 to the mobile object 1. The mobile object 1 generates a route based on the map 2 and performs autonomous driving. The map generation device 100 also generates a map 2 to assist the monitoring device 8 in acquiring the monitoring information and transmits the generated map 2 to the monitoring device 8. The monitoring device 8 acquires monitoring information based on the map 2 and transmits the acquired monitoring information to the map generation device 100. Communication between the map generation device 100 and the mobile object 1 and the monitoring device 8 is performed via a relay device 121. The map generation device 100 and the relay device 121 are connected via a communication network N such as the Internet. The relay device 121, the mobile object 1, and the monitoring device 8 communicate wirelessly.
[0013] -Map- First, we will explain the map generated by the map generating device 100. Figure 2 is an explanatory diagram showing the configuration of the map 2.
[0014] The map 2 includes geographical information 21 and a danger area 3. Specifically, the map 2 includes geographical information 21 and danger area information 22. The danger area information 22 is information about the danger area 3 that is included in the environment and that should be avoided. Furthermore, the map 2 may further include obstacle information 25 and disturbance information 26. The obstacle information 25 is information about the movement of an obstacle. The disturbance information 26 is information about a disturbance to the autonomous driving of the mobile body 1. The map 2 used to acquire monitoring information includes search information 27. The search information 27 is information about the degree of necessity of monitoring each position in the environment.
[0015] The geographical information 21 includes at least one of roads, paths, and features (buildings, trees, rocks, and other objects on land or sea, whether natural or artificial). The geographical information 21 is information related to general maps such as road maps, facility maps, and navigational maps. The geographical information 21 may be static information. The geographical information 21 may change due to changes in roads, demolition of buildings, or weathering of rocks, but is basically permanent information. The geographical information 21 shares a coordinate system with each of the danger area information 22, obstacle information 25, disturbance information 26, and search information 27. In other words, the positions in each of the danger area information 22, obstacle information 25, disturbance information 26, and search information 27 correspond to positions on the geographical information 21.
[0016] The danger area information 22 is information about the position and outer shape of a danger area 3 that is included in the environment and that should be avoided. The danger area information 22 may include at least one of first danger area information 23 and second danger area information 24.
[0017] The first danger area information 23 is information regarding the position and shape of a danger area 3 corresponding to a static obstacle among obstacles contained in the environment. The danger area 3 corresponding to a static obstacle is referred to as a "first danger area 31." A static obstacle is an obstacle that has little movement. Static obstacles may include structures, equipment, and features that exist temporarily or for a limited period of time. For example, static obstacles may include construction sites, accident sites, rough terrain, shallow water, fish farms, dark areas, fog, and localized heavy rain. In the first danger area information 23, since the obstacle in question is static, the position of the first danger area 31 is approximately the same as the location where the obstacle exists, and the shape of the first danger area 31 is approximately the same as the shape of the obstacle, encompassing the obstacle. In addition, static obstacles may also include areas that are legally or customarily impassable.
[0018] The second danger area information 24 is information regarding the position and shape of a danger area 3 corresponding to a dynamic obstacle among obstacles contained in the environment. The danger area 3 corresponding to a dynamic obstacle is referred to as a "second danger area 32." A dynamic obstacle is an obstacle that may move. For example, an object that is inherently moving but accidentally stops, such as a stopped car, is a dynamic obstacle. For example, dynamic obstacles may include people, animals, swarms of insects, vehicles, traffic jams, ships, and drifting objects. In the second danger area information 24, since the target obstacle is dynamic, the position of the second danger area 32 is a position shifted from the location where the obstacle exists in the direction of movement of the obstacle, and the shape of the second danger area 32 has a shape predicted from the direction and amount of movement of the obstacle.
[0019] When the first danger area 31 and the second danger area 32 are not distinguished from each other, they are simply referred to as the "danger area 3."
[0020] The obstacle information 25 is information about the position and movement of dynamic obstacles among obstacles contained in the environment. The obstacle information 25 is information that forms the basis of the second danger area information 24. The second danger area 32 of the second danger area information 22B is generated from the obstacles contained in the obstacle information 25. The obstacle information 25 is also used in the automatic driving of the mobile body 1.
[0021] The disturbance information 26 is information relating to disturbances in the environment. That is, the disturbance information 26 is information relating to disturbances to the mobile object 1 passing through the environment. For example, the disturbances may include wind and currents.
[0022] The search information 27 is information regarding the degree of monitoring necessity for each position in the environment. For example, the degree of monitoring necessity corresponds to the time elapsed since the monitoring information was acquired. The degree of monitoring necessity for a certain position increases according to the time elapsed since the monitoring information for that position was acquired. Immediately after the monitoring information is acquired, the degree of monitoring necessity is low. As time passes since the monitoring information was acquired, the degree of monitoring necessity increases.
[0023] The map 2 is formed by associating at least one of the danger area information 22, obstacle information 25, disturbance information 26, and search information 27 with the geographical information 21. In this example, the map 2 includes the geographical information 21, the first danger area information 23, the second danger area information 24, obstacle information 25, disturbance information 26, and search information 27, each of which is included in a different layer. The positions of the first danger area information 23, the second danger area information 24, the obstacle information 25, the disturbance information 26, and the search information 27 are associated with the positions of the geographical information 21.
[0024] Monitoring Device The monitoring device 8 includes a fixed first monitoring device 8A and a mobile second monitoring device 8B. When the first monitoring device 8A and the second monitoring device 8B are not distinguished, they are simply referred to as "monitoring device 8." Note that in the drawings, the first monitoring device 8A and the second monitoring device 8B are distinguished. The monitoring device has a monitoring sensor that acquires monitoring information within a predetermined range. The monitoring information may include topography, buildings, features, equipment, moving objects, wind direction, tidal currents, etc. The monitoring sensor includes at least one of a camera, LiDAR (Light Detection and Ranging), an infrared sensor, a laser rangefinder, a Doppler LiDAR, and an anemometer. For example, the camera captures still images or video. The laser rangefinder may employ a green laser.
[0025] The first monitoring device 8A is fixedly installed in the environment. The environment refers to the surrounding environment in which the mobile object 1 can move. The first monitoring device 8A is installed, for example, on a building, a street light, or a telephone pole. Alternatively, the first monitoring device 8A is installed on a dedicated structure. The monitoring range of the first monitoring device 8A may be fixed or may be changeable. In this example, the first monitoring device 8A is fixedly installed in the environment and is configured so that its orientation can be changed 360 degrees. In other words, the first monitoring device 8A can change its monitoring range over time and acquire monitoring information for a 360-degree area around it.
[0026] The second monitoring device 8B is a mobile object such as a flying device or a vehicle. In this example, the second monitoring device 8B is a drone as a flying device. The second monitoring device 8B is equipped with the above-mentioned monitoring sensor and can fly freely within the environment to obtain monitoring information of any area within the environment. The second monitoring device 8B is an example of a mobile monitoring device.
[0027] The monitoring device 8 communicates with the map generation device 100 via a relay device 121. The relay device 121 is installed in, for example, a control tower 120. The relay device 121 is connected to the map generation device 100 via a communication network N so as to be able to communicate with the map generation device 100.
[0028] =First Monitoring Device= The first monitoring device 8A continuously acquires monitoring information and periodically transmits the acquired monitoring information to the map generation device 100. The first monitoring device 8A periodically receives the map 2 from the map generation device 100, determines a monitoring area based on the map 2, and acquires monitoring information for the determined monitoring area.
[0029] 3 is a diagram showing a schematic hardware configuration of the first monitoring device 8A. The first monitoring device 8A has a monitoring sensor 81, an actuator 82, a communicator 83, a controller 85, a storage device 86, and a memory 87. The monitoring sensor 81 is at least one of the above-mentioned cameras. In this example, the monitoring sensor 81 is a camera and a LiDAR. The monitoring sensor 81 acquires images and point cloud data of objects within a predetermined monitoring area as monitoring information. The actuator 82 changes the monitoring area of the monitoring sensor 81 by changing the orientation of the monitoring sensor 81. The actuator 82 is, for example, an electric motor. The communicator 83 performs wireless communication with an external device.
[0030] The controller 85 controls the entire first monitoring device 8A. The controller 85 performs various types of arithmetic processing. For example, the controller 85 is formed of a processor such as a CPU (Central Processing Unit). The controller 85 may also be formed of an MCU (Micro Controller Unit), an MPU (Micro Processor Unit), an FPGA (Field Programmable Gate Array), a PLC (Programmable Logic Controller), a system LSI, or the like.
[0031] The memory 86 stores programs executed by the controller 85 and various data. For example, the memory 86 stores a control program. The memory 86 stores location information of the first monitoring device 8A. The memory 86 is formed of a non-volatile memory, a hard disk drive (HDD), a solid state drive (SSD), or the like. The memory 87 temporarily stores data and the like. For example, the memory 87 is formed of a volatile memory.
[0032] 4 is a functional block diagram of the controller 85. The controller 85 realizes various functions by reading out a program from the storage unit 86 into a memory 87 and expanding the program. Specifically, the controller 85 functions as a determiner 88 that determines the monitoring area of the monitoring sensor 81 and an acquirer 89 that acquires monitoring information by the monitoring sensor 81.
[0033] The determiner 88 determines the monitoring area of the monitoring sensor 81 based on the search command from the map generating device 100 and the map 2. The method for determining the monitoring area will be described later.
