Information processing device and information processing method

Abstract

This invention provides an information processing device and information processing method that can extract potential hazardous locations where the conditions of the event data are likely to occur by performing a search based on road features as well as road shapes related to the events that have occurred. [Solution] The information processing device 100 includes a storage unit 102 that stores environmental information including map information, an input unit 103 that inputs event information, a first extraction unit 5 that uses the event information to extract environmental information around the event occurrence point from the environmental information stored in the storage unit 102, a generation unit 6 that uses the event information input by the input unit 103 to generate the movement of the event's moving object, a second extraction unit 7 that extracts environmental information within the range of the movement of the moving object generated by the generation unit 6 from the environmental information stored in the storage unit 102, a specification unit 8 that combines the environmental information extracted by the first extraction unit 5 and the environmental information extracted by the second extraction unit 7 to identify similar locations that are similar to the combined environmental information but different from the event occurrence point, and an output unit 104 that outputs the similar locations identified by the specification unit 8.

Description

Technical Field

[0001] The present invention relates to an information processing apparatus and an information processing method, and more particularly to an information processing apparatus and an information processing method suitable for searching for similar potential dangerous points.

Background Art

[0002] As an example of safe driving support, based on a report or a drive recorder video that summarizes the occurrence status and countermeasures of an accident or a near miss that occurred at a certain point, safe driving education is implemented for the driver. Also, a technique for alerting a driver when driving near a past accident occurrence point is known. By using such safe driving education and techniques, the occurrence of accidents can be prevented in advance. For example, in Patent Document 1, an apparatus is disclosed that collects event information related to road traffic, extracts points having the same road characteristics as potential dangerous points based on road characteristics such as the number of lanes, speed limit, visibility, and presence or absence of sidewalks included in the event information, and notifies the driver.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Utilizing the technology of Patent Document 1, similar potential dangerous points can be extracted from an arbitrary range in a different area from the accident occurrence point based on simple road characteristics such as the number of lanes, speed limit, visibility, and presence or absence of sidewalks. However, as similar potential dangerous points to be extracted, many points having the corresponding road characteristics are extracted, and an alert notification is given to the driver even at a point where the possibility of a similar event occurring is low, which may lead to a decrease in the effect of supporting the driver's safe driving. [[ID= 41]]

[0005] Therefore, the present invention provides an information processing device and an information processing method that can extract potential hazardous locations where the conditions of the event data are likely to occur by performing a search based on the road shape related to the event that occurred, in addition to the road characteristics, when transferring the driver's experience. [Means for solving the problem]

[0006] To solve the above problems, the information processing apparatus according to the present invention is characterized by comprising: a storage unit for storing environmental information including map information; an input unit for inputting event information; a first extraction unit for using the event information to extract environmental information around the location where the event occurred from the environmental information stored in the storage unit; a generation unit for generating the movement of the moving object of the event using the event information input by the input unit; a second extraction unit for extracting environmental information within the range of the movement of the moving object generated by the generation unit from the environmental information stored in the storage unit; a specification unit for combining the environmental information extracted by the first extraction unit and the environmental information extracted by the second extraction unit to identify similar locations that are similar to the combined environmental information but different from the location where the event occurred; and an output unit for outputting the similar locations identified by the specification unit.

[0007] Furthermore, the present invention relates to an information processing apparatus comprising a storage unit for storing environmental information including map information, an input unit, a first extraction unit, a generation unit, a second extraction unit, and a specification unit, wherein the input unit inputs event information, the first extraction unit uses the event information to extract environmental information around the event occurrence point from the environmental information stored in the storage unit, the generation unit uses the event information input from the input unit to generate the movement of the event's moving body, the second extraction unit extracts environmental information within the range of the movement of the moving body generated by the generation unit from the environmental information stored in the storage unit, and the specification unit combines the environmental information extracted by the first extraction unit and the environmental information extracted by the second extraction unit to identify a similar location that is similar to the combined environmental information but different from the event occurrence point. [Effects of the Invention]

[0008] According to the present invention, when transferring the driver's experience, it is possible to provide an information processing device and an information processing method that can extract potential hazardous locations where the conditions of the event data are likely to occur by performing a search based on the road shape related to the event that occurred in addition to the road characteristics. Other issues, configurations, and effects not mentioned above will be clarified by the following description of the embodiments. [Brief explanation of the drawing]

