Road condition data processing device and method

Abstract

To provide a road state data processing device and method.SOLUTION: A road state data processing method of one embodiment is implemented by a processor of a road state data processing device and comprises steps of: collecting sensing data to respective objects on a road from a plurality of sensors on the road; modeling a relation between the respective objects by a graph on the basis of the sensing data to the respective objects; constructing a space index of a grid base to a modeled result; removing overlapped sensing data among pieces of the sensing data to the respective objects included in the space index of the grid base; questioning registration for the respective objects from which the overlapped sensing data is removed to extract an object matching a response to a question; and outputting state recognition data to the object matching the response to the question.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a road condition data processing apparatus and method. The present invention has been supported by the research project of "Development of Advanced Road Condition Recognition Technology for Infrastructure Sensor Base" in the "Autonomous Driving Technology Reform New Business" of the Ministry of Land, Infrastructure and Transport of the Republic of Korea. (Project Specific Number: 1615011990)

Background Art

[0002] Recently, research related to ITS (intelligent transportation system) has been actively carried out and is helpful for constructing a next-generation traffic information system suitable for the information society. By constructing a system capable of sensing the speed of automobiles on the road and traffic information in real time and providing information to drivers, it is possible to have a good influence on the flow of the overall traffic situation.

[0003] The above background art is technical information that the inventor possessed for deriving the present invention or acquired in the process of deriving the present invention, and it cannot necessarily be said that it is a known technology publicly disclosed to the general public before the filing of the present invention.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] One problem of the present invention is to accurately recognize the road condition using data collected from various road infrastructure sensors and promote safe driving of automobiles based on this.

[0006] One objective of this invention is to identify the redundancy of data collected from various road infrastructure sensors, remove unnecessary data, and increase the speed of road condition processing.

[0007] One objective of this invention is to provide real-time situational awareness data necessary for each vehicle through real-time analysis of data collected from various road infrastructure sensors.

[0008] The problems that this invention aims to solve are not limited to those mentioned above. Other problems and advantages of this invention that are not mentioned can be understood from the following description and should be more clearly understood from the embodiments of this invention. Furthermore, it should be understood that the problems that this invention aims to solve and the advantages it offers can be realized by the means and combinations thereof described in the claims. [Means for solving the problem]

[0009] A road condition data processing method according to one embodiment of the present invention is a road condition data processing method performed by a processor of a device that processes road condition data, and includes the steps of: collecting sensing data for each object present on the road from a plurality of sensors provided on the road; modeling the relationships between each object in a graph based on the sensing data for each object; constructing a grid-based spatial index from the results of the graph modeling; removing duplicate sensing data from the sensing data for each object included in the grid-based spatial index; performing a previously registered question on each object from which the duplicate sensing data has been removed, and extracting objects that correspond to the response to the question; and outputting situation recognition data to the objects that correspond to the response to the question.

[0010] A road condition data processing device according to one embodiment of the present invention includes a processor and a memory operably connected to the processor and storing at least one code executed by the processor, wherein the memory can store code that, when executed through the processor, causes the processor to collect sensing data for each object present on the road from a plurality of sensors provided on the road, model the relationships between each object in a graph based on the sensing data for each object, construct a grid-based spatial index on the results of the graph modeling, remove duplicate sensing data from the sensing data for each object included in the grid-based spatial index, perform a previously registered query on each object from which the duplicate sensing data has been removed, extract objects that correspond to the response to the query, and output situation recognition data to the objects that correspond to the response to the query.

[0011] In addition, other methods, other systems for embodying the present invention, and computer-readable recording media on which computer programs for performing the above methods are stored can also be provided.

[0012] Other aspects, features, and advantages beyond those mentioned above will become clear from the following drawings, claims, and detailed description of the invention. [Effects of the Invention]

[0013] According to the present invention, road conditions can be accurately perceived using data collected from various road infrastructure sensors, and based on this, safe driving of automobiles can be promoted.

[0014] Furthermore, by identifying the redundancy of data collected from various road infrastructure sensors and removing unnecessary data, the speed of road condition processing can be increased.

[0015] Furthermore, by providing real-time situational awareness data necessary for each vehicle through real-time analysis of data collected from various road infrastructure sensors, it is possible to promote safe driving.

[0016] The effects of the present invention are not limited to those mentioned above, and other effects not mentioned should be clearly understood by those skilled in the art from the following description. [Brief explanation of the drawing]

[0017] [Figure 1] This is an illustrative diagram of a road condition processing environment including a sensor group, an object group, a road condition data processing device, and a network connecting them to one another, according to this embodiment. [Figure 2] This block diagram is shown to provide a schematic explanation of the configuration of the road condition data processing device according to this embodiment. [Figure 3a] This is an illustrative diagram shown to provide a schematic explanation of the graph modeling process according to this embodiment. [Figure 3b] This is an illustrative diagram shown to provide a schematic explanation of the graph modeling process according to this embodiment. [Figure 3c] This is an illustrative diagram shown to provide a schematic explanation of the graph modeling process according to this embodiment. [Figure 4a] This table shows the attributes of the nodes and edges included in the graph model according to this embodiment. [Figure 4b] This table shows the attributes of the nodes and edges included in the graph model according to this embodiment. [Figure 5a] This is an illustrative diagram shown to illustrate the removal of duplicate sensing data using a graph model and grid-based spatial index according to this embodiment. [Figure 5b] This is an illustrative diagram shown to illustrate the removal of duplicate sensing data using a graph model and grid-based spatial index according to this embodiment. [Figure 6]This is a table showing the connection relationship between the graph model according to this embodiment and the spatial index of the grid base. [Figure 7] This is a block diagram shown for schematically explaining the configuration of a road condition data processing device according to another embodiment. [Figure 8] This is a flowchart for explaining the road condition data processing method according to this embodiment.

