Control of lane change assist function activation

Abstract

The invention relates to a method for controlling a lane change assist function of a vehicle. Following the receipt (200) of a command to activate a lane change assist function to a destination lane, the method comprises the detection (201), based on data from at least one sensor, of at least one vehicle, and the assignment (202) of said at least one detected vehicle to a lane. In addition, driver data is received (203) from a driver monitoring system. A decision (205) to authorize or deny the activation of the lane change assist function is made based on the driver data and the assignment of said at least one detected vehicle to the lane. FIG. 2

Description

Title of the invention: Control of the activation of a lane change assistance function

[0001] The present invention belongs to the field of driver assistance for a motor vehicle, in particular lateral driving assistance for the vehicle.

[0002] It is particularly advantageous in the context of a change of traffic lane.

[0003] The term “vehicle” means any type of vehicle such as a motor vehicle, a moped, a motorcycle, a warehouse storage robot, etc.

[0004] The term "autonomous driving" of an "autonomous vehicle" means any method capable of assisting the driving of the vehicle. This method may consist of partially or fully controlling the vehicle or providing any type of assistance to a person driving the vehicle. Thus, "autonomous driving" covers all levels 1 to 5 of the OICA (International Organization of Motor Vehicle Manufacturers) scale.

[0005] Driving assistance can intervene at different levels of autonomy: from autonomous driving without driver intervention, to assistance with manual driving.

[0006] Driver assistance systems, such as AD AS for "Advanced Driver-Assistance Systems", allow the driver of the vehicle to be assisted or even to fully control certain parameters of the vehicle's steering, such as longitudinal trajectory, speed in particular, and lateral trajectory, turning or changing lanes in particular.

[0007] These systems make it possible to improve driving comfort as well as safety, by taking advantage of data from vehicle sensors, such as radar or lidar data, and / or geolocation devices.

[0008] A driver assistance function helps to assist the lateral movement of the vehicle during a lane change. Such a function is called semi-automated lane change, or SALC, or "Semi-Automated Lane Change" in English.

[0009] Such a function can be activated by the driver using a turn signal, or TI, for "Tum Indicator". It thus indicates the destination lane in order to change lanes from the current lane to the adjacent destination lane.

[0010] The function uses data from the vehicle's sensors to assist lane changing, in particular to accept or reject the activation of the SALC function, and then, if accepted, to assist trajectory control side of the vehicle.

[0011] In particular, sensors such as radars can be arranged at the rear of the vehicle to detect other vehicles travelling behind the vehicle, in the same direction of travel.

[0012] The vehicles thus detected are then assigned to a traffic lane, for example, a right, left or center lane, depending on the number of traffic lanes on the road on which the vehicle is traveling.

[0013] The lanes can be identified by a front camera, then extrapolated to the rear of the vehicle based on vehicle movement data.

[0014] However, every sensor has limitations and / or is liable to make detection errors, particularly when traffic conditions, such as weather conditions in particular, deteriorate, or depending on the road trajectory, particularly at the rear of the vehicle in the event of a turn or significant yaw.

[0015] Errors can lead to the following situations: - Failure to detect a vehicle located in the adjacent destination lane, or an error in the lane assigned to a detected vehicle, for example, if the detected vehicle is assigned to a lane adjacent to the destination lane, while it is actually traveling in the destination lane. Such a situation is called a "false negative," meaning that a vehicle present in the destination lane is not detected or is not assigned as such. The SALC function can then be activated without taking this vehicle into account, which can lead to a dangerous situation, or even a collision; - Detection of a "ghost" object or vehicle, particularly through radar wave reflection off road barriers, in the destination lane, or an error whereby a detected vehicle is mistakenly assigned to the destination lane when it is actually traveling in another lane. Such a situation is a "false positive," meaning that a vehicle is detected and assigned to the destination lane by mistake. In this case, the lane change may be denied, preventing the driver from changing lanes when a safe lane change could be made.

