Method for determining an authorized speed limit at which a motor vehicle is traveling.

Abstract

Method for determining an authorized speed limit where a motor vehicle is traveling. Method for configuring a computer (4) of a motor vehicle (1), the computer incorporating an artificial neural network logic module (43), the method comprising the following steps: - a step of acquiring data, in particular images, - on the basis of the acquired data, a step of detecting events, in particular events in the environment of the motor vehicle, - a step of defining the detected events as input parameters of the artificial neural network, - a step of determining the nature of the traffic lane on which the motor vehicle is traveling and / or the authorized speed limit of the place where the motor vehicle is traveling as output parameters of the artificial neural network. No figure for the abstract.

Description

Title of the invention: Method for determining an authorized speed limit at which a motor vehicle is traveling.

[0001] The invention relates to a method for configuring a computer of a motor vehicle. The invention relates to a method for determining the nature of a traffic lane on which the motor vehicle is traveling. The invention also relates to a method for determining a maximum authorized speed where the motor vehicle is traveling. The invention relates to a computer obtained by implementing the configuration method. The invention also relates to a management system implementing the method for determining the nature of the traffic lane on which the motor vehicle is traveling and / or the method for determining a maximum authorized speed where the motor vehicle is traveling. The invention also relates to a motor vehicle comprising such a system. The invention also relates to a computer program implementing one of the methods mentioned.The invention finally relates to a recording medium on which such a program is recorded.

[0002] On some current motor vehicles, a vehicle system makes it possible to determine the maximum authorized speed at the location where the vehicle is located. Regulatory developments will soon make such a system mandatory on all new vehicles marketed. Furthermore, these systems must be capable of determining and displaying the maximum authorized speed almost permanently.

[0003] Such systems are known, but they require, in addition to the presence of a camera to detect speed limit signs, the use of: - a GPS system to locate the vehicle on the road network, and - a map memory to determine the limitations relating to the different sections of roadways.

[0004] Furthermore, the map memory must be regularly updated to take into account changes in the road traffic network or changes in traffic rules on the road traffic network.

[0005] Known systems are therefore expensive.

[0006] The aim of the invention is to provide a system and a method for determining the nature of a traffic lane on which the motor vehicle is traveling, remedying the above drawbacks and improving the devices and methods known from the prior art. In particular, the invention makes it possible to produce a system and a method which are simple and reliable while using a minimum of resources.

[0007] According to the invention, a method makes it possible to configure a computer of a motor vehicle, the computer incorporating a neural network logic module ar- The process includes the following steps: - a step of acquiring data, in particular images, on the basis of the acquired data, a step of detecting events, in particular events in the environment of the motor vehicle, - a step of defining the detected events as input parameters of the artificial neural network, and - a step of determining the nature of the traffic lane on which the motor vehicle is traveling and / or the authorized speed limit of the place where the motor vehicle is traveling as output parameters of the artificial neural network.

[0008] Events may include events from the following categories: - detection of an explicit speed sign, and / or - detection of an implicit speed sign, and / or - detection of a contextual traffic sign, and / or - detection of a traffic lane structure, and / or - detection of a ground marking, and / or - detection of dynamic information from the motor vehicle, and / or - detection of other objects in the environment of the motor vehicle and their possible dynamics.

[0009] The method may comprise a learning step, the learning step comprising for example: - the operation of the neural network powered by the detected events, and - the operation of a reference system for determining the nature of the road on which the motor vehicle is traveling and / or for determining the authorized speed limit of the place where the motor vehicle is traveling.

[0010] According to the invention, a method makes it possible to determine, by a management system of a motor vehicle, a maximum authorized speed at which the motor vehicle is traveling. The method comprises the following phases: - a phase of using a calculator configured by the implementation of the method according to one of the preceding claims to determine the nature of the traffic lane on which the vehicle is traveling, and - a phase of using the nature of the traffic lane to determine the maximum authorized speed.

[0011] According to the invention, a method makes it possible to determine, by a management system of a motor vehicle, a maximum authorized speed where the motor vehicle is traveling. The method comprises a phase of using a computer configured by the implementation of a method defined previously to determine the authorized speed limit of the place where the motor vehicle is traveling.