Mode for Carrying Out the Invention

[0018] The advantages and features of the present invention, and the methods for achieving them, should become clear by referring to the embodiments described in detail together with the accompanying drawings. However, the present invention is not limited to the embodiments presented below, and can be embodied in various different forms, and it must be understood that the present invention includes any conversions, equivalents, or substitutes included within the spirit and technical scope of the present invention. The embodiments presented below are provided to make the disclosure of the present invention complete and to enable those with ordinary knowledge in the technical field to which the present invention pertains to fully understand the scope of the invention. When it is determined that a specific description of related known technologies in explaining the present invention may obscure the gist of the present invention, the detailed description thereof will be omitted.

[0019] The terms used in this application are merely used for explaining specific embodiments and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly has a different meaning. In this application, terms such as "including" or "having" are intended to specify the presence of features, numbers, steps, operations, components, parts, or combinations thereof described in the specification, and it must be understood that the presence or addition possibility of one or more other features, numbers, steps, operations, components, parts, or combinations thereof is not precluded in advance. Terms such as first, second, etc. can be used to describe various components, but the components should not be limited by the above terms. The above terms are only used for the purpose of distinguishing one component from another.

[0020] Furthermore, in this application, “part” may be a hardware component such as a processor or circuit, and / or a software component executed by a hardware component such as a processor.

[0021] Hereinafter, embodiments of the present invention will be described in detail with reference to the attached drawings. In the description with reference to the attached drawings, identical or corresponding components will be assigned the same drawing number, and redundant explanations will be omitted.

[0022] Figure 1 is an illustrative diagram of a road condition processing environment 1 according to this embodiment, and may include a sensor group 100, an object group 200, a road condition data processing device 300, and a network 400 connecting them to each other.

[0023] Referring to Figure 1, the sensor group 100 is installed on the road and can sense objects present on the road and generate sensing data. In this embodiment, the sensing data may include one or more of the following: the object's location, the object's heading, and the object's speed. Here, depending on the type of object, the object's location may be required to be included in the sensing data, while the object's heading and speed may be selectively included.

[0024] In this embodiment, the sensor group 100 may include a lidar 100_1, a camera 100_2, and a UWB radar 100_3, etc.

[0025] Lidar 100_1 can use laser light to sense objects present on a road and generate sensing data. Lidar 100_1 may include an optical transmitter (not shown), an optical receiver (not shown), and at least one processor (not shown) electrically connected to the optical transmitter and receiver, which processes the received signal and generates data about the object based on the processed signal. Lidar 100_1 can be implemented using either a time-of-flight (TOF) or phase-shift method. Lidar 100_1 can detect objects based on the TOF or phase-shift method using laser light as a medium, and generate sensing data such as the position of the detected object, the distance to the detected object, the relative velocity, and the direction of movement of the object.

[0026] Camera 100_2 can use video to detect objects on the road and generate sensing data. Camera 100_2 may include at least one lens, at least one image sensor (not shown), and at least one processor (not shown) electrically connected to the image sensor, which processes the received signal and generates data about the object based on the processed signal. Camera 100_2 may be at least one of a mono camera, a stereo camera, or an AVM (Around View Monitoring) camera. Camera 100_2 can use various video processing algorithms to generate sensing data such as the position of an object, the distance to a detected object, the relative velocity, and the direction of movement of an object. For example, camera 100_2 can generate sensing data such as the position of an object, the distance to a detected object, the relative velocity, and the direction of movement of an object from the acquired video based on the change in object size over time.

[0027] Radar 100_3 can use radio waves to detect objects on a road and generate sensing data. Radar 100_3 may include an electromagnetic wave transmitter (not shown), an electromagnetic wave receiver (not shown), and at least one processor (not shown) electrically connected to the electromagnetic wave transmitter and receiver, which processes the received signal and generates sensing data for the object based on the processed signal. Radar 100_3 can be implemented as either a pulse radar or a continuous wave radar based on its radio wave emission principle. Among continuous wave radars, Radar 100_3 can be implemented as either an FMCW (frequency modulated continuous wave) or FSK (frequency shift keyong) depending on the signal waveform. On the other hand, Radar 100_3 can be classified differently from each other depending on the detection distance. FM-CW radar is commonly used as a long-range detection radar, with an RF frequency in the 76GHz band and a detection distance range set from 4m to 120m. Furthermore, as a short-range sensing radar, an ultra-wideband (UWB) radar is used, with an RF frequency in the 24GHz band and a sensing range set to 0.1m to 20m. Radar 100_3 can detect objects using electromagnetic waves as a medium, based on the Time of Flight (TOF) method or the phase-shift method, and can generate sensing data such as the position of each object, the distance to the object, the relative velocity, and the direction of movement of the object.

[0028] In this embodiment, a lidar 100_1, a camera 100_2, and a radar 100_3 are disclosed as the sensor group 100, but the invention is not limited to these, and a variety of sensors can be used, such as a sensor for measuring road conditions (not shown), a sensor for measuring visible distance (not shown), a road surface sensor (not shown), and a weather sensor (not shown).

