Cooperative recognition system and cooperative recognition method for autonomous driving assistance

Abstract

According to an embodiment of the present invention, a cooperative recognition system includes: a memory for storing a program for determining driving guidance when a subject vehicle autonomously drives; and a processor which executes the program to receive recognition information indicating a plurality of recognition objects recognized by a plurality of cooperative recognition means within a predetermined distance range based on a position of the subject vehicle from the plurality of cooperative recognition means, estimates a driving intention and a driving trajectory of each of the plurality of recognition objects on the basis of the recognition information, and determines driving guidance for an autonomous driving route of the subject vehicle on the basis of the driving trajectory.

Description

Cooperative Cognition System and Cooperative Cognition Method for Assisted Autonomous Driving

[0001] The present invention relates to a cooperative recognition system and a cooperative recognition method for assisting autonomous driving of a mobility means.

[0002] Autonomous vehicles use various sensors to recognize surrounding objects, fuse them with precise road map information to generate dynamic information, and drive based on this information.

[0003] However, when a large vehicle is ahead of a self-driving car or a long line of traffic is obstructing forward visibility, it can be difficult to accurately perceive surrounding information or objects relative to the detection range of each sensor. Furthermore, the same problem can hinder forward perception in areas with limited visibility, such as curved sections, sections of road with vertical gradients, and intersections.

[0004] It is difficult to solve the problem of securing forward visibility with a stand-alone method based on sensors installed in the autonomous vehicle itself, and this difficulty causes the behavior of the autonomous vehicle to become unnatural.

[0005] The purpose of the present invention is to solve the above problems, and to provide a cooperative recognition system and a cooperative recognition method that enable cooperative recognition of an autonomous vehicle to be performed using not only an autonomous vehicle but also other cognitive means.

[0006] In order to achieve the above purpose, according to an embodiment of the present invention, a cooperative recognition system includes a memory storing a program for determining driving guidance when a target vehicle autonomously drives, and a processor for executing the program to receive recognition information indicating a plurality of recognition objects recognized within a predetermined distance range based on a position of the target vehicle from a plurality of cooperative recognition means and the target vehicle, estimating a driving intention and a driving trajectory of each of the plurality of recognition objects based on the recognition information, and determining driving guidance for an autonomous driving path of the target vehicle based on the driving trajectory.

[0007] The processor determines one dynamic situation information by collecting the recognition object information received from the plurality of cooperative recognition means, and the driving guidance may include longitudinal driving guidance, lateral driving guidance, and driving trajectory guidance.

[0008] The processor may compare a first driving trajectory of one object among the plurality of recognition objects and a second driving trajectory of the target object, and if a difference between a first expected time point at which the one object is located at a first point intersecting the second driving trajectory among the first driving trajectories and a second expected time point at which the target object is located at the first point is less than or equal to a predetermined conflict threshold, the processor may predict that the one object and the target object are in conflict, and if the processor predicts that the one object and the target object are in conflict, the processor may generate driving guidance for the target vehicle to avoid the conflict.

[0009] The above-described plurality of cooperative recognition means may include at least one of road infrastructure including a plurality of first sensor devices installed on a target road on which the target vehicle is driving, at least one drone including a plurality of second sensor devices and deployed in a target area requiring situational information recognition from the target vehicle or the road infrastructure, and at least one surrounding vehicle located around the target vehicle and including a plurality of third sensor devices.

[0010] Each of the plurality of first to third sensor devices may include at least one of a camera, LiDAR, and RADAR.

[0011] Each of the plurality of cooperative recognition means can generate a wireless communication message including an identifier (ID) and type of each corresponding means, latitude, longitude, and altitude above sea level of a location of the corresponding means, information on a corresponding recognition object among the plurality of recognition objects, and a driving intention of the corresponding recognition object.

[0012] The driving intention may include a longitudinal driving intention including at least one of constant speed driving, acceleration driving, deceleration driving, and stopping, a lateral driving intention including at least one of lane keeping, left lane change, and right lane change, and an intersection passage intention including at least one of straight ahead, left turn, right turn, and U-turn.

[0013] The information of the corresponding recognition object may include the time at which the corresponding recognition object is recognized, the ID of the corresponding recognition object, the type of the corresponding recognition object, the longitude and latitude of the location of the corresponding recognition object, the altitude above sea level of the location of the corresponding recognition object, the distance from the corresponding recognition means among the plurality of cooperative recognition means to the corresponding recognition object, the lateral and longitudinal speeds of the corresponding recognition object, the movement azimuth of the corresponding recognition object, the lateral and longitudinal accelerations of the corresponding recognition object, the vertical acceleration of the corresponding recognition object, the yaw rate of the corresponding recognition object, the detection area in which the corresponding recognition object is recognized, the driving lane in which the corresponding recognition object is driving, the width, length, and height of the corresponding recognition object, and the past movement trajectory information of the corresponding recognition object.

[0014] The type of the corresponding cognitive object may be at least one of an unclassified vehicle, a passenger car, a truck, a bus, a pedestrian, a bicycle, a motorcycle, a PM (Personal Mobility), a CAV (Connected Autonomous Vehicle), an AV (Non-connected Autonomous Vehicle), a CHV (Conventional Human-Driven Vehicle), a CV (Connected Vehicle), a baby stroller, a wheelchair, an animal propellant, a trailer chassis, an animal, and a driving obstacle.

[0015] The above detection area may be at least one of a general driving lane, a CAV exclusive lane, a merging acceleration lane, a merging main lane edge lane, a dividing deceleration lane, a dividing main lane edge lane, a weaving section, an unprotected left-turn waiting lane, a left-turn lane, a right-turn lane, a straight lane, a U-turn lane, two or more combined lanes among straight, left, right, and U-turn, a crosswalk, a roundabout entrance ramp, a roundabout exit ramp, a roundabout circling lane, a pedestrian path, an undivided road for pedestrians and vehicles, a two-lane road in both directions, a Hi-pass tollgate entrance ramp, a Hi-pass tollgate exit ramp, a manned tollgate entrance ramp, a manned tollgate exit ramp, and an emergency stop.

[0016] According to another embodiment of the present invention, a cooperative recognition method performed by a processor includes a step of recognizing a plurality of recognition objects within a predetermined range based on a location of the target vehicle by a plurality of cooperative recognition means and a target vehicle, a step of receiving recognition information indicating the plurality of recognition objects from the plurality of cooperative recognition means and the target vehicle, a step of estimating a driving intention and a driving trajectory of each of the plurality of recognition objects based on the recognition information by the processor, and a step of determining a driving guidance for an autonomous driving path of the target vehicle based on the driving trajectory.

[0017] The method may further include: a step of comparing a first driving trajectory of one object among the plurality of recognition objects and a second driving trajectory of the target object; a step of predicting that the one object and the target object are in conflict if a difference between a first expected time point at which the one object is located at a first point intersecting the second driving trajectory among the first driving trajectories and a second expected time point at which the target object is located at the first point is less than a predetermined conflict threshold; and a step of generating driving guidance for the target vehicle to avoid the conflict if the one object and the target object are predicted to be in conflict.

[0018] According to another embodiment of the present invention, a cooperative recognition system includes a situation recognition unit that receives recognition information indicating a plurality of recognition objects recognized within a predetermined distance range based on a position of a target vehicle from a plurality of cooperative recognition means and a target vehicle, a driving intention analysis unit that estimates a driving intention of each of the plurality of recognition objects based on the recognition information, a driving trajectory estimation unit that estimates a driving trajectory of each of the plurality of recognition objects based on the driving intention of each of the plurality of recognition objects, and a driving guidance determination unit that determines driving guidance for an autonomous driving path of the target vehicle based on the driving trajectory of each of the plurality of recognition objects.

[0019] The driving guidance determining unit may compare a first driving trajectory of one object among the plurality of recognition objects and a second driving trajectory of the target object, and if a difference between a first expected time point at which the one object is located at a first point intersecting the second driving trajectory among the first driving trajectories and a second expected time point at which the target object is located at the first point is less than or equal to a predetermined conflict threshold, the driving guidance may be generated so that the one object and the target object are in conflict, and if the one object and the target object are predicted to be in conflict, the driving guidance may be generated so that the target vehicle performs a maneuver to avoid the conflict.

[0020] According to the present invention, the driving of an autonomous vehicle can be controlled based on object information recognized by sensors equipped in the vehicle as well as sensors equipped in other recognition means.

[0021] According to the present invention, an autonomous vehicle can recognize road condition information outside the recognition range of a sensor equipped in the vehicle in advance and utilize it as information for more efficient driving.

[0022] According to the present invention, improved situational awareness and smooth vehicle driving are possible through drone cooperative recognition or vehicle-to-vehicle cooperative recognition even in areas where an Edge RSU (Road Side Unit) for infrastructure guidance is not installed.

[0023] According to the present invention, safe autonomous driving is possible compared to independent autonomous driving of a vehicle.

[0024] According to the present invention, even in blind spots where autonomous vehicle sensors cannot recognize curved roads, vertical lines (uphill and downhill), obstacles, etc., cooperative recognition enables safe and efficient autonomous driving, reduces fuel consumption, and provides passengers with comfortable driving.

[0025] The collaborative system between vehicles, road infrastructure, and drones allows for recognition of a wider range of situations and objects than a stand-alone vehicle approach.

[0026] Figure 1 is a collaborative cognition concept diagram for explaining collaborative cognition among target vehicles, road infrastructure, and drones.

[0027] Figure 2 is a conceptual diagram explaining drone-based collaborative cognition.

[0028] FIG. 3 is a block diagram of a collaborative cognitive system according to one embodiment of the present invention.

[0029] Figure 4 is a block diagram schematically illustrating the detailed configuration of the processor illustrated in Figure 3.

[0030] Figure 5 is a conceptual diagram for explaining a conflict represented by one embodiment of the present invention.

[0031] Figures 6, 7, 8, 9, 10, 11, 12, and 13 are conceptual diagrams for explaining driving guidance in situations where various guidance provision is required.

