Determining the absence of risk of collision between vehicles at an intersection

The telematics system uses intersection graphs and 2D projection to accurately assess collision risks by considering actual lane positions and vehicle trajectories, addressing false detections at intersections with structures.

FR3149120A1Pending Publication Date: 2024-11-29VALEO COMFORT & DRIVING ASSISTANCE
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
FR2023005064
Authority / Receiving Office
FR · FR
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-05-23
Publication Date
2024-11-29

AI Technical Summary

Technical Problem

Existing telematics systems inaccurately detect collision risks between vehicles due to variations in road altitude, particularly at intersections with structures like bridges or tunnels, leading to false collision detections.

Method used

A telematics system uses a roadside unit to send a graph representing the intersection to vehicles, allowing for the determination of collision risks by considering actual traffic lanes and eliminating false positives through route history and position verification using a 2D projection method.

Benefits of technology

The method effectively rules out erroneous collision detections by accounting for actual lane positions and vehicle trajectories, ensuring accurate collision risk assessment at intersections with structures.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 00000000_0000_ABST
    Figure 00000000_0000_ABST
Patent Text Reader

Abstract

We propose a method implemented by a telematics system in a vehicle. The vehicle travels on a road leading to an intersection. The intersection includes a roadside unit. The roadside unit includes memory on which a graph representing the intersection is stored. The graph comprises arcs connected to a central node. Each arc represents a traffic lane leading to the intersection. The method includes providing S10 a collision risk detection at the intersection between the vehicle and another vehicle. The method includes sending S20 from the roadside unit to the vehicle the stored graph. The method includes determining S30 the absence of a collision risk between the vehicle and the other vehicle from the graph. This constitutes an improved vehicle-to-vehicle collision detection solution. [Fig. 1]
Need to check novelty before this filing date? Find Prior Art

Description

Title of the invention: Determination of absence of risk of collision between vehicles at an intersection Technical field

[0001] The present disclosure relates to a method implemented by a telematics system in a vehicle and a telematics system for a vehicle. Technical background

[0002] Today, there are vehicles equipped with telematics systems that allow them to exchange information with each other and with other roadside equipment. For example, these telematics systems may use V2X (Vehicle-to-Everything) communication technology, which allows a vehicle to exchange information with any other system using this technology. For example, a V2X telematics system may exchange information with telematics systems in other vehicles, roadside equipment such as intersection lights or systems incorporating cameras, or with the smartphones of passers-by.

[0003] The information exchanged makes it possible in particular to detect a risk of collision between two vehicles. For example, the two vehicles can exchange location information, and, by calculating the trajectories of the vehicles, the telematics system can detect a risk of collision between the vehicles when their trajectories cross.

[0004] However, the detected collision risks may be erroneous. Indeed, the trajectories calculated for two vehicles may cross while the vehicles are traveling on respective lanes that do not cross. This situation may occur in particular in the presence of a bridge or a tunnel. Indeed, in the presence of such a structure, the trajectories of vehicles arriving at the structure at the same time, but which pass over each other, may cross when they are calculated by projection onto the same 2D plane, whereas in reality the two vehicles do not cross. Variations in road altitude therefore induce false detections of collisions between vehicles. This may occur in particular at an intersection, for example when a bridge or a tunnel is present near this intersection.

[0005] There is therefore a need to improve collision detection between vehicles. Summary

[0006] For this purpose, a method is proposed implemented by a telematics system in a vehicle. The vehicle is traveling on a road leading to an intersection. The intersection comprises a roadside unit. The roadside unit comprises a memory on which is stored a graph representing the intersection. The graph comprises arcs connected to a central node. Each arc represents a traffic lane leading to the intersection. The method comprises providing a detection of risk of collision at the intersection between the vehicle and another vehicle. The method comprises sending by the roadside unit to the vehicle of the stored graph. The method comprises determining the absence of risk of collision between the vehicle and the other vehicle from the graph.

[0007] Each arc may connect the central node to a respective node representing a position on the lane represented by the arc upstream of the intersection. Determining the absence of a risk of collision may include sending, by the other vehicle, a position and a travel history of the other vehicle. Determining the absence of a risk of collision may include checking the absence of the other vehicle on a lane represented by an arc connected to the central node from the sent position and travel history.

