Automatic driving control device and automatic driving control method

Abstract

The present application relates to an automatic driving control device and an automatic driving control method, the method including collecting travel information of a host vehicle and at least one other vehicle traveling automatically, determining a travel intention of the other vehicle based on the travel information of the other vehicle, predicting a travel route of the other vehicle based on the travel intention of the other vehicle, and determining a travel route of the host vehicle based on the predicted travel route of the other vehicle.

Description

[0001] Cross-references to related applications

[0002] This application claims the benefit of Korean Patent Application No. 10-2020-0141305, filed on October 28, 2020, the entire contents of which are incorporated herein by reference. Technical Field

[0003] This invention relates to an automatic driving control device and an automatic driving control method. Background Technology

[0004] Generally speaking, autonomous vehicles refer to vehicles that can automatically drive to their destination without the need for a driver to operate the accelerator pedal, steering wheel, brake pedal, etc., while recognizing road conditions and the surrounding environment.

[0005] Traditional autonomous vehicles use information about the location and dynamics of another vehicle to calculate a predicted route for that vehicle, and then identify that predicted route based on a precise map to output the driving path of the other vehicle.

[0006] However, because vehicles actually driving on the road need to change their routes based on the driving conditions and intentions of neighboring vehicles and the current state of traffic lights, the predicted route of another vehicle based solely on information about its location and dynamics may differ from its actual route. Therefore, when vehicles are automatically driving along routes predicted using conventional methods in congested or crowded traffic conditions, it is difficult to maintain autonomous driving mode due to frequent false warnings and warning failures. Summary of the Invention

[0007] This invention relates to an autonomous driving control device and an autonomous driving control method. Specific embodiments relate to an autonomous driving control device and method capable of determining the driving intention of another vehicle to respond to various traffic environments during autonomous driving.

[0008] Therefore, embodiments of the present invention provide an autonomous driving control device and an autonomous driving control method that substantially eliminate one or more problems caused by the limitations and disadvantages of related technologies.

[0009] The embodiments of the present invention provide an autonomous driving control device and an autonomous driving control method that can improve the accuracy of predicting the driving routes of nearby vehicles in various traffic environments.

[0010] Specifically, embodiments of the present invention provide an autonomous driving control device and method capable of predicting the driving routes of neighboring vehicles by analyzing distances, traffic flow, dynamic characteristics, and infrastructure information between other vehicles existing within a search area and by determining the driving intentions of vehicles based on the mutual influence between vehicles.

[0011] However, the embodiments of the present invention are not limited to the above embodiments, and those skilled in the art will clearly understand other embodiments not mentioned herein through the following description.

[0012] An autonomous driving control method according to an embodiment of the present invention may include: collecting driving information of the vehicle itself and driving information of at least one other vehicle, determining the driving intention of the other vehicle based on the result of the driving information of the other vehicle with a predetermined index applied, predicting the driving route of the other vehicle based on the driving intention of the other vehicle, and determining the driving route of the vehicle itself based on the predicted driving route of the other vehicle.

[0013] Furthermore, an autonomous driving control device according to an embodiment of the present invention may include: a first determiner, a second determiner, and a driving controller; the first determiner is configured to collect driving information of the autonomously driving vehicle and driving information of at least one other vehicle present near the vehicle to determine the traffic environment; the second determiner is configured to determine the driving intention of a vehicle of interest traveling directly in front of the vehicle in the same lane as the vehicle based on the driving information of the other vehicles; the driving controller is configured to predict the driving routes of the other vehicles based on the driving intentions of the other vehicles and the driving information of the other vehicles, and determine the driving route of the vehicle based on the predicted driving routes of the other vehicles. Attached Figure Description

[0014] Embodiments of the invention are illustrated in the accompanying drawings, which are included and form part of this application to provide a further understanding of the invention, and together with the description, serve to explain the principles of the embodiments of the invention. In these drawings:

[0015] Figure 1 This is a schematic block diagram of an automatic driving control device according to an embodiment of the present invention;

[0016] Figure 2 It is shown Figure 1 A schematic block diagram showing the configuration of other vehicle determinants;

[0017] Figure 3 This is a control flowchart of another vehicle determiner according to an embodiment of the present invention;

[0018] Figure 4A , Figure 4B , Figure 5A and Figure 5B This is a schematic diagram illustrating a method for determining lane space allowance according to an embodiment of the present invention;

[0019] Figure 6 This is a schematic diagram illustrating an infrastructure-based determination method according to an embodiment of the present invention;

[0020] Figure 7A and Figure 7B This is a schematic diagram illustrating a method for determining the bias value of other vehicles according to an embodiment of the present invention;

[0021] Figure 8A and Figure 8B This is a schematic diagram illustrating other methods for determining the direction of travel of a vehicle according to embodiments of the present invention;

[0022] Figures 9 to 12 This is a schematic diagram illustrating a method for determining the driving intention of another vehicle according to an embodiment of the present invention;

[0023] Figure 13 and Figure 14 This is a schematic diagram illustrating a method for controlling a vehicle based on the driving intention of another vehicle according to an embodiment of the present invention. Detailed Implementation

[0024] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings to enable those skilled in the art to readily implement these embodiments. However, the invention can be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. In the drawings, for clarity, portions irrelevant to the description of the embodiments of the invention will be omitted. Throughout the specification, the same reference numerals refer to the same elements.