[0032] FIG. 14 is an exemplary diagram illustrating a target vehicle equipped with a drone according to one embodiment of the present invention.

[0033] FIG. 15 is an exemplary diagram illustrating the positions at which each of the target vehicle and surrounding vehicles recognizes an object according to one embodiment of the present invention.

[0034] Figure 16 is a flowchart of a cooperative recognition method according to one embodiment of the present invention.

[0035] The present invention can be implemented with various modifications without departing from the spirit or scope of the invention, and may have one or more embodiments. Furthermore, the embodiments described in the "Specific Description for Carrying Out the Invention" and "Drawings" sections of the present invention are examples intended to specifically illustrate the invention and do not limit or restrict the scope of the invention.

[0036] Accordingly, anything that a person with ordinary knowledge in the technical field to which the present invention belongs can easily infer from the “specific contents for carrying out the invention” and “drawings” of the present invention can be interpreted as falling within the scope of the present invention.

[0037] In addition, the size and shape of each component shown in the drawing may be exaggerated for the purpose of explaining the embodiment, and do not limit the size and shape of the invention actually implemented.

[0038] Unless specifically defined, terms used in the specification of the present invention may have the same meaning as commonly understood by a person of ordinary skill in the art to which the present invention pertains.

[0039] Autonomous vehicles utilize various sensors, including cameras (vision), LiDAR, radar, and ultrasound, to recognize surrounding objects. They then merge this information with precise road map data to generate dynamic information, enabling them to navigate based on this information. However, if an autonomous vehicle's forward field of vision is not secured, it may not accurately perceive its surroundings or objects, resulting in unnatural vehicle behavior or difficulty accurately performing evasive maneuvers to avoid obstacles ahead.

[0040] In one embodiment of the present invention, the driving guidance (e.g., driving recommendation guidance) required for autonomous driving of a target vehicle (hereinafter, the "target vehicle") can be determined by utilizing cooperative recognition between road infrastructure (Road Side Unit, RSU), drones, and surrounding vehicles (hereinafter, the "surrounding vehicles") of the target vehicle (hereinafter, the "target vehicle"). Here, the surrounding vehicles need not be autonomous vehicles, but can be any vehicle equipped with sensors that drives around the target vehicle, such as a car or two-wheeled vehicle.

[0041] Hereinafter, collaborative cognition may refer to the use of information perceived by sensors equipped in the target vehicle as well as information perceived by sensors equipped in road infrastructure, drones, and surrounding vehicles in order for the target vehicle to perceive surrounding objects and situations for autonomous driving.

[0042] Hereinafter, for convenience of explanation, the cognitive means that transmit cognitive information to target vehicles such as road infrastructure, drones, and surrounding vehicles are referred to as “multiple collaborative cognitive means.”

[0043] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

[0044] Figure 1 is a collaborative cognition concept diagram for explaining collaborative cognition among target vehicles, road infrastructure, and drones.

[0045] Referring to Figure 1, each autonomous vehicle (e.g., the target vehicle) can communicate with road infrastructure (I2V) through vehicle-to-vehicle (V2V) communication, with surrounding vehicles (V2V) through vehicle-to-vehicle (V2V) communication, and with drones (D2V) communication. Furthermore, each road infrastructure can communicate with other road infrastructure through infrastructure-to-infra (I2I) communication. Road infrastructure, drones, and vehicles can recognize objects such as bicycles, pedestrians, vehicles, and motorcycles.

[0046] The target vehicle can recognize multiple objects located around the target vehicle based on recognition information received from multiple cooperative recognition means.

[0047] Road infrastructure collaborative awareness, where a target vehicle utilizes road infrastructure, allows road infrastructure equipped with equipment such as cameras (vision) and LiDAR to share situational information perceived by the road infrastructure with an autonomous vehicle (e.g., the target vehicle). This situational information may include perception information representing multiple objects collected by the road infrastructure. The target vehicle can drive by referencing the perception information received from the road infrastructure.

[0048] Drone-based collaborative perception, in which a target vehicle uses a drone, involves the target vehicle or road infrastructure launching a drone and deploying it to an area requiring recognition of the target vehicle's direction or situational information. The drone can then be used to recognize surrounding object information and dynamic road condition information. The dynamic road condition information may include perception information indicating multiple objects on and around the road on which the target vehicle is driving. The target vehicle can then navigate by referencing the perception information received from the drone.

[0049] Collaborative perception, where a target vehicle utilizes surrounding vehicles, allows each vehicle to share the perception information it perceives about its surroundings with surrounding vehicles or road infrastructure. The target vehicle can recognize road conditions by combining the perception information received from surrounding vehicles (or from surrounding vehicles via road infrastructure) and drive based on this perceived road condition.

[0050] Figure 2 is a conceptual diagram explaining drone-based collaborative cognition.

[0051] Referring to FIG. 2, the target vehicle can communicate with the drone through D2V to recognize objects such as other vehicles around the target vehicle and accidents ahead.

[0052] FIG. 3 is a block diagram of a collaborative cognitive system according to one embodiment of the present invention.

[0053] Referring to FIG. 3, the cooperative recognition system (1) may include a target vehicle (10), a cooperative recognition server (11), and a plurality of cooperative recognition means (110).

[0054] The plurality of cooperative recognition means (110) may include road infrastructure (20), drones (30), and surrounding vehicles (40). In FIG. 3, the plurality of cooperative recognition means (110) are illustrated as including road infrastructure (20), drones (30), and surrounding vehicles (40), but this is merely an example, and the plurality of cooperative recognition means (110) may include at least one of the road infrastructure (20), drones (30), and surrounding vehicles (40). However, for the convenience of explanation, the plurality of cooperative recognition means (110) are assumed to include road infrastructure (20), drones (30), and surrounding vehicles (40) below.

[0055] The road infrastructure (20) may be an edge RSU installed on a road belonging to a jurisdictional area corresponding to the location where the target vehicle (10) is driving. The road infrastructure (20) may provide infrastructure guidance to vehicles located within a predetermined jurisdictional area based on the installed location.

[0056] The road infrastructure (20) may include a sensor device (21), a communication device (22), and a computing device (23).

[0057] The sensor device (21) may include various sensors capable of object recognition, such as a camera (Vision), LiDAR, and radar (RADAR). The sensor device (21) may be installed on the target road on which the target vehicle (10) is driving.

[0058] The communication device (22) can exchange signals and / or data with the target vehicle (10) and / or drone (30) via wireless communication.

[0059] The computing device (23) may be an Intelligent Edge Unit (IEU) capable of Edge Computing. The computing device (23) may perform real-time analysis on object information representing an object recognized by the sensor device (21) to generate a cooperative perception message (IG_Cooperative Perception Message). The communication device (22) may transmit perception information including recognized object information and situation information to the target vehicle (10) or the cooperative perception server (11) through the cooperative perception message generated by the computing device (23).

[0060] A collaborative perception message may be a message that shares perception information about surrounding dynamic objects or driving intention information of the vehicle with road infrastructure and surrounding vehicles in an autonomous cooperative driving system.

[0061] The computing device (23) can collect all of the object information recognized by the sensor device (21), situation information, information collected through cooperative recognition with a drone (30), and information collected through cooperative recognition with surrounding vehicles (40) to update the forward object information, road situation information, etc. required for autonomous driving of the target vehicle (10) to a local dynamic map (LDM).

[0062] A drone (30) may be an aircraft deployed to a target area requiring situational information recognition from a target vehicle (10) or road infrastructure (20).

[0063] A drone (30) may include a sensor device (31), a communication device (32), and a computing device (33).

[0064] The sensor device (31) may include various sensors capable of object recognition, such as a camera (Vision), LiDAR, and radar (RADAR).

[0065] The communication device (32) can transmit and receive signals and / or data via wireless communication with the target vehicle (10) and / or road infrastructure (20).

[0066] The computing device (33) may be an Intelligent Edge Unit (IEU) capable of Edge Computing. The computing device (33) may perform real-time analysis on object information representing objects recognized during flight to generate a cooperative perception message (IG_Cooperative Perception Message). The communication device (32) may transmit perception information including recognized object information and situation information to the target vehicle (10) or the cooperative perception server (11) and the road infrastructure (20) corresponding to the competent Edge through the cooperative perception message generated by the computing device (33).

[0067] A surrounding vehicle (40) may be a vehicle located within a predetermined transaction range from the location of the target vehicle. The surrounding vehicle (40) may be a vehicle with which the target vehicle (10) has agreed upon cooperative driving guidance. While FIG. 3 illustrates one surrounding vehicle (40) for convenience of explanation, this is for convenience of explanation, and multiple cooperative recognition means (110) may include one or more surrounding vehicles.

[0068] The surrounding vehicle (40) is a concept for explaining vehicle-to-vehicle (V2V) cooperative recognition based on the target vehicle (10), and the description of the surrounding vehicle (40) described below can be similarly applied to the target vehicle (10), and the description of the target vehicle (10) described below can be similarly applied to the surrounding vehicle (40). In addition, although FIG. 3 illustrates one target vehicle (10), this is for convenience of explanation, and the cooperative recognition system (1) can include one or more target vehicles.

[0069] The cooperative recognition system (1) can report each of at least one vehicle located in a corresponding area as a target vehicle (10) based on recognition information received from a plurality of cooperative recognition means (110) and transmit driving guidance to the target vehicle (10).

[0070] The surrounding vehicle (40) may include a sensor device (41), a communication device (42), and a computing device (43).

[0071] The sensor device (41) may include various sensors capable of object recognition, such as a camera (Vision), LiDAR, and radar (RADAR).

[0072] The communication device (42) can exchange signals and / or data with the target vehicle (10) and / or road infrastructure (20) via wireless communication.

[0073] The computing device (43) may be an Intelligent Edge Unit (IEU) capable of Edge Computing. The computing device (43) may perform real-time analysis on object information representing an object recognized by the sensor device (41) to generate a cooperative perception message (IG_Cooperative Perception Message). The communication device (42) may transmit perception information including recognized object information and situation information to the target vehicle (10) or the cooperative perception server (11) and the road infrastructure (20) corresponding to the competent Edge through the cooperative perception message generated by the computing device (43).

