Method and system for personalizing the operation of a device for assisting the driver of a motor vehicle

Description

Technical field of the invention

[0001] The present invention relates to the field of electronic and computer systems on board motor vehicles to improve the comfort and ergonomics of vehicles. The invention relates in particular to a method for managing, by a computer system on board a motor vehicle, the operation of a driving assistance device of the vehicle. The invention also relates to a system implementing such a method. The invention applies, in particular, to passenger cars, utility vehicles, etc. State of the prior art

[0002] It is known that most current motor vehicles are equipped with computer and electronic equipment that provides driver assistance functions, in particular functions that allow the automated performance of maneuvers or autonomous regulation of the speed of vehicles (e.g. adaptive cruise control, intelligent speed adaptation device, device for managing lane change maneuvers, etc.). Generally speaking, such equipment operates by exploiting data generated by detection devices arranged in the vehicles (e.g. camera, radar, lidar, etc.), data on which they rely to provide the driver assistance functions.It is also known that known driver assistance devices that provide functionalities to autonomously support the performance of maneuvers, for example lane change maneuvers, generally perform these maneuvers in a relatively unchanging manner. For example, when performing a lane change maneuver, the vehicle acceleration applied during the maneuver by most known driver assistance devices will often remain the same from one lane change maneuver to the next, regardless of the type of road taken by the vehicle. And, for some users, such tireless repetition of the same sensations during the same maneuvers, regardless of the type of road taken, may be considered too off-putting, or even create a feeling of discomfort.Furthermore, a user who is used to driving slowly may also experience a feeling of lack of comfort and ergonomics when the driving assistance equipment of his vehicle applies, when carrying out maneuvers, accelerations which deviate too much from those which the user usually applies himself.

[0003] US 2019 / 322276 describes a method for limiting the speed of a vehicle traveling in a curve comprising the steps of: receiving several samples of road data, including location points, determining a parameter value for each of a first set of parameters affecting the speed of the vehicle, selecting a vehicle speed model based on the first set of parameter values, determining a parameter value for each of the parameters of a second set affecting the speed of the vehicle, calculating a recommended speed for the vehicle for at least one of the road samples based on the selected model and the parameter values ​​of the second set of parameters affecting the speed of the vehicle and providing an instruction to adapt the speed of the vehicle to the recommended speed at the location point of at least one road sample.

[0004] US 2019 / 171204 describes a method for controlling an automatically driven vehicle switchable between manual driving in which the vehicle is caused to move depending on an operation of an occupant and automatic driving in which driving characteristics during automatic travel are defined and the vehicle is caused to move automatically based on the driving characteristics. The manual driving characteristics during manual driving by the occupant are learned and, when switching from manual driving to automatic driving, automatic driving is performed with the manual driving characteristics maintained for a predefined manual characteristics maintenance time. Summary of the invention

[0005] The invention aims to overcome these drawbacks. The invention aims to provide a method and a system that allow the operation of a driving assistance device for a motor vehicle to be personalized by taking into account the driving habits of the vehicle's users, in particular for autonomously managing the performance of maneuvers. In this way, the invention aims to improve the comfort and ergonomics of motor vehicles.

[0006] These aims are achieved, according to a first object of the invention, by means of a method of managing, by a computer system on board a motor vehicle, the operation of a vehicle driving assistance device, according to claim 1. The method comprises the steps of: (i) determine data characterizing, for each of the traffic lanes of a set of traffic lanes used by the vehicle, the type of traffic lane; (ii) determine data characterizing, for each traffic lane of the set of traffic lanes used by the vehicle, at least one extreme value relating to the kinematics of the vehicle on the traffic lane;iii) feeding a deep learning module with the data characterizing, for each traffic lane of a set of traffic lanes used by the vehicle, the type of traffic lane and with the data characterizing, for each traffic lane of the set of traffic lanes used by the vehicle, at least one extreme value relating to the kinematics of the vehicle on the traffic lane in order to generate data characterizing, for each type of traffic lane, an estimate of at least one extreme value relating to the kinematics of the vehicle; and iv) managing the operation of the driving assistance equipment based on data characterizing a type of traffic lane used by the vehicle at the current time and data characterizing, for each type of traffic lane, an estimate of at least one extreme value relating to the kinematics of the vehicle. ;

[0007] According to an advantageous variant, step i), step ii) and step iv) may comprise a step consisting of interacting with a navigation device of the vehicle and / or with a traffic sign recognition device of the vehicle. According to another variant, the data characterizing, for each traffic lane of the set of traffic lanes taken by the vehicle, at least one extreme value relating to the kinematics of the vehicle on the traffic lane may include data characterizing at least one value of the instantaneous acceleration of the vehicle, data characterizing at least one value of the instantaneous rate of change of the acceleration of the vehicle and data characterizing at least one value of the instantaneous speed of the vehicle.