[0012] The determination method may be performed in real time and / or the steps may be iterated at a fixed time interval and / or at a fixed movement interval of the motor vehicle.

[0013] The method may comprise a phase of displaying the maximum authorized speed.

[0014] The method may comprise a step of not taking into account a detection of a explicit speed limit sign.

[0015] The method may comprise a step of interpreting the meaning of an implicit traffic sign.

[0016] According to the invention, a computer is configured by implementing the method defined previously.

[0017] The invention also relates to a system for managing a motor vehicle, the management system comprising hardware and / or software elements implementing the method defined previously, in particular hardware and / or software elements designed to implement the method defined previously.

[0018] The invention also relates to a system for managing a motor vehicle comprising means for implementing the method defined above.

[0019] According to the invention, a motor vehicle comprises a management system defined previously.

[0020] The invention also relates to a computer program product comprising program code instructions recorded on a computer-readable medium for implementing the steps of the method defined above when said program operates on a computer.

[0021] The invention also relates to a computer program product downloadable from a communications network and / or recorded on a data medium readable by a computer and / or executable by a computer, characterized in that it comprises instructions which, when the program is executed by the computer, cause the latter to implement the method defined previously.

[0022] The invention also relates to a data recording medium, readable by a computer, on which is recorded a computer program comprising program code instructions for implementing the method defined above.

[0023] The invention also relates to a computer-readable recording medium comprising instructions which, when executed by a computer, cause the latter to implement the method defined above.

[0024] The invention also relates to a signal of a data medium, carrying the computer program product defined previously.

[0025] The attached drawing represents, by way of example, an embodiment of a motor vehicle according to the invention, an embodiment of a determination method according to the invention and an embodiment of a configuration method according to the invention.

[0026] [Fig.l] is a schematic representation of an embodiment of a motor vehicle according to the invention.

[0027] [Fig.2] is a flowchart of an embodiment of a determination method according to the invention.

[0028] [Fig. 3] is a flowchart of an embodiment of a configuration method according to the invention.

[0029] [Fig.4] is a diagram of a neural network that can be used to implement the invention.

[0030] An embodiment of a motor vehicle 1 according to the invention is described below with reference to [Fig.l].

[0031] The motor vehicle 1 is for example a passenger motor vehicle or a utility motor vehicle. However, the vehicle may be of any type. The motor vehicle is intended to use the road traffic network.

[0032] The motor vehicle 1 comprises a management system 2 or a system 2 for determining the nature of the traffic lane on which the motor vehicle is traveling and / or for determining a maximum authorized speed where the motor vehicle is traveling, i.e. the maximum authorized speed on the section of traffic lane where the vehicle is traveling.

[0033] The system 2 comprises a set of sensors 3, 5 for acquiring data and a computer 4 for processing the data.

[0034] The system 2 also comprises an information display element 6, such as a screen 6, in particular an element for displaying the nature of the traffic lane on which the motor vehicle is traveling and / or for displaying the authorized speed limit where the motor vehicle is located. Preferably, this display element is provided on a dashboard in the passenger compartment of the motor vehicle.

[0035] Advantageously, the set of sensors comprises at least one camera 3 making it possible to acquire images capable of being processed by the computer 4. The camera 3 is preferably positioned behind the windshield or at the level of a grille at the front of the motor vehicle. The camera is further arranged and configured to take at least shots of the environment in front of the motor vehicle.

[0036] The set of sensors may further comprise sensors 5 capable of providing dynamic information of the motor vehicle, in particular a longitudinal speed sensor of the motor vehicle and / or a yaw rate sensor of the motor vehicle and / or an acceleration sensor making it possible to provide different accelerations of the motor vehicle and / or a sensor of the rotation angle of a steering wheel.

[0037] The determination system 2 comprises all the elements 2, 3, 4, 41, 42, 43, 5, 6 hardware and / or software implementing or governing a method for determining the nature of the traffic lane on which the motor vehicle is traveling and / or for determining a maximum authorized speed where the motor vehicle is traveling and / or a method for configuring the computer 4 incorporating an artificial neural network logic module 43. In particular, the determination device 2 comprises the hardware and / or software elements 41, 42, 43 making it possible to implement the method steps. These different elements may comprise software modules.

