A method for automatically and safely centering a vehicle in a traffic lane.

Abstract

A method for automatically and safely centering a vehicle in a traffic lane. One aspect of the invention relates to a method for automatically and safely centering a vehicle in a traffic lane, comprising the following steps: Activation of an automatic centering function for the vehicle in the traffic lane; If the driver's hands are not detected on the steering wheel for a first period, issuance of at least one alert to the driver; If the driver's hands are not detected on the steering wheel after a second period following the first period, automatic deactivation of the centering function; the first and second periods depending on a set of factors applying to the vehicle and / or the driver.

Description

Title of the invention: Method for the safe automatic centering of a vehicle in a traffic lane. TECHNICAL FIELD OF THE INVENTION

[0001] The technical field of the invention is that of safety functions in vehicles, and more particularly that of trajectory control functions in vehicles.

[0002] The present invention relates to a method for centering a vehicle in a traffic lane and, in particular, to a safe, automatic method for centering a vehicle in a traffic lane. The invention also relates to a vehicle and a computer program for implementing the method. TECHNOLOGICAL BACKGROUND OF THE INVENTION

[0003] The lane centering function is a feature that allows a vehicle to be centered in the lane it is traveling in, without the driver having to apply any torque to the steering wheel. However, this does not relieve the driver of the need to remain vigilant and ready to take back control of the vehicle; therefore, the lane centering function is coupled with a driver's hand detection system. This detection makes it possible to determine when the driver relies too heavily on the lane centering function and removes their hands from the steering wheel, and to implement a series of alerts, culminating in the deactivation of the lane centering function, depending on the duration for which the driver's hands are not on the steering wheel.

[0004] Theoretically, to improve safety, the alert durations should be minimized, reminding the driver to remain fully in control of the vehicle and preventing accidents. However, systems that detect the driver's hands on the steering wheel generate numerous false positives; that is, they detect that the driver's hands are not on the steering wheel when they are. Furthermore, alert durations that are too short would lead to many undue deactivations of the lane centering function, discouraging the driver from using it.

[0005] Alert durations are therefore currently the result of a compromise between a level of service quality and security.

[0006] There is therefore a need to improve the safety aspect of the lane centering function without compromising the performance provided. Summary of the invention

[0007] The invention offers a solution to the problems mentioned above, by making it possible to ensure that the driver is in control of his vehicle when the lane centering function is activated, without increasing the number of undue deactivations of the function.

[0008] A first aspect of the invention relates to a method for the safe automatic centering of a vehicle in a traffic lane, comprising the following steps: • Activation of an automatic vehicle centering function within the traffic lane; • If the driver's hands are not detected on the steering wheel for an initial period, at least one alert will be issued to the driver; • If the driver's hands are not detected on the steering wheel after a second period following the first period, the centering function will be automatically deactivated; the first duration and the second duration depending on a set of factors applying to the vehicle and / or the driver.

[0009] Thanks to the invention, the time during which the driver's hands are not detected before an alert is issued and the time during which the driver's hands are not detected before the lane centering function is deactivated vary according to conditions affecting the vehicle, such as the type of road on which the vehicle is traveling, and conditions affecting the driver, such as the driver's state of distraction. Thus, the durations are reduced and safety enhanced only in situations considered particularly accident-prone, and increased in other cases, to minimize the inconvenience caused by false positives generated by the driver's hand detection system.

[0010] In addition to the characteristics mentioned in the preceding paragraph, the process according to the invention may have one or more complementary characteristics from among the following, considered individually or according to all technically possible combinations.

[0011] According to one embodiment, the set of factors applicable to the vehicle and / or the driver includes the type of road on which the vehicle is traveling, the presence of a bend or intersection on the road, the weather conditions, the brightness, driver distraction, the speed of the vehicle and / or the safety distances observed by the vehicle.

[0012] Thus, the set of factors influencing the times before alert and before deactivation of the centering function are factors identified as particularly accident-prone and accessible to the vehicle.

[0013] According to an embodiment compatible with the previous embodiment, the process according to the invention comprises the following steps: • If the driver's hands are not detected on the steering wheel during the first period, a first alert is issued; • If the driver's hands are not detected on the steering wheel after a third consecutive period following the first period, a second alert will be issued; the third duration depending on the set of factors applying to the vehicle and / or the driver.

[0014] According to a sub-embodiment of the previous embodiment, the first alert is a visual alert and the second alert is an audible and visual alert.

[0015] Thus, the intrusive nature of the alerts increases as the time since the last detection of the driver's hands on the steering wheel increases.