[0025] Throughout this specification, when a part "includes" or "contains" a component, this does not exclude other components, which may be further included unless otherwise stated. The same reference numerals used throughout this specification refer to the same constituent elements.

[0026] According to an embodiment of the present invention, an automated driving control device analyzes the driving strategy of another vehicle, which is determined by considering the distance between neighboring vehicles, the driving intention of neighboring vehicles, vehicle speed, the presence of dangerous vehicles, and surrounding infrastructure, and predicts the driving strategies of all vehicles within a predetermined search area accordingly. Based on the predicted driving strategy of the other vehicle, the automated driving control device, while controlling the driving of its own vehicle, determines the route required for smooth driving, selects a driving strategy for entering the appropriate lane, and controls the driving of its own vehicle. Thus, since the automated driving of the own vehicle is based on the driving intention of the other vehicle, risk determination criteria can be variably applied taking into account the driving intention of the other vehicle, thereby reducing the occurrence of false warnings and warning failures, and reducing unnecessary stopping or deceleration of the own vehicle that may occur due to not considering the future positions of other vehicles. Furthermore, it can respond to the vehicle changing lanes ahead based on the driving intention of other vehicles in various traffic environments (e.g., complex traffic situations where two or more vehicles change lanes simultaneously), and achieve optimal driving control of the own vehicle.

[0027] In the following description, a vehicle driving control device related to embodiments of the present invention will be described with reference to the accompanying drawings. First, the main terms used in this specification and drawings will now be described.

[0028] This vehicle: its own vehicle

[0029] Another vehicle: a vehicle other than this vehicle.

[0030] Nearby vehicles: Vehicles other than this vehicle detected by sensors installed on this vehicle.

[0031] The vehicle in front: the adjacent vehicle traveling directly in front of this vehicle.

[0032] Driving lane: The lane in which this vehicle is currently traveling.

[0033] Target lane: The lane this vehicle intends to enter.

[0034] Vehicles in the target lane: Adjacent vehicles traveling in the target lane

[0035] Figure 1 This is a schematic block diagram of an automatic driving control device according to an embodiment of the present invention.

[0036] refer to Figure 1 According to an embodiment of the present invention, the autonomous driving control device includes: a sensor 100, a transceiver 110, a map transmission module 118, a driving environment determiner 120, an other vehicle determiner 200, and a driving controller 300.

[0037] Sensor 100 can sense one or more neighboring vehicles located in front of, to the side of, or behind the vehicle, and can detect the position, speed, and acceleration of each neighboring vehicle. Sensor 100 may include various sensors mounted on the front, sides, and rear of the vehicle, such as lidar 102, camera 104, and radar 106.

[0038] The lidar 102 can measure the distance between the vehicle and neighboring vehicles. The lidar 102 can emit laser pulses towards neighboring vehicles and measure the arrival time of the laser pulses reflected from the neighboring vehicles to calculate the spatial coordinates of the reflection point, thereby determining the distance to the neighboring vehicles and the shape of the neighboring vehicles.

[0039] Camera 104 can acquire images of the vehicle's surrounding environment via an image sensor. Camera 104 may include an image processor for performing image processing on the acquired images (e.g., noise removal, quality and saturation adjustment, and file compression).

[0040] Radar 106 can measure the distance between the vehicle and nearby vehicles. Radar 106 can emit electromagnetic waves toward nearby vehicles and receive electromagnetic waves reflected from nearby vehicles to determine the distance to nearby vehicles, as well as the direction and altitude of the nearby vehicles.

[0041] Transceiver 110 can receive information used to sense the location of the vehicle and the locations of nearby vehicles. Transceiver 110 may include various devices capable of receiving information used to identify the location of the vehicle, such as vehicle-to-everything (V2X) transceiver 112, controller area network (CAN) transceiver 114, and global positioning system (GPS) 116.

[0042] The map transmission module 118 provides a lane-distinguished, accurate map. The accurate map can be stored in the form of a database (DB), can be automatically and periodically updated via wireless communication or manually updated by the user, and can include information about lane merging segments (e.g., including the location information of the merging segments and the legal speed limit information of the merging segments), road information based on location, road branch information, and intersection information.

[0043] The driving environment determiner 120 can fuse object information about its own vehicle and other vehicles into an accurate map based on information acquired by the sensor 100, the map transmission module 118, and the transceiver 110, and then output the fused object information. The driving environment determiner 120 may include an object fusion module 122, a road information fusion module 124, and a vehicle position recognition module 126.

[0044] The vehicle location identification module 126 outputs the vehicle's precise location information. The vehicle location identification module 126 can compare the information sensed by the sensor 100, the vehicle's GPS information collected by the transceiver 110, and the precise map information provided by the map transmission module 118, and can simultaneously output the vehicle's location information and location identification reliability information.

[0045] The road information fusion module 124 outputs a precise map of the vehicle's surrounding environment. Based on location recognition reliability information and precise map information, the road information fusion module 124 outputs the precise map information of the vehicle's surrounding environment to the object fusion module 122.