[0034] The acquirer 89 acquires monitoring information of the monitoring area determined by the determiner 88 using the monitoring sensor 81. More specifically, the acquirer 89 activates the actuator 82 to adjust the monitoring area of the monitoring sensor 81, and acquires the monitoring information using the monitoring sensor 81. The acquirer 89 stores the acquired monitoring information, time information at the time of acquisition, and position information of the monitoring area in the memory 86. The acquirer 89 transmits the monitoring information, time information, and position information stored in the memory 86 to the map generation device 100.
[0035] =Second Monitoring Device= The second monitoring device 8B continuously acquires monitoring information while moving within the environment, and periodically transmits the acquired monitoring information to the map generation device 100. The second monitoring device 8B periodically receives map 2 from the map generation device 100, determines a monitoring area based on map 2, and acquires monitoring information for the determined monitoring area.
[0036] FIG. 5 is a diagram illustrating a schematic hardware configuration of the second monitoring device 8B. The second monitoring device 8B includes a monitoring sensor 91, an actuator 92, a communicator 93, a position detector 94, a controller 95, a storage device 96, and a memory 97. The monitoring sensor 91 is at least one of the aforementioned cameras. In this example, the monitoring sensor 91 is a LiDAR and a camera. The monitoring sensor 91 acquires point cloud data and still images of objects within a predetermined monitoring area as monitoring information. The actuator 92 moves the monitoring sensor 91 to change the monitoring area of the monitoring sensor 91. The actuator 92 is, for example, an electric motor for multiple (e.g., four) propellers of a drone. The actuator 92 adjusts the drive of the multiple propellers to fly the second monitoring device 8B and change the monitoring area of the monitoring sensor 91. The communicator 93 wirelessly communicates with an external device. The position detector 94 detects the position of the second monitoring equipment 9B. For example, the position detector 94 is a GPS (Global Positioning System) receiver. The position detector 94 may include a gyro sensor or an acceleration sensor.
[0037] The controller 95 controls the entire second monitoring device 8B. The controller 95 performs various types of arithmetic processing. For example, the controller 95 is formed of a processor such as a CPU (Central Processing Unit). The controller 95 may also be formed of an MCU (Micro Controller Unit), an MPU (Micro Processor Unit), an FPGA (Field Programmable Gate Array), a PLC (Programmable Logic Controller), a system LSI, or the like.
[0038] The storage unit 96 stores programs executed by the controller 95 and various data. For example, the storage unit 96 stores a control program. The storage unit 96 is formed of a non-volatile memory, a hard disk drive (HDD), a solid state drive (SSD), or the like. The memory 97 temporarily stores data and the like. For example, the memory 97 is formed of a volatile memory.
[0039] 6 is a functional block diagram of the controller 95. The controller 95 realizes various functions by reading out a program from a storage device 96 into a memory 97 and expanding the program. Specifically, the controller 95 functions as a determiner 98 that determines the monitoring area of the monitoring sensor 91 and an acquirer 99 that acquires monitoring information by the monitoring sensor 91.
[0040] The determiner 98 determines the monitoring area of the monitoring sensor 91 based on the search command from the map generating device 100 and the map 2. The method for determining the monitoring area will be described later.
[0041] The acquirer 99 acquires monitoring information of the monitoring area determined by the determiner 98 using the monitoring sensor 91. More specifically, the acquirer 99 controls the actuator 92 to move the second monitoring equipment 9B and align the monitoring area of the monitoring sensor 91 with the determined monitoring area. The acquirer 99 acquires monitoring information of the monitoring area using the monitoring sensor 91. The acquirer 99 stores the acquired monitoring information, time information at the time of acquisition, and position information of the monitoring area in the memory 96. The position of the monitoring area is the position of the second monitoring equipment 9B detected by the position detector 94. The acquirer 99 transmits the monitoring information, time information, and position information stored in the memory 96 to the map generation device 100.
[0042] Mobile Object The mobile object 1 is, for example, a vehicle. Alternatively, the mobile object 1 may be a ship or an autonomous robot. FIG. 7 is a diagram showing a schematic hardware configuration of the mobile object 1. The mobile object 1 has a monitoring sensor 11, an actuator 12, a communicator 13, a position detector 14, a display device 111, a controller 15, a storage device 16, and a memory 17. The monitoring sensor 11 is at least one of the above-mentioned cameras. In this example, the monitoring sensor 11 is a LiDAR. The monitoring sensor 11 acquires point cloud data of objects within a predetermined monitoring area as monitoring information. The actuator 12 is a driving source of the mobile object 1. For example, the actuator 12 is an electric motor or an engine. The communicator 13 performs wireless communication with an external device. The position detector 14 detects the position of the mobile object 1. For example, the position detector 14 is a GPS (Global Positioning System) receiver. The position detector 14 may include a gyro sensor or an acceleration sensor. When the moving object 1 is a vehicle, the display device 111 is, for example, a car navigation system. The display device 111 appropriately displays a map 2.
[0043] The controller 15 controls the entire moving body 1. The controller 15 performs various types of arithmetic processing. For example, the controller 15 is formed of a processor such as a CPU (Central Processing Unit). The controller 15 may also be formed of an MCU (Micro Controller Unit), an MPU (Micro Processor Unit), an FPGA (Field Programmable Gate Array), a PLC (Programmable Logic Controller), a system LSI, or the like.
[0044] The memory 16 stores programs and various data executed by the controller 15. For example, the memory 16 stores a control program. The memory 16 stores position information of the moving object 1. The memory 16 is formed of a non-volatile memory, a hard disk drive (HDD), a solid state drive (SSD), or the like.
[0045] The memory 17 temporarily stores data, etc. For example, the memory 17 is formed of a volatile memory.
[0046] The controller 15 controls the actuator 12, causing the moving body 1 to drive autonomously. The moving body 1 receives the map 2 from the map generation device 100 via the communicator 13, and generates a route based on the map 2. The controller 15 controls the actuator 12 based on the detection results of the monitoring sensor 11 and the position detector 14, so that the moving body 1 moves along the generated route. The moving body 1 periodically receives the map 2 from the map generation device 100, and updates the route.
[0047] 8 is a functional block diagram of the controller 15. The controller 15 realizes various functions by reading out a program from the storage unit 16 into the memory 17 and expanding the program. Specifically, the controller 15 functions as a route generator 18, an operation executor 19, and a monitoring information acquirer 110.
[0048] The route generator 18 generates a route based on the map 2. A destination is set in advance for the mobile body 1. The route generator 18 generates a route from the current position of the mobile body 1 to the destination based on the map 2. At this time, the route generator 18 refers to the geographical information 21 and the danger area information 22 included in the map 2. The route generator 18 displays the map 2 on the display device 111. For example, the route generator 18 superimposes the danger area information 22 on the geographical information 21 and displays them on the display device 111. The route generator 18 may also superimpose the route on the map 2 and display them on the display device 111.
[0049] The driving executor 19 controls the actuator 12 so that the moving object 1 moves along the generated route. The driving executor 19 controls the actuator 12 so that the position of the moving object 1 detected by the position detector 14 moves along the generated route. When the driving executor 19 determines based on the map 2 that there is an obstacle with which a collision must be avoided, it controls the actuator 12 to perform avoidance driving to avoid a collision with the obstacle.
[0050] The monitoring information acquirer 110 acquires monitoring information using the monitoring sensor 11. The monitoring information acquirer 110 stores the acquired monitoring information, time information at the time of acquisition, and position information of the monitoring area in the memory 16. The position of the monitoring area is the position of the mobile object 1 detected by the position detector 14. The monitoring information acquirer 110 transmits the monitoring information, time information, and position information stored in the memory 16 to the map generation device 100.
[0051] -Map Generation Device- The map generation device 100 periodically collects monitoring information from multiple monitoring devices 8. The map generation device 100 integrates and processes the monitoring information from the multiple monitoring devices 8 to generate a map 2. The map generation device 100 periodically generates the map 2 and transmits it to the mobile object 1 or the monitoring device 8.
[0052] For example, the map generation device 100 is a computer or a group of servers. Fig. 9 is a diagram showing a schematic hardware configuration of the map generation device 100. The map generation device 100 has a controller 61, a storage device 62, a memory 63, and a communicator 64. The communicator 64 performs wireless communication with an external device.
[0053] The controller 61 controls the entire map generating device 100. The controller 61 performs various types of arithmetic processing. For example, the controller 61 is formed of a processor such as a CPU (Central Processing Unit). The controller 61 may also be formed of an MCU (Micro Controller Unit), an MPU (Micro Processor Unit), an FPGA (Field Programmable Gate Array), a PLC (Programmable Logic Controller), a system LSI, or the like.
[0054] The storage unit 62 stores programs executed by the controller 61 and various data. For example, the storage unit 62 stores a control program. The storage unit 62 stores geographical information 21 of the environment. The storage unit 62 is formed of a non-volatile memory, a hard disk drive (HDD), a solid state drive (SSD), or the like. The memory 63 temporarily stores data and the like. For example, the memory 63 is formed of a volatile memory.
[0055] 10 is a functional block diagram of the controller 61. The controller 61 realizes various functions by reading out a program from the storage device 62 into the memory 63 and expanding the program. Specifically, the controller 61 functions as a collector 65 that collects monitoring information, a recognizer 66 that recognizes obstacles in the environment from the monitoring information, and a generator 67 that generates a map 2.