[0029] The object group 200 communicates with the road condition data processing device 300 via the network 400 and can receive condition recognition data for the road in motion from the road condition data processing device 300. In this embodiment, the object group 200 can include automobiles, pedestrians, motorcycles, bicycles, fallen objects, potholes, construction work, etc. In this embodiment, for the sake of explanation, the object will be limited to automobiles. Therefore, the object group 200 can include automobile groups 200_1 to 200_N, as shown in Figure 1. Also, in this embodiment, the terms object group, automobile group, automobile, pedestrian, motorcycle, bicycle, fallen objects, potholes, and construction work can be used to mean "object".

[0030] The road condition data processing device 300 processes sensing data collected from the sensor group 100 to accurately recognize the road conditions and, based on this, can promote the safe operation of the object group 200.

[0031] The road condition data processing device 300 can model the relationships between objects in a graph based on the sensing data for each object. The road condition data processing device 300 can construct a grid-based spatial index from the results of the graph modeling. The road condition data processing device 300 can remove duplicate sensing data from the sensing data for each object included in the grid-based spatial index. The road condition data processing device 300 can perform a previously registered query on each object from which the duplicate sensing data has been removed, and extract objects that correspond to the responses to the queries. The road condition data processing device 300 can output situation recognition data to the objects that correspond to the responses to the queries.

[0032] Network 400 can connect the sensor group 100, the object group 200, and the road condition data processing device 300. Such a network 400 can encompass wired networks such as LANs (local area networks), WANs (wide area networks), MANs (metropolitan area networks), and ISDNs (integrated service digital networks), as well as wireless networks such as wireless LANs, CDMA, Bluetooth®, and satellite communications, but the scope of the present invention is not limited thereto. Furthermore, network 400 can send and receive information using short-range and / or long-range communication. Here, short-range communication can include technologies such as Bluetooth®, RFID (radio frequency identification), IrDA (infrared data association), UWB (ultra-wideband), ZigBee®, and Wi-Fi (wireless fidelity), while long-range communication can include technologies such as CDMA (code division multiple access), FDMA (frequency division multiple access), TDMA (time division multiple access), OFDMA (orthogonal frequency division multiple access), and SC-FDMA (single carrier frequency division multiple access).

[0033] Network 400 may include connections of network elements such as hubs, bridges, routers, and switches. Network 400 may include one or more connected networks, such as a multiplexed network environment, including shared networks like the internet and private networks like secure corporate private networks. Access to Network 400 may be provided via one or more wired or wireless access networks.

[0034] Furthermore, network 400 can support CAN (controller area network) communication, V2X (vehicle to everything) communication, WAVE (wireless access in vehicular environment) communication technologies, and IoT (Internet of Things) networks and / or 5G communication that exchange and process information between distributed components such as things. Here, V2X communication can include communication between vehicles and any individual, such as V2V (vehicle-to-vehicle), which refers to communication between vehicles, V2I (vehicle to infrastructure), which refers to communication between vehicles and eNBs or RSUs (roadside units), V2P (vehicle-to-pedestrian), which refers to communication between vehicles and UEs (Union Equipment) owned by individuals (pedestrians, cyclists, drivers, or passengers), and V2N (vehicle-to-network).

[0035] Figure 2 is a block diagram illustrating the configuration of the road condition data processing device according to this embodiment; Figures 3a to 3c are illustrative diagrams illustrating the graph modeling process according to this embodiment; Figures 4a and 4b are tables showing the attributes of nodes and edges included in the graph model according to this embodiment; Figures 5a and 5b are illustrative diagrams illustrating the removal of duplicate sensing data using the graph model and grid-based spatial index according to this embodiment; and Figure 6 is a table showing the linkage relationship between the graph model and the grid-based spatial index according to this embodiment. In the following description, any parts that overlap with the description of Figure 1 will be omitted.

[0036] Referring to Figures 2 to 6, the road condition data processing device 300 may include a collection management unit 310, a modeling management unit 320, an index management unit 330, a duplicate removal management unit 340, a recognition management unit 350, a database 360, and a control unit 370.

[0037] The data collection management unit 310 can collect sensing data from the sensor group 100 installed on the road to the object group 200 present on the road. In this embodiment, the sensing data may include one or more of the object group 200's location, direction of movement (heading), and speed of movement (speed). For the sake of explanation, the sensor group 100 will be referred to as a sensor and the object group 200 as an object.

[0038] The data collection management unit 310 can register / remove / change the specifications of sensors on the road, as well as store the type and data format of each sensor in the database 360 ​​and manage them. The data collection management unit 310 can also verify which sensor transmitted the collected sensing data and map it in a standard sensor data format. To this end, the data collection management unit 310 can read the data format for the relevant sensor from the database 360 ​​and convert the actually collected sensing data into the standard sensor data format. The data collection management unit 310 can store the standard sensor data format in the database 360 ​​in real time and link it to higher levels (e.g., the modeling management unit 320, the index management unit 330, etc.).

[0039] The modeling management unit 320 can model the relationships between objects in a graph based on sensing data for each object. When graph modeling, the modeling management unit 320 can display each object as a single node on the graph. Furthermore, it can set relationships and display the set relationships as edges, depending on whether a node corresponding to one object may affect one or more nodes corresponding to other objects.

[0040] When the modeling management unit 320 displays the configured relationships as edges, it can determine whether there is a possibility of collision between a node corresponding to one or more other objects and a node corresponding to one of the objects, based on the position, direction, and velocity of the node corresponding to one of the objects. Depending on whether a collision possibility exists, the modeling management unit 320 can display an edge between the node corresponding to one of the objects and the nodes corresponding to one or more other objects.