[0046] The object fusion module 122 outputs the fused object information to other vehicle determiners 200. Based on the information sensed by the sensor 100 and the accurate map information of the vehicle's surrounding environment, the object fusion module 122 fuses the objects into the accurate map and then outputs the fused object information.

[0047] The other vehicle determiner 200 can receive information about objects already fused into a precise map to determine the driving intentions of other vehicles, and the driving controller 300 can consider the driving intentions of other vehicles output from the other vehicle determiner 200 to determine the driving route of its own vehicle and control its driving. The other vehicle determiner 200 can, as... Figure 2 Configure as shown in the block diagram.

[0048] Figure 2 It is shown Figure 1 A block diagram showing an example configuration of another vehicle determiner 200.

[0049] Other vehicle determiner 200 can receive information about objects that have been integrated into a precise map from driving environment determiner 120, and can determine the driving intentions of other vehicles.

[0050] refer to Figure 2 The other vehicle determiner 200 may include a lane-based free space determination module 210, an infrastructure-based determination module 212, a lane-based traffic flow determination module 214, an other vehicle bias value determination module 215, an other vehicle travel direction determination module 216, and an other vehicle comprehensive driving intention determination module 218.

[0051] The lane clearance determination module 210 outputs a vehicle distance index based on the distance between vehicles. The vehicle distance index can be a numerical value, code, or score assigned based on the distance between vehicles.

[0052] The infrastructure-based determination module 212 determines and outputs a vehicle's driving index based on the current traffic light, remaining time, the next traffic light (when V2X is available), and road markings such as bus stops or school zones. The vehicle's driving index can be a numerical value, code, or score assigned based on the range the vehicle can travel without interruption (e.g., deceleration, stopping, or lane changes).

[0053] The lane traffic flow determination module 214 converts the traffic flow speed within the corresponding lane into an index based on the vehicle speed and outputs the index. The lane traffic flow index can be a numerical value, code, or score assigned according to the speed of vehicles in each lane.

[0054] The Other Vehicle Bias Value Determination Module 215 outputs an index related to the degree to which another vehicle deviates from the center of the lane it is traveling in. The Other Vehicle Bias Value Index can be a numerical value, code, or score assigned based on the degree of deviation of the other vehicle from the center of the lane it is traveling in.

[0055] The other vehicle direction determination module 216 outputs an index related to the angle of travel of other vehicles relative to the target lane. The other vehicle direction index can be a value, code, or score assigned based on the angle of travel of other vehicles relative to the target lane.

[0056] The Other Vehicles Comprehensive Driving Intent Determination Module 218 comprehensively considers the indices output by the above five sub-modules and finally outputs the target lane that other vehicles intend to enter.

[0057] Figure 3 This is a control flowchart of another vehicle determiner 200 according to an embodiment of the present invention.

[0058] Other vehicle determiner 200 can receive information about objects that have been integrated into a precise map from the driving environment determiner 120, and can determine the driving intention of another vehicle.

[0059] Therefore, the other vehicle determiner 200 outputs the vehicle distance index based on the vehicle distance (S110).

[0060] Other vehicle determiners 200 determine and output the vehicle's driving index (S112) based on infrastructure such as the current traffic light, remaining time, the next traffic light (when V2X is available) and road markings such as bus stops or school zones.

[0061] Other vehicle determiner 200 converts the traffic flow speed in the corresponding lane into an index based on the speed of the vehicles traveling in each lane, and outputs the index (S114).

[0062] The other vehicle determiner 200 outputs an index (S116) related to the degree to which other vehicles deviate from the center of their lane relative to other vehicles.

[0063] The other vehicle determiner 200 outputs an index (S118) related to the angle of travel of other vehicles relative to the target lane.

[0064] The other vehicle determiner 200 takes into account the index calculated in the above steps and finally outputs the target lane that other vehicles intend to enter (S210).

[0065] Figure 4A , Figure 4B , Figure 5A and Figure 5B This is a schematic diagram illustrating a method for determining lane space availability according to an embodiment of the present invention. To determine the driving strategy of vehicle M, it is necessary to predict the driving intention of vehicle a traveling in front of vehicle M within the lane in which vehicle M is traveling. To predict the driving intention and determine the driving route of target vehicle a, the lane space availability determination module 210 of the other vehicle determiner 200 can output an index based on the lane space availability.

[0066] Figure 4A and Figure 4B The diagram illustrates the available space in each lane under different conditions. As the distance between traveling vehicles increases, the probability of accelerating or maintaining a constant speed to reduce the distance between vehicles increases. This probability can be used as a factor to increase the average speed within the corresponding lane, and also as a factor to increase the probability that the target vehicle a will choose the corresponding lane based on its driving intention.