[0056] The collector 65 collects monitoring information from the first monitoring device 8A and the second monitoring device 8B. In this example, since the mobile unit 1 also has the function of acquiring monitoring information, the collector 65 also collects monitoring information from the mobile unit 1. Specifically, the collector 65 periodically transmits a search command to acquire monitoring information to the first monitoring device 8A, the second monitoring device 8B, and the mobile unit 1 via the communicator 64, along with the map 2. The map 2 at this time includes at least the search information 27. Upon receiving the search command and the map 2, the first monitoring device 8A, the second monitoring device 8B, and the mobile unit 1 acquire the monitoring information and transmit the acquired monitoring information to the map generation device 100. The collector 65 receives the monitoring information from each of the first monitoring device 8A, the second monitoring device 8B, and the mobile unit 1 via the communicator 64 and stores it in the memory 62. Since the search command is periodically transmitted, the monitoring information is also periodically collected.
[0057] The recognizer 66 recognizes obstacles in the environment based on the monitoring information. The recognizer 66 integrates and processes the monitoring information stored in the memory 62 to identify obstacles in the environment. For example, the recognizer 66 synchronizes the monitoring information stored in the memory 62. The recognizer 66 integrates the synchronized monitoring information based on sensor fusion technology to identify obstacles. The recognizer 66 estimates the state of the identified obstacles, estimating their position, speed, type, size, shape, etc. The recognizer 66 classifies the obstacles into first obstacles and second obstacles based on their types, etc. The first obstacles are obstacles that are the subject of the first dangerous area information 23, i.e., static obstacles. The second obstacles are obstacles that are the subject of the second dangerous area information 24, i.e., dynamic obstacles.
[0058] The generator 67 references various types of information to generate the map 2. Specifically, the generator 67 reads out the geographical information 21 stored in the memory 62. The generator 67 generates danger area information 22, obstacle information 25, disturbance information 26, and search information 27 based on the obstacle recognized by the recognizer 66.
[0059] =First Danger Area Information= The generator 67 generates first danger area information 23 based on the first obstacle. FIG. 11 is a schematic diagram showing an example of the geographical information 21. FIG. 12 is a schematic diagram showing an example of the first danger area information 23. In the example of FIG. 11, the geographical information 21 shows roads divided by buildings and the like. In FIG. 12, for reference, the geographical information 21 is shown by a two-dot chain line. The generator 67 sets an area surrounding the obstacle recognized as the first obstacle by the recognizer 66 as the first danger area 31. In the example of FIG. 12, two first danger areas 31a and 31b are set. The first danger area 31a corresponds to the first obstacle identified as a construction site. The first danger area 31b corresponds to the first obstacle identified as rough terrain. The generator 67 stores the positions and outlines of the first danger areas 31a and 31b in the memory 62 as the first danger area information 23.
[0060] At this time, the generator 67 may set an avoidance necessity degree indicating the necessity of avoidance for the first danger area 31. That is, the generator 67 calculates the avoidance necessity degree for the first danger area 31 based on the monitoring information, and sets the avoidance necessity degree for the first danger area 31. The generator 67 sets a high avoidance necessity degree for a first danger area 31 that absolutely needs to be avoided, and a low avoidance necessity degree for a first danger area 31 that should be avoided if possible. For example, the generator 67 sets a high avoidance necessity degree for a first danger area 31a that corresponds to a construction site, and a low avoidance necessity degree for a first danger area 31b that corresponds to rough ground.
[0061] Furthermore, when a sea route, passage, or road is regulated by law or practice, the generator 67 may include an impassable area as the first danger area 31 in the first danger area information 23. For example, when the direction of travel in a certain sea area is determined by law or practice, an area corresponding to a sea route that may be in the opposite direction may be set as the first danger area 31. The map generating device 100 may obtain information on a sea route, passage, or road regulated by law or practice from an external database or the like.
[0062] =Second Danger Area Information= The generator 67 generates second danger area information 24 based on the second obstacle. FIG. 13 is a schematic diagram showing an example of the second danger area information 24. For reference, in FIG. 13, the geographical information 21 is indicated by a two-dot chain line. The generator 67 predicts the behavior of the second obstacle over time based on information such as the speed of the obstacle recognized as the second obstacle by the recognizer 66, and sets the range in which the second obstacle may exist as a second danger area 32. The behavior prediction of the second obstacle is performed using various known techniques. In the example of FIG. 13, four second danger areas 32a, 32b, 32c, and 32d are set. The second danger area 32a corresponds to the second obstacle identified as a moving vehicle. The second danger area 32b corresponds to the second obstacle identified as another moving vehicle. The second danger area 32c corresponds to the third obstacle identified as yet another stopped vehicle. The second danger area 32d corresponds to a second obstacle identified as a moving person. The second danger areas 32a and 32b are relatively wide areas to accommodate moving vehicles. The second danger area 32c is a relatively narrow area to accommodate stationary vehicles, and a uniform degree of avoidance necessity is set for all directions. The second danger area 32d corresponds to a person, and is therefore narrower than the second danger areas 32a and 32b.
[0063] The generator 67 may also set the degree of avoidance necessity for the second danger area 32, similar to the first danger area 31. That is, the generator 67 calculates the degree of avoidance necessity for the second danger area 32 based on the monitoring information, and sets the degree of avoidance necessity for the second danger area 32. For example, the generator 67 assigns a low degree of avoidance necessity to the second danger area 32 corresponding to the second obstacle whose behavior is stable, and assigns a high degree of avoidance necessity to the second danger area 32 corresponding to the second obstacle whose behavior is unstable. When a moving object passing through a no-entry zone is identified as the second obstacle, the generator 67 sets a high degree of avoidance necessity to the second danger area 32 corresponding to the moving object.
[0064] The generator 67 may vary the degree of avoidance necessity within the second danger region 32. In other words, the degree of avoidance necessity within the second danger region 32 may not be uniform. The generator 67 calculates the degree of avoidance necessity within the second danger region 32 based on monitoring information and sets the degree of avoidance necessity within the second danger region 32. The second danger region 32 is set based on a behavior prediction of the second obstacle, so the probability of the second obstacle existing varies even within the second danger region 32. A high degree of avoidance necessity is set for a portion of the second danger region 32 where the probability of the second obstacle existing is high, for example, a portion close to the current position of the second obstacle, and a low degree of avoidance necessity is set for a portion of the second danger region 32 where the probability of the second obstacle existing is low, for example, a portion far from the current position of the second obstacle. In FIG. 13 , three levels of the degree of avoidance necessity are used, with the solid line representing a high degree of avoidance necessity, the dashed-dotted line representing a medium degree of avoidance necessity, and the dashed line representing a low degree of avoidance necessity.
[0065] As an example, the generator 67 assumes that the predicted trajectory of the second obstacle is represented by the following motion model, and obtains a predicted distribution using a Kalman filter, a Bayesian filter, or the like.
[0066] where x t is the state of the second fault at time t, and u t is the input (such as velocity or acceleration) of the second obstacle at time t, θ is a parameter of the motion model, such as mass or viscous force, and ω is a disturbance, such as random noise.
[0067] The generator 67 then uses the following prediction model for each secondary fault:
[0068] where p(x t+1 |θ,α,β) is x t+1 is the posterior distribution of p(x t ,u t |β) is x t is the prior distribution of p(x t+1 |x t ,u t , θ,α) is a stochastic motion model, and α and β are parameters of the motion model.
[0069] In these equations (1) and (2), it is necessary to set the parameters θ, α, and β of the motion model, and the generator 67 estimates the parameters θ, α, and β. For example, the generator 67 estimates various parameters based on the type and shape of the second obstacle. If the second obstacle is a ship, the wave-making resistance, which is the dominant resistance of the ship, can be estimated from the speed and hull shape (Froude number and wave-making resistance coefficient). Therefore, the generator 67 can estimate the parameters θ, α, and β as a function of the ship type and speed.
[0070] Alternatively, the generator 67 estimates the motion model or the distribution of the inputs using data-driven control. For example, the generator 67 uses past trajectory data to estimate a probabilistic motion model and the distribution of the inputs by machine learning (such as a Gaussian process or Bayesian estimation). The generator 67 may calculate the predicted distribution of the second fault by a sampling method such as the Markov Chain Monte Carlo (MCMC) method. Alternatively, the generator 67 approximates the predicted distribution of the second fault by moment matching or the like.
[0071] In addition to the method using a motion model, the generator 67 may predict the trajectory of the second obstacle using a database. The generator 67 may probabilistically predict the trajectory from a database of past trajectories of the second obstacle. The generator 67 may compare the trajectory of the second obstacle with the past trajectories in the database and perform sampling based on likelihood.
[0072] The generator 67 may also set one second danger area 32 based on multiple second obstacles. For example, the generator 67 clusters the second obstacles based on their positions and movement directions, and approximates a cluster speed distribution from the average speed and speed variance of the second obstacles in the cluster by moment matching or the like. The generator 67 may derive a predicted distribution of the cluster from the approximated speed distribution, and set the second danger area 32. For example, in a congested area where multiple second obstacles are concentrated, the setting of the second danger area 32 is simplified.