[0074] The cooperative recognition server (11) may be included in the road infrastructure (20). In FIG. 3, the cooperative recognition server (11) is illustrated as belonging to the road infrastructure (20), but this is for convenience of explanation and the invention is not limited thereto. In one embodiment, the cooperative recognition server (11) may be included in the target vehicle (10), or may be provided in the road infrastructure (20) or a central control server that controls the operation of the target vehicle (10).

[0075] The cooperative perception server (11) can drive an autonomous driving cooperative system that performs cooperative perception using road infrastructure (20), a drone (30), and surrounding vehicles (40). The cooperative perception server (11) can generate a cooperative perception message (IG_Cooperative Perception Message) to share perception information on surrounding dynamic objects or driving intention information of a target vehicle (10) with the road infrastructure (20) and surrounding vehicles (40).

[0076] Road infrastructure (20) can recognize object information and signal information in all directions within a predetermined infrastructure section, such as an intersection. Here, object information may include the object's type, location, speed, status, etc. The road infrastructure (20) can estimate the object's driving intent by considering traffic priority in accordance with road-related laws and provide optimized infrastructure guidance services. The road infrastructure (20) can periodically transmit information and communication messages within a predetermined infrastructure section to a specific target vehicle (10).

[0077] The cooperative recognition server (11) can estimate the driving intention of an object by considering the traffic priority in accordance with road-related laws and provide an optimized infrastructure guidance service. When the target vehicle (10) enters a predetermined infrastructure guidance section, the target vehicle (10) can receive an information communication message from the cooperative recognition server (11) and determine whether to enter the predetermined infrastructure section. The cooperative recognition server (11) can select and accept a message corresponding to the target vehicle (10) among the guidance messages received from a plurality of cooperative recognition means (110).

[0078] The cooperative recognition server (11) can exchange guidance messages for starting cooperative recognition using multiple cooperative recognition means (110) with multiple cooperative recognition means (110).

[0079] The guidance message may include an infrastructure guidance service participation request message (IG_JoinRequestMessage). The infrastructure guidance service participation request message may be a message requesting service participation from a vehicle entering the infrastructure guidance section (e.g., a target vehicle (10), a surrounding vehicle (40)). The infrastructure guidance service participation request message may include a unique number of an infrastructure guidance service group that performs the same infrastructure guidance service, a list of RSU IDs within the group, and types of infrastructure guidance services that can be serviced. The infrastructure guidance service group may be distinguished by a service group unique number (IG_ServiceGroupID). The service group unique number may be a number that groups multiple RSUs that provide the same service together.

[0080] A guidance message may include a guidance provision message (IG_GuidanceMessage). The guidance provision message may be a message for providing an infrastructure guidance service according to one embodiment. The guidance provision message may include a guidance unique number (idGuidance), a unique number of an RSU transmitting guidance information, a unique number of an object that is the subject of guidance (idObj), positional relationship information between idObj and surrounding objects over time (posRelSet), and idObj's driving trajectory guidance.

[0081] When a guidance provision message includes posRelSet, the cooperative recognition server (11) can generate and provide guidance information to objects capable of receiving guidance information among all related objects. An object receiving a guidance provision message including posRelSet must perform cooperative driving with related objects.

[0082] The driving trajectory guidance included in the guidance provision message can be expressed through mission information (missionPtList) or trajectory information (wayPtList). Here, the mission information provides the spatiotemporal goal that the object must perform through a spatiotemporal range and tolerance, and the trajectory information may provide the trajectory that the object must drive spatiotemporally. If the mission information or trajectory information is provided with location relationship information, a process is performed to confirm agreement on the guidance with related surrounding objects capable of cooperative driving, after which driving according to the guidance can be performed.

[0083] A guidance message may include a guidance response message (IG_ResponseMessage). The guidance response message may be a response from a vehicle that received a request to participate in the infrastructure guidance service or a vehicle that directly received the service. The guidance response message may include the response content, a unique guidance number, and a reason for the guidance rejection.

[0084] The target vehicle (10) may include a vehicle control unit (VCU) (12), an automated driving system (ADS) (13), a sensor device (14), a communication device (15), and a computing device (16).

[0085] VCU (12) can control the operation of the vehicle.

[0086] The ADS (13) can generate a plurality of autonomous driving signals to perform autonomous driving using a dynamic information system (LDM) updated by a plurality of cooperative recognition means (110) and transmit them to the VCU (12). The VCU (12) can generate a control signal for controlling the operation of the vehicle based on the driving guidance received from the cooperative recognition server (11) and the autonomous driving signal received from the ADS (13).

[0087] The sensor device (14) may include various sensors capable of object recognition, such as a camera (Vision), LiDAR, and radar (RADAR).

[0088] The communication device (15) can exchange signals and / or data with the target vehicle (10) and / or road infrastructure (20) via wireless communication.

[0089] The computing device (16) may be an Intelligent Edge Unit (IEU) capable of Edge Computing. The computing device (16) may perform real-time analysis on object information representing an object recognized by the sensor device (14) to generate a cooperative perception message (IG_Cooperative Perception Message). The communication device (15) may transmit perception information including recognized object information and situation information to the cooperative perception server (11) and the road infrastructure (20) corresponding to the competent Edge through the cooperative perception message generated by the computing device (16).

[0090] The collaborative cognitive server (11) may include a memory (100), a processor (200), and an input / output unit (300).

[0091] The memory (100) may include volatile memory and / or non-volatile memory. The memory (100) may store, for example, commands or data related to a processor (200), an input / output unit (300), a plurality of cooperative recognition means (110), and surrounding vehicles (40), one or more programs and / or software, an operating system, etc., in order to implement and / or provide operations, functions, etc. provided by the cooperative recognition system (1).

[0092] The program stored in the memory (100) may include a program (hereinafter, “cooperative recognition program”) that determines driving guidance by considering cooperative recognition when the target vehicle drives autonomously. The cooperative recognition program may provide an infrastructure guidance service using road infrastructure (20). Road infrastructures may be divided into infrastructure guidance service groups including multiple RSUs that perform the same infrastructure guidance service. Multiple infrastructure guidance service groups may be expressed as multiple RSU groups including multiple RSUs. The road infrastructure (20) may be one of multiple RSUs included in one infrastructure guidance service group, and the description of the road infrastructure (20) described below may be similarly applied to each of the multiple RSUs.

[0093] The processor (200) executes a cooperative recognition program to receive recognition information indicating a plurality of recognition objects within a predetermined distance range based on the location of the target vehicle (10) from a plurality of cooperative recognition means (110) and the target vehicle (10), estimate the driving intention and driving trajectory of each of the plurality of recognition objects based on the received recognition information, and determine driving guidance for the autonomous driving path of the target vehicle (10) based on the driving trajectory.

[0094] Hereinafter, for convenience of explanation, among the object(s) recognized by each of the sensor device (14), sensor device (21), sensor device (31), and sensor device (41) within their respective corresponding predetermined jurisdictional ranges, objects located within a predetermined distance range based on the location of the target vehicle (10) are referred to as multiple recognition objects, and information indicating the multiple recognition objects is referred to as recognition information.

[0095] The processor (200) can receive recognition information on a plurality of recognition objects among the objects collected through the sensor device (14), the sensor device (21), the sensor device (31), and the sensor device (41) and utilize the information to determine driving guidance. When the processor (200) determines driving guidance, it can generate a driving guidance message indicating the driving guidance and transmit the generated driving guidance message to the target vehicle (10).

[0096] A driving guidance message may include a data frame of mission point information (IG_MissionPoint) representing the driving guidance. The mission point information data frame may express driving mission information based on the guidance. Within the mission point information data frame, the range may be expressed using one of the following methods: minimum / maximum, median / error.

[0097] In the mission point information, the mission point can indicate the temporal range information within which the mission must be performed, and the spatial relative position and speed information within which the mission must be performed within the temporal range.

[0098] Temporal range information is expressed as a minimum (startTime) and maximum (endTime), and the vehicle assigned a mission must complete the mission within that time range. For example, a vehicle assigned a mission can complete the mission at a time corresponding to the midpoint of the time range.

[0099] Spatial relative positions can be divided into lateral and longitudinal directions. Lateral spatial relative positions may include the number of lanes being changed and the direction of the lane change. Here, the vehicle assigned to the task must reach the lateral relative position within a given time range. Longitudinal spatial relative positions may include forward longitudinal distance traveled based on the current position. Here, the vehicle assigned to the task can be considered to have completed the task if it reaches a position between the minimum and maximum longitudinal relative position values ​​within the given time range. For example, the vehicle assigned to the task can reach a position corresponding to the midpoint of the spatial relative position range.

[0100] Speed ​​information may include speed information changed through driving guidance. Here, the assigned vehicle must reach a speed range expressed as a median speed and a tolerance within a given time period. For example, the assigned vehicle may reach the median speed indicated by the speed information.

[0101] The driving guidance message may further include mission point list information (IG_MissionPointList) and mission point guidance information (IG_MPGuidance). The mission point list information may include a list of driving mission information based on the guidance. The mission point guidance information includes guidance information based on driving mission information, and may include initial status information (location, status, time, etc.) that serves as a reference for the mission point, and may include a list of mission points based on the initial status information.

[0102] In addition to this, the guidance message may include information on each of the target vehicle (10), road infrastructure (20), surrounding vehicles (40), and information on transit points.

[0103] The processor (200) can control the operation of each of the memory (100) and the input / output unit (300). The processor (200) may be a computing device. The processor (200) may include at least one of processing devices such as an application specific integrated circuit (ASIC), a digital signal processor (DSP), a programmable logic device (PLD), a field programmable gate array (FPGA), a central processing unit (CPU), a microcontroller, and a microprocessor.

[0104] The processor (200) can provide a cooperative recognition service to the target vehicle (10) using a cooperative recognition server (11).