[0038] The calculator 4 advantageously comprises all or part of the following elements: - a module 41 for processing acquired data, - a module 42 for detecting events, in particular events in the environment of the motor vehicle, - module 43 of artificial neural network logic.

[0039] A mode of execution of the method for determining a maximum authorized speed at which the motor vehicle is traveling is described below with reference to [Fig.2],

[0040] In a first phase E1, the computer 4 configured by the implementation of a configuration method as described below is used to determine the nature of the traffic lane on which the motor vehicle is traveling.

[0041] In a second phase E2, the nature of the traffic lane determined previously is used to determine the maximum authorized speed. Preferably, the nature of the traffic lane can be: - urban road (where the speed limit in France is 50 km / h by default), - extra-urban or rural roads, for example departmental roads, national roads (with the French designation, where the speed limit in France is 80 km / h by default), - expressway (where the speed limit in France is 110 km / h by default), - motorway (where the speed limit in France is 130 km / h by default).

[0042] In particular, this information on the nature of the traffic lane makes it possible to determine, as seen previously, a default speed limit value. Thus, in the absence of recognition of a speed limit sign, it is possible, knowing the nature of the lane on which the vehicle is traveling, to determine by assumption the authorized speed limit value.

[0043] Advantageously, in this phase E2, in a step S70, the authorized speed limit value determined by the system 2 is displayed on the display element 6 by default. However, if an explicit speed limit sign is detected and recognized by the system 2 (for example a 30 km / h limit on an urban road), it is the speed indicated by this sign which is considered and displayed on the display element 6.

[0044] Furthermore, in this phase of use, it is possible to disregard an explicit speed limit sign which has been detected and recognized (or partially recognized) by the management system 2. This is for example the case in the three situations mentioned below: - system 2 detected and recognized an 80 km / h speed limit sign located at the rear of a truck in front of the vehicle equipped with system 2 while vehicle 1 was traveling on an urban road limited to 50 km / h, - system 2 detected and recognized a 50 km / h speed limit sign located on an adjacent lane while vehicle 1 was traveling on a motorway lane limited to 130 km / h, - system 2 detected and recognized a 50 km / h speed limit sign without detecting the display of a sign restricting the speed limit to a particular category of vehicle or to particular weather conditions while vehicle 1 was traveling on a motorway lane limited to 130 km / h.

[0045] In all these cases, the detected and inconsistent speed limits are not considered and are therefore not displayed on the display element 6. On the contrary, in this case, it is: - the default speed limit on the track, or - the latest recognized and consistent limitation information, which is considered and displayed on the display element 6. Thus, the use phase may include a step of not taking into account a detection of an explicit speed limit sign.

[0046] Furthermore, the use phase may include a step of interpreting the meaning of an implicit traffic sign. This is particularly the case for an end of speed limit sign. Such a sign does not indicate the authorized speed limit after passing the sign. However, this authorized speed limit can be determined if the nature of the traffic lane on which the vehicle 1 is traveling is known. This knowledge is obtained by implementing the first phase E1 of the determination method. The authorized speed limit is then the default authorized speed limit on this traffic lane.

[0047] It should also be noted that, in certain use cases, no sign (whether explicit or implicit) is simply present. This is the case, for example, when the vehicle enters a town from a rural road. In such a case, the analysis of images of the environment, in particular of buildings, can make it possible to determine the nature of the traffic lane when no sign is encountered.

[0048] Advantageously, the determination method is executed in real time. By "executed in real time" is meant that the different phases of the method are iterated at reduced time intervals, for example every 20 ms or 10 ms. Alternatively, the different phases of the method can be iterated at a fixed movement interval of the vehicle 1, for example every 50 m or every 100 m. Alternatively, the different steps of the method can be iterated each time a new event is detected, in particular each time new data acquired by the camera lead to an event detection by the module 42.