[0016] According to an embodiment compatible with the previous embodiments, the deactivation step of the centering function includes a substep of emitting a third alert for a fourth duration following the second duration, the fourth duration depending on the set of factors applicable to the vehicle and / or the driver, the third alert being an audible alert.

[0017] Thus, the driver is warned of the deactivation of the centering function.

[0018] According to an embodiment compatible with the embodiment variants In previous versions, the method according to the invention includes a step of determining an accident risk from the set of factors applying to the vehicle and / or the driver, each duration being defined according to the determined accident risk.

[0019] Thus, the more the vehicle is subjected to accident-prone constraints, the higher the risk of accident and the shorter the times are to ensure safety.

[0020] According to a sub-variant of the preceding variant embodiment, the risk of accident is a counter and the determination step includes at least one sub-step of incrementing the counter when a condition relating to a factor of the set of factors applying to the vehicle and / or the driver is verified.

[0021] Thus, the solution is simple and requires little data and few computing resources in the vehicle.

[0022] According to an alternative embodiment of the previous sub-variant embodiment, the determination step includes a sub-step of using a machine learning algorithm providing the risk of accident from the set of factors applying to the vehicle and / or the driver.

[0023] Thus, the determination of the risk of accident is more precise.

[0024] A second aspect of the invention relates to a vehicle comprising means for implementing the method according to the invention.

[0025] A third aspect of the invention relates to a computer program product comprising instructions which, when the program is executed by a computer, lead the latter to implement the steps of the process according to the invention.

[0026] The invention and its various applications will be better understood by reading the following description and examining the accompanying figures. BRIEF DESCRIPTION OF THE FIGURES

[0027] The figures are presented for illustrative purposes only and are in no way limiting of the invention. • The [Fig. 1] is a synoptic diagram illustrating the sequence of steps of a process according to the invention. • Fig. 2 shows a schematic representation of a vehicle on which the process according to the invention is applied, as a function of time. DETAILED DESCRIPTION

[0028] Unless otherwise specified, the same element appearing on different figures has a unique reference.

[0029] The invention relates to a method for centering a vehicle in a traffic lane on which the vehicle is traveling, and this in an automatic manner, that is to say without intervention by a driver of the vehicle, and in a safe manner.

[0030] The sequence of steps of process 100 is illustrated in [Fig.1] and their effects on vehicle 200 as a function of time are illustrated in [Fig.2].

[0031] A first step 101 of the process 100 consists of activating a function implemented on the vehicle 200, allowing the automatic centering of the vehicle 200 in the traffic lane.

[0032] Such activation can be carried out automatically or by the driver of vehicle 200.

[0033] On [Fig.2], the first step 101 is carried out at time t0.

[0034] A second step 102 of the process 100 consists of determining a risk of accident based on a set of factors applying to vehicle 200 and / or the driver of vehicle 200.

[0035] The set of factors applicable to the vehicle 200 and / or the driver includes, for example, a plurality of factors identified as particularly accident-causing, for example from accident data from road safety or reported by vehicles involved in accidents, and more particularly a plurality of factors that can be assessed from data accessible to the vehicle 200, for example from a sensor of the vehicle 200 or by consulting a shared database accessible to the vehicle 200.

[0036] The set of factors applicable to vehicle 200 and / or the driver includes, for example, the type of road on which vehicle 200 is traveling, the presence of a bend or intersection on the road, weather conditions, brightness, driver distraction, vehicle speed 200 and / or safety distances observed by vehicle 200.

[0037] For example, data relating to the type of road on which the vehicle 200 is traveling and the presence of a bend or intersection on the road can be obtained by consulting a map from the location of the vehicle 200 and data relating to driver distraction can be obtained by processing images obtained from an interior camera of the vehicle 200.

[0038] For example, in [Fig.2], vehicle 200 is travelling on a two-way road, this type of road having been identified as particularly accident-prone.

[0039] According to a first embodiment, the second step 102 includes at least one substep 1021 of incrementing a counter corresponding to the risk of accident to be determined.

[0040] The counter is for example initialized to zero as soon as the lane centering function is activated and then incremented as soon as a condition relating to a factor in the set of factors applying to the vehicle 200 and / or the driver is verified.

[0041] The counter is incremented for example if vehicle 200 travels on a two-way road, then if the road has a bend, then if the speed of vehicle 200 exceeds the authorized limit.