[0008] The check may comprise, for each respective node, a calculation of an equation of an ellipse having as its major arc a segment defined between the central node and the respective node, and, a verification that the position and the path history sent are not inside the ellipse from the calculated equation.

[0009] The route history may comprise a set of successive positions occupied by the other vehicle during a past period.

[0010] The roadside unit may be integrated into a traffic signal of the crossing.

[0011] The provision of the collision risk detection may comprise a provision, for each vehicle, of a path of the vehicle projected in a 2D plane. The provision of the collision risk detection may comprise a determination of an encounter on the 2D plane between the path of the vehicle and the path of the other vehicle.

[0012] Providing the routes may include, for at least one of the vehicles, calculating the route of the vehicle from the speed, acceleration, yaw rate, steering angle and / or dimensions of the vehicle, preferably without using the altitude of the vehicle.

[0013] A computer program for a vehicle telematics system is also provided. The computer program comprises instructions which, when the program is executed by a processor, cause the processor to implement the method.

[0014] A computer-readable storage medium on which the computer program is recorded is also provided.

[0015] A vehicle telematics system is also provided. The telematics system includes the storage medium. The system is configured to perform the method. Brief description of the figures

[0016] Non-limiting examples will be described with reference to the following figures:

[0017] [Fig.l] shows a flowchart of an example implementation of the method.

[0018] [Fig.2] and [Fig.3] illustrate examples of determining the absence of risk of collision between the vehicle and another vehicle. Detailed description

[0019] With reference to [Fig.l], a method is proposed implemented by a telematics system in a vehicle. The vehicle is traveling on a road leading to an intersection. The intersection comprises a roadside unit. The roadside unit comprises a memory on which a graph representing the intersection is recorded. The graph comprises arcs connected to a central node. Each arc represents a traffic lane leading to the intersection. The method comprises a provision S10 of a detection of risk of collision at the intersection between the vehicle and another vehicle. The method comprises a sending S20 by the roadside unit to the vehicle of the recorded graph. The method comprises a determination S30 of the absence of risk of collision between the vehicle and the other vehicle from the graph.

[0020] The method provides an improved solution for detecting collisions between vehicles.

[0021] Indeed, the method makes it possible to better rule out detected collision risks that are not real, i.e. those whose detection is erroneous. In particular, the method uses the graph representing the intersection to deduce whether the risk of collision is real or not, taking into account the traffic lanes on which the vehicles are traveling, and the intersections that exist at the intersection. When the other vehicle is not traveling on an arc connected to the central node, the graph makes it possible to deduce that the risk is not real. Thus, in the presence of a structure allowing the simultaneous passage of several vehicles at the same time (a bridge for example), the method will make it possible to rule out the risks of collision detected between vehicles for vehicles arriving on different lanes at the structure.The process uses the graph to eliminate the risk of collision whenever the lanes on which traffic is traveling do not intersect.

[0022] The vehicle that includes the telematics system executing the method may be a so-called EGO vehicle, i.e. one whose behavior in interactions with other vehicles involved in a situation is to be controlled. The vehicle may be any type of vehicle configured to include such a telematics system (for example a car, a truck or a motorcycle). The other vehicle in the method may be a so-called remote vehicle, i.e. another vehicle that is at a given moment in the vicinity of the so-called EGO vehicle.

[0023] The method may be executed while the vehicle is traveling. Steps S10 to S30 may be repeated by the method while the vehicle is traveling. For example, steps S10 to S30 may be repeated each time a new vehicle has a path that meets that of the vehicle (at the same intersection). Steps S10 to S30 are then repeated for this new vehicle. The risk of collision detected is in this case at each repetition a risk of collision at the intersection with the new vehicle encountered. Alternatively or additionally, steps S10 to S30 may be repeated each time the vehicle arrives at a new intersection equipped with a roadside unit (and another vehicle has a path that meets that of the vehicle at this new intersection). The risk of collision detected is in this case at each repetition a risk of collision at the new intersection and with the other vehicle encountered at this new intersection.The details given below apply to each repetition of steps S10 to S30 for each new vehicle encountered at each new intersection.