[0067] The target vehicle 'a' can be identified based on the driving intentions of vehicles 'M' ahead of the target vehicle 'M', and the available space within the lanes can be calculated within a preset search distance. The available space within the lanes can be measured as the distance between the front of the target vehicle 'a' and the rear of other vehicles 'b', 'c', and 'd' ahead of it, and the distance between the front of other vehicles 'b', 'c', and 'd' and the rear of other vehicles 'e' and 'f' ahead of them. The available space within each lane can be calculated as the sum of the available spaces within each lane. Since the available space within each lane corresponds to the distance between vehicles (excluding the vehicle's length), the sum of the distances between vehicles within a lane decreases as the number of vehicles traveling in that lane increases. Conversely, the sum of the distances between vehicles within a lane increases as the number of vehicles traveling in that lane decreases. If there are no vehicles in a lane, the total search distance can be determined as the distance between vehicles.

[0068] The lane clearance determined in the above manner is also related to the degree of vehicle congestion. Lane clearance can be replaced by the number of vehicles per predetermined distance (e.g., number / km) and can be used when calculating the index.

[0069] Figure 5A and Figure 5B The method of assigning an index based on the speed of each vehicle is shown in different situations.

[0070] To account for the speed factor that has the greatest impact on traffic flow in each lane, weights can be assigned based on speed even when the same margin of safety is measured in different lanes. Generally, the higher the speed of vehicle c traveling directly in front of the target vehicle a in the same lane, and vehicles b and d traveling in adjacent lanes, the easier it is to follow. Conversely, the lower the speed of vehicles c, b, and d, the less likely it is to follow. Therefore, the index can be assigned based on the speed of each vehicle, with higher weights for higher speeds.

[0071] For vehicles traveling directly in front of vehicles c, b, and d that are ahead of the target vehicle a, whose driving intention is determined, attention can be reduced. Therefore, weights can be set differently based on the position numbers of vehicles counting from the target vehicle a in the direction of travel.

[0072] Therefore, when another vehicle has a higher speed and is closer to the target vehicle a, which is the intended target, a higher weight can be assigned to that other vehicle. For example, as Figure 5A and Figure 5B As shown, based on traffic flow, 8 to 10 points can be allocated to vehicles b, c, and d that are directly in front of the vehicle a whose driving intention is determined; 5 to 7 points can be allocated to vehicles e and f that are in front of vehicles b, c, and d; and 2 to 4 points can be allocated to vehicles (not shown) that are in front of vehicles e and f.

[0073] However, when a vehicle such as a bus or taxi decelerates in the outermost lane in front of a target vehicle (a) with a driving intention, the probability of that vehicle stopping is higher than that of stopping due to deceleration for other reasons. Therefore, a higher weight can be assigned based on the type of vehicle.

[0074] Figure 6 This is a schematic diagram illustrating an infrastructure-based determination method according to an embodiment of the present invention.

[0075] Generally, when approaching a traffic signal at an intersection, a vehicle tends to change its driving strategy according to the status of the traffic signal while driving. For example, when the traffic signal in front of the vehicle is green and the remaining time is 3 seconds, even if it is predicted based on the driving routes of other vehicles that the vehicle will pass through the intersection in 3 seconds, this prediction may be different from the actual driving of the vehicle. Therefore, information on infrastructure such as traffic signal I1, school area I2, and bus stop I3 needs to be used to predict the driving intentions of other vehicles.

[0076] As Figure 6 shown, the exponential assignment method is simplified into three stages, such that the exponents for other vehicles with predicted stop times of T1, T2, and T3 (T1 < T2 < T3) are set to S1 points, S2 points, and S3 points (S1 < S2 < S3), respectively. However, the exponential assignment method can be implemented in any of various other ways.

[0077] In addition, higher exponents are assigned to locations or sections where other vehicles can drive more smoothly, so that the future driving conditions of other vehicles can be predicted more accurately when determining the driving intentions of other vehicles.

[0078] On the other hand, when it is predicted that other vehicles are driving slowly due to a stop signal, school area, or bus stop, a smaller exponent can be assigned to the driving intentions of other vehicles based on this.

[0079] Generally, since the road shoulder or the outermost lane is an area where vehicle parking and stopping frequently occur, many ordinary drivers tend to reduce the vehicle speed to the average driving speed when driving in the corresponding lane. Therefore, a relatively small exponent can be applied to the corresponding lane.

[0080] Figure 7A And Figure 7B are schematic diagrams for explaining a method for determining the deviation value of other vehicles according to an embodiment of the present invention.

[0081] Figure 7A shows a case where the center line CL' of the target vehicle a for determining the driving intention deviates to the left from the center line CL of the driving lane of the vehicle M, Figure 7B shows a case where the center line CL' of the target vehicle a for determining the driving intention deviates to the right from the center line CL of the driving lane of the vehicle M.

[0082] As Figure 7AAs shown, when the driving intention determines that the target vehicle a is skewed to the left, the time required for it to enter the right lane may increase, but the time required to enter the left lane may decrease. Typically, for economic reasons, vehicles tend to travel in a state skewed towards the target lane before entering it, prior to entering the target lane. Therefore, the driving intention index associated with lane change can increase proportionally to the lane skewed by the driving intention determining the target vehicle a, along with its skew value.

[0083] Even when the driving intention of target vehicle a enters the lane in the opposite direction to the lane that the driving intention of target vehicle a has already deviated from, its driving intention is recalculated every frame based on the real-time bias value, so that the latest bias information can be used to determine the updated driving intention.