[0105] The processor (200) can control the overall operation of the cooperative cognitive system (1) by executing one or more programs stored in the memory (100).

[0106] For example, the processor (200) can execute a cooperative recognition program stored in the memory (100) to generate driving guidance for a plurality of autonomous driving, and transmit the generated driving guidance to the VCU (12) to control the driving of the target vehicle (10).

[0107] The VCU (12) can change the driving speed of the target vehicle (10), change lanes, perform evasive maneuvers, etc. based on driving guidance received from the cooperative recognition server (11) when performing multiple control operations for autonomous driving of the target vehicle (10). The processor (200) can analyze the driving intention of each of the multiple recognition objects based on the recognition information received from the input / output unit (300), estimate each driving trajectory, and determine the driving guidance of the target vehicle (10) based on the estimated driving trajectory.

[0108] The input / output unit (300) can receive recognition information from at least one of the target vehicle (10), road infrastructure (20), drone (30), and surrounding vehicle (40).

[0109] Each of the target vehicle (10), road infrastructure (20), drone (30), and surrounding vehicles (40) can transmit perception information representing recognized surrounding objects to the cooperative perception server (11) through a cooperative perception message (IG_Cooperative Perception Message). The cooperative perception message can include data frame information representing surrounding object perception information. The surrounding object perception information can include perception information on a plurality of recognized perception objects. The cooperative perception message is a wireless communication message and can include an identifier (ID) and type of a cooperative perception means transmitting the message, latitude, longitude, and altitude above sea level of a location of the cooperative perception means, perception object information recognized by a corresponding cooperative perception means among a plurality of perception objects, and driving intention of the perception object.

[0110] The information of the cognitive object included in the cooperative cognitive message may include two or more elements from among the time of recognizing the corresponding cognitive object, the ID of the corresponding cognitive object, the type of the corresponding cognitive object, the longitude, latitude, and altitude above sea level of the location of the corresponding cognitive object, the distance from the means for transmitting the corresponding cooperative cognitive message among the plurality of cooperative cognitive means (110) to the cognitive object, the lateral and longitudinal speeds of the corresponding cognitive object, the moving azimuth of the corresponding cognitive object, the lateral and longitudinal accelerations of the corresponding cognitive object, the vertical accelerations of the corresponding cognitive object, the yaw rate of the corresponding cognitive object, the detection area in which the corresponding cognitive object was recognized, the driving lane in which the corresponding cognitive object is driving, the width, length, and height of the corresponding cognitive object, and the past movement trajectory of the corresponding cognitive object. Hereinafter, a data frame representing the cognitive object information included in the cooperative cognitive message will be described for each data element through examples.

[0111] Data elements of cognitive object information may include object recognition time (secMark DSecond). The object recognition time may be synchronized with UTC (minutes) time between information-sharing objects such as a target vehicle (10), road infrastructure (20), a drone (30), and surrounding vehicles (40).

[0112] The data element of the recognition object information may include a temporary identifier (Temporary ID). Here, the temporary identifier may be an identifier indicating an object recognized by a means for transmitting a corresponding cooperative recognition message among a plurality of cooperative recognition means (110). The cooperative recognition system (1) may assign a temporary identifier to each dynamic object tracked by the sensor device (14), the sensor device (21), the sensor device (31), and the sensor device (41), and may maintain the same temporary identifier as the identifier of the corresponding object unless tracking of the object indicated by the temporary identifier fails.

[0113] The data element of the recognition object information may include an object type. Each of the plurality of cooperative recognition means (110) may classify the recognized object type into unclassified vehicles, passenger cars, trucks, buses, pedestrians, bicycles, motorcycles, two-wheeled vehicles, PMs (Personal Mobility, electric kickboards, electric wheels, etc.), CAVs (Connected Autonomous Vehicles), AVs (Non-connected Autonomous Vehicles), CHVs (Conventional Human-Driven Vehicles), CVs (Connected Vehicles), baby strollers, wheelchairs, animal propellants (horses, etc.), trailer chassis, animals, driving obstacles, etc. Each of the plurality of cooperative recognition means (110) may include a data element indicating the type of the classified object in a cooperative recognition message.

[0114] Data elements of the recognition object information may include an object distance. The object distance may represent the distance from a reference point (ego) such as a target vehicle (10), road infrastructure (20), drone (30), or surrounding vehicle (40) to the recognized object.

[0115] Data elements of cognitive object information may include the lateral velocity (SpeedX) of the cognitive object.

[0116] Data elements of cognitive object information may include the longitudinal speed (SpeedY) of the cognitive object.

[0117] Data elements of the cognitive object information may include the heading of the cognitive object.

[0118] The data elements of the cognitive object information may include four-directional acceleration information (AccelerationSet4Way) including accelerations in the longitudinal, lateral, and vertical directions of the cognitive object, and angular velocities (Yaw rates).

[0119] Data elements of the cognitive object information may include a detection area. The detection area may include a general driving lane, a CAV-only lane, a merging acceleration lane, a merging main lane edge lane, a classified deceleration lane, a classified main lane edge lane, a weaving section, an unprotected left-turn waiting lane, a left-turn lane, a right-turn lane, a straight lane, a U-turn lane, a combined lane including two or more of straight, left, right, and U-turn, a crosswalk, a roundabout entrance ramp, a roundabout exit ramp, a roundabout turning lane, a pedestrian path, an undivided road for pedestrians and vehicles, a two-lane road in both directions, a Hi-pass tollgate entrance ramp, a Hi-pass tollgate exit ramp, a manned tollgate entrance ramp, a manned tollgate exit ramp, an emergency stop, etc.

[0120] Data elements of cognitive object information may include lane ID information for the lane in which the cognitive object is driving. Lane ID information may include identifiers such as numbers that indicate the lane in which the cognitive object is driving.

[0121] Data elements of cognitive object information may include the size of the cognitive object (Vehicle Size). The size of the cognitive object may represent planar dimensions such as the width and length of the cognitive object.

[0122] A data element of cognitive object information may include the height of the cognitive object.

[0123] Data elements of the recognition object information may include a past movement trajectory (Path History) of the recognition object. The past movement trajectory of the recognition object may include a path record such as an initial position of the recognition object, a GNSS status of the recognition object, and a plurality of way points indicating the past movement trajectory. Here, the initial position of the recognition object may indicate the position of the corresponding object at a reference point in time when a corresponding cooperative recognition means among the plurality of recognition objects started to recognize the corresponding object. In addition, the past movement trajectory may include location information including each time the corresponding object was recognized and the path, latitude, and altitude above sea level of the corresponding object at each time.

[0124] Additionally, the collaborative recognition message may optionally further include data frame information indicating the driving intent of a corresponding object among multiple objects. Below, each data element of the data frame indicating the driving intent of the recognition object included in the collaborative recognition message is described.

[0125] The driving intention of the cognitive object included in the collaborative cognitive message may include longitudinal driving intention, lateral driving intention, and intersection driving intention.

[0126] Longitudinal driving intentions may include constant speed driving, acceleration driving, deceleration driving, and stopping.

[0127] Lateral driving intentions may include staying in lane, changing to the left lane, and changing to the right lane.

[0128] Intersection traffic intent may include going straight, turning left, turning right, and turning U-turn.

[0129] Each of the sensor device (14), the sensor device (21), the sensor device (31), and the sensor device (41) can recognize corresponding surrounding objects and transmit recognition information about the objects to the target vehicle (10) or the cooperative recognition server (11) through the corresponding communication device among the communication device (15), the communication device (22), the communication device (32), and the communication device (42). The recognition information received by the cooperative recognition server (11) can include recognition information about a plurality of objects located around the target vehicle (10) among the objects recognized by the sensor device (14), the sensor device (21), the sensor device (31), and the sensor device (41).

[0130] Figure 4 is a block diagram schematically illustrating the detailed configuration of the processor illustrated in Figure 3.

[0131] Referring to FIG. 4, the processor (200) may include a situation recognition unit (201), a recognition object matching unit (202), a driving intention analysis unit (203), a driving trajectory determination unit (204), and a driving guidance determination unit (205).

[0132] The situation recognition unit (201) can recognize the situations of multiple recognition objects within a predetermined distance range based on the location of the target vehicle (10) by analyzing the recognition information received from the input / output unit (300) by multiple cooperative recognition means (110) and the target vehicle (10). Here, the situation of the recognition object can include the type, location, speed, acceleration, etc. of the recognition object.

[0133] The cognitive object matching unit (202) can synchronize time between a plurality of cognitive objects indicated by cognitive information and a target vehicle (10), and can collect overlapping information about one object among the plurality of cognitive objects and match it to cognitive information about one object.

[0134] The driving intention analysis unit (203) can estimate the driving intention of each of a plurality of recognition objects based on recognition information. The driving intention analysis unit (203) can estimate the driving intention (or moving intention) of each of a plurality of recognition objects for a paved and striped road and an intersection. The driving intention of each of the plurality of recognition objects can include a longitudinal driving intention, a lateral driving intention, an intersection passage intention, etc.

[0135] The driving intention analysis unit (203) can estimate the longitudinal driving intention using the trajectory (IG_WPGuidance) information of a corresponding recognition object among multiple recognition objects. For example, the longitudinal driving intention may be constant speed, acceleration, deceleration, or stopping.

[0136] Lateral driving intent can be an intent to change lanes, turn, etc. Lane change intent can be to stay in lane, change direction (left / right), or change lane number. Intersection passage intent can be one of the following: going straight, turning left, turning right, making a U-turn, entering an intersection, or exiting an intersection.

[0137] Intersection traffic intent can include whether pedestrians, two-wheeled vehicles, etc. will drive through the intersection and, if so, their driving speed. Intersection traffic intent can include going straight, turning left, turning right, or making a U-turn.

[0138] The driving trajectory determination unit (204) can estimate the driving trajectory of each of the plurality of recognition objects based on the driving intention of each of the plurality of recognition objects to determine an expected driving trajectory. The driving trajectory can include locations where each of the plurality of recognition objects is expected to be located at a predetermined cycle for a predetermined time period after the current point in time.