[0024] The provision S10 of the collision risk detection may comprise a provision SI 1, for each vehicle, of a path of the vehicle projected in a 2D plane. The path of a vehicle may represent on the 2D plane a simulation of the movement that the vehicle will follow taking into account a current situation of the vehicle (for example taking into account its position, its speed and / or the angle of the steering wheel). The 2D plane may represent the space around the collision. For example, the 2D plane may be a map representing a projection on a 2D plane of the lanes around the collision.

[0025] The path of each vehicle may represent the movement of the vehicle during a predetermined future duration (for example, a duration of less than 10 minutes and / or greater than 10 seconds, for example, a duration of approximately 1 minute). The method may calculate the length of the path and its orientation at regular time intervals. For example, the method may comprise repeating the calculation of the path (for example, the length and orientation of the path) of each vehicle (for example, every 100 milliseconds). At each repetition, the method may determine whether there is an encounter between the new paths calculated for the two vehicles. For one of these repetitions, the method may determine that there is an encounter between the path of the vehicle and the path of the other vehicle (determination of step S12).For each vehicle, the path that is provided in step S10 is therefore the path that is calculated for the vehicle at the time of the repetition for which it is determined that there is an encounter. The method can calculate the path of the so-called EGO vehicle from the same parameters as for calculating the path of the other vehicle (speed, acceleration, yaw rate, steering angle and / or dimensions of the vehicle). The path can represent the movement of the vehicle over a predetermined length (for example . example a length less than 10 kilometers and / or greater than 100 meters, for example a length of approximately 300 meters). The length of the route may depend on the parameters used for its calculation (for example, the length may increase proportionally with the speed).

[0026] The path may represent the spatial movement of the vehicle. For example, the path may represent the movement of the vehicle on the 2D plane during the entire predetermined duration. Alternatively, the path may represent the spatial and temporal movement of the vehicle. In this case, the path may also represent the evolution in time of the movement of the vehicle.

[0027] Each path may comprise a line on the 2D plane, for example materializing the evolution of the center of the vehicle (for example the center of gravity). Alternatively, each path may materialize the movement of a footprint of the vehicle on the ground in the 2D plane. In this case, the path may comprise a set of points each representing a space on the ground occupied by the vehicle at a given moment. For example, the path may comprise a line with a thickness materializing the movement of the entire vehicle (i.e. considering its width), or two lines starting from two ends of the front of the vehicle.

[0028] Alternatively, the route of the other vehicle may be calculated. For example, the provision SI 1 may comprise a calculation of the route of the other vehicle from one or more parameters of the other vehicle. For example, the route calculation may be a function of the speed, acceleration, yaw rate, steering angle and / or dimensions of the other vehicle. Preferably, the route calculation may not use the altitude of the other vehicle. For example, the calculated route may be in a 2D plane, i.e. may represent the evolution of the footprint of the other vehicle in a 2D plane. The parameters may have been sent by the other vehicle. In this case, the provision SU may comprise, before the calculation, a reception of these parameters by the telematics system, and the calculation may be done on the received parameters. For example, the parameters may be sent by the other vehicle during a V2X exchange.The transmission and reception of these parameters may be achieved using V2X communication technology. For example, the other vehicle may also include a telematics system, and the telematics systems of the two vehicles may be configured to regularly exchange values ​​for these parameters with each other (e.g., every 100 milliseconds) when they are within a certain distance of each other (e.g., a distance less than or equal to 1 kilometer, e.g., less than or equal to 600 meters or 300 meters). The values ​​of these parameters used for calculating the other vehicle's route may be the values ​​sent by the other vehicle during one of these V2X exchanges.

[0029] As for the other vehicle, the route of the vehicle equipped with the system implementing the method may be directly provided by the vehicle (for example by another system of the vehicle), or may be calculated by the telematics system. For example, the route calculation may be based on the same parameters as for the other vehicle (speed, acceleration, yaw rate, steering angle and / or vehicle dimensions). The vehicle dimensions may have been measured on the vehicle, for example before the execution of the method. The other parameters (for example speed, acceleration, yaw rate, steering angle and position) may be provided by other computers (for example a speed sensor or a GPS).