[0016] There is therefore a need to improve driving comfort and / or safety within the framework of a lane change assistance function of a vehicle.

[0017] The present invention improves the situation.

[0018] To this end, a first aspect of the invention relates to a method for controlling a lane-changing assistance function of a vehicle, known as an ego-vehicle, the method comprising: - receipt of a command to activate the lane change assistance function, for a change from a current lane to a de- lane destination; - detection, based on sensor data, of at least one vehicle in a rear position relative to said ego-vehicle, and assignment of each detected vehicle to a traffic lane; - receiving driver data from a driver monitoring module; - deciding whether or not to allow the activation of the lane change assistance function, based on driver data and the assignment of said at least one detected vehicle to a traffic lane.

[0019] Thus, the invention makes it possible to use driver data representative of driver behavior, particularly alertness, or driver fitness, to authorize or deny the activation of the lane change assistance function for changing lanes to the destination lane. This driver data thus enhances the interpretation of sensor data, thereby improving safety and / or driving comfort during lane changes.

[0020] According to some embodiments, the driver data may include an indication of the driver consulting a rearview mirror, and the decision to allow or not to allow the activation of the lane change assistance function may depend on the indication of the driver consulting the rearview mirror.

[0021] Such an indication makes it possible to condition the activation of the lane change assist function in certain driving situations, when the driver is attentive and checks the rearview mirror. Safety associated with lane changes is thus improved. In addition, the decision to authorize or deny activation may depend on comparing the duration of the rearview mirror check with a given threshold duration. Activation of the lane change assist function may, in particular, be conditional upon the rearview mirror being checked for a longer period than the given threshold duration.

[0022] In addition, the decision to allow or not the activation of the lane change function may include the comparison of driver data and the assignment of said at least one vehicle to the lane, with conditions of at least one predefined decision-making rule, and at least one predefined decision-making rule may include a condition dependent on the driver's indication of consulting the rearview mirror.

[0023] The use of predefined decision-making rules makes it possible to deal with driving situations identified as potentially problematic or dangerous.

[0024] In addition, said at least one vehicle can be detected with a probability of Given detection, at least one predefined decision-making rule may include a condition relating to a comparison between the probability of vehicle detection and a detection threshold value.

[0025] This makes it possible to take into account the uncertainty in detecting a vehicle. Such uncertainty can, in particular, be compensated for by greater driver vigilance, which can be assessed from driver data.

[0026] In addition or alternatively, said at least one detected vehicle can be assigned to the traffic lane with a given probability of assignment, at least one predefined decision-making rule may include a condition relating to a comparison between the probability of assignment of the vehicle and a threshold value of assignment.

[0027] It is thus made possible to take into account the uncertainty of assigning the traffic lane to the detected vehicle, in order in particular to compensate for such uncertainty according to the vigilance of the driver which is indicated by the driver data received from the driver monitoring system.

[0028] In addition or alternatively, at least one predefined decision-making rule may include a first condition relating to a first comparison between a distance separating the ego-vehicle from said at least one vehicle detected and assigned to the destination lane, and a first longitudinal distance threshold value, and may include a second condition relating to a second comparison of said distance with a second longitudinal distance threshold value, the second longitudinal distance threshold value being less than the first longitudinal distance threshold value.

[0029] Thus, it is made possible, through a comparison with two threshold values ​​of distance, to classify the distance with the vehicle detected at the rear into at least three distance intervals, and to make a decision based on the interval in which the vehicle detected at the rear is located.

[0030] In addition, the decision-making rule comprising the first and second conditions may allow the activation of the lane change assistance function if the rearview mirror consultation indication shows that the driver is consulting the rearview mirror and if said distance is between the second longitudinal distance threshold value and the first longitudinal distance threshold value.

[0031] Thus, when the detected vehicle is in the destination lane at an intermediate distance from the ego-vehicle, activation of the lane change assistance function may still be permitted if it is determined that the driver is being vigilant, i.e. when checking his rearview mirror.