[0049] Preferably, the following rules are also implemented: - When a change in the nature of the traffic lane is determined, the speed limit authorized by default is displayed on the lanes of the new nature determined. - When determining an authorized speed limit different from the default authorized speed limit on roads of the determined nature, this different speed is displayed if it satisfies one of the following two conditions (i) the speed difference is reasonable (for example, a direct change from a speed limit of 130 km / h to 50 km / h directly is not reasonable), or (ii) a particular circumstance justifies a large difference in speed limit (for example, a roadworks zone, a motorway exit or variable speed signage on a light panel. - The display of the default authorized speed limit is returned if (i) an end of speed limit sign is detected or (ii) a change of road is detected, in particular after an intersection or a roundabout (detectable via a sign, an analysis of the steering wheel angle, a yaw angle or vehicle dynamics) or (iii) another sign is detected and indicates the return to the default speed limit or indicates a new exceptional limit.

[0050] As seen previously, the determined nature of the road advantageously makes it possible to define upper speed limits which should not be displayed on an explicit sign. For example, if the nature of the road is "urban road", one should not encounter a sign indicating an authorized speed limit of 130 km / h. This makes it possible, for example, to filter out aberrant speed limits. For example, a 120 km / h limit sign positioned at the rear of a heavy vehicle (over 12t) can thus be discriminated against, so that the management system will not take this sign into account if the motor vehicle is on an urban road. In another example, on a motorway, the management system can detect non-existent or erroneous signs and can consider certain traffic signs as town entrance signs.Again, knowing the nature of the traffic lane, certain detections can be discriminated and excluded from the decision-making logic.

[0051] A mode of execution of the first phase El is described below in detail with reference to [Fig.2]. This mode of execution of the first phase can be seen as a mode of execution of a method for determining the nature of the traffic lane on which the motor vehicle is traveling. The first phase El comprises the following steps.

[0052] In a first step S10, the system 2 acquires data, in particular images. This data is provided by the camera 3 and by all of the other sensors 5.

[0053] In a second step S20, the data are processed by the system 2, in particular by the computer 4 of the system 2, in particular by the module 41. This processing includes all the processes for processing the electrical signals supplied by the cameras 3 and other sensors 5. This processing may include image processing, in particular signal processing making it possible to put the signals into optimized forms to carry out the third detection step S30 which follows.

[0054] In the third step S30, using the acquired data, the system 2, in particular the module 42, detects events, in particular events occurring in the environment of the motor vehicle or state events of the motor vehicle.

[0055] For example, events may fall into the following categories: - detection of an explicit speed sign (e.g. 130 km / h speed limit sign), and / or - detection of an implicit speed sign (e.g. end of speed limit sign, sign indicating the motorway nature of the road or the express nature of the road), and / or - detection of a contextual traffic sign (any other sign or any other characteristic of a sign such as a background color of a sign), and / or - detection of a traffic lane structure (such as a roundabout, a bend, a structure such as a bridge), and / or - detection of a road marking (such as a continuous white line, a broken white line, a pedestrian crossing), and / or - detection of dynamic information from the motor vehicle 1 (such as speed, steering angle of the steered wheels, acceleration), and / or - detection of other objects in the environment of the motor vehicle 1 and their possible dynamics (such as a pedestrian, a cyclist, a motorcyclist, another motor vehicle, a building).

[0056] In a fourth step S40, the events are supplied to the artificial neural network logic module 43 and processed by this module. At the output of this module, the nature of the traffic lane on which the motor vehicle is traveling is obtained, this nature being determined by the artificial neural network logic module 43.

[0057] Artificial neural network means a machine learning model that can be represented by a parametric function comprising a succession linear operations (multiplication by a scalar commonly called "weights") and / or non-linear operations (application of a so-called "activation" function, for example Sigmoid, Heaviside, Hyperbolic Tangent, Relu, etc.), these operations being grouped into "layers". A number of predefined artificial neural network architectures can be used to create an artificial neural network, including multilayer perceptrons (in English Multilayer Perceptron abbreviated MLP), convolutional neural networks (in English Convolutional Neural Network abbreviated CNN), or recurrent neural networks (in English Recurrent Neural Network abbreviated RNN).

[0058] As an alternative to the mode of execution of the determination method described previously with reference to [Fig.2], the method comprises a single phase in which the computer 4 configured by the implementation of a configuration method as described below is used to directly determine the maximum authorized speed at which the motor vehicle is traveling.

[0059] In such a phase, steps S10 to S30 are identical to those described previously.