[0030] The provision S10 may then comprise a determination of an encounter on the 2D plane between the path of the vehicle and the path of the other vehicle. For example, the determination may comprise a deduction of the encounter when the intersection of the path of the vehicle and the path of the other vehicle is non-zero. In this case, the determination may comprise a calculation of the intersection of the two paths. The determination may deduce therefrom the location of the risk of collision (the location of the intersection). When the paths are lines, the intersection may be a point of intersection between the lines. When the paths are lines, the intersection may be a surface (the area of ​​intersection of the two lines).

[0031] The determination of the encounter can consider only the spatial movement of the two vehicles. In this case, each path can represent the only spatial movement of the vehicle. The calculation of the paths (and therefore of the collision points which are the meeting points of the calculated paths) takes into account several dynamic parameters (speed, acceleration), which make it possible to know the evolution in space and time. Thus, the evolution in time is calculated every 100 milliseconds.

[0032] The collision risk detection provided is at the intersection. This means that the collision is detected near the intersection. In other words, the meeting between the paths of the two vehicles is located near the intersection. For example, the collision is within a certain perimeter around the center of the intersection, for example a circular perimeter around the intersection (with a diameter for example less than or equal to 500 meters, for example less than or equal to 300 meters). The collision may for example be at a distance less than or equal to the distance of the arcs, that is to say be at the same distance or closer to the center of the intersection than the points on the lanes materialized by the respective nodes of the arcs which represent them.

[0033] The graph may represent all the lanes that arrive at the intersection. For each lane of the intersection, the graph may include a respective arc connected to the central node (materializing the center of the intersection). The central node may represent the center of the intersection. Each arc of the graph may connect the central node to a respective node representing a position on the track represented by the arc. Each arc may include a first end connected to the central node and a second end connected to a respective node. The respective node may represent a position at the center of the track represented by the arc. The position represented by the respective node may be upstream of the intersection. For example, for each arc, the respective node may represent a position located at a distance less than or equal to 500 meters from the center of the intersection (for example, a distance less than or equal to 300 meters from the intersection).

[0034] The graph may correspond to a portion of a general graph representing all the lanes of a territory in which the vehicle is traveling. For example, the general graph may represent all the lanes of a region or country in which the vehicle is traveling.

[0035] Each arc of the graph can be oriented. The orientation of the arc can represent the direction of travel of the vehicle on the lane that the arc represents. Each arc of the graph can be oriented towards the central node (the lanes represented by the graph can always be oriented towards the traffic light). Each lane can correspond to a portion of road on which a vehicle can travel. In the case of a road comprising several traffic lanes in the same direction, the graph can comprise a respective arc for each lane of the road. Alternatively, the same arc can represent the several traffic lanes going in the same direction. In this case, the graph can then comprise for each arc a variable representing the number of lanes on the road.

[0036] The graph may be sent before the vehicle arrives at the intersection. For example, the roadside unit may continuously send the graph that is stored in its memory to each vehicle that arrives at the intersection. For example, the roadside unit may send the graph when the vehicle is at a distance less than or equal to 1 kilometer from the intersection and / or greater than or equal to 150 meters from the intersection (for example at a distance of approximately 300 meters from the intersection). The graph may have been previously stored in the memory before the execution of the method. For example, the graph may have been stored just before the execution of the method. In this case, the graph may have been downloaded by the roadside unit (for example from a remote server), and then stored in its memory.Alternatively, the graph may have been recorded on the roadside unit's memory well before the vehicle arrived (e.g. at the time of its construction).

[0037] The graph may have been determined taking into account a current configuration of the tracks at the crossing. In the event of a change in the configuration of the tracks at the crossing, the graph may have been updated to take into account this change in the configuration of the crossing (for example, addition of a track and / or removal of a track). pressure of a channel).

[0038] After sending the graph S20, the method comprises determining S30 the absence of risk of collision between the vehicle and the other vehicle from the graph. The determination S30 is based on the sent graph, i.e. the absence of risk of collision is deduced from a reading of the graph.