[0073] =Obstacle Information= The generator 67 generates obstacle information 25 based on the obstacle. FIG. 14 is a schematic diagram showing an example of the obstacle information 25. For reference, in FIG. 14, the geographical information 21 is indicated by a two-dot chain line. The generator 67 stores the position, speed, type, size, and shape of the obstacle recognized as the second obstacle by the recognizer 66 in the memory 62 as obstacle information 25. The generator 67 may also store the speed variance of the second obstacle in the memory 62 as obstacle information 25. Furthermore, the generator 67 may also store the parameters of the motion model used to determine the second danger area 32 in the memory 62 as obstacle information 25 of the corresponding second obstacle.
[0074] In the example of Fig. 14, the obstacle information 25 includes four second obstacles. Fig. 14 shows the position and speed of each second obstacle. Second obstacle 41a is a moving vehicle. Second obstacle 41b is another moving vehicle. Second obstacle 41c is yet another stopped vehicle. Second obstacle 41d is a moving person.
[0075] =Disturbance Information= The generator 67 generates the disturbance information 26. FIG. 15 is a schematic diagram showing an example of the disturbance information 26. Note that in FIG. 15, for reference, the geographical information 21 is indicated by a two-dot chain line. The generator 67 generates the disturbance information 26 based on information related to the disturbance from the moving body 1 or the monitoring device 8. The disturbance is, for example, wind direction and wind speed. The disturbance may also be a tidal current. The generator 67 expresses the disturbance information 26 as a vector field as shown in FIG. 15.
[0076] For example, if the mobile object 1 or the monitoring device 8 has a Doppler LiDAR as a monitoring sensor, the mobile object 1 or the monitoring device 8 can measure wind direction and wind speed in the environment. The mobile object 1 or the second monitoring device 8B measures wind direction and wind speed at each point while moving through the environment. The first monitoring device 8A measures wind direction and wind speed at the point where it is installed. The map generating device 100 collects position information, wind direction, and wind speed of the measurement point from the mobile object 1 or the monitoring device 8. The monitoring sensor that detects wind direction and wind speed is not limited to an anemometer, but may also be a camera such as an infrared camera. The infrared camera can capture steam or smoke in the environment, and wind direction and wind speed can be derived based on the steam or smoke.
[0077] Alternatively, if the moving object 1 or the second monitoring device 8B has a disturbance observer, the generator 67 may collect the disturbance determined by the disturbance observer from the moving object 1 or the second monitoring device 8B.
[0078] If the disturbance includes a tidal current, an underwater drone may be employed as the second monitoring device 8B. The generator 67 may determine the tidal current based on the disturbance determined by a disturbance observer of the underwater drone or the water-resistant speed determined by a Doppler Velocity Log (DVL).
[0079] Since there are only a limited number of points in the environment where disturbances can be detected, the generator 67 may estimate a vector field in the environment from a plurality of actually obtained disturbance vectors using a model obtained by machine learning. Alternatively, the generator 67 may interpolate the disturbance vectors at each point in the environment from a plurality of actually obtained disturbance vectors using a Gaussian process or the like.
[0080] =Search Information= The generator 67 generates the search information 27. FIG. 16 is a schematic diagram showing an example of the search information 27. FIG. 17 is an explanatory diagram showing changes in the degree of monitoring necessity. Note that in FIG. 16, the geographical information 21 is indicated by a two-dot chain line for reference. The generator 67 generates the search information 27 based on the monitoring information from the monitoring device 8. The search information 27 is information regarding the degree of monitoring necessity for each position in the environment. The generator 67 sets the degree of monitoring necessity for each position according to the elapsed time since the monitoring information for each position was obtained. The longer the time since the monitoring information was obtained, the higher the degree of monitoring necessity becomes. In FIG. 16, the degree of monitoring necessity is represented by the density of dots. Areas with a high density of dots have a high degree of monitoring necessity, and areas with a low density of dots have a low degree of monitoring necessity.
[0081] For example, the generator 67 sets the degree of monitoring necessity φ(q,t) of each position q in the environment Q as a function of the position q and time t. The degree of monitoring necessity φ is set as shown in the following formula so that it gradually increases with the elapsed time since the monitoring information was acquired and decreases once the monitoring information is acquired. The search information 27 includes the degree of monitoring necessity φ(q,t) or a partial derivative of the degree of monitoring necessity φ(q,t) of formula (3).
[0082] where τ d (q) is the time constant for the decrease in the monitoring necessity degree φ at the position q. u (q) is the time constant for the increase in the monitoring necessity degree φ at the position q.
[0083] The vertical axis of Figure 17 represents the degree of monitoring necessity φ, and the horizontal axis represents the position q. The second monitoring device 8B moves from left to right along the horizontal axis while acquiring monitoring information. The degree of monitoring necessity φ of the position q for which monitoring information has been acquired by the second monitoring device 8B is low, and the degree of monitoring necessity φ of the position q for which monitoring information has not yet been acquired is high. The degree of monitoring necessity φ of the position q for which monitoring information has already been acquired increases over time.
[0084] Time constant τ d (q), τ u (q) may be a different value depending on the position q. For example, at a position q with a large number of accidents in the past, 1 / τ u(q) is larger than other positions q. u (q) may be set.
[0085] The generator 67 basically changes the degree of monitoring necessity φ in accordance with equation (3). However, the generator 67 may also change the degree of monitoring necessity φ in a special way, independent of equation (3), in response to a specific event. For example, the generator 67 may predict the trajectory of the moving object 1 and set the degree of monitoring necessity φ at a position q on the predicted trajectory high. This increases the degree of monitoring necessity φ on the future trajectory of the moving object 1. When an unconfirmed moving object with an unknown ID or the like is detected, the generator 67 may also set the degree of monitoring necessity φ around the unconfirmed moving object high so as to follow the unconfirmed moving object. Note that the method of increasing the degree of monitoring necessity φ does not directly change the degree of monitoring necessity φ, but rather increases the time constant τ u (q) may be increased. This increases the rate at which the degree of monitoring necessity φ at a specific position q increases, and the degree of monitoring necessity φ increases at an earlier stage.
[0086] -Monitoring Process- Next, the monitoring process in the driving assistance system 1000 will be described. The driving assistance system 1000 executes a monitoring process to collect monitoring information for each position in the environment. In the monitoring process, the map generation device 100 transmits a search command to the monitoring device 8, and the monitoring device 8 acquires the monitoring information and transmits it to the map generation device 100. FIG. 18 is a flowchart of the monitoring process. The map generation device 100 and the monitoring device 8 execute the monitoring process in accordance with the flowchart in FIG. 18. Note that the multiple monitoring devices 8 periodically exchange position information with each other and are aware of each other's positions.
[0087] First, in step S101, the map generation device 100 transmits a search command and a map 2 for search to the monitoring devices 8. The map 2 for search includes at least search information 27. In this example, the map 2 also includes geographical information 21. The map 2 is common to all monitoring devices 8. The map generation device 100 generates the search information 27 based on the currently acquired monitoring information. Specifically, the generator 67 calculates the monitoring necessity degree φ using equation (3) based on the elapsed time t since the monitoring information at each position q in the environment was acquired, and generates the search information 27. The generator 67 reads the geographical information 21 from the memory 62 and generates a map 2 including the geographical information 21 and the search information 27. The map generation device 100 transmits the search command and the map 2 to all monitoring devices 8. At this time, the map generation device 100 may also transmit the search command and the map 2 to a mobile object 1 that has a function of acquiring monitoring information.
[0088] When the search command and map 2 are transmitted, the monitoring device 8 receives the search command and map 2 in step S201. In step S202, each monitoring device 8 determines a monitoring area, which is an area for which monitoring information should be acquired, based on the search information 27. Each monitoring device 8 determines a monitoring area based on the search information 27 so that the monitoring necessity degree φ in the environment is reduced overall. The search information 27 includes the monitoring necessity degree φ(q,t) or a partial derivative of the monitoring necessity degree φ(q,t) in equation (3). In detail, the monitoring device 8 determines a monitoring area so that the overall necessity degree J shown in the following equation is reduced, and calculates the speed (specifically, speed vector) u of the monitoring device 8. i As a result, the monitoring area is set to an area in the environment where the degree of monitoring necessity φ is high.
[0089] where p i is the position vector of the monitoring device 8, and u i is the velocity vector of the monitoring device 8, and C is a constant.
[0090] At this time, the monitoring device 8 determines the monitoring area by performing cooperative control with the other monitoring devices 8. Specifically, the second monitoring device 8B is aware of the positions of the first monitoring device 8A and the other second monitoring devices 8B. The second monitoring device 8B sets monitoring information for an area with a high level of monitoring necessity φ that is located relatively close to itself. If there is another monitoring device 8 whose distance to an area with a high level of monitoring necessity φ is shorter than the distance from itself, the second monitoring device 8B leaves the acquisition of monitoring information for that area to the other monitoring device 8. Since the first monitoring device 8A cannot move, it determines the monitoring area within its own monitorable range. In this way, the monitoring device 8 determines the monitoring area in cooperation with the other monitoring devices 8.
[0091] Once the monitoring area is set, the monitoring device 8 acquires monitoring information in step S203. Specifically, the first monitoring device 8A aligns the orientation of the monitoring sensor 81 (e.g., the FOV (Field of View) of the LiDAR) with the monitoring area and acquires monitoring information using the monitoring sensor 81. The second monitoring device 8B moves so that the FOV of the monitoring sensor 81 aligns with the monitoring area and acquires monitoring information using the monitoring sensor 81. At this time, the second monitoring device 8B references the geographical information 21 and generates a route so as to avoid colliding with buildings or features.