[0060] On the other hand, step S40 is different. Indeed, the events remain supplied to the artificial neural network logic module 43 and processed by this module. On the other hand, at the output of this module, the authorized speed limit at which the motor vehicle is traveling is directly obtained, this speed limit being determined by the artificial neural network logic module 43. In this variant, the module 43 will therefore have been configured differently.

[0061] A mode of execution of a method for configuring the computer 4 incorporating the module 43 of logic with artificial neural network 200 is described below with reference to [Fig. 3]. This method can be implemented by a configuration system similar to the system 2, but additionally integrating: - a reference device capable of determining at any time in a very reliable manner the nature of the traffic lane on which the motor vehicle is traveling and / or the authorized speed limit of the place where the motor vehicle is traveling (This system advantageously includes a means of geolocating the vehicle and a map of the road traffic network), and - a device for modifying the logic of the neural network.

[0062] In a first step SI 10, the configuration system acquires data, in particular images. This data is provided by the camera 3 and by all of the other sensors 5.

[0063] In a second step S120, the data are processed by the configuration system, in particular by the computer 4, in particular of the module 41, of the configuration system.

[0064] In a third step S130, using the data, the configuration system detects events, in particular events occurring in the environment of the motor vehicle or state events of the motor vehicle.

[0065] For example, events may fall into the following categories: - detection of an explicit speed sign (e.g. 130 km / h speed limit sign), and / or - detection of an implicit speed sign (e.g. end of speed limit sign, sign indicating the motorway nature of the road or the express nature of the road), and / or - detection of a contextual traffic sign (any other sign or any other characteristic of a sign such as a background color of a sign), and / or - detection of a traffic lane structure (such as a roundabout, a bend, a structure such as a bridge), and / or - detection of a road marking (such as a continuous white line, a broken white line, a pedestrian crossing), and / or - detection of dynamic information from the motor vehicle 1 (such as speed, steering angle of the steered wheels, acceleration), and / or - detection of other objects in the environment of the motor vehicle 1 and their possible dynamics (such as a pedestrian, a cyclist, a motorcyclist, another motor vehicle, a building).

[0066] In a fourth step S140, the detected events are defined as input parameters of the artificial neural network and the events are supplied to the artificial neural network logic module 43 and processed by this module. At the output of this module, the nature of the traffic lane on which the motor vehicle is traveling and / or the authorized speed limit of the place where the motor vehicle is traveling are obtained, this nature or this speed being determined by the artificial neural network logic module 43. This is the output parameter(s) of the artificial neural network.

[0067] [Fig.4] represents an example of an artificial neural network 200 that can be used in the context of the invention. The neural network comprises an input layer 210, hidden layers 220 and an output layer 230.

[0068] In order to be processed by the neural network, the events are transmitted to the artificial neural network in the form of a vector of labels. In other words, the artificial neural network takes, as input, an array of values ​​(or vector of labels) each of whose indices corresponds to a particular event, for example: - an index may correspond to the detection of a speed limit sign at 80km / h; - another index for detecting a speed limit sign at 70km / h etc. (we can therefore have an index for each detectable sign); - other clues may correspond in the same way to each of the detectable explicit and implicit signs.

[0069] The value of the table at index i may then take a first value (for example 1) following a positive detection of the specific event associated with index i, and a second value (for example 0) in the absence of a detection of the specific event associated with index i.

[0070] We can also define a lifetime associated with each detection, corresponding to the duration during which a detection is represented in the label vector (by a 1 according to the preceding example), a vector being generated at each iteration of the determination method (every 10ms or 20ms etc.). The value of the table at index i is set to 0 when the lifetime is reached. The event is then expired.

[0071] Thus, at each iteration of the determination method, we can have a vector of labels corresponding to the current state of the vehicle and grouping together all of the events detected and which have not yet expired.