[0009] [Figure 1] This is a schematic diagram of the information processing device according to Embodiment 1 of the present invention. [Figure 2] This is a functional block diagram of the information processing device according to Embodiment 1 of the present invention. [Figure 3] Figure 2 shows a functional block diagram of the first extraction unit and an example of information acquired by the first extraction unit. [Figure 4] Figure 2 shows a functional block diagram of the generation unit and an example of information acquired by the generation unit. [Figure 5] Figure 4 is a functional block diagram of the mobile body motion analysis unit that constitutes the generation unit. [Figure 6] Figure 2 shows a functional block diagram of the second extraction unit and an example of information acquired by the second extraction unit. [Figure 7] Figure 2 shows a functional block diagram of a specific part and an example of a road shape template using that specific part. [Figure 8] This is a flowchart showing the overall processing flow of the information processing device according to Embodiment 1 of the present invention. [Figure 9] This is a functional block diagram of the information processing device according to Embodiment 2 of the present invention. [Figure 10] Figure 9 shows a functional block diagram of the traffic information analysis unit and an example of the traffic information analysis results performed by the traffic information analysis unit. [Figure 11] Figure 9 is a functional block diagram of a specific part. [Figure 12]FIG. 11 is a diagram showing an example of a road shape template by an event occurrence related road shape acquisition unit constituting a specific part shown in FIG. 11 and a traffic information template by a traffic information analysis unit.

MODE FOR CARRYING OUT THE INVENTION

[0010] In this specification, the term "event" means an event such as an accident or a near miss that occurred at a certain location. Hereinafter, embodiments of the present invention will be described with reference to the drawings.

EXAMPLE

[0011] FIG. 1 is a schematic configuration diagram of an information processing apparatus according to Embodiment 1 of the present invention. As shown in FIG. 1, the information processing apparatus 100 has a function of extracting, from a different region, a point where an event similar to an event that occurred in the past may occur based on the environment information database 1 and the event information database 2 held in the data storage unit 200. When performing driving education in real time during a driver's operation, the point output from the information processing apparatus 100 is displayed when the driving vehicle approaches a similar point or the like based on the direct or indirect output of the vehicle sensor 300.

[0012] The information processing apparatus 100 shown in FIG. 1 is a computer including a processing unit 101 such as a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), or a GPGPU (General-purpose computing on graphics processing units), a storage unit 102 such as a semiconductor memory, an input unit 103 such as a removable semiconductor memory or a touch panel, an output unit 104 such as a display unit such as a liquid crystal display or an organic EL display and / or an audio output unit that outputs voice guidance or a message by voice, and a communication unit 105 that communicates with the vehicle sensor 300 and the data storage unit 200. Then, by executing a predetermined program by the processing unit 101, each function described later is realized.

[0013] Figure 2 is a functional block diagram of an information processing device according to Embodiment 1 of the present invention. As shown in Figure 2, the information processing device 100 includes a communication unit 105 that can input environmental information and event information stored in the environmental information database 1 and event information database 2 which constitute the data storage unit 200, a storage unit 102, an input unit 103, a first extraction unit 5, a generation unit 6, a second extraction unit 7, a specification unit 8, and an output unit 104. Here, the first extraction unit 5, the generation unit 6, the second extraction unit 7, and the specification unit 8 constitute the processing unit 101. Furthermore, the first extraction unit 5, the generation unit 6, the second extraction unit 7, and the specification unit 8 are realized when a processor such as a CPU reads various programs pre-stored in ROM (Read Only Memory) (not shown), expands them into RAM (Random Access Memory) which is a main memory device (not shown), and the processor executes the various programs.

[0014] The environmental information database 1, which constitutes the data storage unit 200 shown in Figure 2, is a database of maps and other data showing road structures consisting of nodes and links. The event information database 2 is a database of explanatory text describing the movements of related vehicles, people, and other moving objects, road width, angle, presence and location of crosswalks, visibility, surrounding structures, and other environmental conditions when traffic accidents or near misses occur, as well as sensor data such as cameras and LiDAR (Light Detection And Ranging) acquired when the event occurs. The data storage unit 200 is implemented by a server such as a cloud. The storage unit 102 stores the latitude and longitude of road nodes, link length, width, and the latitude and longitude of traffic facilities such as crosswalks and traffic lights from the environmental information database 1 via the communication unit 105. The input unit 103 accesses the event information database 2 via the communication unit 105, selects predetermined event information such as traffic accidents and near misses that it wants to search for as similar potential hazard locations from the event information database 2, and inputs it to the first extraction unit 5 and the generation unit 6. The first extraction unit 5 extracts the latitude and longitude of the location where the event occurred from the event information input by the input unit 103, and obtains the surrounding road shape of the location where the event occurred from the environmental information stored in the storage unit 102. The generation unit 6 analyzes the movement of the moving object from the event information input by the input unit 103 and generates the trajectory that the moving object passed through. The second extraction unit 7 extracts the road shape around the trajectory of the moving object generated by the generation unit 6 from the environmental information stored in the storage unit 102. The identification unit 8 obtains the road shape related to the event occurrence by integrating the environmental information extracted by the first extraction unit 5 and the second extraction unit 7, and searches for environmental information with a shape similar to the obtained road shape from the environmental information stored in the storage unit 102. The output unit 104 outputs the location having the road shape related to the event occurrence that the identification unit 8 has searched for. The details of each function of the first extraction unit 5, generation unit 6, second extraction unit 7, identification unit 8, and output unit 104 will be described below.