[0084] Figure 8A and Figure 8B This is a schematic diagram illustrating other methods for determining the direction of travel of a vehicle according to embodiments of the present invention.

[0085] Figure 8A This illustrates a scenario where the target vehicle a's body is offset to the left relative to the centerline of the lane in which the vehicle M is traveling, indicating a driving intention. Figure 8B This illustrates a scenario where the target vehicle a's body is deflected to the right relative to the centerline of the driving lane of the vehicle M, indicating a driving intention.

[0086] Because the driving intention determines that the target vehicle a's direction of travel is at a greater angle toward the left lane, the time required for it to enter the right lane may increase, but the time required to enter the left lane may decrease. Generally, for economic reasons, vehicles tend to move toward the target lane before entering it.

[0087] Therefore, the indices A, B, and C can be set according to the angle θ at which the target vehicle a travels towards the target lane relative to the centerline of the driving lane, based on the driving intention. Here, the values ​​of indices A, B, and C, as well as the curve shape of the (2n+1)th order (n∈N) graph, can be set as tuning parameters.

[0088] For example, the driving intention index associated with lane change can increase proportionally to the angle +θ that the target vehicle a is traveling toward the target lane. Even when the target vehicle a enters the lane in the opposite direction to the lane it is already traveling at an angle -θ, its driving intention is recalculated every frame based on the real-time direction of travel, thus allowing the latest bias information to be used to determine the updated driving intention.

[0089] Figures 9 to 12This is a schematic diagram illustrating a method for determining the driving intention of other vehicles according to an embodiment of the present invention.

[0090] Figure 9 The first diagram in the diagram shows the margin of safety and weight settings within the lane. Figure 9 The second diagram illustrates the conversion of the index based on the available space and weight within the lane. Figure 9 The third table shows the method for calculating the total score based on indices according to lane space and weights, bias values, travel angles, and infrastructure.

[0091] refer to Figure 9 The first diagram illustrates how, starting with the target vehicle 'a' determined by its driving intention, the available space within a lane is measured within a preset search distance. Weights can be applied to the available space within the lane and can be converted into an exponent. These weights can be assigned based on the vehicle's speed and the position number of the vehicle counting along the direction of travel starting with the target vehicle 'a' determined by its driving intention.

[0092] Based on traffic flow, 8 to 10 points of weight can be allocated to vehicles b, c, and d that are closest to the target vehicle a, whose driving intention is determined; 5 to 7 points of weight can be allocated to vehicles e and f that are the second closest to the target vehicle a; and 2 to 4 points of weight can be allocated to vehicles ahead of vehicles e and f. Alternatively, if there are no vehicles ahead, the surplus space can be defined as the space up to the end of the search area, and 2 to 4 points of weight can be allocated to this surplus space. Weights can be applied to the surplus space between vehicles and can be converted into exponents, such as... Figure 9 The second diagram in the diagram is shown.

[0093] For example, when the route from the target vehicle a with a determined driving intention to the vehicle c in the same lane that is directly in front of the target vehicle a is called "node 3", and the margin of space in between is measured to be 4m, the weights for speed and position can be set to 9. In this case, the index of node 3 can be set to 13 points.

[0094] When the route from the target vehicle a with a defined driving intention to vehicle b, which is located in the right lane ahead of the target vehicle a, is designated as "node 5", and the margin of safety in between is measured to be 5m, the weights for speed and position can be set to 9. In this case, the index of node 5 can be set to 14.

[0095] When the route from the target vehicle a with the determined driving intention to the vehicle d in the left lane ahead of the target vehicle a is called "node 1", and the margin of space between them is measured to be 10m, the weights of speed and position can be set to 10. In this case, the index of node 1 can be set to 20 points.

[0096] When the route from vehicle c to vehicle f, which is located in front of vehicle c, is called "node 4", and the margin of space in between is measured to be 7m, the weights for speed and position can be set to 7. In this case, the index of node 4 can be set to 14.

[0097] When the route from vehicle b to vehicle e, which is located in front of vehicle b, is called "node 6", and the margin of space in between is measured to be 5m, it can be set to 6 based on the weights of speed and position. In this case, the index of node 6 can be set to 11.

[0098] When there are no vehicles ahead of vehicle d, the route to the end of the search area is called "node 2", and the margin of space in between is measured to be 12m. The weights for speed and position can be set to 7. In this case, the index of node 2 can be set to 19.

[0099] When there are no vehicles ahead of vehicle e, the route to the end of the search area is called "node 7", and the margin of space in between is measured to be 2m, the weights for speed and position can be set to 2. In this case, the index of node 7 can be set to 4.

[0100] refer to Figure 9 In the table, to determine the driving intention of target vehicle a, a total score can be calculated based on the index of lane space and weight, the index based on the bias value, the index based on the travel angle, and the index based on the infrastructure.

[0101] The index used to determine driving intention can be calculated based on the driving intention to determine the expected route of the target vehicle a, i.e., the case of staying in the current lane (nodes 3 and 4), changing to the left lane (nodes 1 and 2), or changing to the right lane (nodes 5 and 6).