[0042] According to a second embodiment, the second step 102 includes a substep 1022 of using a machine learning algorithm, or "machine learning" in English, providing the risk of accident from the set of factors applying to the vehicle 200 and / or the driver, that is to say that the output of the algorithm is the risk of accident when it is provided as input the set of factors applying to the vehicle 200 and / or the driver.

[0043] The machine learning algorithm is, for example, an artificial neural network previously trained to provide an accident risk from a set of factors applying to the vehicle 200 and / or the driver.

[0044] A third step 103 of the process 100 is implemented if the driver's hands are not detected on the steering wheel for a duration, referred to as the first duration Dl, following the activation of the lane centering function during the first step 101.

[0045] Detection is carried out for example via one or more capacitive sensors integrated into the steering wheel of vehicle 200.

[0046] The first duration Dl is defined according to the set of factors applying to the vehicle 200 and / or the driver, and more particularly the risk of accident determined from the set of factors applying to the vehicle 200 and / or the driver during the second step 102.

[0047] In the first embodiment, the first duration DI depends on the final value of the counter and has, for example, a first value if the counter is greater than a first threshold, a second value if the counter is greater than a second threshold, itself less than the first threshold, a third value if the counter is greater than a third threshold, itself less than the second threshold, and a fourth value otherwise.

[0048] In the second embodiment, the first duration DI depends on the risk of accident provided by the machine learning algorithm and has, for example, a first value if the risk of accident is greater than a first threshold, a second value if the risk of accident is greater than a second threshold, itself less than the first threshold, a third value if the risk of accident is greater than a third threshold, itself less than the second threshold, and a fourth value otherwise.

[0049] The first value is for example equal to 5 seconds, the second value is for example equal to 5 seconds, the third value is for example equal to 10 seconds and the fourth value is for example equal to 14 seconds.

[0050] On [Fig.2], the first duration DI extends between the instants t0 and tl.

[0051] The third step 103 consists of issuing an alert, called the first alert, to destination of the driver of vehicle 200.

[0052] The first alert is for example a visual alert, the visual alert consisting for example of the display of a message on a screen of the vehicle 200, the message being for example "Keep your hands on the steering wheel".

[0053] A fourth step 104 of the process 100 is implemented if the driver's hands are not detected on the steering wheel for a duration called the third duration D3, following the first duration DI.

[0054] The third duration D3 is defined according to the set of factors applying to the vehicle 200 and / or the driver, and more particularly the risk of accident determined from the set of factors applying to the vehicle 200 and / or the driver during the second step 102.

[0055] In the first embodiment, the third duration D3 depends on the final value of the counter and has, for example, a first value if the counter is greater than a first threshold, a second value if the counter is greater than a second threshold, itself less than the first threshold, a third value if the counter is greater than a third threshold, itself less than the second threshold, and a fourth value otherwise.

[0056] In the second embodiment, the third duration D3 depends on the accident risk provided by the machine learning algorithm and, for example, has a first value if the accident risk is greater than a first threshold, a a second value if the risk of accident is greater than a second threshold, itself less than the first threshold, a third value if the risk of accident is greater than a third threshold, itself less than the second threshold, and a fourth value otherwise.

[0057] The first value is for example equal to 0 seconds, the second value is for example equal to 5 seconds, the third value is for example equal to 10 seconds and the fourth value is for example equal to 14 seconds.

[0058] On [Fig.2], the third duration D3 extends between the instants t1 and t2.

[0059] The fourth step 104 consists of issuing an alert, called a second alert, to destination of the driver of vehicle 200.

[0060] The second alert is for example a visual and audible alert, the alert consisting for example of the display of a message on a screen of the vehicle 200, the message being for example "Keep your hands on the steering wheel" and the emission of an audible signal.

[0061] A fifth step 105 of the process 100 is implemented if the driver's hands are not detected on the steering wheel for a period, called the second period D2, following the first period DI and including the third period D3.

[0062] The second duration D2 is defined according to the set of factors applying to the vehicle 200 and / or the driver, and more particularly the risk of accident determined from the set of factors applying to the vehicle 200 and / or the driver during the second step 102.

[0063] In the first embodiment, the second duration D2 depends on the final value of the counter and has, for example, a first value if the counter is greater than a first threshold, a second value if the counter is greater than a second threshold, itself less than the first threshold, a third value if the counter is greater than a third threshold, itself less than the second threshold, and a fourth value otherwise.

[0064] In the second embodiment, the second duration D2 depends on the risk of accident provided by the machine learning algorithm and has, for example, a first value if the risk of accident is greater than a first threshold, a second value if the risk of accident is greater than a second threshold, itself less than the first threshold, a third value if the risk of accident is greater than a third threshold, itself less than the second threshold, and a fourth value otherwise.