[0139] The driving guidance decision unit (205) can determine driving guidance for the autonomous driving path of the target vehicle (10) based on the driving trajectory of each of the plurality of recognition objects.

[0140] The driving guidance decision unit (205) can determine whether a situation requiring guidance provision has occurred around the target vehicle (10) based on cognitive information. Here, situations requiring guidance include situations where there is a risk of vehicle-to-vehicle or vehicle-to-person accidents, road construction or lane closures, falling objects on the road, recognizing a parked vehicle on the road ahead, recognizing a slow-moving vehicle on the road ahead, situations where bad weather has occurred, situations where a conflict is expected between a vehicle on the main road and a vehicle entering a ramp on a merging road, situations where one wants to merge at the end of a queue for a lane for proceeding in a specific direction (e.g., a pocket lane for turning left / right, a lane for exiting a ramp, etc.) on a certain range of roads approaching a road classification section or a pocket lane for turning, or situations where one wants to proceed while avoiding the queue, situations where one wants to avoid a collision or pass by minimizing acceleration or deceleration on a certain range of roads approaching an intersection such as an unsignalized intersection or an unprotected left-turn intersection, situations where one wants to avoid a collision or pass by minimizing acceleration or deceleration when entering or exiting a roundabout, situations where one must overtake a preceding vehicle on a two-lane road, situations where one must give way to an approaching emergency vehicle, and other situations on the road. This may include situations requiring right-of-way designation, etc. For example, the driving guidance decision unit (205) may determine a situation in which the target vehicle (10) is expected to collide with at least one of a plurality of recognition objects as a situation requiring guidance provision. A situation requiring guidance provision may include a situation in which a conflict with at least one of the plurality of recognition objects is not expected, but a driving recommendation is required to ensure that the target vehicle (10) drives optimally so as not to impede traffic flow, or to optimize fuel consumption to minimize deceleration and acceleration.

[0141] In this way, when a cooperative recognition message is expressed as data for sharing information on a situation requiring guidance provision, elements of the data for sharing information on a situation requiring guidance provision may include a sharing ID for a situation requiring guidance provision, an ID for a type of situation requiring guidance provision, a recognition time for a situation requiring guidance provision, an ID indicating an object that caused a situation requiring guidance provision (hereinafter referred to as a cause object), a type of cause object, longitude and latitude of the cause object, a reference longitude and latitude coordinate system, lateral and longitudinal moving speeds of the cause object, respective accelerations, vertical acceleration of the cause object, yaw rate of the cause object, a lane in which the cause object is located, width of the cause object, length of the cause object, height of the cause object, and a record of the cause object's driving path.

[0142] The Guidance Provision Need Context Information Sharing ID may be an identifier to distinguish the Guidance Provision Need Context Information Sharing data from other data.

[0143] The Guidance Provision Needed Situation Type ID may be an identifier to distinguish the type of Guidance Provision Needed Situation triggered from other types of Guidance Provision Needed Situations.

[0144] The time of recognition of a situation requiring guidance provision may indicate the time at which a corresponding means among multiple cooperative recognition means (110) recognizes a recognition object.

[0145] The risk-causing object ID, which indicates the object that caused the situation requiring guidance provision, may be an identifier for distinguishing the corresponding cognitive object from other objects among multiple cognitive objects.

[0146] The causal object type can indicate the type of the corresponding cognitive object.

[0147] The longitude and latitude of the cause object can represent the longitude and latitude of the location of the corresponding cognitive object.

[0148] The reference latitude and longitude coordinate system can represent a reference coordinate system for indicating the location of a cognitive object.

[0149] The transverse and longitudinal movement velocities of the causal object can represent the transverse and longitudinal velocities of the corresponding cognitive object.

[0150] The transverse and longitudinal accelerations of the causal object can represent the transverse and longitudinal accelerations of the corresponding cognitive object, respectively.

[0151] The vertical acceleration of the causal object can represent the vertical acceleration of the corresponding perceptual object.

[0152] The causal object yaw rate can represent the yaw rate of the corresponding cognitive object.

[0153] The lane where the causal object is located can indicate the detection area that recognized the corresponding cognitive object.

[0154] The cause object width, the risk trigger object length, and the cause object height may represent the width, length, and height of the corresponding cognitive object.

[0155] The causal object driving path record can represent the past movement trajectory of the corresponding cognitive object.

[0156] The driving guidance determined by the driving guidance determination unit (205) may include longitudinal driving guidance, lateral driving guidance, and driving trajectory guidance. The driving guidance determined by the driving guidance determination unit (205) may include lane change guidance for lane changes of the target vehicle (10) and driving speed guidance for the driving speed of the target vehicle (10). Here, the driving speed guidance may be longitudinal guidance, and the lane change guidance may be longitudinal guidance.

[0157] The driving trajectory guidance may include the order and coordinates of multiple waypoints that the vehicle must pass through, and the range of passage times of the recognized object at each of the multiple waypoints.

[0158] Lane change guidance may include information sharing ID for the situation where guidance provision is required, current driving lane, lane recommended for change, starting point location of the lane change guidance application section, and ending point location of the lane change guidance application section.

[0159] The sharing ID for the situation information requiring guidance provision may indicate the situation requiring guidance provision that caused the provision of guidance. The current driving lane may indicate the lane in which the target vehicle (10) is currently driving. The change recommended lane may indicate the lane in which the driving guidance decision unit (205) recommends the target vehicle (10) to change from the current driving lane. The positions of the start and end points of the lane change guidance application section may be positions indicating the point in time when the target vehicle (10) is located in a predetermined range of road sections to which the lane change guidance is applied, and the end point when the target vehicle (10) leaves the predetermined range of road sections to which the lane change guidance is applied.

[0160] Driving speed guidance may include a guidance provision need situation information sharing ID, a driving speed guidance application section start location, a driving speed guidance application section end location, a longitudinal recommended minimum driving speed, a longitudinal recommended maximum driving speed, etc.

[0161] The starting and ending points of the driving speed guidance application section may be the points at which the target vehicle (10) is located within a predetermined range of road sections to which the driving speed guidance is applied, and the ending point at which the target vehicle (10) leaves the predetermined range of road sections to which the driving speed guidance is applied. The longitudinal recommended minimum driving speed and maximum driving speed may represent the lowest and highest values ​​of the longitudinal speeds indicated by the driving speed guidance. Here, when the driving guidance determination unit (205) sets the minimum / maximum driving speed of the driving speed guidance to 0, it can be seen that the target vehicle (10) is recommended to stop.

[0162] The processor (200) receives recognition information about a plurality of recognition objects in the surroundings based on cooperative recognition with road infrastructure (20), cooperative recognition with drones (30), cooperative recognition with surrounding vehicles (40), etc., and can perform the operations described below by utilizing the information. Here, the cooperative recognition server (11) including the processor (200) that performs cooperative recognition with road infrastructure (20), cooperative recognition with drones (30), cooperative recognition with surrounding vehicles (40), etc., may be included in the road infrastructure (20) as illustrated in FIG. 3, or may be included in other road infrastructure or target vehicles (10). As described above, for the convenience of explanation, the cooperative recognition server (11) is assumed to be included in the road infrastructure (20).

[0163] The processor (200) can generate conflict prediction information between the recognized object and the target vehicle (10) and establish a conflict avoidance plan. Here, "conflict" means that the time difference between the expected movement trajectories of each object, such as the recognized object and the target vehicle (10), intersects with a certain value (t conflict ) can be expressed in the following cases. Hereinafter, the concept of conflict is explained with reference to Fig. 5.

[0164] Figure 5 is a conceptual diagram for explaining a conflict represented by one embodiment of the present invention.

[0165] In Fig. 5, a vehicle (Vehicle_101) may represent a surrounding vehicle (40), and a vehicle (Vehicle_102) may represent a target vehicle (10).

[0166] Referring to FIG. 5, the driving trajectory determination unit (204) can determine the moving trajectory of the vehicle (Vehicle_101) as the trajectory (TRJ_1) based on the current position (P10) of the vehicle (Vehicle_101). The driving trajectory determination unit (204) can determine the moving trajectory of the vehicle (Vehicle_102) as the trajectory (TRJ_2) based on the current position (P20) of the vehicle (Vehicle_102).

[0167] The driving guidance decision unit (205) determines the point (t) at which the vehicle (Vehicle_101) is located at the point (P11) on the trajectory (TRJ_1). 101 ) and the point (P21) on the trajectory (TRJ_2) at which the vehicle (Vehicle_102) is located (t 102 ), comparing the time point (t 101 ) and time point (t 102 ) difference between (|t 101 -t 102 |) is a given conflict threshold (t conflict ) below, it can be predicted that the vehicle (Vehicle_101) and the vehicle (Vehicle_102) are in conflict.

[0168] The driving guidance decision unit (205) can determine whether a conflict occurs between objects including a plurality of recognition objects and a target vehicle (10).

[0169] The driving guidance determination unit (205) can estimate the driving trajectory of each of the plurality of recognition objects and the driving trajectory of the target object. The driving guidance determination unit (205) compares the first driving trajectory of one object among the plurality of recognition objects and the second driving trajectory of the target vehicle (10), and if the difference between the first expected time point at which one object is located at the first point where the second driving trajectory intersects with the first driving trajectory and the second expected time point at which the target vehicle (10) is located at the first point is less than or equal to a predetermined conflict threshold, it can be predicted that one object and the target vehicle (10) are in conflict.

[0170] If it is predicted that the target vehicle (10) and one of the plurality of recognition objects are in conflict, the driving guidance determination unit (205) can generate driving guidance to avoid the conflict by decelerating, accelerating, stopping, changing lanes, etc. for the target vehicle (10) so as to resolve the conflict prediction information. The driving guidance determination unit (205) can transmit the driving guidance to the VCU (12) so that the target vehicle (10) can avoid the conflict.