[0039] In examples, the determination S30 may comprise a verification of the absence of a lane at the intersection on the side of the vehicle (left side or right side) from which the other vehicle is arriving. For example, the determination S30 may comprise a determination of the side of the vehicle (left side or right side) from which the other vehicle is arriving (for example from the trajectories of the two vehicles, and / or from a sending of its position and its position history by the other vehicle, for example by V2X exchange), then, a verification of the absence of an arc at the intersection on this side. For example, the verification may comprise an identification of all the arcs arriving at the intersection (for example except the one on which the vehicle is traveling). For example, the verification may comprise an identification of one or more arcs arriving at the intersection (in addition to the arc on which the vehicle is traveling).Then the check may include checking that no arc is located on the other vehicle's side.

[0040] Verifying that no arc is located on the other vehicle side may be based on the position of the respective node of each arc. For example, the verification may include calculating the position of the respective node of each arc, and then verifying that the calculated position of each respective node is not on the other vehicle side. For example, the verification may include, for each arc of the intersection, calculating the Cartesian position relative to the vehicle of the respective node of the arc, and then verifying that the calculated Cartesian position is not located on the other vehicle side.

[0041] In examples, the determination S30 may comprise a sending S31, by the other vehicle, of a position and a travel history of the other vehicle, then, a check S32 of the absence of circulation of the other vehicle on a lane represented by an arc connected to the central node. These steps S31 and S32 may be executed after the execution of the verification that no arc is located on the side of the other vehicle (for example when this verification fails).

[0042] The sending S31 of the position and the route history of the other vehicle can be carried out at the time of a V2X exchange between the two vehicles. For example, this V2X exchange can be carried out before the detection risk is detected (at the same time as the values ​​of the parameters of the other vehicle for the route calculation are provided for example, i.e. during the same exchange). Alternatively, the sending S31 can be carried out just after the risk is detected. For example, the other vehicle can be configured to continuously send its position and route history to the vehicle (via successive V2X exchanges for example).

[0043] The route history sent may comprise a set of successive positions occupied by the other vehicle during a past period. The set of successive positions may have been recorded by the other vehicle. For example, the other vehicle may be configured to regularly record its position on a memory, and to send the recorded positions to other vehicles (during V2X exchanges with these other vehicles for example). The number of successive positions sent may be less than or equal to 10 and / or greater than or equal to 2. For example, the number of successive positions may be greater than or equal to 3.

[0044] After sending S31 of the position and the travel history, the determination S30 may comprise a check S32 of the absence of circulation of the other vehicle on a lane represented by an arc connected to the central node from the position and the travel history sent. The check S32 may comprise a verification, for each arc connected to the central node (for example except the one on which the vehicle is traveling), of the absence of the other vehicle near the lane. For example, the verification may comprise a determination of a perimeter around the lane then a verification of the absence of the other vehicle in the perimeter. The absence of the other vehicle may be verified from the location sent by the other vehicle and its travel history.

[0045] The check S32 for the absence of circulation of the other vehicle on a lane represented by an arc is now discussed in general. When only one arc is connected to the central node (other than the one on which the vehicle is traveling), the determination may perform this check only once for this single arc. When several arcs are connected to the central node (in addition to the one on which the vehicle is traveling), the method may repeat this check for each of these arcs. In examples, the method may perform this check only for the arc located on the side of the other vehicle. For example, when at least one arc is not located on the side of the other vehicle, the method may not perform the check for this at least one arc.

[0046] In examples, the check S32 may comprise a verification that the other vehicle is not located within a given perimeter around the lane represented by the arc (e.g., an ellipse around the arc). For example, the check S32 may comprise a calculation S33 of an equation of an ellipse having as its major arc a segment defined between the central node and the respective node. For example, the calculation may comprise a calculation of the position of the nodes that are located at the ends of the arc (the central node and the respective node of the arc). The calculation may then comprise determining a segment between the calculated positions of these end nodes. The calculation of the positions of the nodes may comprise a transformation of the coordinates GPS of the nodes in Cartesian coordinates relative to the vehicle (for example by exploiting the angle of rotation between the arcs of the intersection). The calculation may include a calculation of the position in Cartesian coordinates relative to the vehicle of the center of the determined segment. The control may then include a calculation of the equation of an ellipse having as its major arc the determined segment and as its minor arc a predefined distance (for example the width of the track).