[0015] FIG. 3 is a diagram showing a functional block diagram of the first extraction unit shown in FIG. 2 and an example of acquired information by the first extraction unit. As shown in the upper diagram of FIG. 3, the first extraction unit 5 includes an event occurrence location latitude / longitude information acquisition unit 51 that acquires latitude / longitude information of the location where an event has occurred, an event occurrence location feature point acquisition unit 52 that acquires feature points indicating road information such as intersections, turning angles, and dead ends that are most relevant to the occurrence of the event, and an event occurrence location surrounding road shape acquisition unit 53 that acquires the surrounding road shape of the location where the event has occurred. As shown in the lower diagram of FIG. 3, the acquired information 54 by the first extraction unit 5 includes the latitude / longitude information 55 acquired by the event occurrence location latitude / longitude information acquisition unit 51, the feature points 56, 57a to 57d acquired by the event occurrence location feature point acquisition unit 52, the feature points 57a to 57d connected to the feature point 56, and the road shape 58 acquired by the event occurrence location surrounding road shape acquisition unit 53.

[0016] In the first extraction unit 5, first, the event occurrence location latitude / longitude information acquisition unit 51 acquires the latitude / longitude information 55 of the location where the event has occurred from the event information input from the input unit 103. Note that the method for acquiring the latitude / longitude information 55 by the event occurrence location latitude / longitude information acquisition unit 51 may be to analyze the text data included in the event information, or to use the information acquired from vehicle sensors 300 such as GPS (Global Positioning System) included in the event information, and is not particularly limited.

[0017] Next, the event occurrence location feature point acquisition unit 52 uses the latitude and longitude information 55 acquired by the event occurrence location latitude and longitude information acquisition unit 51 and the environmental information stored in the storage unit 102 to extract the nearest feature point 56 from the latitude and longitude information 55 acquired from the event occurrence location latitude and longitude information acquisition unit 51 as road information expected to be related to the event occurrence. In this embodiment, the above feature point is a point that represents a road intersection or endpoint, for example, an intersection, a corner, a dead end, etc., but is not particularly limited as long as it relates to road features. The method of extracting the feature point 56 in the event occurrence location feature point acquisition unit 52 may be a method of analyzing the text data of the event information acquired from the input unit 103 and extracting the latitude and longitude information of a location expected to be highly related to the event occurrence as a feature point, or a method of extracting the nearest feature point from the above latitude and longitude information, and is not particularly limited.

[0018] Finally, the event occurrence point surrounding road shape acquisition unit 53 acquires feature points 57a to 57d connected to the feature point 56 acquired by the event occurrence point feature point acquisition unit 52, and acquires the structure consisting of each feature point and the road information connecting them as the surrounding road shape 58. The method for selecting the connecting feature points 57a to 57d is not particularly limited and may include selecting feature points adjacent to feature point 56 or selecting feature points based on a predetermined distance threshold from feature point 56. Through the above processing, the first extraction unit 5 can acquire the road shape around the event occurrence point, and in this example, it can acquire the road shape 58 acquired by the event occurrence point surrounding road shape acquisition unit 53 as road shape information for a crossroads.

[0019] Figure 4 shows a functional block diagram of the generation unit shown in Figure 2 and an example of information acquired by the generation unit. As shown in the upper part of Figure 4, the generation unit 6 has a mobile body motion analysis unit 61 that analyzes the movement of a moving body before and after an event such as turning left or right, and a mobile body trajectory generation unit 62 that generates a series of travel trajectories of the moving body before and after the event. The lower left part of Figure 4 shows an example of a mobile body trajectory 63 generated by the generation unit 6. The mobile body trajectory 63 has a self-vehicle trajectory 64 generated by the mobile body trajectory generation unit 62 and a trajectory of another vehicle 65 generated by the mobile body trajectory generation unit 62. First, in the generation unit 6, the mobile body motion analysis unit 61 analyzes the text data contained in the event information input by the input unit 103 using LLM (Large language Models), etc., to identify moving bodies such as vehicles and people involved in the event that occurred, and analyzes the movement of each moving body before and after the event. Here, movement refers to general behavior of a moving body such as going straight, turning left, or turning right. Next, the mobile body trajectory generation unit 62 generates a mobile body trajectory based on the motion analyzed by the mobile body motion analysis unit 61, representing the path the mobile body took to reach the latitude and longitude information 55 of the event occurrence point, and finally outputs it as mobile body trajectory information. In the example of a mobile body trajectory 63 generated by the generation unit 6 shown in the lower left figure of Figure 4, there are two mobile bodies involved in the event: the own vehicle and another vehicle. The mobile body trajectory generation unit 62 outputs the own vehicle trajectory 64 generated by the mobile body trajectory generation unit 62, and the other vehicle trajectory 65 generated by the mobile body trajectory generation unit 62, which is a trajectory in which the other vehicle makes two left turns, as mobile body trajectories 63 generated by the generation unit 6. Here, as shown in the lower center figure of Figure 4, only the "straight" self-vehicle trajectory 64 generated by the mobile body trajectory generation unit 62 is stored in, for example, the memory unit 102. Furthermore, as shown in the lower right diagram of Figure 4, the other vehicle tracks 65 generated by the moving vehicle track generation unit 62 are stored in the memory unit 102 in chronological order, for example, as "straight," "left turn," "straight," "left turn," and "straight."