[0065] The first value is for example equal to 30 seconds, the second value is for example equal to 35 seconds, the third value is for example equal to 40 seconds and the fourth value is for example equal to 44 seconds.

[0066] On [Fig.2], the second duration D2 extends between the instants t1 and t3.

[0067] The fifth step 105 consists of deactivating the centering function, and thus giving control back to the driver of vehicle 200.

[0068] Deactivation is performed automatically.

[0069] In [Fig.2], the deactivation is carried out at time t3.

[0070] The fifth step 105 may include a substep 1051 of issuing an alert, called a third alert, during a fourth duration D4 following the second duration D2.

[0071] The fourth duration D4 is for example defined according to the set of factors applying to the vehicle 200 and / or the driver, and more particularly according to the risk of accident determined from the set of factors applying to the vehicle 200 and / or the driver during the second step 102.

[0072] In the first embodiment, the fourth duration D4 depends on the final value of the counter and has, for example, a first value if the counter is greater than a first threshold, a second value if the counter is greater than a second threshold, itself less than the first threshold, a third value if the counter is greater than a third threshold, itself less than the second threshold, and a fourth value otherwise.

[0073] In the second embodiment, the fourth duration D4 depends on the risk of accident provided by the machine learning algorithm and has, for example, a first value if the risk of accident is greater than a first threshold, a second value if the risk of accident is greater than a second threshold, itself less than the first threshold, a third value if the risk of accident is greater than a third threshold, itself less than the second threshold, and a fourth value otherwise.

[0074] The first value is for example equal to 5 seconds, the second value is for example equal to 5 seconds, the third value is for example equal to 5 seconds and the fourth value is for example equal to 5 seconds.

[0075] The third alert is, for example, an audible alert and consists of the emission of a sound signal.

[0076] On [Fig.2], the fourth duration D4 extends between the instants t3 and t4.

Claims

Demands

1. A method (100) for the safe automatic centering of a vehicle (200) in a traffic lane, comprising the following steps: • Activation (101) of an automatic centering function of the vehicle (200) in the traffic lane; • If the driver's hands are not detected on the steering wheel for a first duration (D1), emission (103, 104) of at least one alert to the driver; • If the driver's hands are not detected on the steering wheel after a second duration (D2) following the first duration (D1), automatic deactivation (105) of the centering function; the method (100) being characterized in that the first duration (D1) and the second duration (D2) depend on a set of factors applying to the vehicle (200) and / or the driver.

2. A method (100) according to claim 1, wherein the set of factors applicable to the vehicle (200) and / or the driver includes the type of road on which the vehicle (200) is traveling, the presence of a bend or intersection on the road, the weather conditions, the brightness, driver distraction, the speed of the vehicle (200) and / or the safety distances observed by the vehicle (200).

3. A method (100) according to any one of the preceding claims, comprising the following steps: • If the driver's hands are not detected on the steering wheel during the first duration (D1), emission (103) of a first alert; • If the driver's hands are not detected on the steering wheel after a third duration (D3) following the first duration (D1), emission (104) of a second alert; the third duration (D3) depending on the set of factors applicable to the vehicle (200) and / or the driver.

4. Method (100) according to the preceding claim, wherein the first alert is a visual alert and the second alert is an audible and visual alert.

5. A method (100) according to any one of the preceding claims, wherein the step (105) of deactivating the centering function comprises a substep (1051) of emitting a third alert during a fourth duration (D4) following the second duration (D2), the fourth duration (D4) depending on the set of factors applicable to the vehicle (200) and / or the driver, the third alert being an audible alert.

6. A method (100) according to any one of the preceding claims, comprising a step (102) of determining an accident risk from the set of factors applicable to the vehicle (200) and / or the driver, each duration (D1, D2, D3, D4) being defined according to the determined accident risk.

7. A method (100) according to the preceding claim, wherein the risk of accident is a counter and the determination step (102) includes at least one substep (1021) of incrementing the counter when a condition relating to a factor in the set of factors applying to the vehicle (200) and / or the driver is met.

8. Method (100) according to claim 7, wherein the determination step (102) comprises a substep (1022) of using a machine learning algorithm providing the risk of accident from the set of factors applying to the vehicle (200) and / or the driver.

9. Vehicle (200) comprising means for implementing the method (100) according to any one of the preceding claims.

10. Product computer program comprising instructions which, when the program is executed by a computer, cause the computer to carry out the steps of the process (100) according to any one of claims 1 to 8.

Images