[0047] After the calculation S33, the check S32 may comprise a verification S34 that the position and the path history sent are not inside the ellipse. For example, the verification S34 may comprise a verification that a predetermined number of points of the path history are not inside the ellipse (for example the last 5 points, or preferably the last 3 points of the history). The verification S34 may fail when at least one of the position and the points of the history is inside the ellipse.

[0048] The verification S34 may comprise a transformation, in the Cartesian coordinates relative to the vehicle, of the position and the predetermined number of points in the travel history of the other vehicle. The verification S34 may then comprise a verification that the position of the other vehicle and the predetermined number of points in a travel history of the other vehicle are not within the ellipse calculated in the space defined relative to the vehicle (i.e., using the coordinates of the points and the equation of the ellipse in this space defined relative to the vehicle).

[0049] After the determination S30 of the absence of risk of collision, the method may comprise a cancellation of a normally programmed alert in the event of a risk of collision with another vehicle (for example a visual or audible alert intended for the driver of the vehicle). Alternatively or additionally, the method may comprise a cancellation of the execution of a collision avoidance method with the other vehicle (such as emergency braking or a trajectory modification for the vehicle). The method may comprise a deprogramming of the alert and / or of the execution of the collision avoidance method.

[0050] Examples will now be described with reference to Figures 2 and 3.

[0051] [Fig. 2] illustrates a first example of determining the absence of risk of collision between the vehicle and another vehicle. In this first example, three lanes 101, 102, 103 lead to the intersection 100. The graph 120 representing the intersection 100 therefore comprises three arcs 121, 122, 123 representing these three lanes 101, 102, 103 which lead to the intersection. Each of the arcs 121, 122, 123 is connected to a central node 124 (“b”) and to a respective node (“a”, “d” and “c” for the arcs 121, 122, 123 respectively). Each respective node represents the center of the lane at a certain distance from the center of the intersection. The vehicle 141 travels on the lane 101 and includes a telematics system implementing the process.

[0052] Another lane 104 is located near the intersection. This other lane 104 passes next to the intersection 100 but does not lead to the intersection 100. For example, this other lane 104 is located on a bridge passing over the lane 103. Alternatively, this other lane 104 is located in a tunnel passing under the lane 103. Another vehicle (the vehicle 142) is traveling on this other lane 104.

[0053] The intersection comprises a first roadside unit integrated in the traffic light 160. The roadside unit of the traffic light 160 comprises a memory on which the graph 120 representing the intersection 100 is recorded. Optionally, the intersection may comprise a second unit integrated in the traffic light 161 (for example for vehicles arriving from the lane 102). Alternatively, a single roadside unit may be used for all the lanes.

[0054] The method implemented by the telematics system of the vehicle 141 comprises the provision S10 of a collision risk detection at the intersection between the vehicle 141 and the other vehicle 142. The collision risk detection is induced by the fact that the trajectories of the two vehicles 141, 142 (respectively 181 and 182), which are projected in the same 2D plane, meet. The collision detection method, which is based on the projected trajectories of the vehicles, therefore detects a risk of collision between the two vehicles. However, in reality, the vehicles will not collide. Indeed, the real trajectories of the two vehicles are not at the same altitude because the lane 104 on which the vehicle 142 is traveling passes above or below the lane on which the vehicle 141 is traveling (the lane 101 which becomes the lane 103 after the intersection) because of the bridge or the tunnel.The detected collision risk is therefore erroneous since, although their trajectories meet, vehicle 141 will not collide with vehicle 142.

[0055] Before arriving at the intersection, the method comprises sending S20 by the roadside unit 160 to the vehicle 141 of the graph 120 which is recorded on the memory of the roadside unit 160. For example, the graph can be sent to the vehicle 141 when it arrives at the location materialized by the respective node “a” (for example approximately 500 or 300 meters before the traffic light 160). Sending the graph comprises sending information recorded on the intersection. The information sent can comprise the GPS position of one or more nodes of the graph (for example at least nodes “b” and “c”). The information sent can comprise the number of lanes and the width of the lanes.