[0041] Referring to Figures 3a to 3c, the graph modeling process according to this embodiment can be described as follows: The modeling management unit 320 sets the relationships between objects based on sensing data collected from the collection management unit 310, which includes one or more of the object's position, object's movement direction, and object's movement speed, and can model them in a graph as shown in Figure 3c.

[0042] Figure 3a shows a roundabout according to one embodiment, where a lidar 100_1, camera 100_2, and radar 100_3 are installed, and various objects (e.g., pedestrians, cars, potholes, etc.) are located on the road.

[0043] Figure 3b shows the display of sensing data collected from lidar 100_1, camera 100_2, and radar 100_3 on a map. Here, L1 to L5 can represent objects 1 to 5 detected by lidar 100_1. C1 to C4 can represent objects 1 to 5 detected by camera 100_2. U1 to U4 can represent objects 1 to 4 detected by radar 100_3. P1 can represent pothole 1 on the road. Also, in Figure 3b, arrows can indicate the direction and speed of movement of each object.

[0044] In this embodiment, since at least three sensors are present on the road, each sensor can generate different sensing data for the same object. In other words, the sensing data generated by each sensor may overlap for the same object.

[0045] The modeling management unit 320 can set and display each object as a node in the graph during graph modeling, and set and display the relationships between each object as edges. Any object becomes a node on the graph, and relationships between objects can be set when an object may affect other objects. For this purpose, the modeling management unit 320 can calculate the probability of collision (e.g., predicted collision time or predicted approach time) based on the current velocity, direction, and position. The modeling management unit 320 can set any relationship and periodically update the relationships, as long as the probability of collision is not infinite.

[0046] In Figure 3b, when we look at the direction and velocity of C2's movement, there is no possibility of collision with C4 or C3. However, when we look at the direction and velocity of C2's movement, there is a possibility of collision with L5, and in such a case, an edge can be set between C2 and L5. Furthermore, the modeling management unit 320 records the predicted collision time for both C2 and L5 as an edge attribute, and can update the edge attribute whenever the positions of C2 and L5 are updated.

[0047] Figure 4a is a table showing the attributes of nodes in the graph model, and Figure 4b is a table showing the attributes of edges in the graph model.

[0048] Referring to Figure 4a, Node ID can indicate a graph node identifier. Trajectory can indicate trajectory data for the past 5 objects, based on a time period unit (e.g., 0.1 seconds), where location, heading, and speed were detected. Sensor Type is the type of sensor used to detect the object, where 0 indicates a camera, 1 indicates a lidar, and 2 indicates a radar. Location can indicate the object's x and y positions. Heading can indicate the object's direction of movement. Speed ​​can indicate the object's speed. AoI is Age of Information, and can indicate the difference between the situational recognition data output time of the cognitive management unit 350 and the cognitive execution estimation time. In this embodiment, this time difference can be used as an important basis when determining the validity of certain data. Object Type is the type of object, where 0 indicates a car, 1 indicates a pedestrian, 2 indicates a motorcycle, 3 indicates a bicycle, 4 indicates a fallen object, 5 indicates a pothole, and 6 indicates construction. Here, for static objects such as fallen objects, potholes, and construction, the speed and direction of movement do not need to exist. The MBR is the minimum bounding rectangle of an object and can contain four coordinate values ​​(x_left, y_upper, x_right, y_lower).

[0049] Referring to Figure 4b, the Edge ID can indicate the graph edge identifier. The TTC can indicate the time to collision. The Deadline is the Deadline to Inform Message, and it can indicate the critical time within which information transmission must be completed to nodes connected at the edge, taking into account the AoI (Area of ​​Interest) and the time to collision.

[0050] Referring back to Figure 2, the index management unit 330 can construct a grid-based spatial index for the graph modeling results generated by the modeling management unit 320. Figure 5a shows the result of constructing a grid-based spatial index for the graph modeling results of Figure 3c. In this embodiment, the grid-based spatial index can consist of 16 cells, and the number of cells is changeable.

[0051] In this embodiment, the index management unit 330 can determine the size of the cells constituting the grid-based spatial index based on the sensing error for each of the multiple sensors and the speed limit set on the road. In particular, the size of a single cell can be smaller as the sensing error decreases and the speed limit decreases. Since it is possible to construct the grid-based spatial index and determine whether or not an object is accurately located in any one of the cells, it can be seen that the risk of collision decreases even if the size of the cells is small. The reason for determining the size of the cells as described above in this embodiment is that by considering only the objects contained in a specific cell and the cells adjacent to that specific cell, all candidate objects for duplicate removal can be compared.

[0052] The duplicate removal management unit 340 can remove duplicate sensing data from the sensing data for objects included in the grid-based spatial index.

[0053] The duplicate removal management unit 340 can set one or more existing objects as duplicate object candidates by comparing the location data of a new object detected in any one of the multiple cells that constitute the grid-based spatial index with the location data of one or more existing objects contained in the aforementioned one cell and in cells adjacent to that one cell.

[0054] As one embodiment, when setting duplicate object candidates, the duplicate removal management unit 340 can detect the first cell where the new object is located among a plurality of cells constituting the spatial index of the grid base, based on sensing data for the new object. Next, the duplicate removal management unit 340 can detect the locations of one or more existing objects located in the first cell and in cells adjacent to the first cell. Next, the duplicate removal management unit 340 can calculate the difference between the location data of the new object in the first cell and the location data of one or more existing objects included in the first cell and in cells adjacent to the first cell as a first distance value. Here, the first distance value may mean the actual physical distance between each object. The duplicate removal management unit 340 can set one or more existing objects whose first distance value is less than or equal to a first threshold as duplicate object candidates.