[0032] According to some embodiments, the method may further include identifying a vehicle type for said at least one detected vehicle. The vehicle detection threshold, the vehicle assignment threshold, the first threshold value of longitudinal distance and / or the second threshold value of longitudinal distance can be determined based on the type of vehicle identified for said at least one vehicle detected.

[0033] Thus, the decision-making rules are adapted to the type of vehicle detected, which improves the accuracy of decision-making and the safety associated with changing lanes.

[0034] A second aspect of the invention relates to a computer program comprising instructions for implementing the method according to the first aspect of the invention, when these instructions are executed by a processor.

[0035] A third aspect of the invention relates to a control module for a lane change assistance function of a vehicle, known as an ego-vehicle, comprising: - an interface capable of receiving a command to activate a lane change assistance function, for a change from a current lane to a destination lane; - a processor capable of detecting, based on sensor data, at least one vehicle in a rear position relative to said ego-vehicle, and of assigning each detected vehicle to a traffic lane; - another interface capable of receiving driver data from a driver monitoring module. The processor is also capable of deciding whether or not to allow the activation of the lane change assistance function, based on driver data and the assignment of each detected vehicle to a traffic lane.

[0036] Other features and advantages of the invention will become apparent from an examination of the detailed description below, and the accompanying drawings in which:

[0037] [Fig-1] illustrates a vehicle according to an embodiment of the invention;

[0038] [Fig.2] is a diagram illustrating the steps of a control process for a function of changing the traffic lane of a vehicle, according to embodiments of the invention;

[0039] [Fig.3a] illustrates a first driving situation of a vehicle according to modes of realization of the invention;

[0040] [Fig.3b] illustrates a second situation of driving a vehicle according to modes realization of the invention;

[0041] [Fig.3c] illustrates a third driving situation of a vehicle according to modes of realization of the invention;

[0042] [Fig.4] illustrates the structure of a control device for an assistance function lane change of a vehicle, according to embodiments of the invention.

[0043] Fig. 1 presents a vehicle 100, or ego-vehicle 100, according to an embodiment of the invention.

[0044] The ego-vehicle 100 includes a control device 101. The control device 101 can be an ECU type element, for "Electronic Control Unit" in English, in charge of the centralized control of the vehicle 100. The control device 101 can include several modules in charge of piloting or communicating with different elements of the motor vehicle.

[0045] The vehicle 100 may further include a human-machine interface, HMI, 102 capable of enabling the control device 101, and other entities of the vehicle 100, to exchange information with the driver or with passengers of the vehicle 100.

[0046] No restrictions are attached to the HMI 102, which includes any element capable of receiving a command from the user, whether this command relates to driving the vehicle or controlling interior equipment, or of transmitting information to the user. To this end, the HMI 102 may include a screen, such as a touchscreen, a set of one or more buttons, a loudspeaker, a microphone, an instrument panel capable of displaying one or more illuminated pictograms of predefined shapes, a steering wheel vibration system, etc.

[0047] Preferably, the HMI 102 is capable of transmitting visual and / or audible data to the vehicle driver. Additionally, the driver or passengers can input information via the HMI 102 for transmission to the ECU 101 control unit via a button, touch control, voice command, or any other type of user input.

[0048] The vehicle 100 may further include a turn signal activation interface 106 capable of receiving a turn signal activation command, so as to transmit such a command to the ECU 101, which is thus capable of controlling a lighting / signaling device to trigger a turn signal. Such an interface 106 for receiving a turn signal activation command may be capable of receiving a right turn signal activation command and a left turn signal activation command. For example, it may be a rotary lever, which the user can turn in either direction to generate a right or left turn signal activation command.

[0049] The ego-vehicle 100 may include a driving assistance module 104, also called AD AS, for "Advanced Driver Assistance Systems" in English, capable of performing at least one driver assistance function, based in particular on data from sensors 103.1 and 103.2 of vehicle 100, and possibly other sensors not shown in [Fig.1].