[0072] In an alternative embodiment, relative to the neural network, with regard to the panel detection events, the labels (valued 0 or 1) can be replaced by weightings which can take other values. - A first weighting is advantageously defined which is associated with the aging of the detection and decreases with this aging of the detection. The first detection event of the panel causes the panel aging parameter to take a weighting first equal to 1, then which decreases with the distance traveled since the instant of detection. For example, in a case of a maximum age of 10 km, the weighting decreases linearly to reach a value equal to 0 after 10 km. - A second weighting is advantageously defined which is associated with the detection occurrences of the sign. Each time a sign is seen, the time of its occurrence is kept in memory. The second weighting is given according to the number of occurrences in a predetermined distance (for example 2km, which can vary depending on the country and the speed of the vehicle). In an example where a maximum number of occurrences equal to 5 is chosen, each detection of the sign occurring within the predetermined distance increases the second weight by 1 / 5.

[0073] With respect to the neural network, with regard to structure (or infrastructure) detection events, a label (input parameter) without weighting is preferably applied (in other words, a value of 1 without impact of the aging) for each of the following events: - nature of the traffic lane (urban, extra-urban, expressway, motorway): a label is defined for each nature with a label value of 1 when the actual nature of the lane corresponds to the label. This label is not an input parameter of the neural network when it constitutes an output parameter of the neural network. - number of traffic lanes on the road (1 lane, 2 lanes, 3 lanes, 4 lanes and more): a label is defined for each number of lanes with a label value of 1 when the number corresponds to the number of traffic lanes. - lane delimitation (continuous line, broken line, double line, shoulder, wall, bollards, parked vehicles): a label is defined for each situation with a label value of 1 when the situation corresponds to the lane delimitation type. - roundabouts, tolls, intersections are preferably managed like the signs mentioned above.

[0074] However, it will be possible to add to these events (or labels) a numerical value comparable to a rate of occurrence of the different events over a given distance. In this case, it will be possible, for example, to define, in addition to the labels, the following input parameters of the artificial neural network: - occurrence of the number of queues over a predefined distance (for example as a percentage over the last two kilometers). - occurrence of boundaries over a predefined distance (e.g. as a percentage over the last two kilometers).- occurrence of roundabouts, tolls, intersections etc. (e.g. as a percentage over the last two kilometers).

[0075] Relating to the neural network, with respect to the dynamic information of the vehicle for each of the following events: - average speed range: we preferably define a label per speed range interval [0—30], [30—50], ... - traffic jam: we preferably define a label to characterize a traffic jam situation - steering angle range: preferably define a label per steering angle range interval of [-100 — 100], [-200—200],... - we also preferably define an aging parameter for ground marking detection events (such as arrows, stop lines, pedestrian crossings, pictograms, writing on the ground, lying police officers.

[0076] Relative to the neural network, with regard to the other objects in the environment of the motor vehicle 1 and their possible dynamics for each of the following events: - object type (car, truck, cyclist, pedestrian): we define a label for each type with a parameter value for the number of objects and an aging parameter, - speed of objects: we define a parameter for characterizing speeds.

[0077] A parameter is also preferably defined for the n (for example n=2) previously determined limit speeds.

[0078] The size of the neural network may depend on the number of panels available in a given country.

[0079] A single neural network can be made that would be valid for a set of countries (e.g. all of Europe) and that would accept all relevant parameters in Europe.

[0080] The input layer of the neural network can have a size equal to the sum of the number of parameters evoked for a region or a country.

[0081] If multiple neural networks are configured for different geographic areas, the relevant neural network is activated based on the vehicle position. The vehicle position may be detected by the system 2 or may be input by a user.

[0082] The hidden layers of the neural network are configurable to maximize the performance of the system without having a large network that would consume memory storage resources and computing resources. For example, the maximum number of perceptrons per layer is less than the number of parameters of the input layer. More preferably, the total number of layers does not exceed 10 to 12.

[0083] In parallel with step S140, in a step S150, the nature of the traffic lane on which the motor vehicle is traveling and / or the authorized speed limit of the place where the motor vehicle is traveling is determined using the reference device.

[0084] In a test step S160, it is determined whether the results obtained during steps S140 and S150 are identical.

[0085] If this is the case, we loop back to step S110. If this is not the case, we move on to a step S170 in which we indicate to the artificial neural network the output value that should have been determined. The logic of the artificial neural network is modified accordingly. We thus carry out the training of the neural network.