[0055] In another embodiment, when setting duplicate object candidates, the duplicate removal management unit 340 can determine a first point where the new object is located based on sensing data for the new object. The duplicate removal management unit 340 can detect the locations of one or more existing objects located within a predetermined distance from the first point. The duplicate removal management unit 340 can calculate the difference between the location data of the new object and the location data of one or more existing objects located within a predetermined distance from the first point as a first distance value. Here, the first distance value may mean the actual physical distance between each object. The duplicate removal management unit 340 can set one or more existing objects whose first distance value is less than or equal to a first threshold as duplicate object candidates.

[0056] The duplicate removal management unit 340 can determine a candidate duplicate object as the final duplicate object by comparing the trajectory data of the candidate duplicate object with the trajectory data of the new object.

[0057] When determining the final duplicate object, the duplicate removal management unit 340 can extract a first point group located within a pre-set time period from a three-dimensional coordinate system based on the position, direction, and velocity included in the sensing data of the duplicate object candidate. The duplicate removal management unit 340 can also extract a second point group located within a pre-set time period from a three-dimensional coordinate system based on the position, direction, and velocity included in the sensing data of the new candidate.

[0058] The duplicate removal management unit 340 can calculate the difference between the first point group and the second point group as a second distance value. Here, the second distance value may be a distance value indicating the similarity between the data, rather than the actual physical distance. In this embodiment, the method for calculating the difference may be to match the data of the first point group and the data of the second point group with data from the same time and calculate the difference. The data of the first point group and the second point group may include five data points prior to the time t when the position data of the new object and the duplicate candidate object was confirmed. For example, if measurements are taken in 0.1-second increments, these may be trajectory data at t-0.1, t-0.2, t-0.3, t-0.4, and t-0.5. Also, the first point group and the second point group may be data within the same coordinate system. The duplicate removal management unit 340 can determine one or more duplicate object candidates whose second distance value is less than or equal to the second threshold as the final duplicate object.

[0059] The duplicate removal management unit 340 can remove either the sensing data for the last duplicate object or the sensing data for the new object.

[0060] In this embodiment, because positional errors may occur depending on the type of sensor, if the same object is detected by different sensors, they may be recognized as different objects. When the same object is recognized as different objects, unnecessary data duplication and misperception of the situation may occur, making duplication necessary. To perform duplication quickly, it is necessary to quickly find the existing object that is in the position most similar to the new object. Sequentially comparing all objects is impossible due to the resulting slowdown.

[0061] To quickly perform duplicate removal, the grid-based spatial index constructed by the index management unit 330 can be used. That is, it can calculate which cell in the grid-based spatial index the location of a new object falls into and compare its similarity with existing objects in that cell.

[0062] Referring to Figure 5b, the cells where the new object 510, "New," is located are cells 1 and 4, where existing objects U4(520) and U3(530) are located. Here, since the first distance value, which is the difference between the position data of the new object 510 and the position data of the existing objects U4(520) and U3(530), is less than or equal to the first threshold, existing objects U4(520) and U3(530) can be set as duplicate object candidates. Here, the first distance value may mean the actual physical distance between each object. Subsequently, the difference between the trajectory data of the new object 510 (first point group) and the trajectory data of the existing objects U4(520) and U3(530) (second point group) can be calculated as the second distance value. The second distance value may not be the actual physical distance but a distance value indicating the similarity between the data. One or more of the existing objects U4(520) and U3(530) whose second distance value is less than or equal to the second threshold can be determined as the final duplicate object.

[0063] When comparing similarity by approaching objects contained in each cell using a grid-based spatial index, they can be approached in a structure like the table in Figure 6. Figure 6 is a table showing the linkage between the graph model and the grid-based spatial index according to this embodiment. The ID of an object contained in any one cell is stored in the grid-based spatial index, and the node attributes and adjacent node list for that object can be quickly obtained in the graph structure corresponding to the object ID.

[0064] The cognitive management unit 350 can perform a previously registered query on objects from which duplicate sensing data has been removed, and extract objects that match the response to the query. The cognitive management unit 350 can perform a query on objects whose sensing data is updated at a pre-set time unit (e.g., 0.1 seconds) to check for the existence of objects whose collision probability exceeds a threshold. As a result of the query, the cognitive management unit 350 can extract objects whose collision probability exceeds a threshold as objects that match the response to the query.

[0065] The cognitive management unit 350 can output situational awareness data for objects that correspond to the response to a question. The cognitive management unit 350 can output warning data for objects that correspond to the response to a question, warning of the possibility of a collision.

[0066] In this embodiment, the graph model can update the attributes of the graph nodes and edges in real time whenever a new object is recognized. At this time, the cognitive management unit 350 can extract the road condition recognition results through simple questioning of the graph.

[0067] Here, the question could include, for example, a question to find a node whose predicted collision time with an adjacent node is less than or equal to a threshold value of 1. This could be a question to determine whether the collision risk of a given node is greater than or equal to a threshold value of 1. Alternatively, the question could include, for example, a question to find a node adjacent to a node adjacent to a node whose predicted collision time with an adjacent node is less than or equal to a threshold value of 1. This could be a question to determine whether the collision risk of an adjacent node to a given node is greater than or equal to a threshold value of 1. In this embodiment, for the sake of explanation, only two example questions have been described, but a variety of questions can be registered in the database 360.