[0050] No restrictions are attached to vehicle 100, in particular regarding the level of driving autonomy permitted by the vehicle. The AD AS 104 module can assist the driver in piloting the vehicle, particularly for determining or maintaining speed or acceleration, or ensuring autonomous vehicle operation without driver input. Driver assistance thus also encompasses automatic vehicle control.

[0051] According to the invention, the AD AS 104 module is at least capable of performing a semi-automated traffic lane change function, also called SALC function for "Semi Automated Lane Change".

[0052] Such a function may use data from at least one first sensor 103.1 located at the rear of the vehicle and capable of detecting one or more vehicles traveling behind the vehicle 100, in particular those traveling in the same direction. The first sensor 103.1 may be a radar sensor, or it may be a camera facing the rear of the vehicle. In the following, it is considered, for illustrative purposes only, that the first sensor 103.1 is a radar sensor.

[0053] The vehicle 100 may further include at least one second sensor 103.2 capable of identifying traffic lanes of the road on which the vehicle 100 is traveling, for example the traffic lanes in front of the vehicle 100. The sensor 103.2 may be a camera facing the front of the vehicle 100 associated with an analysis module capable of detecting traffic lanes, identifying a current traffic lane and one or more adjacent lanes, from the sensor images by the camera.

[0054] Thus, the activation of the SALC function is conditional upon the detection of at least one other vehicle travelling on the same road, on the detection of traffic lanes, and on the assignment of a traffic lane to each vehicle thus detected.

[0055] The sensor 103.1 can also be capable of acquiring position and speed data from at least one other vehicle traveling on the road. For example, for each other vehicle, the sensor 103.1 is capable of determining: - the distance between said other vehicle and vehicle 100; - the speed of the other vehicle.

[0056] The vehicle 100 may further include a driver monitoring system 105, also called DMS for "Driver Monitoring System", which may include in particular a camera directed towards the driver inside the vehicle, in particular towards the driver's face, as well as an analysis module capable of analyzing the images captured by the camera of the DMS 105.

[0057] Figure 2 is a diagram illustrating the steps of a method for controlling a lane change assistance function of vehicle 100, in particular for controlling the SALC function described above. Such a method can be implemented in the AD AS 104 module described above, or alternatively in the ECU 101. Alternatively, some steps are implemented in the ECU 101 and other steps are implemented in the AD AS 104 module. The following illustrates an example of an implementation in the AD AS 104 module. uniquely.

[0058] At step 200, a command to activate the SALC function is received by the AD AS module 104. Such a command may originate from a driver command on the HMI 102. Alternatively, such an activation command may be automatically generated by the ECU module 101 upon receipt of a signal from the turn signal activation interface 106. The command identifies a destination lane, which is the lane adjacent to the right of the current lane if the right turn signals are activated, and which is the lane adjacent to the left of the current lane if the left turn signals are activated.

[0059] At a step 201, the AD AS 104 module detects at least one other vehicle travelling on the same road as vehicle 100, in the same direction of travel, and behind vehicle 100, based on data from the first sensor 103.1.

[0060] At step 202, the AD AS 104 module assigns each vehicle detected in step 201 to a traffic lane. Step 202 may include a preliminary identification of the traffic lanes based on data from the second sensor 103.2, for example. Such traffic lane identification may be implemented continuously, particularly for use by other driver assistance functions implemented by the AD AS 104 module.

[0061] Thus, each detected vehicle can be assigned to the current lane of vehicle 100 or to one of the other lanes, including the destination lane.

[0062] As detailed below, uncertainties may be associated with steps 201 to 202 due to limitations of sensors 103.1 and 103.2, and / or due to weather conditions. Vehicle detection in step 201 may also include evaluating a probability of detection (Pdet), which can be a value between 0 and 1, with 1 being strictly greater than 0. 1 indicates a level of total certainty, while 0 indicates total uncertainty regarding the detected vehicle. The probability of detection (Pdet) thus indicates the reliability associated with vehicle detection. Several methods for evaluating such a probability (Pdet) are known and are not further detailed in this description.