[0092] When the mobile object 1 is performing autonomous driving, the mobile object 1 having the function of acquiring monitoring information acquires the monitoring information necessary for autonomous driving. In other words, the mobile object 1 acquires the monitoring information regardless of the monitoring necessity degree φ.
[0093] Thereafter, in step S204, the monitoring device 8 transmits the acquired monitoring information to the map generating device 100.
[0094] In step S102, the map generating device 100 receives the monitoring information. The map generating device 100 collects the monitoring information from a plurality of monitoring devices 8 in the environment.
[0095] When the map generation device 100 receives the monitoring information, in step S103, it updates the search information 27. Specifically, the map generation device 100 decreases the partial derivative of the monitoring necessity degree φ(q,t) of a point for which monitoring information has been acquired according to the upper equation of equation (3), and increases the partial derivative of the monitoring necessity degree φ(q,t) of a point for which monitoring information has not been acquired according to the lower equation of equation (3). In this way, the search information 27 is updated.
[0096] Thereafter, the map generating device 100 uses the updated search information 27 to repeat the process from step S101.
[0097] In this way, the search information 27 is updated so that the degree of monitoring necessity φ increases according to the time elapsed since the monitoring information was acquired, and the multiple monitoring devices 8 preferentially acquire monitoring information for areas with a high degree of monitoring necessity φ based on the updated search information 27. As a result, the number of points in the environment that are left unmonitored for long periods of time decreases, and the monitoring information in the environment is kept fresh.
[0098] -Map Generation Process- Next, a description will be given of driving assistance in the driving assistance system 1000. The driving assistance system 1000 executes a map generation process that generates a map 2 for driving assistance based on monitoring information in the environment. In the map generation process, the map generation device 100 generates the map 2 based on the monitoring information.
[0099] The map 2 for driving assistance includes at least geographical information 21 and danger area information 22. In this example, the map 2 includes first danger area information 23 and second danger area information 24 as the danger area information 22. The map 2 also includes obstacle information 25 and disturbance information 26. The map 2 is common to all mobile objects 1. The map generation device 100 may generate the map 2 in a format including all of the geographical information 21, danger area information 22, obstacle information 25, and disturbance information 26. Alternatively, the map generation device 100 may generate multiple types of maps 2 including different information, such as a map 2 including the geographical information 21 and danger area information 22, a map 2 including the geographical information 21 and obstacle information 25, or a map 2 including disturbance information 26. In the following description, the map 2 includes the geographical information 21, the first danger area information 23, the second danger area information 24, the obstacle information 25, and the disturbance information 26.
[0100] The map generating device 100 periodically collects monitoring information from the monitoring device 8 and periodically updates the map 2. The map generating device 100 periodically transmits the map 2 to the mobile object 1. The collection period of the monitoring information, the update period of the map 2, and the transmission period of the map 2 may or may not be synchronized with each other.
[0101] --Autonomous Driving--Next, the autonomous driving of the mobile object 1 will be described. The mobile object 1 receives the map 2 from the map generation device 100, generates a route based on the map 2, and performs autonomous driving according to the generated route. The mobile object 1 also performs obstacle avoidance operations as necessary. Figure 19 is a flowchart of the autonomous driving. The mobile object 1 performs autonomous driving according to the flowchart of Figure 19.
[0102] First, in step S301, the mobile object 1 receives a map 2. In this example, the map 2 includes geographical information 21, first danger area information 23, second danger area information 24, obstacle information 25, and disturbance information 26.
[0103] In step S302, the moving body 1 generates a route 5 based on the map 2. More specifically, a destination 51 is set for the moving body 1. In addition, a waypoint 52 may also be set for the moving body 1. The route generator 18 generates the entire route 5 from the current location to the destination 51 based on the geographical information 21 and the first danger area information 23. If the waypoint 52 is set, the route generator 18 generates a route 5 that travels from the current location to the destination 51 via the waypoint 52. At this time, the route generator 18 generates the route 5 taking into account the influence of disturbances based on the disturbance information 26.
[0104] FIG. 20 is a schematic diagram of a map 2 in which first danger area information 23 is superimposed on geographical information 21. In this example, a destination 51 and a waypoint 52 are set. The route generator 18 of the mobile object 1 basically generates a route 5 from the current location to the destination based on the geographical information 21. The route generator 18 first generates a route 5 as shown by the solid line in FIG. 20. The route generator 18 references the geographical information 21 to generate a route that does not collide with features, etc. If roads, passages, or routes are set in the environment, the route generator 18 generates a route that follows the roads, passages, or routes. The route generator 18 also references disturbance information 26 to generate a route 5 that avoids areas of strong wind. For example, the route generator 18 generates a route using an A-star search algorithm, a hybrid A-star search algorithm, or an RRT-star search algorithm.
[0105] If a first danger area 31 exists on or near the generated route 5, the route generator 18 modifies the route 5. In the example of Fig. 18, the route 5 generated only from the geographical information 21 passes over the first danger area 31a and near the first danger area 31b. Therefore, the route generator 18 modifies the route 5 so as to bypass the first danger area 31a and move away from the first danger area 31b, as shown by the dashed dotted line. At this time, the route generator 18 also refers to the geographical information 21 to modify the route 5 so as to avoid interference with features, etc.
[0106] However, when a degree of avoidance necessity is set for the first danger area 31, the path generator 18 may determine, based on the degree of avoidance necessity, whether to detour around the first danger area 31, how much distance to leave between the path 5 and the first danger area 31, and the like. For example, a relatively high degree of avoidance necessity is set for the first danger area 31a, and a relatively low degree of avoidance necessity is set for the first danger area 31b. In this case, the path generator 18 may correct the path 5 so as to detour around the first danger area 31a and not detour around the first danger area 31b. Alternatively, the path generator 18 may correct the path 5 so as to make a large detour around the first danger area 31a and a small detour around the first danger area 31b.
[0107] Next, in step S303, the route generator 18 locally modifies the route 5. Specifically, the route generator 18 modifies the route 5 based on the second danger area information 24. FIG. 21 is a schematic diagram of a map 2 in which the first danger area information 23 and the second danger area information 24 are superimposed on the geographical information 21. The route generator 18 determines, based on the second danger area information 24, whether a second danger area 32 exists on or near the route 5 within a predetermined first range R1 from the current location of the mobile object 1. In the example of FIG. 21 , the second danger area 32e is located on the route 5 within the first range R1. The route generator 18 modifies the route 5 so as to detour around the second danger area 32e. Specifically, the route generator 18 sets a local destination (waypoint) 53 on the route 5 outside the first range R1, and modifies the route 5 so as to detour around the second danger area 32e while heading toward the local destination 53, as indicated by the dashed dotted line. At this time, the route generator 18 also refers to the geographical information 21 and the first danger area information 23, and modifies the route 5 so as to avoid interference with features and the first danger area 31. Furthermore, the route generator 18 may refer to disturbance information 26. For example, the route generator 18 may search for an energy-efficient route that takes into account the dynamics of the moving object 1 and disturbances.
[0108] Although the second danger area 32a is also located on the route 5, the route generator 18 does not perform any correction to the route 5 corresponding to the second danger area 32a because the second danger area 32a is outside the first range R1.
[0109] For the first danger area 31, the path generator 18 generates the path 4 by taking into account the first danger area 31 inside and outside the first range R1, whereas for the second danger area 32, the path generator 18 generates the path 4 by taking into account only the second danger area 32 within the first range R1. In other words, since the first danger area 31 hardly fluctuates over time, the path generator 18 considers the first danger area 31 that is relatively far from the moving object 1. On the other hand, since the second danger area 32 may fluctuate over time, the path generator 18 considers only the area that is relatively close to the moving object 1.
[0110] Furthermore, when a degree of avoidance necessity is set for the second danger area 32, the path generator 18 may determine, based on the degree of avoidance necessity, whether to detour around the second danger area 32, and how much distance to leave between the path 5 and the second danger area 32. For example, when a second danger area 32 corresponding to a second obstacle exists for a mobile body passing through a restricted area, a relatively high degree of avoidance necessity is set for the second danger area 32. In this case, the path generator 18 may modify the path 5 so as to detour around the second danger area 32 with higher priority or by a larger amount than around other second danger areas 32.
[0111] Furthermore, when an avoidance necessity degree is set within the second danger area 32, the path generator 18 can determine which parts of the second danger area 32 must be avoided at all costs and which parts should be avoided if possible. For example, the path generator 18 may modify the path 5 so as to make a large detour around parts of the second danger area 32 for which a high avoidance necessity degree is set, and a small detour around parts of the second danger area 32 for which a low avoidance necessity degree is set. Alternatively, the path generator 18 may modify the path 5 so as to detour around parts of the second danger area 32 for which a high avoidance necessity degree is set, and pass through parts for which a low avoidance necessity degree is set.
[0112] Once the route 5 is generated, the moving body 1 executes movement in step S304. The operation executor 19 performs feedback control of the actuator 12 so as to move along the route 5 while referring to the detection results of the monitoring sensor 11 and the position detector 14.
[0113] Next, in step S305, the moving body 1 determines whether or not an avoidance operation is necessary based on the map 2. The driving executor 19 determines whether or not an obstacle exists against which a collision must be avoided based on the obstacle information 25 on the map 2. Fig. 22 is a schematic diagram of the map 2 in which the behavior of the obstacle predicted based on the obstacle information 25 is superimposed on the geographical information 21. In Fig. 22, first danger area information 23 and second danger area information 24 are also superimposed.