[0056] After sending the graph 120, the method comprises the determination S30 of the absence of risk of collision between the vehicle 141 and the other vehicle 142 from the sent graph 120. In this first example, the determination S30 comprises the verification of the absence of a lane at the crossing on the side of the vehicle 141 (right side in this example) from which the other vehicle 142 is arriving. For example, the determination S30 may comprise determining that the other vehicle 142 is arriving to the right of the vehicle 141. For example, the determination of the side may be based on the trajectories 181 and 182 of the two vehicles 141 and 142, and / or the sending of its position and its position history 143 by the other vehicle 142 (by V2X exchange). The determination S30 may then comprise verifying that there is no arc at the intersection on this determined side. For example, the verification may comprise identifying all the arcs (102, 103) arriving at the intersection (except the one on which the vehicle is traveling 101). Then, the verification comprises verifying that none of these arcs 102 and 103 is located on the side of the vehicle 141 from which the other vehicle 142 is arriving (the right side in this example).

[0057] The verification that no arc of these arcs 102 and 103 is located on the right side of the vehicle 141 is based on the position of the respective nodes “c” and “d” of the two arcs 102 and 103. The verification comprises a calculation of the respective position of the two respective nodes “c” and “d”, then, a verification that the calculated position of each respective node “c” and “d” is not on the right side of the vehicle 141. For this, the verification comprises, for each of the arcs 102 and 103, the calculation of the Cartesian position relative to the vehicle 141 of the respective node of the arc (respectively “c” and “d”), then, the verification that the calculated Cartesian position is not located on the side of the other vehicle 142.

[0058] [Fig. 3] illustrates a second example of determining the absence of risk of collision between the vehicle and another vehicle. In this second example, the situation of the intersection 200 differs from that of the intersection 100 in that one of the lanes is on the side of the other vehicle 142 (the lane 202 is now on the right side of the vehicle 141). The first verification previously discussed therefore fails (the verification of the absence of a lane at the intersection on the side of the vehicle 141). In this second example, the determination S30 therefore comprises a second verification. The second verification can be carried out in addition to the first, or without this first prior verification, i.e. first when the risk of collision is detected.

[0059] In this second example, the determination S30 comprises the sending S31, by the other vehicle 142, of its position and its route history 143, then, the checking S32 of the absence of circulation of the other vehicle 142 on a route represented by an arc connected to the central node 124. In this second example, the graph sent is the graph 220, and the verification is carried out on this graph 220.

[0060] In examples, these steps S31 and S32 may be performed after performing the verification that no arc is located on the side of the other vehicle 142 previously discussed. In other examples, these steps may be performed in first place (without first carrying out the verification that no arc is located on the side of the other vehicle 142 previously discussed).

[0061] After the sending S31 of the position and the travel history 143 by the other vehicle 142, the determination S30 comprises the checking S32 of the absence of circulation of the other vehicle 142 on a lane represented by an arc connected to the central node 124 from the position and the travel history 143 sent.

[0062] In this example, the method performs this check S32 only for the arc 202 located on the side of the other vehicle 142. Indeed, the arc 103 is not located on the side of the other vehicle 142, and the vehicle 141 is traveling on the arc 101. The graph 220 therefore does not include any other arc which is both on the side of the other vehicle and also which represents a lane on which the vehicle is not traveling.

[0063] The check S32 of the arc 103 comprises the verification that the other vehicle 142 is not located within a given perimeter around the lane represented by the arc (an ellipse in this example). The check S32 comprises a calculation S33 of an equation of an ellipse 210 having as its major arc a segment 212 defined between the central node 124 and the respective node 225 of the arc 222. For example, the calculation may comprise a calculation of the position of the nodes which are located at the ends of the arc (the central node “b” and the respective node “c” of the arc 222). The calculation may then comprise the determination of a segment 212 between the calculated positions of these end nodes. The calculation of the positions of the nodes may include a transformation of the GPS coordinates of the nodes into Cartesian coordinates relative to the vehicle (for example using the GPS position of the vehicle called EGO, as well as the positions of the nodes, at least "b" and "c" for example).The calculation may include a calculation of the position in Cartesian coordinates relative to the vehicle of the center of the determined segment. The control may then include a calculation of the equation of the ellipse 210 having as its major arc 212 the determined segment and as its minor arc 211 a predefined distance (for example approximately equal to the width of the lane 202).