[0063] Similarly, assigning a detected vehicle to a traffic lane in step 202 may also include evaluating a Pass lane assignment probability, which can be a value between 0 and 1, being strictly greater than 0. 1 indicates a level of total certainty, while 0 indicates total uncertainty regarding the traffic lane assigned to the detected vehicle. Several methods for evaluating such a Pass probability are known and are not further detailed in this description.

[0064] At a step 203, which can be implemented in parallel with steps 201 and 202, or before step 201, or after step 202, but before step 204 described below, the The AD AS 104 module obtains driver data from the DMS 105. No restrictions are attached to this driver data, which includes any descriptive data about driver behavior, including the driver's face, and / or the driver's physical or physiological state. Preferably, the driver data identifies whether the driver is checking the rearview mirror. More specifically, the driver data can identify the driver's gaze position. The method of obtaining such data is known and is not described further here. It can, in particular, be obtained by the image analysis module of the DMS 105 camera.

[0065] At step 204, the AD AS 104 module determines a decision to authorize or not the SALC function, based on the data obtained in steps 201 to 203, namely, based on: - the assignment of a traffic lane to at least one detected vehicle; - driver data from the DMS 105 system.

[0066] Thus, taking driver data into account in the decision-making process regarding the authorization of SALC function activation improves the decision-making process, since sensor data is reinforced by driver behavior data. In particular, SALC function activation can be authorized when driver data indicates that the driver is checking the rearview mirror. Conversely, SALC function activation can be denied in case of uncertainty regarding the detected vehicle or lane assignment, when driver data indicates that the driver is not checking the rearview mirror.

[0067] The following general decision-making rule can be applied by the AD AS 104 module: activation of the SALC function is permitted only if no vehicle is reliably detected in the destination lane at a distance from vehicle 100 less than a threshold longitudinal distance value Dmin. A vehicle is reliably detected in the destination lane if it is detected with a high probability of detection Pdet, greater than a given probability of detection threshold, and is assigned to the destination lane with a high probability of assignment Pass, greater than a given probability of assignment threshold. The threshold longitudinal distance value Dmin can be fixed or can preferably be calculated based on the speed Vego of vehicle 100 and the speed Varr of the vehicle detected in the destination lane.In particular, the higher the detected vehicle speed Varr in the destination lane, the greater the threshold longitudinal distance value Dmin.

[0068] When Varr is less than Vego - Vseuill, the threshold distance value Dmin can be equal to a minimum distance pMinDist. Vseuill is a first speed threshold predetermined, which may depend on Vego, and which may be equal to 10 km / h for example.

[0069] When Varr is between Vego - Vthreshold 1 and Vego, the threshold distance value Dmin varies linearly as a function of Varr, between pMinDist and Vego*TgMin, TgMin being a minimum duration, which may depend on Vego.

[0070] When Varr is between Vego and Vego + Vthreshold 2, the threshold distance value can vary linearly with Varr, between Vego*TgMin and a pSupDist value greater than Vego*TgMin. pSupDist can depend on Vego.

[0071] When Varr is greater than Vego + Vthreshold 2, Dmin can vary beyond pSupDist, for example as a function of the square of Varr.

[0072] Other rules for calculating Dmin can be applied according to the invention.

[0073] According to the invention, decision-making rules, implemented by the AD AS 104 module and further taking into account driver data, are given below, it being understood that this set of rules is not exhaustive, and that the decision to authorize or not the SALC function may include other decision-making rules including other conditions, not described in this application.

[0074] Each decision-making rule detailed below includes a decision, namely to allow or deny, corresponding to a set of at least one condition. If all the conditions of the decision-making rule are met, then the rule's decision is applied. Decision-making rules are preferably mutually exclusive, so that two contradictory decisions cannot be made for the same detected vehicle. If two vehicles are detected, the decision-making rules are applied independently to the two vehicles, and in the event of a conflict in the decision-making—that is, if one rule for the first vehicle leads to acceptance, and another rule for the second vehicle leads to denial—the AD AS 104 module can refuse to activate the SALC function.