[0114] Specifically, the driving executor 19 determines whether an obstacle exists within a predetermined second range R2 from the current location of the mobile object 1 based on the obstacle information 25. The second range R2 is narrower than the first range R1. If an obstacle exists within the second range R2, the driving executor 19 predicts the behavior of the obstacle and determines whether the obstacle is likely to interfere with the mobile object 1. In the example of FIG. 21 , obstacles 41e and 41f exist within the second range R2. Specifically, the driving executor 19 estimates the predicted trajectories of the obstacles 41e and 41f based on the positions, speeds, types, sizes, shapes, or parameters of a motion model of the obstacles 41e and 41f. The driving executor 19 may estimate the predicted trajectories of the obstacles 41e and 41f using parameters of a motion model, etc., in the same manner as the generator 67.
[0115] If there is a possibility that the obstacle 41e or 41f in the second range R2 will interfere with the path 5, the driving executor 19 performs an avoidance operation based on the obstacle information 25. The driving executor 19 feedback-controls the position of the moving object 1 so as to avoid the predicted trajectory of the obstacle 41e or 41f. For example, the driving executor 19 controls the moving object 1 using a model predictive control (MPC) or a control barrier function (CBF). For example, the moving object 1 moves off the path 5 to avoid the obstacle 41e or 41f. The moving object 1 stops or decelerates to avoid collision with the obstacle 41e or 41f. At this time, the path generator 18 also refers to the geographical information 21 and the first danger area information 23 to move the moving object 1 so as to avoid interference with features, etc. and the first danger area 31. Alternatively, the motion executor 19 may calculate a feasible operation amount that achieves both disturbance compensation and avoidance, taking into account the influence of the disturbance, and operate the actuator 12 with the calculated operation amount.
[0116] When the avoidance operation is completed or when there is no obstacle that may cause interference within the second range R2, in step S307, the driving executor 19 determines whether or not the moving body 1 has reached the destination 51. When the moving body 1 has not reached the destination 51, the moving body 1 returns to step S301 and repeats the process from step S301. When the moving body 1 has reached the destination 51, the moving body 1 ends the automatic driving.
[0117] In this way, in the automated driving, the mobile body 1 generates and modifies the route 5 based on the map 2 received from the map generation device 100, and performs avoidance operations as necessary, thereby autonomously moving to the destination 51 without interfering with obstacles or the like. By periodically receiving the updated map 2, the mobile body 1 can generate and modify the route 5 and perform avoidance operations based on the latest map 2.
[0118] In such a driving assistance system 1000, danger area information 22 is obtained based on monitoring information within the environment, and a map 2 including geographical information 21 and the danger area information 22 is generated. By using the map 2, a route 5 can be easily generated. More specifically, if only the detection result of an obstacle is added to the geographical information 21 as monitoring information, it is necessary to identify the obstacle and predict the behavior of the obstacle when generating the route 5. However, the map generation device 100 obtains the danger area 3 based on the monitoring information and generates the danger area information 22. By generating the route 5 based on the geographical information 21 while regarding the danger area 3 as an area to be avoided, it is possible to generate a route 5 that avoids interference with obstacles in the environment. Since there is no need to identify the obstacle or predict the behavior of the obstacle, the route 5 can be easily generated.
[0119] Since the map generation device 100 generates the danger area 3 corresponding to the movement of the obstacle, by generating the route 5 taking the danger area 3 into consideration, it is possible to generate the route 5 that can avoid interference with the dynamic obstacle. More specifically, the map generation device 100 predicts the dynamic behavior of the obstacle and generates the second danger area 32 that includes the predicted behavior of the obstacle. In other words, the predicted behavior of the dynamic obstacle is expressed in the form of the second danger area 32. By generating the route 5 so as to bypass the second danger area 32, it is possible to easily form the route 5 that avoids interference with the dynamic obstacle without having to predict the behavior of the obstacle when generating the route 5.
[0120] Furthermore, the map generating device 100 classifies the danger areas 3 into a first danger area 31 corresponding to a static obstacle and a second danger area 32 corresponding to a dynamic obstacle. This allows the danger areas 3 to be appropriately reflected in the generation of the route 5. Specifically, by classifying the danger areas 3 into the first danger area 31 and the second danger area 32, information regarding the movement of the obstacle is added to the danger areas 3. This allows the danger areas 3 to be appropriately selected and appropriately reflected when generating the route 5. For example, the second danger area 32 corresponds to a dynamic obstacle and may change over time. Therefore, it is preferable that the second danger area 32 be reflected in a relatively short-term route plan. On the other hand, the first danger area 31 corresponds to a static obstacle and therefore hardly changes over time. Therefore, it is preferable that the first danger area 31 be reflected in a relatively long-term route plan. By appropriately using the first danger area 31 and the second danger area 32 to generate the route 5, it is possible to form a route 5 that appropriately avoids obstacles.
[0121] Furthermore, the map generating device 100 calculates the degree of avoidance necessity of the danger area based on the monitoring information and sets the degree of avoidance necessity for the danger area. This allows the danger area to be appropriately reflected in the generation of the route 5. Specifically, it is possible to determine whether or not to reflect the danger area in the generation of the route 5 based on the degree of avoidance necessity. Various constraints are imposed when generating the route 5, and the danger area is one of the constraints. It is possible to determine the priority of the danger area when reflecting it in the generation of the route 5 based on the degree of avoidance necessity of the danger area.
[0122] Furthermore, the map generating device 100 calculates the degree of necessity of avoidance within the danger area based on the monitoring information and sets the degree of necessity of avoidance within the danger area. In other words, different degrees of necessity of avoidance can be assigned within the danger area. This allows the danger area to be appropriately reflected in the generation of the route 5. In more detail, for one danger area, it is possible to determine which parts must be avoided at all costs and which parts should be avoided if possible based on the degree of necessity of avoidance within the danger area.
[0123] The map generating device 100 obtains obstacle information 25 and adds the obstacle information 25 to the map. This allows the mobile body 1 to perform the necessary avoidance action when moving along the route 5. Even if a route 5 that avoids obstacles can be generated in the route plan, obstacles that need to be avoided may appear during actual movement. Therefore, by adding the obstacle information 25 to the map 2 in addition to the dangerous area, the mobile body 1 can perform the avoidance action based on the obstacle information 25.
[0124] The map generating device 100 obtains disturbance information 26 and adds the disturbance information 26 to the map 2. This makes it possible to generate a route 5 taking into account disturbances in addition to the danger area 3. The mobile object 1 may be affected by disturbances when it moves. By generating the route 5 based on the map 2, it is possible to generate a route 5 taking into account the effects of disturbances.
[0125] The map generating device 100 obtains search information 27 relating to the degree of monitoring necessity for each position in the environment and adds the search information 27 to the map 2. This allows the monitoring device 8 to determine the area in the environment from which to obtain monitoring information, based on the search information 27. As a result, the map generating device 100 can appropriately collect monitoring information in the environment.
[0126] Furthermore, the monitoring necessity level is set according to the time elapsed since the monitoring information was acquired at each location in the environment. This determines the priority of the areas from which monitoring information is acquired based on the time elapsed since the monitoring information was acquired. This allows the overall monitoring information in the environment to be kept as up-to-date as possible.
[0127] Furthermore, when determining an area from which monitoring information is to be acquired based on the search information 27, the monitoring device 9 performs cooperative control with other monitoring devices 9. Specifically, a monitoring device 9 located near an area where monitoring is highly necessary acquires monitoring information for that area. The monitoring device 9 cooperates with the other monitoring devices 9 to appropriately share the acquisition of monitoring information within the environment. As a result, the overall monitoring information within the environment can be kept as up-to-date as possible.
[0128] Other Embodiments As described above, the above-described embodiments have been described as examples of the technology disclosed in the present application. However, the technology of the present disclosure is not limited to these embodiments and can be applied to embodiments in which modifications, substitutions, additions, omissions, etc. are made as appropriate. Furthermore, the components described in the above-described embodiments can be combined to create new embodiments. Furthermore, the components described in the accompanying drawings and detailed description may include not only components essential for solving the problem, but also components that are not essential for solving the problem in order to exemplify the technology. Therefore, the fact that these non-essential components are described in the accompanying drawings or detailed description should not be interpreted as immediately determining that these non-essential components are essential.
[0129] The moving body 1 and the monitoring device 8 in the driving assistance system 1000 are not limited to the above examples. For example, the moving body 1 is not limited to a vehicle, but may be an autonomous robot (e.g., a carrier pallet), a ship, or an aircraft (e.g., a drone), etc. The second monitoring device 8B may be a vehicle or an underwater drone.
[0130] The number of moving objects 1 and the number of monitoring devices 8 included in the driving assistance system 1000 are not limited to the example in Fig. 1 and are arbitrary. The number of first monitoring devices 8A and the number of second monitoring devices 8B are also arbitrary. For example, one of the first monitoring device 8A and the second monitoring device 8B may be omitted.
[0131] Communication between the map generation device 100 and the mobile body 1 and the monitoring device 8 is not limited to communication via the relay device 121. Furthermore, the map generation device 100 does not need to be installed in a fixed location. For example, the map generation device 100 may be mounted on a mobile body and move together with the mobile body. For example, the map generation device 100 may be mounted on a ship. The map generation device 100 may collect monitoring information from surrounding drones or other ships, generate a map 2 based on the monitoring information, and distribute the map 2 to surrounding mobile bodies 1.