[0008] According to the invention, step ii) comprises a step of determining whether the absolute value of the instantaneous rate of change of the acceleration of the vehicle is non-zero.

[0009] According to the invention, step i) and step iv) comprise a step consisting of determining data characterizing an identifier of a driver of the vehicle.

[0010] According to another variant, the step of determining data characterizing an identifier of a driver of the vehicle can be carried out by interacting with a human-machine interface of the vehicle.

[0011] Furthermore, the invention also relates to a system for managing the operation of a driving assistance device of a motor vehicle, the system comprising at least one information processing unit, comprising at least one processor, and a data storage medium configured to implement a method as described above.

[0012] Furthermore, the invention also relates to a program comprising program code instructions for executing the steps of a method as described above when said program is executed on a computer and / or a processor.

[0013] Furthermore, the invention also relates to a medium usable in a computer on which a program as described above is recorded.

[0014] Finally, the invention also relates to a motor vehicle comprising a system as described above. Brief description of the figures

[0015] Other characteristics and advantages of the invention will appear on examining the detailed description below, and the attached drawings, in which: [ Fig. 1 ] is a functional diagram of a system according to the invention; and [ Fig. 2 ] is a flowchart illustrating the steps of a method according to the invention. Detailed description of the invention

[0016] According to the invention, a system 100 for managing the operation of a driving assistance device of a motor vehicle is a computer system, represented schematically in Figure 1 , which comprises an information processing unit 101, comprising one or more processors, a data storage medium 102, at least one input and output interface 103, allowing the reception of data and the transmission of data, and a deep learning module 104, which comprises a neural network, for example a multi-layer perceptron.

[0017] According to certain embodiments, the system 100 according to the invention is embedded in a motor vehicle (e.g. car, bus, heavy goods vehicle, etc.) and it is hosted on one or more of the computers, electronic control units or other telematics boxes of the vehicle. According to other embodiments, the system 100 according to the invention is hosted on a computer of a motor vehicle and it interacts via its input and output interface 103 with a computer of a driving assistance device of a motor vehicle. According to the preferred embodiment, the system 100 according to the invention is an integral part of a computer of a driving assistance device of a motor vehicle.Consequently, whatever the embodiment of the invention, the system 100 according to the invention is always able to interact, via its input and output interface 103, not only with a driving assistance device of a motor vehicle, but also, via such a driving assistance device, with any other device which, in a conventional manner, equips and / or interacts with a driving assistance device of a current motor vehicle.

[0018] However, according to the invention, a driving assistance device for a motor vehicle comprises at least one detection device (e.g. lidar, radar, camera, ultrasonic sensor, inertial unit, accelerometer, etc.) for perceiving the driving environment and one or more dedicated calculators, computers and / or processors, which, depending on raw data generated by the detection device(s), can control the operation of certain components of the vehicle which govern its movement (e.g. electronic power steering, electronic trajectory corrector, ABS, cruise control, etc.).In addition, to implement certain steps of the method according to the invention described below, a driving assistance device for a motor vehicle according to the invention also comprises a traffic sign recognition device, a human-machine interface, which comprises a touch keyboard, a facial recognition device and / or a fingerprint recognition device, and it comprises dedicated hardware (e.g. connectors) and software means for interacting with a navigation device of a motor vehicle, which comprises a receiver interacting with a satellite positioning system. Thus, by interacting with or forming an integral part of such a driving assistance device for a motor vehicle, the system 100 according to the invention is in particular able to assume the guidance of the vehicle when carrying out maneuvers.Furthermore, thanks to these means, the system 100 according to the invention can also determine data characterizing, for each of the traffic lanes of a set of traffic lanes taken by a vehicle, the type of traffic lane, or data characterizing, for each traffic lane of the set of traffic lanes taken by the vehicle, at least one extreme value relating to the kinematics of the vehicle on the traffic lane (e.g. speed, acceleration, rate of variation of the acceleration of the vehicle, etc.).