[0102] When the intended travel of target vehicle a is determined to be changing from the current lane to the left lane, the indices of nodes 1 and 2 can be used. Furthermore, since the centerline CL' of target vehicle a is deviated to the left from the centerline CL of the driving lane, the deviation value index can be set to 15 points. Additionally, based on the travel angle + θ of the target vehicle a's body relative to the centerline of the driving lane, the travel direction index can be set to 15 points. Information about infrastructure such as traffic lights or bus stops can be used to set the infrastructure index to 15 points.

[0103] When the intended travel of target vehicle a is determined to remain in the current lane, the indices of nodes 3 and 4 can be used. Furthermore, since the centerline CL' of target vehicle a is determined to be to the left of the centerline CL of the travel lane, the deviation value index can be set to 3 points. Additionally, based on the travel angle +θ of the target vehicle a's body relative to the centerline of the travel lane, the intended travel direction index can be set to 0 points. Based on information about infrastructure such as traffic lights or bus stops, the infrastructure index can be set to 15 points.

[0104] When the intended travel of target vehicle a is determined to be changing from the current lane to the right lane, the indices of nodes 5, 6, and 7 can be used. Additionally, since the centerline CL' of target vehicle a is deviated to the left from the centerline CL of the driving lane, and the lane is changing to the right lane, the deviation value index can be set to 0. Furthermore, based on the travel angle +θ of the target vehicle a's body relative to the centerline of the driving lane, the travel direction index can be set to -15. Based on information about infrastructure such as traffic lights or bus stops, the infrastructure index can be set to 10.

[0105] When the anticipated driving intention determines that target vehicle a changes from the current lane to the left lane, the total score of the index calculated as described above is 84. When the anticipated driving intention determines that target vehicle a remains in the current lane, the total score of the index calculated as described above is 45. When the anticipated driving intention determines that target vehicle a changes from the current lane to the right lane, the total score of the index calculated as described above is 24. Therefore, it can be determined that the probability of changing from the current lane to the left lane (which is assigned the highest total score) is the highest.

[0106] Figure 10 It is used to explain the basis Figure 9 The diagram illustrates how the index calculation method programs the equation used to calculate the total score.

[0107] Figure 10The “Dnn / Vnn” described in the text is a mathematical representation of the driving intention determining the distance and traffic flow between target vehicle a and each of other vehicles b, c, d, e, and f. The index, which is the result of mathematically transforming the driving intention determining the distance and traffic flow between target vehicle a and each of other vehicles b, c, d, e, and f, can be substituted into Equation 1 below to calculate the final total score.

[0108] D (LaneNum)，N The score is the distance between vehicle N and vehicle N-1 in lane (LaneNum).

[0109] V (LaneNum)，N The score of the traffic flow for the Nth vehicle in lane (LaneNum).

[0110] I (LaneNum) The score for the driving of another vehicle in lane (LaneNum) relative to the infrastructure.

[0111] B LaneNum The score of the other vehicle's bias relative to lane number (LaneNum).

[0112] H LaneNum(θ) The score for the other vehicle's travel angle relative to lane number (LaneNum).

[0113] W D W V W B W H W I Regarding the weighting of the above scores

[0114] Equation 1:

[0115]

[0116] Here, when the value of "NextLaneNum" is determined to be high enough to cause a collision based on the time to collision (TTC), it can be finally determined that the driving intention of the target vehicle a will not change lanes.

[0117] The driving intention prediction method described above can be used to predict the driving intentions of all vehicles in front of the vehicle in the left lane, right lane, and same lane, thereby determining the driving route of each vehicle at the next time point.

[0118] When it is predicted that another vehicle will remain in the current lane, PID control based on the distance to the tracking point ahead can be executed. When it is predicted that another vehicle will change lanes, it is necessary to determine the vehicle's driving strategy in response to the lane change to the corresponding lane link. When the time required for the lane change is T, the corresponding vehicle can be considered to be in the "NextLane" after T seconds. That is, the position changes according to the expected lane change route.

[0119] The required time T may depend to a large extent on the direction of travel of the other vehicle, its current speed, and its deviation value, and can be based on, for example, Figure 11 The table of pre-calculated values ​​shown is used to calculate appropriate values ​​through interpolation or mathematical modeling.

[0120] Or, such as Figure 12 As shown, the above parameters can be learned based on deep learning parameters, and the required time can be calculated using time-series prediction methods such as LSTM or CNN.

[0121] Ultimately, the driving intention of another vehicle can be determined by the target lane and the time points at which other vehicles are located in that target lane. Furthermore, through a single determination and N sample observations, the driving intention of another vehicle and the time required to change lanes can be calculated as reliable data.

[0122] Figure 11 and Figure 12 A calculation method that can be used to predict the driving intention of another vehicle according to an embodiment of the present invention is shown.

[0123] Figure 11 This diagram shows the results of modeling the required time using a mesh plot, based on the direction of travel and vehicle speed, when the driving intention determines the bias value of target vehicle a to be 0.8m and 0. A uniform speed lane change model can be calculated using a required time plot or a 3D polygon path model, where the crossing time for travel angle and vehicle speed is shown.

[0124] Figure 12 This is a schematic diagram illustrating an implementation scheme that uses deep learning to process input data such as direction of travel, current speed, and bias value to calculate the time required for another vehicle to change lanes.