[0132] The map 2 does not need to include all of the above-mentioned information. For example, the map 2 may include only one of the first danger area information 23 and the second danger area information 24 as the danger area information 22. The map 2 may not include at least one of the obstacle information 25 and the disturbance information 26.
[0133] The map 2 for search includes at least search information 27. The map 2 for search does not necessarily include the geographical information 21.
[0134] The methods of obtaining and expressing the various pieces of information included in the map 2 are not limited to the above examples. For example, the second danger area 32 is expressed as three overlapping circles with different diameters, but is not limited to this. For example, the second danger area 32 may be expressed as an oval or a square with a smooth outline.
[0135] The flowcharts are merely examples. Steps in the flowcharts may be changed, replaced, added, omitted, etc. as appropriate. The order of steps in the flowcharts may also be changed, and serial processing may be performed in parallel.
[0136] The functions performed by the components described herein may be implemented in circuitry or processing circuitry, including general-purpose processors, application-specific processors, integrated circuits, ASICs (Application Specific Integrated Circuits), a CPU (a Central Processing Unit), conventional circuits, and / or combinations thereof, programmed to perform the described functions. A processor includes transistors and other circuits and is considered a circuit or processing circuit. A processor may also be a programmable processor that executes a program stored in a memory.
[0137] In this specification, a circuit, unit, or means is hardware that is programmed to realize or performs the described functions, which may be any hardware disclosed herein or any hardware known to be programmed to realize or perform the described functions.
[0138] If the hardware is a processor considered to be a type of circuitry, the circuit, means, or unit is a combination of hardware and software used to configure the hardware and / or processor.
[0139] The technology of the present disclosure can be summarized as follows.
[0140] [1] The map generating device 100 includes a collector 65 that collects environmental monitoring information, and a generator 67 that generates a map 2 for assisting the autonomous driving of a mobile body 1 based on the monitoring information. The generator 67 determines dangerous areas 3 that should be avoided in the environment based on the monitoring information, and generates the map 2 that includes geographical information 21 and the dangerous areas 3.
[0141] According to this configuration, the route 5 for automated driving can be easily generated by using the map 2. In other words, when generating the route 5, there is no need to identify obstacles detected as monitoring information or predict the behavior of the obstacles. When generating the route 5 based on the geographical information 21, the dangerous area 3 is regarded as an area to be avoided, and the route 5 is generated accordingly, so that the route 5 can be easily generated while avoiding interference with obstacles in the environment. As a result, the computational load, i.e., the amount of processing, when generating the route 5 can be reduced.
[0142] [2] In the map generating device 100 described in [1], the generator 67 determines a danger area 3 corresponding to the movement of an obstacle included in the environment based on the monitoring information.
[0143] According to this configuration, it is possible to generate a route 5 that can avoid interference with dynamic obstacles by generating the route 5 taking into consideration the danger area 3. In other words, since it is not necessary to predict the behavior of obstacles when generating the route 5, it is possible to further reduce the calculation load, i.e., the amount of processing.
[0144] [3] In the map generating device 100 according to [1] or [2], the generator 67 determines the danger area 3 corresponding to the movement of a set of multiple obstacles included in the environment based on the monitoring information.
[0145] According to this configuration, it is possible to generate the route 5 by regarding multiple obstacles as one danger area 3. As a result, it is possible to further reduce the computational load, i.e., the amount of processing, when generating the route 5.
[0146] [4] In the map generating device 100 described in any one of [1] to [3], the generator 67 calculates the degree of avoidance necessity of the danger area 3 based on the monitoring information, and adds the degree of avoidance necessity to the map 2.
[0147] According to this configuration, it is possible to prioritize the dangerous areas 3 based on the degree of necessity of avoidance when generating the route 5. As a result, it is possible to flexibly generate the route 5, for example, by determining whether or not to avoid the dangerous area 3, or by how far away from the dangerous area 3.
[0148] [5] In the map generating device 100 described in any one of [1] to [4], the generator 67 calculates the degree of avoidance necessity within the danger area 3 based on the monitoring information, and adds the degree of avoidance necessity to the map 2.
[0149] According to this configuration, different degrees of necessity of avoidance can be assigned within the danger area 3. In this case, it is possible to determine which parts of one danger area 3 should be avoided preferentially and which parts should not, based on the degree of necessity of avoidance. As a result, the danger area 3 can be appropriately reflected in the generation of the route 5.
[0150] [6] In the map generating device 100 described in any one of [1] to [5], the generator 67 classifies the danger area 3 into a first danger area 31 corresponding to a static obstacle and a second danger area 32 corresponding to a dynamic obstacle.
[0151] This configuration makes it possible to know whether the danger area 3 changes over time. Therefore, the danger area 3 can be appropriately reflected in the generation of the route 5. For example, since the first danger area 31 does not change over time, the route 5 is generated taking into account the first danger area 31, which is relatively far away. On the other hand, since the second danger area 32 may change over time, the route 5 is generated taking into account only the second danger area 32, which is relatively close. By generating the route 5 using the first danger area 31 and the second danger area 32 appropriately, it is possible to form a route 5 that appropriately avoids obstacles.
[0152] [7] In the map generating device 100 described in any one of [1] to [6], the generator 67 obtains information on the movement of an obstacle that forms the basis of the danger area 3, i.e., obstacle information 25, based on the monitoring information, and adds the obstacle information 25 to the map 2.
[0153] According to this configuration, autonomous driving is performed using the obstacle information 25 as necessary. Even if a route 5 that avoids the dangerous area 3 is generated, an obstacle that needs to be avoided may appear during the actual movement of the mobile object 1. The mobile object 1 can determine the appearance of such an obstacle based on the obstacle information 25. If there is a possibility of a collision, the mobile object 1 can also perform an avoidance operation based on the obstacle information 25. In other words, by including the obstacle information 25 in the map 2, the mobile object 1 can take more urgent measures against the obstacle.
[0154] [8] In the map generating device 100 described in any one of [1] to [7], the generator 67 obtains information on disturbances in the automatic driving of the moving body 1, i.e., disturbance information 26, based on the monitoring information, and adds the disturbance information 26 to the map 2.
[0155] According to this configuration, the route 5 can be generated taking into consideration disturbances in addition to the danger area 3. The mobile object 1 may be affected by disturbances when it moves. By generating the route 5 based on the map 2, the route 5 can be generated taking into consideration the effects of disturbances.
[0156] [9] In the map generating device 100 described in any one of [1] to [8], the generator 67 sets a monitoring necessity degree for each position in the environment based on the monitoring information, and adds the monitoring necessity degree to the map 2.
[0157] According to this configuration, the points in the environment from which monitoring information is to be acquired can be determined based on the degree of necessity for monitoring. In other words, by using the map 2, monitoring information can be collected from necessary points.
[0158]
[10] In the map generating device 100 described in [9], the generator 67 sets the degree of necessity of monitoring according to the time elapsed since the monitoring information was acquired at each position in the environment.
[0159] With this configuration, the location within the environment from which monitoring information is to be acquired can be determined based on the amount of time that has elapsed since the monitoring information was acquired. For example, monitoring information from locations where a long amount of time has elapsed since the monitoring information was acquired can be acquired preferentially. This allows the overall monitoring information within the environment to be kept as up-to-date as possible.
[0160]
[11] A driving assistance system 1000 includes a mobile body 1 that performs autonomous driving, and a map generating device 100 that generates a map 2 to assist the autonomous driving of the mobile body 1. The map generating device 100 has a collector 65 that collects environmental monitoring information, and a generator 67 that generates the map 2 based on the monitoring information. The generator 67 determines dangerous areas 3 that should be avoided in the environment based on the monitoring information, generates the map 2 including geographical information 21 and the dangerous areas 3, and transmits the map 2 to the mobile body 1. The mobile body 1 generates a route 5 for autonomous driving based on the geographical information 21 and the dangerous areas 3.
[0161] According to this configuration, the mobile body 1 can easily generate the route 5 by using the map 2. In other words, when generating the route 5, the mobile body 1 does not need to identify obstacles detected as monitoring information or predict the behavior of the obstacles. When generating the route 5 based on the geographical information 21, the mobile body 1 generates the route 5 by identifying the danger area 3 as an area to be avoided, thereby easily generating the route 5 that avoids interference with obstacles in the environment. As a result, the computational load, i.e., the amount of processing, of the mobile body 1 when generating the route 5 can be reduced.
[0162]
[12] In the driving assistance system 1000 described in
[11] , the generator 67 determines the degree of avoidance necessity of the danger area 3 based on the monitoring information and adds the degree of avoidance necessity to the map 2, and the moving body 1 generates the route 5 based on the degree of avoidance necessity in addition to the geographical information 21 and the danger area 3.
[0163] According to this configuration, the mobile body 1 can prioritize and handle the danger areas 3 based on the degree of necessity of avoidance, rather than treating all danger areas 3 uniformly, when generating the route 5. For example, the mobile body 1 can determine whether to avoid the danger area 3 or how far to move away from the danger area 3 based on the degree of necessity of avoidance, and generate the route 5 that flexibly reflects the danger areas 3.