[0020] Figure 5 is a functional block diagram of the mobile motion analysis unit that constitutes the generation unit shown in Figure 4. As shown in Figure 5, the mobile motion analysis unit 61 that constitutes the generation unit 6 has an image analysis unit 611, a GPS position analysis unit 612, and a text analysis unit 613. The mobile motion analysis unit 61 analyzes the movement of a mobile object before and after an event related to the occurrence of an event, based on the event information input by the input unit 103. However, the mobile motion analysis unit 61 does not necessarily need to have all of these functions; it may use multiple functions in combination or just one function, as long as it can acquire the movement of the mobile object. The image analysis unit 611 assumes that the input unit 103 has input camera images attached to the vehicle or security camera images, and performs image recognition such as object recognition and road detection on each image to analyze the movement of the vehicle and other vehicles. The GPS position analysis unit 612 assumes that the input unit 103 has input the GPS positions of the vehicle and other vehicles, and analyzes the movement of the vehicle and other vehicles from the GPS data. The text analysis unit 613 assumes that text data describing the traffic event has been input to the input unit 103, and analyzes the actions of the vehicle and other vehicles from the text data using LLM or the like. The method used by the mobile body motion analysis unit 61 to analyze the actions of mobile bodies is not particularly limited to methods that can estimate the actions of mobile bodies caused by the event information, as well as the general behavior and order of each mobile body, in addition to methods that analyze image information, GPS information, and text data individually or in combination.

[0021] Figure 6 shows a functional block diagram of the second extraction unit shown in Figure 2 and an example of information acquired by the second extraction unit. As shown in the upper part of Figure 6, the second extraction unit 7 has a mobile body trajectory map information fitting unit 71 that fits the trajectory of the mobile body onto the map, and a mobile body operating range road shape acquisition unit 72 that acquires the road shape around the trajectory of the mobile body, which is the range that was not extracted by the first extraction unit 5. As shown in the lower part of Figure 6, the information 73 acquired by the second extraction unit 7 includes the vehicle's own track 74 adapted by the mobile vehicle track map information adaptation unit 71, the other vehicle's track 75 adapted by the mobile vehicle track map information adaptation unit 71, latitude and longitude information 76 acquired by the event occurrence point latitude and longitude information acquisition unit 51 (Figure 3) which constitutes the first extraction unit 5, a feature point 77 acquired by the event occurrence point feature point acquisition unit 52 (Figure 3), feature points 78a to 78d connected to feature point 77, of which feature point 78d is a feature point on the mobile vehicle track, feature points 79a to 79b adjacent to the feature point on the mobile vehicle track, and road shape 710 acquired by the mobile vehicle operating range road shape acquisition unit 72.

[0022] In the second extraction unit 7, first, the mobile body trajectory map information adaptation unit 71 searches for a road shape that realizes the trajectory of the mobile body generated by the generation unit 6 from among the surrounding environment information of the location where the event occurred, which is stored in the storage unit 102, and adapts the trajectory of the mobile body to that road shape. At this time, the surrounding environment information of the location where the event occurred is stored as a structure consisting of nodes and links, and the trajectory of the mobile body is also stored as a structure having length and direction. Feature points on that trajectory are extracted. Next, the mobile body operating range road shape acquisition unit 72 acquires the structure consisting of the feature points on the trajectory extracted by the mobile body trajectory map information adaptation unit 71 and the feature points adjacent to those feature points as the road shape 710 acquired by the mobile body operating range road shape acquisition unit 72. In the second extraction unit 7, through the processing described above, it is possible to obtain road shapes around the vehicle's own track 74 and other vehicle tracks 75, which are adapted by the mobile vehicle track map information adaptation unit 71 and are highly likely to be related to the occurrence of events that could not be extracted by the first extraction unit 5. In this example, it is possible to obtain road shape information that branches into two from feature point 78d connected to feature point 77, such as the road shape 710 obtained by the mobile vehicle operating range road shape acquisition unit 72.

[0023] Figure 7 shows a functional block diagram of the specific unit shown in Figure 2 and an example of a road shape template created by the specific unit. As shown in the upper part of Figure 7, the specific unit 8 includes an event occurrence-related road shape acquisition unit 81 that acquires the road shape at the location where the event occurred, and a similar road shape location search unit 82 that searches for road shapes from other regions where similar events are likely to occur. The lower part of Figure 7 shows an example of a road shape 83 acquired by the event occurrence-related road shape acquisition unit 81, and a road shape template 84 that shows image features patterned from the road shape acquired by the event occurrence-related road shape acquisition unit 81.