[0064] After the calculation S33, the control S32 includes the verification S34 that the position and the first three points of the route history 143 sent are not inside the ellipse 210. The verification S34 fails when at least one of the position of the other vehicle and the last three points of the history is inside the ellipse 210. In this example, the verification S34 succeeds since the position of the vehicle 142 and the first three points of the history 143 are all outside the ellipse 210.

[0065] Verification S34 comprises the transformation, in the Cartesian coordinates relating to the vehicle 141, of the position and the first three points of the route history 143 of the other vehicle. Verification S34 then comprises the verification that the position of the other vehicle 142 and the first three points are not located in the ellipse 210 calculated in the space defined relative to the vehicle (i.e. using the coordinates of the points and the equation of the ellipse in this space defined relative to the vehicle 141).

[0066] After the determination S30 of the absence of risk of collision, the method comprises the cancellation of a normally programmed alert in the event of a risk of collision with another vehicle (for example a visual or audible alert intended for the driver of the vehicle). Alternatively or additionally, the method comprises the cancellation of the execution of a collision avoidance method with the other vehicle (such as emergency braking or a trajectory modification for the vehicle). The method comprises a deprogramming of the alert and / or of the execution of the collision avoidance method.

Claims

Claims

1. Method implemented by a telematics system in a vehicle, the vehicle traveling on a lane leading to an intersection, the intersection comprising a roadside unit, the roadside unit comprising a memory on which is recorded a graph representing the intersection, the graph comprising arcs connected to a central node, each arc representing a traffic lane leading to the intersection, the method comprising: • providing a detection of risk of collision at the intersection between the vehicle and another vehicle; • sending by the roadside unit to the vehicle of the recorded graph; and • determining the absence of risk of collision between the vehicle and the other vehicle from the graph.

2. Method according to claim 1, in which each arc connects the central node to a respective node representing a position on the lane represented by the arc upstream of the crossing, the determination of the absence of risk of collision comprising: • a sending, by the other vehicle, of a position and a travel history of the other vehicle; and • a check of the absence of circulation of the other vehicle on a lane represented by an arc connected to the central node from the position and the travel history sent.

3. A method according to claim 2, wherein the check comprises, for each respective node: • a calculation of an equation of an ellipse having as its major arc a segment defined between the central node and the respective node; and • a verification that the position and the path history sent are not inside the ellipse from the calculated equation.

4. A method according to claim 2 or 3, wherein the path history comprises a set of successive positions occupied by the other vehicle during a past period.

5. A method according to any one of claims 1 to 4, wherein the roadside unit is integrated into a traffic signal of the intersection.

6. A method according to any one of claims 1 to 5, wherein providing the collision risk detection comprises: • providing, for each vehicle, a path of the vehicle projected in a 2D plane; and • determining an encounter on the 2D plane between the path of the vehicle and the path of the other vehicle.

7. A method according to claim 6, wherein providing the paths comprises, for at least one of the vehicles, calculating the path of the vehicle from the speed, acceleration, yaw rate, steering angle and / or dimensions of the vehicle, preferably without using the altitude of the vehicle.

8. A computer program for a vehicle telematics system comprising instructions which, when the program is executed by a processor, cause the latter to implement the method according to any one of claims 1 to 7.

9. A computer-readable storage medium on which the computer program according to claim 8 is recorded.

10. A vehicle telematics system comprising the storage medium of claim 9, the system being configured to perform the method of any one of claims 1 to 7.

Citation Information

Patent Citations

  • Autonomous transportation system and methods

    US20190088148A1

  • Cellular network-based assisted driving method and traffic control unit

    US20200258389A1

  • Methods and systems for lane changes using a multi-corridor representation of local route regions

    US20220340201A1

  • Systems and methods for determining road safety

    US20220397402A1