[0068] Risk factors can be searched through this type of graph-based questioning, and situational awareness data can only be extracted by continuously performing the aforementioned questioning; therefore, the continuous questioning processing technique can be applied. In the continuous questioning processing technique, once a question is registered in database 360, each time a new object is recognized, it checks if there is an object that matches the response to the question, and if a matching object exists, situational awareness data can be immediately output to that object.

[0069] Database 360 ​​can store general data collected, processed, generated, and output by the road condition data processing device 300. In one embodiment, database 360 ​​can store sensing data collected from multiple sensors at predetermined intervals, graph modeling results, node and edge attribute update results, grid-based spatial index construction results, duplicate sensing data detection and removal results, various questions, and situation awareness data output results.

[0070] The control unit 370 can control the operation of the entire road condition data processing device 300 as a kind of central processing device. The control unit 370 can include any kind of device capable of processing data, such as a processor. Here, "processor" may mean a data processing device built into hardware, for example, having a physically structured circuit to perform a function expressed in code or instructions contained in a program. Examples of such data processing devices built into hardware include microprocessors, central processing units (CPUs), processor cores, multiprocessors, ASICs (application-specific integrated circuits), FPGAs (field programmable gate arrays), etc., but the scope of the present invention is not limited thereto.

[0071] Figure 7 is a block diagram illustrating the configuration of a road condition data processing device according to another embodiment. In the following description, any parts that overlap with the descriptions of Figures 1 to 6 will be omitted. Referring to Figure 7, the road condition data processing device 300 according to another embodiment may include a processor 380 and a memory 390.

[0072] In this embodiment, the processor 380 can process the functions performed by the data collection management unit 310, modeling management unit 320, index management unit 330, deduplication management unit 340, recognition management unit 350, database 360, and control unit 370 disclosed in Figure 2.

[0073] Such a processor 380 can control the operation of the entire road condition data processing device 300. Here, "processor" may mean a data processing device built into hardware, for example, having a physically structured circuit to perform a function expressed by code or instructions contained in a program. Examples of such data processing devices built into hardware include microprocessors, central processing units (CPUs), processor cores, multiprocessors, ASICs (application-specific integrated circuits), FPGAs (field programmable gate arrays), etc., but the scope of the present invention is not limited thereto.

[0074] The memory 390 is operablely connected to the processor 380 and can store at least one code in association with the operations performed by the processor 380.

[0075] Furthermore, the memory 390 can perform the function of temporarily or permanently storing data processed by the processor 380. Here, the memory 390 may include magnetic storage media or flash storage media, but the scope of the present invention is not limited thereto. Such memory 390 may include internal memory and / or external memory, and may include volatile memory such as DRAM, SRAM, or SDRAM; non-volatile memory such as OTPROM (one-time programmable ROM), PROM, EPROM, EEPROM, mask ROM, flash ROM, NAND flash memory, or NOR flash memory; flash drives such as SSDs, CF (compact flash) cards, SD cards, Micro-SD cards, Mini-SD cards, Xd cards, or memory sticks; or storage devices such as HDDs.

[0076] Figure 8 is a flowchart illustrating the road condition data processing method according to this embodiment. In the following explanation, any parts that overlap with the explanations for Figures 1 to 7 will be omitted.

[0077] Referring to Figure 8, in step S810, the road condition data processing device 300 can collect sensing data on objects present on the road from multiple sensors installed on the road.

[0078] At stage S820, the road condition data processing device 300 can model the relationships between objects in a graph based on sensing data for the objects.

[0079] The road condition data processing device 300 can display each object as a single node on the graph. The road condition data processing device 300 can set relationships and display these relationships as edges, depending on whether a node corresponding to one of the objects may affect one or more nodes corresponding to other objects.

[0080] In this embodiment, when the road condition data processing device 300 displays relationships as edges, it can determine whether there is a possibility of collision between a node that corresponds to one or more other objects and a node that corresponds to one of the objects, based on the position, direction, and speed of the node that corresponds to one of the objects. Depending on whether a possibility of collision exists, the road condition data processing device 300 can display the area between the node that corresponds to one of the objects and the nodes that correspond to one or more other objects as edges.

[0081] At step S830, the road condition data processing device 300 can construct a grid-based spatial index based on the results of the graph modeling. When constructing the grid-based spatial index, the road condition data processing device 300 can determine the size of the cells that make up the grid-based spatial index based on the sensing error for each of the multiple sensors and the speed limit set on the road. Here, the size of the cells may be smaller as the sensing error is smaller and the speed limit is lower.

[0082] At stage S840, the road condition data processing device 300 can remove duplicate sensing data from the sensing data for objects included in the grid-based spatial index.

[0083] The road condition data processing device 300 can set one or more existing objects as duplicate object candidates by comparing the location data of a new object detected in any one of the multiple cells that constitute the grid-based spatial index with the location data of one or more existing objects contained in any one of the aforementioned cells and in cells adjacent to any one of the aforementioned cells.

[0084] As one embodiment, when setting duplicate object candidates, the road condition data processing device 300 can detect the first cell where the new object is located among a plurality of cells constituting the spatial index of the grid base, based on sensing data for the new object. The road condition data processing device 300 can detect the location of one or more existing objects located in the first cell. The road condition data processing device 300 can calculate the difference between the location data of the new object in the first cell and the location data of one or more existing objects included in the first cell and cells adjacent to the first cell as a first distance value. The road condition data processing device 300 can set one or more existing objects whose first distance value is less than or equal to a first threshold as duplicate object candidates.