[0171] The driving guidance decision unit (205) can determine whether a situation requiring guidance provision has occurred around the target vehicle (10) if it is determined that no conflict has occurred between objects.

[0172] If it is determined that a situation requiring guidance provision has occurred, the driving guidance determination unit (205) can generate driving guidance to support the situation requiring guidance provision. The driving guidance determination unit (205) can transmit the driving guidance to the VCU (12) so that the target vehicle (10) can resolve the situation requiring guidance provision.

[0173] The driving guidance decision unit (205) is configured to determine a risk of a vehicle-to-vehicle or vehicle-to-person accident due to an object or situation in an area outside the recognition range of the sensor device (14) or the sensor device (41), a road construction or lane closure situation, a situation where an object falls on the road, a situation where a parked vehicle is recognized on the road ahead, a situation where a slow-moving vehicle is recognized on the road ahead, a situation where bad weather occurs, a situation where a collision is expected between a main road vehicle and a vehicle entering a ramp on a merging road, a situation where a lane for proceeding in a specific direction (e.g., a pocket lane for a left / right turn, a lane for exiting a ramp, etc.) on a road within a predetermined range approaching a road classification section or a pocket lane for turning, or a situation where a vehicle wants to merge at the end of a queue for proceeding in a specific direction on a road within a predetermined range approaching an intersection such as an unsignalized intersection or an unprotected left-turn intersection, a situation where a collision is avoided or acceleration or deceleration is minimized when entering or exiting a roundabout, a situation where a two-way two-lane road is configured to determine a risk of a vehicle accident or a traffic accident, a situation where a vehicle is expected to merge at the end of a queue for proceeding in a specific direction on a road within a predetermined range approaching a road classification section or a pocket lane for turning, or a situation where a vehicle wants to proceed while avoiding the queue ..., such as an unsignalized intersection or an unprotected left-turn intersection, or a situation where a vehicle is expected to merge at the end of a queue for proceeding in a specific direction on a road within a predetermined range approaching a road, such as an unsignalized intersection In situations where it is necessary to overtake a preceding vehicle on the road, give way to an approaching emergency vehicle, or other situations where the right of way on the road is required, a driving guidance capable of decelerating, accelerating, stopping, changing lanes, and performing evasive maneuvers can be generated by predicting in advance, so that the target vehicle (10) can take actions such as decelerating, accelerating, stopping, changing lanes, and performing evasive maneuvers in advance.

[0174] Figures 6, 7, 8, 9, 10, 11, 12, and 13 are conceptual diagrams for explaining driving guidance in situations where various guidance provision is required.

[0175] Referring to FIG. 6, when it is determined that there is a slow-moving vehicle or a parked vehicle in front of the driving lane of the target vehicle (10) based on recognition information received from a plurality of cooperative recognition means (110), the driving guidance determination unit (205) can generate driving guidance that allows the vehicle to change lanes in advance and pass through a situation requiring guidance provision in front without conflicting with the slow-moving vehicle or parked vehicle in front.

[0176] Referring to FIG. 7, if it is determined that driving in a specific lane is dangerous due to an accident, construction, fallen objects, flooding, etc. in front of the driving lane of the target vehicle (10) based on the recognition information received from a plurality of cooperative recognition means (110), the driving guidance determination unit (205) may change the driving lane of the target vehicle (10) to a lane that can be safely driven in advance, or generate driving guidance that allows the target vehicle (10) to drive on an alternative route. Here, the alternative route may be a first alternative route representing an efficient driving alternative for traffic flow or a second alternative route representing an eco-friendly driving alternative. Here, the first alternative route may include a route that includes a minimum number of congested or slow-moving sections among a plurality of routes from a starting point to a destination on which the target vehicle (10) autonomously drives. Additionally, the second alternative route may include a route that includes a minimum driving distance and / or a maximum number of automobile-only roads to minimize fuel consumption of the target vehicle (10) among multiple routes from a starting point to a destination where the target vehicle (10) autonomously drives.

[0177] Even when no conflict occurs between objects and no situation requiring guidance provision occurs, the driving guidance determination unit (205) can determine whether a first alternative route exists and, if so, generate driving guidance indicating the first alternative route. If no first alternative route exists, the unit can determine whether a second alternative route exists and, if so, generate driving guidance indicating the second alternative route. If no second alternative route exists, the cooperative recognition program can be terminated without generating a separate driving guidance.

[0178] Referring to FIG. 8, when a target vehicle (10) approaches a road junction and determines that a collision is expected at the road junction based on recognition information received from a plurality of cooperative recognition means (110), a plurality of driving guidance determination units (205) can generate driving guidance to cause the target vehicle (10) to accelerate, decelerate, or change lanes to avoid the collision.

[0179] In Fig. 8, the driving guidance generated by the driving guidance determination unit (205) may be such that, when the vehicle (10_c1) is the target vehicle (10), the target vehicle (10) driving on the main road may recognize another vehicle (10_c2) entering the ramp early through recognition information indicating cooperative recognition, and the target vehicle (10) may change lanes to an upper lane (e.g., from lane 2 to lane 1) to avoid the other vehicle (10_c1).

[0180] Alternatively, in FIG. 8, the driving guidance generated by the driving guidance determination unit (205) may be such that, when the vehicle (10_c2) is the target vehicle (10), the target vehicle (10) entering the ramp may recognize another vehicle (10_c1) approaching from the main line early through recognition information indicating cooperative recognition, and may adjust the entry speed of the target vehicle (10) into the main line in advance to prevent a collision, thereby ensuring safety and reducing fuel consumption.

[0181] Referring to FIG. 9, if it is determined that the road classification section or pocket turn lane that the target vehicle (10) intends to enter is saturated based on the recognition information received from a plurality of cooperative recognition means (110), the driving guidance determination unit (205) may recognize the last vehicle in the queue of other vehicles in advance and generate driving guidance that allows the target vehicle (10) to join the end of the queue. Here, the target vehicle (10) may be a vehicle that intends to exit the road through a ramp or enter a left-right turn lane.

[0182] Referring to FIG. 10, when a target vehicle (10) approaches an unprotected left-turn intersection, a driving guidance determination unit (205) can generate driving guidance based on recognition information received from a plurality of cooperative recognition means (110) to enable the target vehicle (10) to safely perform an unprotected left turn at the unprotected left-turn intersection or to safely drive without colliding with other vehicles performing an unprotected left turn at the intersection.

[0183] In Fig. 10, the driving guidance generated by the driving guidance determination unit (205) may be such that, when the vehicle (10_e1) is the target vehicle (10), it recognizes another vehicle (10_e2) approaching an unprotected left turn intersection based on recognition information and allows the target vehicle (10) to safely perform an unprotected left turn.

[0184] Alternatively, in FIG. 10, the driving guidance generated by the driving guidance determination unit (205) may, when the vehicle (10_e2) is the target vehicle (10), recognize from a distance another vehicle (10_e1) that is attempting to make an unprotected left turn based on recognition information, and reduce the target vehicle's (10) approach speed to the unprotected left turn intersection so as not to cause a conflict with the other vehicle (10_e1) making an unprotected left turn.

[0185] Referring to FIG. 11, when a target vehicle (10) approaches a non-signalized intersection, a driving guidance determination unit (205) can generate driving guidance to prevent the target vehicle (10) from colliding with other vehicles at the non-signalized intersection based on recognition information received from a plurality of cooperative recognition means (110).

[0186] In Fig. 11, the road infrastructure (20) may provide the cooperative recognition server (11) with the traffic priority and passage time category, or, if the road infrastructure (20) is not installed within a predetermined range based on the non-signalized intersection, the surrounding vehicle (40) may provide the cooperative recognition server (11) with recognition information for object cooperative recognition. The situation recognition unit (201) may recognize the moving direction and speed of multiple recognition objects, such as other vehicles or pedestrians, around the non-signalized intersection based on signals received from the road infrastructure (20) or the surrounding vehicle (40). The driving intention analysis unit (203) and the driving trajectory determination unit (204) may analyze each driving intention based on the moving direction and speed of each of the multiple recognition objects, and determine each driving trajectory based on each driving intention. The driving guidance decision unit (205) can generate driving guidance by determining the driving lane, driving speed, etc. of the target vehicle (10) based on the driving trajectories of each of the other objects expected to collide with the target vehicle (10) at an unsignalized intersection.

[0187] In Fig. 11, the driving guidance generated by the driving guidance determination unit (205) may be such that, when the vehicle (10_f1) is the target vehicle (10), it recognizes another vehicle (10_f2) approaching the non-signalized intersection based on recognition information and allows the target vehicle (10) to safely make a left turn at the non-signalized intersection.

[0188] Alternatively, in FIG. 11, the driving guidance generated by the driving guidance determination unit (205) may, when the vehicle (10_f2) is the target vehicle (10), recognize from a distance another vehicle (10_f1) attempting to pass through the non-signalized intersection based on recognition information, and reduce the target vehicle's (10) approach speed to the non-signalized intersection so as not to cause a conflict with the other vehicle (10_f1) making a left turn at the non-signalized intersection.

[0189] Referring to FIG. 12, when it is determined that there is a Vulnerable Road User (VRU) crossing or attempting to cross a non-signalized crosswalk or non-signalized crosswalk in front of the target vehicle (10) based on recognition information received from a plurality of cooperative recognition means (110), the driving guidance determination unit (205) can determine driving guidance that allows the target vehicle (10) to drive without conflict with the VRU.

[0190] In Fig. 12, when receiving recognition information indicating a VRU approaching a non-signalized crosswalk from a plurality of cooperative recognition means (110), the situation recognition unit (201) can measure the moving direction, speed, etc. of the VRU based on the recognition information. The driving intention analysis unit (203) and the driving trajectory determination unit (204) can analyze the driving intention of the VRU based on the moving direction and speed of the VRU, and determine the driving trajectory and the expected time to complete crossing the non-signalized crosswalk of the VRU based on each driving intention. The driving guidance determination unit (205) can generate driving guidance by determining the driving lane, driving speed, etc. of the target vehicle (10) based on the driving trajectory of another VRU that is expected to collide with the target vehicle (10) at the non-signalized crosswalk.