[0075] A first decision-making rule can be applied by the AD AS 104 module: to refuse activation of the SALC (decision-making) function if the driver data indicates that the driver is not looking in the rearview mirror (first condition) and if there is uncertainty about the presence of a vehicle detected from behind (second condition). There is uncertainty about the presence of a vehicle detected from behind if the probability of detection (Pdet) of the vehicle is less than a first detection threshold value (Pdsl). The first decision-making rule makes it possible to avoid activating the SALC function in case of uncertainty about detection, as the driver data indicates that the driver has not consulted their rearview mirror. This avoids any risk-taking when the data from the sensors is unreliable and is not compensated for by the driver's vigilance.

[0076] A second decision-making rule can be applied by the AD AS module 104: Deny activation of the SALC (decision-making) function if driver data indicates that the driver is not looking in the rearview mirror (first condition), and if a vehicle is assigned to a lane other than the destination lane with uncertainty about the lane assignment (second condition), i.e., with certainty about the lane assignment below a given threshold. Specifically, the uncertainty may correspond to a probability of the vehicle being assigned lane (Pass) that is lower than a predetermined assignment threshold (Pasl), or it may correspond to the vehicle being too close to the destination lane, for example, at a lateral distance less than a given first lateral distance threshold (Dlatl).

[0077] Fig. 3a illustrates a first driving situation in which the second decision rule can be applied.

[0078] In the first driving situation, vehicle 100 is traveling on a road 300 comprising a right lane 301, a center lane 302, and a left lane 303. The current lane of vehicle 100 is the right lane 301. The driver of the vehicle has activated their right turn signal, and the destination lane is therefore the center lane 302. In the first situation, a first vehicle 310.1 or a second vehicle 310.2 is detected behind vehicle 100. The first vehicle 310.1 can be assigned to the current lane 301, but with a low degree of certainty due to its proximity to the destination lane 302. Similarly, the second vehicle 310.2 can be assigned to the left lane 303, but with a low degree of certainty due to its proximity to the destination lane 302.Thus, if at least one detected vehicle is at a distance from the destination lane 302 which is less than the first lateral distance threshold value Dlatl, the application of the second rule by the AD AS 104 module implies the refusal to activate the SALC function if the driver data indicates that the driver is not consulting the rearview mirror.

[0079] Referring again to [Fig. 2], a third decision-making rule can be applied by the AD AS 104 module: to allow activation of the SALC function (decision-making) if the driver data indicates that the driver is consulting the rearview mirror (first condition), and if there is uncertainty about the presence of a vehicle detected behind the ego-vehicle 100 (second condition). There is uncertainty about the presence of a vehicle detected behind if, for each detected vehicle, the probability of detection Pdet is less than a second detection threshold value Pds2, the second detection threshold value Pds2 preferably being greater than the first detection threshold value Pds1 of the first decision-making rule. The third rule can include the additional condition that the driver turns the steering wheel at least partially toward the destination lane to allow activation of the SALC function.