[0024] In the specific unit 8, first, the event occurrence-related road shape acquisition unit 81 integrates the surrounding road shape extracted by the first extraction unit 5 and the road shape within the operating range of the mobile body extracted by the second extraction unit 7 to acquire the road shape related to the event occurrence. At this time, each road shape is represented by a structure consisting of nodes and links, and is acquired as a composite structure. The compositing method is to combine the road shapes so that identical nodes overlap based on the coordinates of the nodes included in each road shape. In this example, the road shape 58 acquired by the event occurrence point surrounding road shape acquisition unit 53 (Figure 3) of the intersection extracted by the first extraction unit 5 and the road shape 710 acquired by the mobile body operating range road shape acquisition unit 72 (Figure 6) of the fork extracted by the second extraction unit 7 can be combined to obtain the road shape 83 acquired by the event occurrence-related road shape acquisition unit 81. Next, a road shape template 84 including road edge information, etc., is created from the acquired road shape 83. The method for creating the road shape template 84 is not particularly limited and includes methods such as creating an image as the road shape template 84 in which the road shape is drawn within a rectangle and the road shape is represented as edge information, as in this example; creating the road shape template 84 by cropping an aerial photograph of the vicinity of the event occurrence from the above environmental information and latitude and longitude information 55; or creating the road shape template 84 by treating the road shape template 84 as a matrix corresponding to the size of the rectangle surrounding the road shape and distinguishing between road areas and crosswalk areas based on the values ​​of the elements in the matrix, thereby creating information in which the road shape is represented numerically as the road shape template 84. Next, the similar road shape location search unit 82 extracts environmental information for a predetermined area from the environmental information stored in the storage unit 102, divides it into multiple road shape templates using the same method as the road shape template 84 creation method, and searches for similar road shapes between these and the road shape template 84 acquired by the event occurrence related road shape acquisition unit 81 using a general pattern matching method with images and strings as features. Furthermore, when creating multiple road shape templates 84 from predetermined environmental information, the size of the road shape template 84 is not particularly limited; for example, it may be set based on the size of the road shape template 84 acquired by the event occurrence-related road shape acquisition unit 81.Furthermore, the method for determining a predetermined area from environmental information is not particularly limited and may include methods such as the user specifying a desired area to search for similar potential hazard locations using a GUI (Graphical User Interface), determining the area based on information such as the user's daily traffic area, or determining the area based on the driving route from the user's current location to the destination. The method for dividing the environmental information is to divide the environmental information so that it is of the same size as the road shape template 84 acquired by the event occurrence-related road shape acquisition unit 81. Through the processing described above, the identification unit 8 can search for locations with road shapes similar to the event occurrence location from within other regions.

[0025] Figure 8 is a flowchart showing the overall processing flow of an information processing device according to Embodiment 1 of the present invention. As shown in Figure 8, first, in step S1, the input unit 103 of the information processing device 100 selects an event for which it wants to search for a location where a similar event occurs, and extracts one event information from the event information database 1. Next, in step S2, the first extraction unit 5 of the information processing device 100 obtains the road shape around the event occurrence location from the environmental information database 1. Next, in step S3, the generation unit 6 of the information processing device 100 analyzes the movement of a moving object related to the event occurrence from the event information database 2 and generates the trajectory of the moving object. Next, in step S4, the second extraction unit 7 of the information processing device 100 uses the generated trajectory of the moving object to obtain the road shape around the trajectory of the moving object from the environmental information database 1. Next, in step S5, the identification unit 8 of the information processing device 100 synthesizes the road shapes obtained in steps S2 and S4 to obtain the road shape related to the event occurrence. Next, in step S6, the identification unit 8 specifies the range in which it wants to search for locations where similar events occur, and searches the environmental information database 1 for locations with similar road shapes. Finally, in step S7, the identification unit 8 determines whether processing has been completed for all the events to be searched for. If the result of the determination is that processing has not been completed for all items, the process returns to step S1 and is repeated. On the other hand, if the result of the determination is that processing has been completed for all items, the process is terminated.

[0026] As described above, according to this embodiment, when transferring the driver's experience, it is possible to provide an information processing device and information processing method that can extract potential dangerous locations where the conditions of the event data are likely to occur by performing a search based on the road shape related to the event that occurred in addition to the road characteristics. Furthermore, by providing warnings to drivers at locations where similar incidents are likely to occur, the effectiveness of safe driving support can be improved. [Examples]

[0027] Figure 9 is a functional block diagram of an information processing device according to Embodiment 2 of the present invention. As shown in Figure 9, the information processing device 100a according to this embodiment differs from Embodiment 1 in that it includes a traffic information analysis unit 91, and a driver database 92 and a vehicle database 93. In the following, the same reference numerals are used for the same components as in Embodiment 1.