[0086] Furthermore, when at step S160, it is determined that the results of steps S140 and S150 are very often identical, for example identical to within a percentage during a test period of a few kilometers traveled with the vehicle, it can be considered that the computer 4, in particular the artificial neural network logic module 43, is correctly configured. From then on, the logic defined in the artificial neural network logic module 43 can be saved. saved, then recorded on other calculators 4 which are intended to equip management systems 2 and vehicles 1 produced in series.

[0087] Preferably, the training of the neural network is done massively “offline” over hundreds of thousands of kilometers (in real driving of a motor vehicle), which is called “training data set”. Then, the performance of the neural network is judged by testing it during a validation procedure (“validation data set”). If the results are conclusive, the neural network is deployed on the computers of mass-produced vehicles so that it can be used in real time.

[0088] The solutions described above have the advantage of not using GPS resources to locate the motor vehicle and of not using mapping storing in memory a model of traffic lane networks and the authorized speed limits on each section of these lanes.

[0089] The solutions described make it possible to maximize the number of indices collected, thus making it possible to secure and make more reliable the determination of the nature of the traffic lane on which the motor vehicle is traveling and, consequently, to secure and make more reliable the determination of the authorized speed limit where the motor vehicle is located.

Claims

Claims

1. Method for configuring a computer (4) of a motor vehicle (1), the computer incorporating an artificial neural network logic module (43), the method comprising the following steps: • a step of acquiring data, in particular images, • on the basis of the acquired data, a step of detecting events, in particular events in the environment of the motor vehicle, • a step of defining the detected events as input parameters of the artificial neural network, • a step of determining the nature of the traffic lane on which the motor vehicle is traveling and / or the authorized speed limit of the place where the motor vehicle is traveling as output parameters of the artificial neural network.

2. Configuration method according to the preceding claim, characterized in that the events include events of the following categories: • detection of an explicit speed sign, and / or • detection of an implicit speed sign, and / or • detection of a contextual sign, and / or • detection of a traffic lane structure, and / or • detection of a road marking, and / or • detection of dynamic information from the vehicle to the car, and / or • detection of other objects in the environment of the vehicle to the car and their possible dynamics.

3. Configuration method according to claim 1 or 2, characterized in that the method comprises a learning step, the learning step comprising for example: • the operation of the neural network powered by the detected events, and • the operation of a reference system for determining the nature of the road on which the motor vehicle is traveling and / or for determining the authorized speed limit of the place where the motor vehicle is traveling.

4. Method for determining, by a management system (2) of a motor vehicle (1), a maximum authorized speed at which the motor vehicle is traveling, the method comprising the following phases: • a phase of using a computer configured by the implementation of the method according to one of the preceding claims to determine the nature of the traffic lane on which it is traveling, and • a phase of using the nature of the traffic lane to determine the maximum authorized speed.

5. Method for determining, by a management system (2) of a motor vehicle (1), a maximum authorized speed where the motor vehicle is traveling, the method comprising a phase of using a computer configured by the implementation of the method according to one of claims 1 to 3 to determine the authorized speed limit of the place where the motor vehicle is traveling.

6. Determination method according to claim 4 or 5, characterized in that the determination method is executed in real time and / or in that the steps are iterated at a fixed time interval and / or at a fixed movement interval of the motor vehicle.

7. Determination method according to one of claims 4 to 6, characterized in that it comprises a phase of displaying the maximum authorized speed.

8. Determination method according to one of claims 4 to 7, characterized in that the method comprises a step of not taking into account a detection of an explicit speed limit sign.

9. Determination method according to one of claims 4 to 8, characterized in that the method comprises a step of interpreting the meaning of an implicit traffic sign.

10. Calculator (4) configured by implementing the method according to one of claims 1 to 3.

11. Management system (2) of a motor vehicle (1), the management system comprising hardware and / or software elements (3, 4, 5, 6) implementing the method according to one of claims 4 to 9, in particular hardware elements (3, 4, 5, 6) and / or software designed to implement the method according to one of the preceding claims.

12. Motor vehicle (1) comprising a management system (2) according to the preceding claim.

13. A computer program product comprising program code instructions recorded on a computer-readable medium for implementing the steps of the method according to any one of claims 1 to 9 when said program is running on a computer.

14. Data recording medium (4), readable by a computer, on which is recorded a computer program comprising program code instructions for implementing the method according to one of claims 1 to 9.

Images