[0191] Similarly, if it is determined that there is a road in front of the target vehicle (10) where a pedestrian path and a road are not distinguished and a VRU is crossing or attempting to cross the road based on recognition information received from a plurality of cooperative recognition means (110), the driving guidance determination unit (205) can determine driving guidance that allows the target vehicle (10) to drive without conflict with the VRU.

[0192] Referring to FIG. 13, when a target vehicle (10) approaches a roundabout, a driving guidance determination unit (205) can generate driving guidance to prevent the target vehicle (10) from colliding with a vehicle turning within the roundabout based on recognition information received from a plurality of cooperative recognition means (110).

[0193] The driving guidance decision unit (205) can exchange lane change messages with a plurality of cooperative recognition means (110) to seek agreement between a host object (Host Object, HO, e.g., target vehicle (10)) that is trying to change lanes and a remote object (Remote Object, RO, e.g., surrounding vehicle (40)). In the lane change message, the RO is an object that is expected to have a high possibility of collision when the HO changes lanes by currently occupying or will occupy a space in the target lane to which the HO is trying to move in the future, and may be a following object when the HO completes the lane change.

[0194] The lane change message may include a lane change yield request message (IG_AS_LaneChangeYeildRequestMessage). The lane change yield request message may be a message in which the HO, which wishes to change lanes, requests the RO to yield. The lane change yield request message may include the unique number of the guidance that serves as the basis for seeking agreement, the unique number of the agreement seeking request, the unique number of the HO, the unique number of the RO, the unique number of the lane to which the HO intends to move, the HO's driving path, etc. Here, the unique number of the lane to which the HO intends to move must be the same as the lane in which the RO is currently driving.

[0195] The lane change message may include a response message (IG_AS_LaneChangeYieldResponseMessage) to a lane change yield request. The response message to the lane change yield request may be a message in which the RO responds to the HO's lane change yield request. The response message to the lane change yield request may include the agreement seeking request unique number, the HO's unique number, the RO's unique number, the unique number of the lane the HO is trying to move to, and the response type (approval, rejection, etc.).

[0196] The driving guidance decision unit (205) can exchange intersection priority messages with multiple cooperative recognition means (110) to seek agreement between an HO (e.g., target vehicle (10)) and an RO (e.g., surrounding vehicle (40)) that want to pass through an intersection first. In the intersection priority message, RO refers to an object that is expected to have a high possibility of collision when the HO passes through the intersection by currently occupying or will occupy the space of the intersection that the HO wants to pass through in the future, and RO can indicate an object that enters the intersection after the HO completes passing through the intersection. Here, when the RO is 2 or more, the driving guidance decision unit (205) must perform an agreement seeking process with each of the 2 or more ROs.

[0197] The intersection priority message may include an intersection priority yield request message (IG_AS_CrossingYieldRequestMessage). The intersection priority yield request message may be a message from an HO seeking priority through an intersection requesting the RO to yield. The intersection priority yield request message may include the unique number of the guidance that serves as the basis for seeking agreement, the unique number of the agreement seeking request, the unique number of the HO, the unique number of the RO, and the unique number of the intersection. The unique number of the intersection may be an identifier to distinguish it from other intersections nearby.

[0198] The intersection priority message may include a response message (IG_AS_CrossingYieldResponseMessage) for a request to yield to an intersection priority. The response message for a request to yield to an intersection priority may be a message in which the RO responds to the request to yield to an intersection priority from the HO. The response message for a request to yield to an intersection priority may include a unique number for the agreement-seeking request, a unique number for the HO, a unique number for the RO, and a response type (approval, rejection, etc.).

[0199] The driving guidance decision unit (205) can generate a driving guidance message indicating the generated driving guidance and transmit it to the VCU (12). The VCU (12) can drive the target vehicle (10) according to the driving guidance indicated by the driving guidance message received from the cooperative recognition server (11).

[0200] Below, the operation of the cooperative recognition server (11) determining driving guidance based on cooperative recognition with the drone (30) is described.

[0201] A drone (30) is installed on a target vehicle (10) and / or road infrastructure (20) and can be deployed to a target area requiring cooperative recognition under the control of the road infrastructure (20) and / or the target vehicle (10).

[0202] If there is a road infrastructure (20) in the jurisdiction area corresponding to the location where the target vehicle (10) is driving, the drone (30) equipped on the target vehicle (10) and / or the road infrastructure (20) can be deployed to the target area under the control of the road infrastructure (20).

[0203] If there is no road infrastructure (20) in the jurisdiction area corresponding to the location where the target vehicle (10) is driving, the drone (30) equipped on the target vehicle (10) can be deployed to the target area.

[0204] The target area can be determined by the cooperative recognition server (11) equipped in each of the road infrastructure (20) or the target vehicle (10) based on the need for improving the situational awareness performance for cooperative recognition, expanding the recognition range, etc. For example, the cooperative recognition server (11) can determine the target area when the proportion of large vehicles is high, when congestion occurs, when an event occurs, when a Dynamic Driving Task (DDT) fallback situation occurs, etc. The DDT fallback situation is an essential part of a driving automation system of level 3 or higher, and can be performing DDT after a failure of the autonomous driving system (ADS) or driving that achieves the minimum risk condition.

[0205] FIG. 14 is an exemplary diagram illustrating a target vehicle equipped with a drone according to one embodiment of the present invention.

[0206] Referring to Fig. 14, when the target vehicle (10) is equipped with a drone (30), the drone (30) equipped in the target vehicle (10) can be ejected through the bonnet or roof of the target vehicle (10) when cooperative recognition is required.

[0207] In Fig. 14, a drone (30) for cooperative recognition is mounted inside the bonnet of a vehicle and the drone (30) is shown protruding above the bonnet of the vehicle for takeoff.

[0208] Below, the operation of the cooperative recognition server (11) determining driving guidance based on cooperative recognition with surrounding vehicles (40) is described.

[0209] Each of the sensor device (14) and the sensor device (41) can recognize an object from a predetermined area indicating the relative position in eight directions centered on the target vehicle (10) and the corresponding vehicle among the surrounding vehicles (40), and transmit recognition information including the recognized object information and situation information to the cooperative recognition server (11) through a cooperative recognition message.

[0210] FIG. 15 is an exemplary diagram illustrating the positions at which each of the target vehicle and surrounding vehicles recognizes an object according to one embodiment of the present invention.

[0211] Referring to FIG. 15, the eight directions centered on the target vehicle (10) or surrounding vehicle (40) may include the left front (R0), front (R1), right front (R2), left (R3), right (R4), left rear (R5), rear (R6), and right rear (R7) of the target vehicle (10) or surrounding vehicle (40).

[0212] Referring to FIG. 15, the relative positions of the eight directions of the target vehicle (10) or the surrounding vehicles (40) can be expressed as points where the center points of the smallest rectangles containing the vehicles are located in corresponding directions among the eight directions. Among the eight directions, the left front (R0), left (R3), and left rear (R5) may be areas on the left lane based on the lane in which the target vehicle (10) or the surrounding vehicles (40) are driving. The front (R1) and rear (R6) may be areas on the lane in which the target vehicle (10) or the surrounding vehicles (40) are driving. The right front (R2), right (R4), and right rear (R7) may be areas on the right lane based on the lane in which the target vehicle (10) or the surrounding vehicles (40) are driving.

[0213] When a surrounding vehicle (40) crosses laterally in front of a target vehicle (10) at an intersection or the like, the surrounding vehicle (40) can be recognized as being located at the left front (RO), front (R1), and right front (R2) of the target vehicle (10). Similarly, when a surrounding vehicle (40) crosses laterally in the rear of the target vehicle (10), the surrounding vehicle (40) can be recognized as being located at the left rear (R5), rear (R6), and right rear (R7) of the target vehicle (10).

[0214] In this way, the relative position between objects such as vehicles can be expressed as an index of one of the eight relative positions or a combination of two or more indices of the eight relative positions.

[0215] The target vehicle (10) or surrounding vehicles can recognize objects from each of the eight directional areas. Below, the operation of the cooperative recognition server (11) determining driving guidance based on cooperative recognition with road infrastructure (20) is described.

[0216] The road infrastructure (20) can recognize an object using a sensor device (21). The computing device (23) of the road infrastructure (20) can collect all of the object information recognized by the sensor device (21), situation information, information collected through cooperative recognition with a drone (30), and information collected through cooperative recognition with surrounding vehicles (40), update the front object information, road situation information, etc. required for autonomous driving of the target vehicle (10) to a dynamic information system (LDM), and generate a cooperative recognition message based on the updated information. The road infrastructure (20) can transmit the cooperative recognition message to a cooperative recognition server (11) within its jurisdiction through a communication device (22).

[0217] Figure 16 is a flowchart of a cooperative recognition method according to one embodiment of the present invention.

[0218] In the following, any redundant description of the above-described cooperative recognition system (1) may be omitted.

[0219] Referring to Fig. 16, when the road infrastructure (20) recognizes an object (S101), it can transmit recognition information indicating the recognized object to the cooperative recognition server (11) (S102).

[0220] When a drone (30) recognizes an object (S103), it can transmit recognition information indicating the recognized object to the cooperative recognition server (11) (S104).

[0221] When a surrounding vehicle (40) recognizes an object (S105), it can transmit recognition information indicating the recognized object to the cooperative recognition server (11) (S106).

[0222] When the target vehicle (10) recognizes an object (S107), it can transmit recognition information indicating the recognized object to the cooperative recognition server (11) (S108).

[0223] When the cooperative recognition server (11) receives recognition information from road infrastructure (20), drones (30), and surrounding vehicles (40), the situation recognition unit (201) analyzes the recognition information and can recognize the situation of multiple recognition objects within a predetermined distance range based on the location of the target vehicle (10) (S201).