[0028] As shown in Figure 9, the information processing device 100a according to this embodiment includes a communication unit 105, a storage unit 102, an input unit 103, a first extraction unit 5, a generation unit 6, a second extraction unit 7, a specific unit 8, an output unit 104, and a traffic information analysis unit 91, which can input environmental information and event information stored in the environmental information database 1 and event information database 2 that constitute the data storage unit 200. Here, the first extraction unit 5, the generation unit 6, the second extraction unit 7, the specific unit 8', and the traffic information analysis unit 91 constitute the processing unit 101. Furthermore, the first extraction unit 5, the generation unit 6, the second extraction unit 7, the specific unit 8', and the traffic information analysis unit 91 are realized when a processor such as a CPU reads various programs pre-stored in ROM (Read Only Memory) (not shown), expands them into RAM (Random Access Memory) which is a main memory device (not shown), and the processor executes the various programs.

[0029] Based on the road shape around the event occurrence point extracted by the road shape acquisition unit 53 (Figure 3) which constitutes the first extraction unit 5, the road shape within the operating range of a moving object extracted by the road shape acquisition unit 72 which constitutes the operating range of a moving object, traffic information analyzed by the traffic information analysis unit 91, the driver database 92, and the driver's vehicle database 93, the identification unit 8' searches for similar locations. In Figure 9, the environmental information database 1, event information database 2, storage unit 102, input unit 103, first extraction unit 5, generation unit 6, second extraction unit 7, and output unit 9 have the same or similar functions as in Embodiment 1. The traffic information analysis unit 91 analyzes the presence and location of traffic facilities such as traffic signals and pedestrian crossings, and analyzes the movement of surrounding moving objects that are not directly related to the event. The driver database 92 is a database that holds unique driving rules for drivers using the information processing device 100a, such as truck drivers taking routes that minimize right turns and make left turns as much as possible, and novice drivers taking routes that avoid narrow roads and only use main roads. The driver database 93 is a database that holds driver information such as the type of vehicle driven by the driver using the information processing device 100a and whether or not it has ADAS (Advanced Driver-Assistance Systems) functions. The identification unit 8', which constitutes the information processing device 100a, has the function of using the input information to search for locations with similar road shapes in different regions. The traffic information analysis unit 91 and the identification unit 8' will be described below.

[0030] Figure 10 shows a functional block diagram of the traffic information analysis unit shown in Figure 9 and an example of the traffic information analysis results by the traffic information analysis unit. As shown in the upper part of Figure 9, the traffic information analysis unit 91 analyzes the movements of surrounding moving objects unrelated to the event before and after the event, as well as the presence and location of traffic facilities such as signals and pedestrian crossings, from the event information input by the input unit 103. The traffic information analysis unit 91 has an image analysis unit 911 and a text analysis unit 912. However, the traffic information analysis unit 91 does not necessarily need to have all of these functions; it may use multiple functions in combination or just one function as long as it can acquire the movement of moving objects. The image analysis unit 911 assumes that the input unit 103 has received camera images from its own vehicle or security camera images, and performs image recognition such as object recognition and road detection on each image to analyze the movement of surrounding moving objects and the location of traffic facilities. The text analysis unit 912 assumes that text data describing the situation of a traffic event has been input to the input unit 103, and analyzes the movement of surrounding moving objects and the location of traffic equipment from the text data using LLM or the like. The method used by the traffic information analysis unit 91 to analyze the movement of surrounding moving objects and the location of traffic equipment is not particularly limited to methods that can estimate the general behavior of moving objects and the location of traffic equipment based on event information, in addition to methods that analyze image information and text data individually or in combination. The traffic information analysis unit 91 uses the image analysis unit 911 and the text analysis unit 912 to output traffic information analysis results 913 showing the movement of surrounding moving objects and the location of traffic equipment, for example, as shown in the lower figure of Figure 10.

[0031] Figure 11 is a functional block diagram of the specific unit shown in Figure 9, and Figure 12 is a diagram showing an example of a road shape template by the event occurrence-related road shape acquisition unit and a traffic information template by the traffic information analysis unit that constitute the specific unit shown in Figure 11. As shown in Figure 11, the specific unit 8' includes an event occurrence-related road shape acquisition unit 81' that acquires the road shape of the location where an event occurred, a similar road shape location search unit 82 that searches for road shapes from other regions where similar events are likely to occur, and a final similar road shape location selection unit 85 that selects a location where a more similar event is likely to occur by utilizing the driver database 92 and the vehicle database 93.