[0019] According to the invention, all the elements described above therefore contribute to enabling the system 100 according to the invention to implement, on board a motor vehicle, a method for managing the operation of a vehicle driving assistance device, as described below in connection with the Figure 2 .

[0020] According to a first step 201 of the method according to the invention, the system 100 according to the invention determines data characterizing, for each of the traffic lanes of a set of traffic lanes used by the vehicle, the type of traffic lane. To do this, the system 100 according to the invention proceeds by first determining data characterizing an identifier of a driver of the vehicle and, to do this, it interacts with the human-machine interface of the vehicle. Indeed, when a driver of the vehicle sits in the driving position, he is preferentially invited, for example by means of a visual message broadcast on a screen of the interface and / or by means of an audible message broadcast by a loudspeaker, to identify himself, for example by means of the touch keyboard, the facial recognition device and / or the fingerprint recognition device of the human-machine interface.When the driver identifies himself, or when he is automatically identified by the facial recognition device, data characterizing an identifier of a driver of the vehicle are generated by the human-machine interface and these data are extracted by the system 100 according to the invention, which records them on its data storage medium 102 in order to constitute a user profile for the identified driver. Then, when the driver of the vehicle is identified, the system 100 according to the invention determines, in connection with the driver who has been identified, the data characterizing, for each of the traffic lanes of a set of traffic lanes used by the vehicle, the type of traffic lane.In other words, the system 100 according to the invention determines at each instant, by interacting for this purpose with the navigation system and / or with the traffic sign recognition equipment of the vehicle, the type of traffic lane taken by the vehicle (e.g. motorway, national road, departmental road, etc.). Thus, by this first step 201 of the method according to the invention, the system 100 according to the invention actually constitutes data which characterizes, in connection with an identified driver, types of traffic lane taken.

[0021] According to a second step 202 of the method according to the invention, the system 100 according to the invention determines data characterizing, for each traffic lane of the set of traffic lanes taken by the vehicle, at least one extreme value relating to the kinematics of the vehicle on the traffic lane. In other words, for each traffic lane taken by the vehicle, which the system 100 according to the invention determines during the previous step of the method, the system 100 according to the invention determines at least one extreme value relating to the kinematics of the vehicle on the traffic lane.Indeed, for each traffic lane taken by the vehicle, the system 100 according to the invention determines, by interacting for example with the navigation equipment and / or with the driving assistance equipment of the vehicle, data characterizing at least one value of the instantaneous acceleration of the vehicle, data characterizing at least one value of the instantaneous rate of variation of the acceleration of the vehicle and data characterizing at least one value of the instantaneous speed of the vehicle.More specifically, when it determines that the absolute value of the instantaneous rate of change of the acceleration of the vehicle changes from a zero value to a non-zero value, which corresponds to a situation in which the vehicle performs a maneuver during which the speed of the vehicle increases or decreases, the system 100 according to the invention then determines data characterizing the maximum acceleration reached by the vehicle during the maneuver, data characterizing the maximum instantaneous rate of change of the acceleration reached by the vehicle during the maneuver as well as data characterizing the instantaneous speed of the vehicle at the end of the maneuver.Thus, by this second step of the method according to the invention, the system 100 according to the invention actually constitutes data which characterizes, for maneuvers carried out, in connection with a traffic lane of a particular type and an identified driver, the driving behavior of the driver in terms of acceleration and speed of the vehicle during the maneuvers.

[0022] According to a third step 203 of the method according to the invention, the system 100 according to the invention feeds the deep learning module 104 with the data characterizing, for each traffic lane of a set of traffic lanes taken by the vehicle, the type of traffic lane, which were determined during the first step 201, and with the data characterizing, for each traffic lane of the set of traffic lanes taken by the vehicle, at least one extreme value relating to the kinematics of the vehicle on the traffic lane, which were determined during the second step 202.More specifically, during this third step, the system 100 according to the invention preferably carries out training of the neural network of the deep learning module 104 by using the aforementioned data to do this, in order to generate data characterizing, for each type of traffic lane, an estimate of at least one extreme value relating to the kinematics of the vehicle, preferably an estimate of a maximum acceleration value and an estimate of a maximum value of the instantaneous rate of variation of the acceleration of the vehicle. In other words, by this third step 203, the system 100 according to the invention advantageously constitutes, for a plurality of types of traffic lanes previously taken by the vehicle, data which characterize the usual driving behavior of an identified driver, in particular in terms of acceleration applied when performing maneuvers.