[0224] The cognitive object alignment unit (202) can align each of a plurality of cognitive objects (S202).

[0225] The driving intention analysis unit (203) can estimate the driving intention of each of a plurality of recognition objects based on recognition information (S203).

[0226] The driving trajectory determination unit (204) can estimate the driving trajectory of each of the plurality of recognition objects based on the driving intention of each of the plurality of recognition objects and determine the expected driving trajectory (S204).

[0227] The driving guidance decision unit (205) can determine whether a conflict is expected to occur between objects including the plurality of recognition objects and the target vehicle (10) based on the driving trajectories of each of the plurality of recognition objects (S205).

[0228] If it is determined that no conflict will occur at step S205, the driving guidance decision unit (205) can determine whether a situation requiring guidance provision has occurred (S206).

[0229] If it is determined that a situation requiring guidance provision has not occurred in step S206, the driving guidance decision unit (205) can determine whether there is a first alternative route that represents an efficient travel alternative for traffic flow (S207).

[0230] If the first alternative route does not exist at step S207, the driving guidance decision unit (205) can determine whether a second alternative route representing an environmentally friendly travel alternative exists (S208).

[0231] If a second alternative route does not exist at step S208, the driving guidance decision unit (205) may terminate the cooperative recognition program without deciding on driving guidance.

[0232] If it is determined that a conflict will occur in step S205, if it is determined that a situation requiring guidance provision has occurred in step S206, if a first alternative route exists in step S207, or if a second alternative route exists in step S208, the driving guidance determination unit (205) can determine each corresponding driving guidance (S209).

[0233] Driving guidance may include lane change guidance and / or driving speed guidance to avoid a conflict if it is determined that a conflict will occur in step S205.

[0234] Driving guidance may include lane change guidance and / or driving speed guidance to avoid situations requiring guidance provision when it is determined that a situation requiring guidance provision has occurred in step S206.

[0235] The driving guidance may include guidance to cause the target vehicle (10) to drive along the first alternative route if a first alternative route exists in step S207.

[0236] If a second alternative route exists at step S208, guidance may be included to cause the target vehicle to drive on the second alternative route.

[0237] The driving guidance decision unit (205) can generate a driving guidance message indicating driving guidance (S210) and transmit it to the target vehicle (10) (S211).

[0238] The target vehicle (10) can drive according to the driving guidance indicated by the driving guidance message through the VCU (12) (S301).

[0239] Meanwhile, the disclosed embodiments may be implemented in the form of a recording medium storing computer-executable instructions. The instructions may be stored in the form of program code, and when executed by a processor, may generate program modules to perform the operations of the disclosed embodiments. The recording medium may be implemented as a computer-readable recording medium.

[0240] Computer-readable storage media include all types of storage media that store instructions that can be deciphered by a computer. Examples include read-only memory (ROM), random access memory (RAM), magnetic tape, magnetic disks, flash memory, and optical data storage devices.

[0241] While the embodiments of the present invention have been described above, the present invention is not limited to the above embodiments, and various modifications may be made within the scope of the detailed description and the attached drawings, as long as the scope of the invention is not deviated from or the effects are not impaired. Furthermore, it is obvious that such embodiments fall within the scope of the present invention.

Claims

1. Memory storing a program that determines driving guidance when the target vehicle drives autonomously; and A processor that executes the above program to receive recognition information indicating a plurality of recognition objects recognized within a predetermined distance range based on the position of the target vehicle from a plurality of cooperative recognition means and a target vehicle, estimates the driving intention and driving trajectory of each of the plurality of recognition objects based on the recognition information, and determines driving guidance for an autonomous driving path of the target vehicle based on the driving trajectory. Collaborative cognitive system.

2. In paragraph 1, The above processor, By collecting the cognitive object information received from the above-mentioned multiple cooperative cognitive means, a single dynamic situation information is determined, The above driving guidance is, Including longitudinal driving guidance, lateral driving guidance, and driving trajectory guidance, Collaborative cognitive system.

3. In paragraph 1, The above processor, A method of comparing a first driving trajectory of one object among the plurality of cognitive objects and a second driving trajectory of the target object, and predicting that the one object and the target object are in conflict when the difference between the first expected time point at which the one object is located at a first point intersecting the second driving trajectory among the first driving trajectories and the second expected time point at which the target object is located at the first point is less than a predetermined conflict threshold value, and generating driving guidance for the target vehicle to avoid the conflict when predicting that the one object and the target object are in conflict. Collaborative cognitive system.

4. In paragraph 1, The above multiple cooperative recognition means are: Road infrastructure including a plurality of first sensor devices installed on a target road on which the target vehicle is driven; At least one drone comprising a plurality of second sensor devices and deployed in a target area requiring situational information recognition from the target vehicle or the road infrastructure; and At least one of the surrounding vehicles, located around the target vehicle and including a plurality of third sensor devices, Collaborative cognitive system.

5. In paragraph 4, Each of the plurality of first to third sensor devices, including at least one of a camera, LiDAR, and RADAR, Collaborative cognitive system.

6. In paragraph 4, Each of the above multiple collaborative recognition means, Generating a wireless communication message including an identifier (ID) and type of each corresponding means, latitude, longitude, and altitude above sea level of the location of the corresponding means, information on a corresponding recognition object among the plurality of recognition objects, and a driving intention of the corresponding recognition object. Collaborative cognitive system.

7. In paragraph 6, The above driving intention is, Longitudinal driving intention including at least one of constant speed driving, acceleration driving, deceleration driving, and stopping; Lateral driving intention including at least one of lane keeping, left lane change, and right lane change; and Including an intersection traffic intent that includes at least one of straight ahead, left turn, right turn, and U-turn; Collaborative cognitive system.

8. In paragraph 6, Information of the corresponding cognitive object above The corresponding recognition object comprises two or more of the following: a time of recognizing the corresponding recognition object, an ID of the corresponding recognition object, a type of the corresponding recognition object, longitude and latitude of the location of the corresponding recognition object, an altitude above sea level of the location of the corresponding recognition object, a distance from a corresponding recognition means among the plurality of cooperative recognition means to the corresponding recognition object, a lateral and longitudinal speed of the corresponding recognition object, a moving azimuth of the corresponding recognition object, a lateral and longitudinal acceleration of the corresponding recognition object, a vertical acceleration of the corresponding recognition object, an angular velocity (yaw rate) of the corresponding recognition object, a detection area in which the corresponding recognition object is recognized, a driving lane in which the corresponding recognition object is driving, a width, a length, a height of the corresponding recognition object, and a past movement trajectory of the corresponding recognition object. Collaborative cognitive system.

9. In paragraph 8, The type of cognitive object corresponding to the above is, At least one of the following: unclassified vehicles, passenger cars, trucks, buses, pedestrians, bicycles, motorcycles, PM (Personal Mobility), CAV (Connected Autonomous Vehicle), AV (Non-connected Autonomous Vehicle), CHV (Conventional Human-Driven Vehicle), CV (Connected Vehicle), baby strollers, wheelchairs, animal propellants, trailer chassis, animals, and driving obstacles. Collaborative cognitive system.

10. In paragraph 8, The above detection area is, At least one of a general driving lane, a CAV exclusive lane, a merging acceleration lane, a merging main lane edge lane, a dividing main lane deceleration lane, a dividing main lane edge lane, a weaving section, an unprotected left-turn waiting lane, a left-turn lane, a right-turn lane, a straight lane, a U-turn lane, a combined lane including two or more of straight, left, right, and U-turn, a crosswalk, a roundabout entrance ramp, a roundabout exit ramp, a roundabout circling lane, a pedestrian path, an undivided road for pedestrians and vehicles, a two-lane road in both directions, a Hi-pass tollgate entrance ramp, a Hi-pass tollgate exit ramp, a manned tollgate entrance ramp, a manned tollgate exit ramp, and an emergency stop. Collaborative cognitive system.

11. A collaborative cognitive method performed by a processor, A step in which a plurality of cooperative recognition means and a target vehicle recognize a plurality of recognition objects within a predetermined range based on the location of the target vehicle; A step in which a processor receives recognition information representing the plurality of recognition objects from the plurality of cooperative recognition means and the target vehicle; A step in which the processor estimates the driving intention and driving trajectory of each of the plurality of recognition objects based on the recognition information; and A step of determining driving guidance for an autonomous driving path of the target vehicle based on the driving trajectory, How to recognize collaboration.

12. In paragraph 11, A step of comparing a first driving trajectory of one of the plurality of cognitive objects and a second driving trajectory of the target object; A step of predicting that the one object and the target object are in conflict when the difference between the first expected time point at which the one object is located at the first point intersecting the second driving trajectory among the first driving trajectories and the second expected time point at which the target object is located at the first point is less than a predetermined conflict threshold value; and Further comprising a step of generating driving guidance for the target vehicle to avoid the collision when the one object and the target object are predicted to collide. How to recognize collaboration.

13. A situation recognition unit that receives recognition information indicating a plurality of recognition objects recognized within a predetermined distance range based on the location of the target vehicle from a plurality of cooperative recognition means and a target vehicle; A driving intention analysis unit that estimates the driving intention of each of the plurality of recognition objects based on the above recognition information; A driving trajectory estimation unit that estimates the driving trajectory of each of the plurality of cognitive objects based on the driving intention of each of the plurality of cognitive objects; and A driving guidance determination unit that determines driving guidance for an autonomous driving path of the target vehicle based on the driving trajectory of each of the plurality of cognitive objects. A collaborative cognitive system including:

14. In paragraph 13, The above driving guidance decision unit is, A method of comparing a first driving trajectory of one object among the plurality of cognitive objects and a second driving trajectory of the target object, and predicting that the one object and the target object are in conflict when the difference between the first expected time point at which the one object is located at a first point intersecting the second driving trajectory among the first driving trajectories and the second expected time point at which the target object is located at the first point is less than a predetermined conflict threshold value, and generating driving guidance for the target vehicle to avoid the conflict when predicting that the one object and the target object are in conflict. Collaborative cognitive system.

Images