[0032] As shown in Figure 12, the road shape template 84, which shows image features obtained by patterning the road shape acquired by the event occurrence-related road shape acquisition unit 81', includes the road shape 83 acquired by the event occurrence-related road shape acquisition unit 81'. First, the identification unit 8' uses the event occurrence-related road shape acquisition unit 81' to integrate the surrounding road shape extracted by the first extraction unit 5 and the road shape within the mobile body's operating range extracted by the second extraction unit 7, similar to Example 1, to acquire the road shape 83 related to the event occurrence and create a road shape template 84 that includes road edge information, etc. Next, it adds a traffic information template obtained by the traffic information analysis unit 91, which includes the operating paths of surrounding mobile bodies and the locations of traffic facilities such as traffic lights. Then, the similar road shape location search unit 82 searches for similar road shapes using a general pattern matching method that uses images and strings as features. Finally, the final similar road shape location selection unit 85 uses the driver database 92 and the vehicle database 93 to select locations where a more similar event is likely to occur. As a selection method, when using the driver database 92, one method is to select an appropriate driving route for the driver based on the driver database 92 and search for locations with similar road shapes within the range of that driving route. When using the driver's vehicle database 93, when the event occurrence-related road shape acquisition unit 81' outputs the road shape template 84, information on vehicles where an event occurred, previously extracted from the event information database 2 using LLM, etc., is added to the road shape template 84. By comparing this with the information on the driver's vehicle using the information processing device 100a, similar road shape locations where similar events are likely to occur in the driver's vehicle are selected. In the example shown in the lower left of Figure 12, the road shape template 84 is treated as a matrix corresponding to the size of the rectangle surrounding the road shape, and information in which the road shape is expressed numerically is created as the road shape template 84 by distinguishing the road area, the crosswalk area, etc., based on the values ​​of the elements in the matrix. Similarly, the traffic information template is treated as a matrix corresponding to the size of the rectangle surrounding the movement path of the surrounding moving object obtained by the traffic information analysis unit 91, and information in which the location of the surrounding moving object and traffic equipment such as traffic lights is expressed numerically is created as the traffic information template by distinguishing the location of the surrounding moving object and traffic equipment such as traffic lights based on the values ​​of the elements in the matrix.

[0033] As described above, according to this embodiment, in addition to the effects of Embodiment 1, the driving education device that transfers the driver's experience can output similar road shapes by narrowing down to locations where the situation of the event data can be easily reproduced, by performing a search based on the road shape related to the event, the movement of surrounding moving objects, and the location of traffic facilities. This makes it possible to extract potential hazardous locations with a higher accident prevention effect.

[0034] It should be noted that the present invention is not limited to the embodiments described above, and various modifications are included. For example, the embodiments described above are described in detail to make the present invention easier to understand, and are not necessarily limited to those having all the configurations described. Furthermore, it is possible to replace parts of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of one embodiment. [Explanation of Symbols]

[0035] 1…Environmental Information Database 2…Event Information Database 5...First extraction part 6...Generation section 7...Second extraction section 8,8'…Specific part 51...Event Occurrence Location Latitude and Longitude Information Acquisition Unit 52...Event occurrence location feature point acquisition unit 53...Road shape acquisition section around the incident location 54…Information obtained by the first extraction unit 55,76… Latitude and Longitude Information 56... Feature points acquired by the event occurrence location feature point acquisition unit 52 57a, 57b, 57c, 57d... Feature points connected to feature point 56 58...Road shape acquired by the road shape acquisition unit 53 around the event occurrence point 61...Mobile motion analysis section 62…Mobile trajectory generation unit 63...Trajectory of the moving object generated by the generation unit 6 64... Vehicle trajectory generated by the mobile trajectory generation unit 62 65...Other vehicle tracks generated by the mobile vehicle track generation unit 62 71... Mobile object trajectory map information compatibility section 72...Road shape acquisition unit for mobile body operation range 73... Information obtained by the second extraction unit 74...The vehicle's track adapted by the mobile vehicle track map information adaptation unit 71 75...Other vehicle tracks adapted by the mobile vehicle track map information adaptation unit 71 77... Feature points acquired by the event occurrence location feature point acquisition unit 52 78a, 78b, 78c, 78d... Feature points connected to feature point 77 79a, 79b, 79c, 79b... Feature points adjacent to feature points on the moving object's trajectory 81,81' ...Road shape acquisition section related to event occurrence 82…Similar road shape point search section 83...Road shape acquired by the road shape acquisition unit 81' related to the occurrence of an event 84...Road shape template 85...Final similar road shape point selection section 91…Traffic information analysis department 92... Driver Database 93... Driver Database 100,100a… Information Processing Device 101... Processing Section 102...Storage section 103...Input section 104...Output section 105... Communications Department 200...Data storage unit 300... Vehicle Sensor 611...Image Analysis Department 612...GPS position analysis section 613…Text analysis department 710...Road shape acquired by the road shape acquisition unit 72 within the range of motion of the mobile body 911...Image Analysis Department 912…Text analysis department 913…Traffic information analysis results

Claims

1. A memory unit that stores environmental information including map information, An input section for inputting event information, A first extraction unit extracts environmental information about the area surrounding the location where the event occurred from the environmental information stored in the storage unit, using the event information. A generation unit generates the movement of the moving object of the event using the event information input by the input unit, A second extraction unit extracts the environmental information within the range of motion of the mobile body generated by the generation unit from the environmental information stored in the storage unit, A special unit combines the environmental information extracted by the first extraction unit with the environmental information extracted by the second extraction unit, and identifies similar locations that are similar to the combined environmental information but different from the location where the event occurred. An information processing apparatus comprising an output unit that outputs the similar locations identified by the identification unit.