[0085] In another embodiment, the road condition data processing device 300 can determine a first point where a new object is located based on sensing data for the new object when setting up duplicate object candidates. The road condition data processing device 300 can detect the locations of one or more existing objects located within a predetermined distance from the first point. The road condition data processing device 300 can calculate the difference between the location data of the new object and the location data of one or more existing objects located within a predetermined distance from the first point as a first distance value. The road condition data processing device 300 can set one or more existing objects whose first distance value is less than or equal to a first threshold as duplicate object candidates.

[0086] The road condition data processing device 300 can determine a candidate duplicate object as the final duplicate object by comparing the trajectory data of the candidate duplicate object with the trajectory data of the new object. The road condition data processing device 300 can extract a first group of points located in a pre-set time period from a three-dimensional coordinate system based on the position, direction, and speed included in the sensing data of the candidate duplicate object. The road condition data processing device 300 can extract a second group of points located in a pre-set time period from a three-dimensional coordinate system based on the position, direction, and speed included in the sensing data of the new candidate. The road condition data processing device 300 can calculate the difference between the first group of points and the second group of points as the second distance value. The road condition data processing device 300 can determine one or more candidate duplicate objects whose second distance value is less than or equal to the second threshold as the final duplicate object.

[0087] The road condition data processing device 300 can remove one of the sensing data for the last duplicate object and the sensing data for the new object.

[0088] At stage S850, the road condition data processing device 300 can perform a previously registered query on objects from which duplicate sensing data has been removed, and extract objects that correspond to the response to the query.

[0089] When the road condition data processing device 300 extracts objects that correspond to the response to a question, it can perform a series of questions to check for the existence of objects whose collision probability exceeds a threshold, targeting objects whose sensing data is updated at pre-set time intervals. As a result of the series of questions, the road condition data processing device 300 can output situation recognition data as objects whose collision probability exceeds a threshold, indicating that these objects correspond to the response to the question.

[0090] At stage S860, the road condition data processing device 300 can output situation recognition data to an object that corresponds to the response to the question. Here, the road condition data processing device 300 can output warning data to an object that corresponds to the response to the question, warning that there is a possibility of collision.

[0091] The embodiments of the present invention described above can be embodied in the form of computer programs executable on a computer through a variety of components, and such computer programs can be recorded on a computer-readable medium. In this case, the medium may include magnetic media such as hard disks, floppy disks and magnetic tapes, optical recording media such as CD-ROMs and DVDs, magneto-optical media such as floptical disks, and hardware devices specially configured to store and execute program instructions, such as ROM, RAM, and flash memory.

[0092] On the other hand, the computer program described above may be specifically designed and configured for the present invention, or it may be publicly known and usable by those skilled in the field of computer software. Examples of computer programs include not only machine code, such as that produced by a compiler, but also high-level language code that can be executed by a computer using an interpreter or the like.

[0093] In the specification of this invention (especially in the claims), the use of the term "above" and similar demonstrative pronouns may be singular or plural. Furthermore, where a range is described in this invention, it includes inventions to which individual values ​​belonging to the above range are applied (unless otherwise stated), and is equivalent to describing each individual value constituting the above range in the detailed description of the invention.

[0094] Unless otherwise explicitly stated, the steps constituting the method according to the present invention may be performed in any order. The present invention is not necessarily limited to the order in which the steps are described. The use of all examples or exemplary terms (e.g., etc.) in the present invention is simply for the purpose of detailing the invention and, as not limited by the claims, the scope of the present invention is not limited by the examples or exemplary terms. Furthermore, those skilled in the art will see that the design conditions and factors can be constructed within the scope of the claims or their equivalents with various modifications, combinations, and changes.

[0095] Therefore, the concept of the present invention should not be limited to the embodiments described above, and it can be said that not only the claims described later, but also all scope equivalent to or equivalently modified from these claims, falls within the scope of the concept of the present invention. [Explanation of symbols]

[0096] 100: Sensor Group 200: Object Group 300: Road condition data processing device 400: Network

Claims

1. A method for processing road condition data, performed by a processor of a device that processes road condition data, A step of collecting sensing data for each object present on the road from multiple sensors installed on the road; A step in which the relationships between the aforementioned objects are modeled in a graph based on sensing data for each of the aforementioned objects; The next step is to construct a grid-based spatial index based on the results modeled in the aforementioned graph; A step of removing duplicate sensing data from the sensing data for each object included in the spatial index of the grid base; The steps include: conducting a query to confirm the existence of objects whose collision probability with adjacent objects exceeds a threshold, using each of the objects from which the duplicate sensing data has been removed, and extracting objects that correspond to the response to the query; and The step includes outputting contextual awareness data that warns of a potential collision with an object corresponding to the response to the aforementioned question; Road condition data processing method.

2. The stage of modeling using the aforementioned graph is, The step of displaying each of the aforementioned objects as one node on the graph; and The steps include: setting up relationships and displaying those relationships as edges, depending on whether a node corresponding to any one of the aforementioned objects may affect nodes corresponding to one or more other objects; The road condition data processing method according to claim 1.

3. The step of displaying at the aforementioned edge is, A step of determining whether there is a possibility of collision with one or more other objects based on the position, direction, and velocity of the node that corresponds to any one of the aforementioned objects; and Depending on the existence of the aforementioned possibility of collision, the step of displaying an edge between the node corresponding to one of the aforementioned objects and the nodes corresponding to one or more other objects; The road condition data processing method according to claim 2.