[0023] And, according to a fourth step 204 of the method according to the invention, the system 100 according to the invention manages the operation of the driving assistance equipment as a function of data characterizing a type of traffic lane taken by the vehicle at the current time and data characterizing, for each type of traffic lane, an estimate of at least one extreme value relating to the kinematics of the vehicle. To do this, the system 100 according to the invention determines the data characterizing a type of traffic lane taken by the vehicle at the current time by proceeding as in the first step 201 of the method, i.e. by interacting with the navigation system and / or with the traffic sign recognition equipment of the vehicle.Then, to determine the data characterizing, for each type of traffic lane, an estimate of an extreme value relating to the kinematics of the vehicle, the system 100 according to the invention interacts with the human-machine interface of the vehicle to determine, as during the first step 201 of the method, an identifier of the driver of the vehicle at the current time. On the basis of this data, the system 100 according to the invention then acquires, in connection with the identified driver, the data characterizing, for each type of traffic lane, an estimate of an extreme value relating to the kinematics of the vehicle, which were determined during the third step 203 of the method. Finally, it is by using this data that the system 100 according to the invention manages the operation of the driving assistance equipment of the vehicle, in particular that it manages the operation of the driving assistance equipment when performing maneuvers.In other words, by this fourth step 204 of the method, the system 100 according to the invention uses the data which have been previously generated by the deep learning module 104 in connection with an identified driver to manage the operation of the driving assistance equipment when carrying out maneuvers. Advantageously, the system 100 according to the invention is therefore able to manage the carrying out of maneuvers in a manner which faithfully reproduces the usual driving behavior of the driver, in particular in terms of applied acceleration.

[0024] Consequently, under the terms of the method and system according to the invention described above, a solution is provided for personalizing the operation of a driving assistance device of a motor vehicle by taking into account the driving habits of the users of the vehicle, in particular when autonomously taking charge of the performance of maneuvers. By this means, the method and system according to the invention therefore significantly improve the comfort and ergonomics of motor vehicles.

Claims

1. Method for managing, by a computer system (100) on board a motor vehicle, the operation of a vehicle driving assistance device, the method comprising the steps of: i) determining data characterizing an identifier of a driver of the vehicle and data characterizing, for each of the traffic lanes of a set of traffic lanes used by the vehicle, the type of traffic lane; ii) determining, once it is established that the absolute value of the instantaneous rate of change of the acceleration of the vehicle is non-zero, data characterizing, for each traffic lane of the set of traffic lanes used by the vehicle, at least one extreme value relating to the kinematics of the vehicle on the traffic lane; iii) feeding a deep learning module (104) with data characterizing, for each traffic lane of a set of traffic lanes used by the vehicle, the type of traffic lane and with data characterizing, for each traffic lane of the set of traffic lanes used by the vehicle, at least one extreme value relating to the kinematics of the vehicle on the traffic lane in order to generate data characterizing, for each type of traffic lane, an estimate of at least one extreme value relating to the kinematics of the vehicle; and iv) managing the operation of the driving assistance equipment based on data characterizing an identifier of a driver of the vehicle, data characterizing the type of traffic lane used by the vehicle at the current time and data characterizing, for each type of traffic lane, an estimate of an extreme value relating to the kinematics of the vehicle.

2. Method according to claim 1, characterized in that step i), step ii) and step iv) comprise a step consisting of interacting with a navigation device of the vehicle and / or with a traffic sign recognition device of the vehicle.

3. Method according to one of the preceding claims, characterized in that the data characterizing, for each traffic lane of the set of traffic lanes used by the vehicle, at least one extreme value relating to the kinematics of the vehicle on the traffic lane include data characterizing at least one value of the instantaneous acceleration of the vehicle, data characterizing at least one value of the instantaneous rate of change of the acceleration of the vehicle and data characterizing at least one value of the instantaneous speed of the vehicle.

4. Method according to claim one of the preceding claims, characterized in that the step of determining data characterizing an identifier of a driver of the vehicle is carried out by interacting with a human-machine interface of the vehicle.

5. A system (100) for managing the operation of a driving assistance device of a motor vehicle, the system comprising at least one information processing unit (101), comprising at least one processor, and a data storage medium (102) configured to implement a method according to any one of the preceding claims.

6. A computer program comprising program code instructions for executing the steps of a method according to any one of claims 1 to 4 when said program is executed on a computer.

7. A medium usable in a computer, on which a program according to claim 6 is recorded.

8. A motor vehicle, comprising a system according to claim 5.