[0164]
[13] In the driving assistance system 1000 described in
[11] or
[12] , the generator 67 determines the degree of avoidance necessity within the danger area 3 based on the monitoring information and adds the degree of avoidance necessity to the map 2, and the moving body 1 generates the route 5 based on the degree of avoidance necessity in addition to the geographical information 21 and the danger area 3.
[0165] According to this configuration, different degrees of necessity of avoidance can be assigned within the danger area 3. In this case, the mobile body 1 can determine which parts of one danger area 3 should be avoided with priority and which parts should not, based on the degrees of necessity of avoidance. As a result, the mobile body 1 can generate a route 5 that flexibly reflects the danger area 3.
[0166]
[14] In the driving assistance system 1000 described in any one of
[11] to
[13] , the generator 67 classifies the danger area 3 into a first danger area 31 corresponding to a static obstacle and a second danger area 32 corresponding to a dynamic obstacle, and the moving body 1 generates the route 5 taking into account the second danger area 32 within a predetermined first range R1 from the current location and the first danger area 31 within the first range R1 and outside the first range R1.
[0167] This configuration allows the mobile body 1 to know whether the danger area 3 changes over time. Therefore, the mobile body 1 can appropriately reflect the danger area 3 in the generation of the route 5. Because the first danger area 31 does not change over time, the mobile body 1 generates the route 5 by taking into account the first danger area 31 inside and outside the first range R1. On the other hand, because the second danger area 32 may change over time, the mobile body 1 generates the route 5 by taking into account only the second danger area 32 within the first range R1. This is because if the distant second danger area 32 were taken into account, the second danger area 32 may have already moved by the time the mobile body 1 reaches the second danger area 32, and such consideration in the route generation would be wasted. By classifying the danger area 3 into the first danger area 31 and the second danger area 32, the mobile body 1 can generate the route 5 by appropriately reflecting the necessary danger area 3.
[0168]
[15] In the driving assistance system 1000 described in any one of
[11] to
[14] , the generator 67 obtains information on the movement of an obstacle underlying the danger area 3, i.e., obstacle information 25, based on the monitoring information, and adds the obstacle information 25 to the map 2. The moving body 1 then performs an avoidance operation to avoid a collision with the obstacle based on the obstacle information 25.
[0169] According to this configuration, the mobile body 1 can utilize the obstacle information 25 as necessary during autonomous driving. Even if a route 5 that avoids the dangerous area 3 is generated, an obstacle that needs to be avoided may appear during the actual movement of the mobile body 1. The mobile body 1 can determine the appearance of such an obstacle based on the obstacle information 25. If there is a possibility of a collision, the mobile body 1 can also perform an avoidance operation based on the obstacle information 25. In other words, by including the obstacle information 25 in the map 2, the mobile body 1 can take more urgent measures against the obstacle.
[0170]
[16] In the driving assistance system 1000 described in any one of
[11] to
[15] , the generator 67 obtains information on disturbances in the autonomous driving of the moving body 1, i.e., disturbance information 26, based on the monitoring information, and adds the disturbance information 26 to the map 2. The moving body 1 generates the route 5 based on the disturbance information 26 in addition to the geographical information 21 and the danger area 3.
[0171] According to this configuration, the moving body 1 can generate the route 5 by taking into consideration disturbances in addition to the danger area 3. The moving body 1 may be affected by disturbances when it moves. By generating the route 5 based on the map 2, the moving body 1 can generate the route 5 by taking into consideration the effects of disturbances.
[0172]
[17] The driving assistance system 1000 according to any one of
[11] to
[16] further comprises a monitoring device 8 that acquires the monitoring information within the environment and transmits the monitoring information to the map generating device 100, wherein the generator 67 sets a monitoring necessity degree for each position within the environment based on the monitoring information, adds the monitoring necessity degree to the map 2, and transmits the map 2 to the monitoring device 8, and the monitoring device 8 determines an area within the environment from which the monitoring information is to be acquired based on the monitoring necessity degree.
[0173] With this configuration, the monitoring device 8 can determine the points in the environment from which monitoring information should be acquired based on the degree of necessity for monitoring. In other words, the map generating device 100 can cause the monitoring device 8 to acquire monitoring information for necessary points by using the map 2.
[0174]
[18] In the driving assistance system 1000 described in
[17] , the monitoring device 8 includes a plurality of mobile monitoring devices, i.e., second monitoring devices 8B, and the second monitoring devices 8B determine the area within the environment from which the monitoring information is to be obtained based on the monitoring necessity level and the positions of the other second monitoring devices 8B.
[0175] With this configuration, the monitoring device 8 can determine the points in the environment from which monitoring information should be acquired based on the amount of time that has elapsed since the monitoring information was acquired. For example, the monitoring device 8 preferentially acquires monitoring information from points where a long amount of time has elapsed since the monitoring information was acquired. This allows the map generation device 100 to keep the overall monitoring information in the environment as up-to-date as possible.
[0176] 1000 Driving assistance system 100 Map generation device 1 Mobile body 2 Map 21 Geographical information 25 Obstacle information (information on the movement of an obstacle) 26 Disturbance information (information on a disturbance) 3 Danger area 31 First danger area 32 Second danger area 5 Route 65 Collector 67 Generator 8 Monitoring device 8B Second monitoring device (mobile monitoring device)
Claims
1. A map generating device comprising: a collector that collects environmental monitoring information; and a generator that generates a map for assisting automated driving of a mobile object based on the monitoring information, wherein the generator determines dangerous areas in the environment that should be avoided based on the monitoring information, and generates the map that includes geographical information and the dangerous areas.
2. A map generating device according to claim 1, wherein the generator determines the danger area corresponding to the movement of an obstacle contained in the environment based on the monitoring information.
3. A map generating device according to claim 2, wherein the generator determines the danger area corresponding to the movement of a set of multiple obstacles contained in the environment based on the monitoring information.
4. A map generating device according to claim 1, wherein the generator determines the degree of necessity of avoidance of the dangerous area based on the monitoring information, and adds the degree of necessity of avoidance to the map.
5. A map generating device according to claim 1, wherein the generator determines the degree of necessity of avoidance within the danger area based on the monitoring information, and adds the degree of necessity of avoidance to the map.
6. A map generating device according to claim 1, wherein the generator classifies the danger areas into first danger areas corresponding to static obstacles and second danger areas corresponding to dynamic obstacles.
7. A map generating device according to claim 1, wherein the generator determines information about the movement of an obstacle underlying the danger area based on the monitoring information, and adds the information about the movement of the obstacle to the map.
8. A map generating device according to claim 1, wherein the generator obtains information about a disturbance in the autonomous driving of the mobile body based on the monitoring information, and adds the information about the disturbance to the map.
9. A map generating device according to claim 1, wherein said generator sets a degree of monitoring necessity for each position in the environment based on said monitoring information, and adds said degree of monitoring necessity to said map.
10. A map generating device according to claim 9, wherein the generator sets the degree of necessity of monitoring according to the time elapsed since the monitoring information was acquired at each position in the environment.
11. A driving assistance system comprising: a mobile body that performs autonomous driving; and a map generation device that generates a map to assist the autonomous driving of the mobile body, wherein the map generation device has a collector that collects environmental monitoring information and a generator that generates the map based on the monitoring information, wherein the generator determines dangerous areas in the environment that should be avoided based on the monitoring information, generates the map including geographical information and the dangerous areas, and transmits the map to the mobile body, and wherein the mobile body generates a route for autonomous driving based on the geographical information and the dangerous areas.
12. A driving assistance system as described in claim 11, wherein the generator determines the degree of avoidance necessity of the dangerous area based on the monitoring information and adds the degree of avoidance necessity to the map, and the mobile body generates the route based on the degree of avoidance necessity in addition to the geographical information and the dangerous area.
13. A driving assistance system as described in claim 11, wherein the generator determines the degree of avoidance necessity within the danger area based on the monitoring information and adds the degree of avoidance necessity to the map, and the mobile body generates the route based on the degree of avoidance necessity in addition to the geographical information and the danger area.
14. A driving assistance system as described in claim 11, wherein the generator classifies the danger areas into first danger areas corresponding to static obstacles and second danger areas corresponding to dynamic obstacles, and the moving body generates the route taking into account the second danger areas within a predetermined range from the current location, and taking into account the first danger areas both within and outside the range.
15. A driving assistance system as described in claim 11, wherein the generator obtains information regarding the movement of an obstacle underlying the danger area based on the monitoring information, adds the information regarding the movement of the obstacle to the map, and the mobile body performs an avoidance operation to avoid a collision with the obstacle based on the information regarding the movement of the obstacle.
16. A driving assistance system as described in claim 11, wherein the generator obtains information regarding disturbances to the autonomous driving of the mobile body based on the monitoring information and adds the information regarding the disturbances to the map, and the mobile body generates the route based on the information regarding the disturbances in addition to the geographical information and the dangerous areas.
17. A driving assistance system as described in claim 11, further comprising a monitoring device that acquires the monitoring information within the environment and transmits the monitoring information to the map generation device, wherein the generator sets a monitoring necessity level for each position within the environment based on the monitoring information, adds the monitoring necessity level to the map, and transmits the map to the monitoring device, and the monitoring device determines an area within the environment from which to acquire the monitoring information based on the monitoring necessity level.
18. A driving assistance system as described in claim 17, wherein the monitoring device includes a plurality of mobile monitoring devices, and the mobile monitoring devices determine the area within the environment from which to obtain the monitoring information based on the degree of monitoring necessity and the positions of the other mobile monitoring devices.