2. In the information processing apparatus according to claim 1, The first extraction unit is, An event occurrence location latitude and longitude information acquisition unit acquires the latitude and longitude information of the location where the event occurred, An event occurrence location feature point acquisition unit acquires feature points indicating road information of the intersection, corner, or dead end most relevant to the occurrence of the event, An information processing device characterized by having an event occurrence location surrounding road shape acquisition unit that acquires the surrounding road shape of the location where the event occurred.

3. In the information processing apparatus according to claim 2, The second extraction unit is, A mobile object trajectory map information matching unit that matches the trajectory of a mobile object onto a map, An information processing device characterized by having a mobile body operating range road shape acquisition unit that acquires the road shape around the trajectory of a mobile body, which is the range that was not extracted by the first extraction unit.

4. In the information processing apparatus described in claim 3, The generating unit is A mobile body motion analysis unit analyzes the movement of a moving object before and after an event occurs, An information processing apparatus characterized by having a moving object trajectory generation unit that generates a moving object trajectory, which is a series of moving object trajectories before and after an event occurs.

5. In the information processing apparatus according to claim 4, The aforementioned mobile body motion analysis unit is an information processing device characterized by performing motion analysis of a mobile body from event information obtained from an event information database by image analysis, GPS location information analysis, or text analysis.

6. In the information processing apparatus according to claim 5, The mobile body trajectory map information matching unit searches for a road shape that realizes the mobile body trajectory obtained from the mobile body trajectory generation unit and environmental information, and adjusts the length and direction of the mobile body trajectory to match the road shape to match the trajectory, thereby matching the trajectory.

7. In the information processing apparatus according to claim 6, The information processing device is characterized in that the mobile body operating range road shape acquisition unit acquires a structure consisting of feature points on the trajectory extracted by the mobile body trajectory map information matching unit and feature points adjacent to those feature points as the mobile body operating range road shape.

8. In the information processing apparatus according to claim 7, The specified part is, An event occurrence-related road shape acquisition unit acquires the road shape at the location where the event occurred, An information processing device characterized by having a similar road shape location search unit that searches for road shapes from other regions where similar events are likely to occur.

9. In the information processing apparatus according to claim 8, The event occurrence-related road shape acquisition unit is characterized by integrating the surrounding road shape extracted by the first extraction unit and the road shape within the operating range of the moving object extracted by the second extraction unit by overlapping the same nodes contained in each road shape, thereby acquiring the road shape related to the event occurrence.

10. In the information processing apparatus according to claim 9, The information processing device is characterized in that the event occurrence-related road shape acquisition unit creates a road shape template that includes the road shape and is represented by an edge image, an aerial photograph of the vicinity of the event occurrence, or a pattern matrix.

11. In the information processing apparatus according to claim 10, An information processing device further comprising a traffic information analysis unit that analyzes the presence or location of traffic facilities, including traffic signals or pedestrian crossings, and analyzes the movement of surrounding moving objects not directly related to the event.

12. An information processing method for an information processing apparatus comprising a storage unit for storing environmental information including map information, an input unit, a first extraction unit, a generation unit, a second extraction unit, and a specification unit, The aforementioned input unit inputs event information, The first extraction unit uses the event information to extract environmental information about the area surrounding the event occurrence point from the environmental information stored in the storage unit. The generation unit generates the movement of the moving object of the event using the event information input from the input unit. The second extraction unit extracts the environmental information within the range of motion of the moving body generated by the generation unit from the environmental information stored in the storage unit. An information processing method characterized in that the identifying unit combines the environmental information extracted by the first extraction unit and the environmental information extracted by the second extraction unit, and identifies similar locations that are similar to the combined environmental information but different from the location where the event occurred.

13. In the information processing method described in claim 12, The first extraction unit includes an event occurrence location latitude and longitude information acquisition unit, an event occurrence location feature point acquisition unit, and an event occurrence location surrounding road shape acquisition unit. The latitude and longitude information acquisition unit for the location where the event occurred acquires the latitude and longitude information of the location where the event occurred. The aforementioned event occurrence location feature point acquisition unit acquires feature points indicating road information of the intersection, corner, or dead end most relevant to the occurrence of the event. An information processing method characterized in that the unit for acquiring the road shape around the location where the event occurred acquires the road shape around the location where the event occurred.

14. In the information processing method described in claim 13, The second extraction unit comprises a mobile body trajectory map information matching unit and a mobile body operating range road shape acquisition unit. The aforementioned mobile object trajectory map information matching unit matches the trajectory of the mobile object onto the map, The information processing method is characterized in that the moving body operating range road shape acquisition unit acquires the road shape around the trajectory of the moving body, which is the range that was not extracted by the first extraction unit.

15. In the information processing method described in claim 14, The generation unit comprises a mobile body motion analysis unit and a mobile body trajectory generation unit. The aforementioned mobile body motion analysis unit analyzes the motion of the mobile body before and after the occurrence of an event, The information processing method is characterized in that the moving body trajectory generation unit generates a moving body trajectory, which is a series of trajectories of a moving body before and after an event occurs.

Images