4. The step of constructing the spatial index of the aforementioned grid base is, The step includes determining the size of the cells constituting the spatial index of the grid base based on the sensing error for each of the plurality of sensors and the speed limit set on the road, The aforementioned cell is The smaller the sensing error, and the lower the speed limit, the smaller the magnitude of the error. The road condition data processing method according to claim 1.

5. The step of removing the duplicate sensing data is as follows: A step of setting one or more existing objects as duplicate object candidates by comparing the position data of a new object detected in any one of the multiple cells constituting the spatial index of the grid base with the position data of one or more existing objects contained in any one of the cells and any one of the cells adjacent to that cell; The step of determining the candidate duplicate object as the final duplicate object by comparing the trajectory data of the candidate duplicate object with the trajectory data of the new object; and The step of removing one of the sensing data for the last duplicate object and the sensing data for the new object; The road condition data processing method according to claim 1.

6. The step of setting the aforementioned object as a candidate for duplicate object is: A step of detecting a first cell in which the new object is located among a plurality of cells constituting the spatial index of the grid base, based on sensing data for the new object; Steps include detecting the location of one or more existing objects located in the first cell; A step of calculating a first distance value by taking the difference between the position data of a new object present in the first cell and the position data of one or more existing objects contained in the first cell and cells adjacent to the first cell; and The step of setting one or more existing objects whose first distance value is less than or equal to a first threshold as candidates for duplicate objects; The road condition data processing method according to claim 5.

7. The step of determining the final duplicate object is, Steps include extracting a first point group located within a predetermined time period from a three-dimensional coordinate system based on the position, direction, and velocity included in the sensing data of the aforementioned duplicate object candidates; A step of extracting a second point group located within a predetermined time period from a three-dimensional coordinate system based on the position, direction, and velocity included in the sensing data of the aforementioned new object; A step of calculating the difference between the first point group and the second point group as the second distance value; and The process includes the step of determining one or more candidate duplicate objects whose second distance value is less than or equal to a second threshold as the final duplicate object; The road condition data processing method according to claim 6.

8. The step of setting the aforementioned object as a candidate for duplicate object is: A step of determining a first point where the new object is located, based on sensing data for the new object; A step of detecting the location of one or more existing objects located within a predetermined distance from the first point; A step of calculating a first distance value as the difference between the position data of the new object and the position data of one or more existing objects located within a predetermined distance from the first point; and The step of setting one or more existing objects whose first distance value is less than or equal to a first threshold as candidates for duplicate objects; The road condition data processing method according to claim 5.

9. The step of extracting objects that correspond to the answers to the aforementioned questions is: A step of conducting a series of questions to confirm the existence of objects whose collision probability exceeds a threshold, targeting the objects whose sensing data is updated at pre-set time intervals; and The process includes the step of extracting objects whose collision probability exceeds a threshold as a result of the aforementioned series of questions, as objects that correspond to the responses to the questions; The road condition data processing method according to claim 5.

10. The aforementioned situational awareness data is warning data that warns of the possibility of a collision. The road condition data processing method according to claim 9.

11. A computer-readable recording medium on which a computer program is stored that causes a computer to perform any one of the methods of claims 1 to 10 using a computer.

12. A device for processing road condition data, Processor; and A memory operably connected to the processor and storing at least one code executed by the processor; When the memory is executed through the processor, the processor will Multiple sensors installed on the road collect sensing data for each object present on the road. Based on the sensing data for each of the aforementioned objects, the relationships between the aforementioned objects are modeled in a graph. A grid-based spatial index is constructed from the results of the modeling in the aforementioned graph. Among the sensing data for each object included in the spatial index of the grid base, duplicate sensing data is removed. For each of the objects from which the duplicate sensing data has been removed, a query is performed to confirm the existence of objects whose collision probability with adjacent objects exceeds a threshold, and objects corresponding to the response to the query are extracted. Save the triggering code so that it outputs situational awareness data warning of a potential collision with an object corresponding to the response to the aforementioned question. Road condition data processing device.

13. The memory uses the processor to When removing the duplicate sensing data, the position data of a new object sensed in any one of the multiple cells constituting the spatial index of the grid base is compared with the position data of one or more existing objects contained in any one of those cells and in cells adjacent to any one of those cells, thereby setting one or more existing objects as candidates for duplicate objects. The trajectory data of the duplicate object candidate and the trajectory data of the new object are compared to determine the duplicate object candidate as the final duplicate object. Save the code that triggers the removal of one of the sensing data for the last duplicate object and the sensing data for the new object. The road condition data processing device according to claim 12.

14. The memory uses the processor to When setting a candidate for a duplicate object, based on sensing data for the new object, the first cell in which the new object is located is detected among the multiple cells constituting the spatial index of the grid base. The position of one or more existing objects located in the first cell is detected, The difference between the position data of a new object in the first cell and the position data of one or more existing objects contained in the first cell and cells adjacent to the first cell is calculated as the first distance value. Save the code that triggers setting one or more existing objects whose first distance value is less than or equal to a first threshold as duplicate object candidates. The road condition data processing device according to claim 13.

15. The memory uses the processor to When determining the final duplicate object, a first point group located within a predetermined time period is extracted from a three-dimensional coordinate system based on the position, direction, and velocity included in the sensing data of the candidate duplicate object. Extract a second point group located within a predetermined time period from a three-dimensional coordinate system based on the position, direction, and velocity included in the sensing data of the aforementioned new object. The difference between the first point group and the second point group is calculated as the second distance value. The code that triggers the determination of one or more duplicate object candidates whose second distance value is less than or equal to the second threshold is the final duplicate object. The road condition data processing device according to claim 14.

Images