[0125] Figure 13 and Figure 14 This is a schematic diagram illustrating a method for controlling a vehicle based on the driving intention of another vehicle according to an embodiment of the present invention.

[0126] Figure 13 and Figure 14This is a schematic diagram illustrating a method for changing the strategy of the vehicle according to an embodiment of the present invention. Specifically, Figure 13 This demonstrates a method for controlling the vehicle's movement when the vehicle in front changes lanes to the left. Figure 14 This demonstrates a method for controlling the vehicle's movement while the vehicle in front remains in the same lane.

[0127] refer to Figure 13 The vehicle f traveling directly in front of vehicle M can be the target vehicle whose driving intention is determined. When it is determined that the distance between vehicles in the left lane is relatively long, the traffic flow in the left lane is relatively smooth, and the target vehicle f whose driving intention is determined is moving to the left, it can be predicted that the corresponding vehicle f will change to the left lane.

[0128] Therefore, it is possible to control vehicle M to remain in the current lane or change to the right lane.

[0129] refer to Figure 14 The vehicle f traveling directly in front of vehicle M can be the target vehicle for determining its driving intention. When it is determined that the distance between vehicles in the lane in which vehicle M is traveling is neither too short nor too long, the traffic flow in the corresponding lane is relatively smooth, and the target vehicle f for determining its driving intention is in a normal driving state, it can be predicted that the corresponding vehicle f will remain in the current lane.

[0130] Therefore, it is possible to control vehicle M to remain in the current lane or to change to the left lane, which is a longer distance between vehicles.

[0131] As described above, embodiments of the present invention can predict the routes of other vehicles in a search area based on the distance between adjacent vehicles, the driving intentions of adjacent vehicles, vehicle speed, the presence of dangerous vehicles and surrounding infrastructure, and determine the driving route of the vehicle based on the predicted routes of other vehicles, thereby enabling the vehicle to react to vehicles changing lanes ahead and maintain optimal driving conditions in various traffic environments (e.g., complex traffic situations where two or more vehicles change lanes simultaneously).

[0132] Specifically, the embodiments of the present invention select the vehicle closest to the vehicle in the same lane as the vehicle of interest, analyze the distance, traffic flow, dynamic characteristics and infrastructure information between other vehicles in the search area of ​​the vehicle of interest, and determine the driving intention of the vehicle based on the mutual influence between vehicles, thereby improving the accuracy of predicting the driving routes of other vehicles, including the vehicle of interest.

[0133] Furthermore, by variably applying risk assessment criteria while taking into account the driving intentions of other vehicles, the occurrence of false warnings and warning failures can be reduced, as can the occurrence of stops or decelerations that may occur due to the failure to consider the future positions of other vehicles, thereby improving ride comfort.

[0134] It is evident from the above description that the automatic driving control device and automatic driving control method according to at least one embodiment of the present invention, configured as described above, can improve the accuracy of predicting the driving routes of nearby vehicles in various traffic environments.

[0135] Specifically, embodiments of the present invention can predict the driving routes of neighboring vehicles by analyzing the distances, traffic flows, dynamic characteristics, and infrastructure information between other vehicles existing within the search area, and by determining the driving intentions of vehicles based on the mutual influence between vehicles.

[0136] Furthermore, embodiments of the present invention can reduce the occurrence of false warnings and warning failures by variably applying risk determination criteria by taking into account the driving intentions of other vehicles, and can reduce the occurrence of stops or decelerations that may occur due to not taking into account the future positions of other vehicles, thereby improving ride comfort.

[0137] However, the effects achievable through the embodiments of the present invention are not limited to those described above, and those skilled in the art will clearly understand other effects not mentioned herein through the above description.

[0138] Embodiments of the present invention can be implemented as code that can be written on a computer-readable recording medium and thus read by a computer system. Computer-readable recording media include various recording devices in which data that can be read by a computer system is stored. Examples of computer-readable recording media include hard disk drives (HDDs), solid-state drives (SSDs), silicon disk drives (SDDs), read-only memory (ROM), random access memory (RAM), read-only optical disc drives (CD-ROMs), magnetic tape, floppy disks, and optical data storage devices.

[0139] It will be apparent to those skilled in the art that various changes in form and detail may be made without departing from the spirit and essential characteristics of the embodiments of the invention described herein. Therefore, the detailed description above is not intended to be construed as limiting the invention in all respects, but is to be considered as exemplary. The scope of the invention should be determined by a reasonable interpretation of the appended claims, and all equivalent modifications made without departing from the invention should be included in the appended claims.

Claims

1. An automatic driving control method, the method comprising: Collect driving information of the self-driving vehicle and at least one other vehicle; Determine the driving intentions of other vehicles based on their driving information; Predict the driving routes of other vehicles based on their driving intentions; The vehicle's route is determined based on the predicted routes of other vehicles. in: Determining the driving intentions of other vehicles includes: determining the driving intentions of vehicles of interest that are traveling directly in front of the vehicle in the same lane as the vehicle in question; Determining the driving intention of a vehicle of interest includes: predicting the lane that the vehicle of interest intends to enter based on at least one of the following: the position and speed information of the vehicle of interest and other vehicles, the bias value information of the vehicle of interest, or the direction of travel information of the vehicle of interest. Determining the driving intention of the vehicle under observation includes: An index related to lane clearance is calculated based on the distance between the vehicle in question and another vehicle. Information based on infrastructure is used to calculate indices related to the vehicle's driving status. Calculate the index related to the speed of vehicles in each lane; Calculate an index related to the degree of deviation of the vehicle of interest from the center of the lane in which the vehicle of interest is traveling; Calculate and monitor indices related to the vehicle's travel angle; By comprehensively determining the calculated indices, the lane that the vehicle in question intends to enter can be predicted.

2. The method according to claim 1, wherein, The driving information includes at least one of the following: the position information of the vehicle and other vehicles, the speed information of the vehicle and other vehicles, the acceleration information of the vehicle and other vehicles, the deviation value information of other vehicles, the travel angle information of other vehicles, precise map information, or information containing infrastructure such as traffic lights, bus stops, and school zones.

3. The method according to claim 1, wherein, Determining the driving intentions of other vehicles includes: predicting the lanes that other vehicles intend to enter based on the results of applying a predetermined index to the driving information of other vehicles.

4. The method according to claim 1, wherein, The calculation of indices related to lane space includes: Measure the distance between the vehicle in question and the other vehicle. A predetermined weight is applied based on the speed of the other vehicle; A predetermined weight is applied based on the position number of the other vehicle, which is counted along the direction of travel starting from the vehicle of interest.

5. The method according to claim 4, wherein, The higher the speed of the other vehicle, the greater the weight applied; the closer the other vehicle is to the vehicle of interest, the greater the weight applied.

6. The method according to claim 1, wherein, Infrastructure-based information calculations related to vehicle driving status include indices that are calculated when the probability of a vehicle's speed decreasing is higher.

7. The method according to claim 1, wherein, Calculating an index related to the degree of deviation of the vehicle of interest from the center of the lane in which the vehicle of interest is traveling includes applying a higher index to the lane in which the vehicle of interest is deviating.

8. The method according to claim 1, wherein, Calculating the index related to the travel angle of the vehicle in focus includes applying a higher index to the vehicle in focus traveling at a greater angle relative to its centerline toward the lane it is traveling in.

9. The method of claim 1, further comprising predicting the lane with the highest total score of the calculated index as the lane that the vehicle of interest intends to enter.

10. A non-transitory computer-readable recording medium having a program recorded thereon for performing the method according to claim 1.

11. An automatic driving control device, the device comprising: The first determiner is configured to collect driving information of the autonomously driving vehicle and at least one other vehicle present in the vicinity of the vehicle to determine the traffic environment; The second determiner is configured to determine the driving intention of the vehicle of interest that is driving directly in front of the vehicle in the same lane as the vehicle, based on the driving information of other vehicles. as well as The driving controller is configured to predict the driving routes of other vehicles based on their driving intentions and driving information, and to determine the driving route of the vehicle itself based on the predicted driving routes of other vehicles. The second determiner includes: The lane surplus space determination module is configured to calculate an index related to the lane surplus space based on the determination result of the first determiner and the distance between the vehicle of interest and other vehicles. The infrastructure-based determination module is configured to calculate indices related to the vehicle's driving status based on infrastructure information. The lane traffic flow determination module is configured to calculate an index related to the speed of vehicles in each lane. Other vehicle bias value determination module, which is configured to calculate an index related to the degree of bias of the vehicle of interest relative to the center of the lane in which the vehicle of interest is traveling; The module for determining the direction of travel of other vehicles is configured to calculate an index related to the travel angle of the vehicle of interest; and The other vehicle driving intention determination module is configured to predict the lane that the vehicle of interest intends to enter by comprehensively determining the calculated index.

12. The automatic driving control device according to claim 11, wherein, The driving information includes at least one of the following: the position information of the vehicle and other vehicles, the speed information of the vehicle and other vehicles, the acceleration information of the vehicle and other vehicles, the deviation value information of other vehicles, the travel angle information of other vehicles, precise map information, or information containing infrastructure such as traffic lights, bus stops, and school zones.

13. The automatic driving control device according to claim 11, wherein, The lane clearance determination module is configured to: measure the distance between the vehicle of interest and another vehicle, apply a predetermined weight based on the speed of the other vehicle, and apply a predetermined weight based on the position number of the other vehicle counted from the vehicle of interest along the forward direction, in order to calculate an index related to the lane clearance.

14. The automatic driving control device according to claim 11, wherein, The infrastructure-based determination module is configured such that, based on infrastructure information, the higher the probability that the infrastructure reduces the speed of other vehicles, the smaller the calculated exponent.

15. The automatic driving control device according to claim 11, wherein, The other vehicle bias value determination module is configured to apply a higher index to the lane in which the vehicle of interest is biased towards the center of the lane in which the vehicle of interest is traveling.

16. The automatic driving control device according to claim 11, wherein, The other vehicle travel direction determination module is configured to apply a higher index to the vehicle of interest so that it travels at a greater angle relative to the centerline of the vehicle of interest toward the lane it is traveling in.

Images