[0080] The AD AS 104 module can further apply the fourth decision-making rule The following condition is required: to allow activation of the SALC (decision-making) function if driver data indicates that the driver is checking the rearview mirror (first condition) and if a vehicle is detected and assigned to the destination lane (second condition), at a distance from the vehicle 100 that is less than the threshold distance value Dmin (third condition), referred to as the first threshold longitudinal distance value Dmin, but greater than a second threshold longitudinal distance value Dmin' (fourth condition), where Dmin' is less than Dmin. The second threshold longitudinal distance value Dmin' can depend on the speed of the vehicle 100 and the speed of the vehicle detected in the destination lane. As with Dmin, Dmin' can vary depending on different Varr value ranges. - Dmin' is equal to a fixed low value pMinDist' if Varr is less than Vego -Vseuill, with pMinDist' less than or equal to pMinDist; - first linear dependence of Dmin' on Varr, with a first coefficient of variation, when Varr is between Vego -Vthreshold 1 and Vego. The second threshold distance value Dmin' thus varies between pMinDist' and Vego*TgMin', TgMin' being a minimum duration, which can depend on Vego and which can be less than TgMin; - second linear dependence of Dmin' on Varr, with a second coefficient of variation, greater than the first coefficient of variation, when Varr is between Vego and Vego + Vthreshold 2. The second threshold distance value Dmin' can be between Vego*TgMin and a value pSupDist' greater than Vego*TgMin', t greater than or equal to pSupDist. pSupDist' can depend on Vego; - dependence of Dmin' on the square of Varr, when Varr is greater than Vego + Vthreshold 2: Dmin' can vary beyond pSupDist'.

[0081] The AD AS module can apply the fourth rule in the second situation of illustrated driving with reference to [Fig.3b].

[0082] In [Fig.3b], the vehicle 100 is travelling on a road 320 comprising two traffic lanes, namely a right lane 321, which is the current lane of the vehicle 100, and a left lane 322, which is the destination lane of the vehicle 100 when the driver activates the left turn signals.

[0083] A vehicle 330 is detected and assigned to the destination lane 322. In particular, the vehicle 330 can be detected and assigned with certainty to the destination lane 322, that is to say with high probabilities of detection and assignment Pdet and Pass, in particular higher than the threshold values ​​Pdsl and Pasl described previously.

[0084] The detected vehicle 330 is at a distance 323 from the ego-vehicle 100, the distance 323 being less than a first distance 324 equal to the first threshold value of longitudinal distance Dmin, and greater than a second distance 325 equal to the second threshold value of longitudinal distance Dmin'.

[0085] Thus, in the second situation of [Fig.3b], the AD AS 104 module can allow the activation of the SALC function if the driver data indicates that the driver is consulting the rearview mirror.

[0086] The fourth rule may include the additional condition that the driver turn the steering wheel at least partially towards the destination lane to allow activation of the SALC function.

[0087] Referring again to [Fig. 2], the AD AS 104 module can further apply the following fifth decision-making rule: activation of the SALC function is validated (decision-making) if the driver data indicates that the driver is checking the rearview mirror (first condition) and if a vehicle is detected and assigned to the destination lane (second condition), but with uncertainty regarding the assignment to the destination lane (third condition). For example, the uncertainty may correspond to a probability of assignment Pass lower than the aforementioned first assignment threshold value Pasl, or to a second assignment threshold value Pas2 different from Pasl. Alternatively, or additionally, the uncertainty may correspond to a lateral distance of the vehicle less than a given second lateral distance threshold value Dlat2 from a lane adjacent to the destination lane.The second lateral distance threshold value Dlat2 can be equal to the first lateral distance threshold value Dlatl used in the second rule described above. Alternatively, Dlatl and Dlat2 are equal.

[0088] The AD AS 104 module can apply the fifth rule in a third situation of driving illustrated with reference to [Fig.3c].

[0089] In the third driving situation of [Fig.3c], the vehicle 100 is travelling on a road 340 comprising three lanes, namely a right lane 341, which is the current lane of the vehicle 100, a central lane 342 and a left lane 343.

[0090] The driver of vehicle 100 indicates a change of lane towards the destination lane, which is the central lane 342, by activating the left turn signals of vehicle 100.

[0091] A vehicle 350.1 or 350.2 is detected and assigned to the center lane 342, which is the destination lane. Both vehicles 350.1 and 350.2 are illustrated in [Fig. 3c], although the third driving situation occurs when one of the vehicles 350.1 and 350.2 is detected and assigned with uncertainty to the destination lane.

[0092] As shown in [Fig. 3c], vehicle 350.1 is close to the right lane 341, which is a lane adjacent to the destination lane 342. An uncertainty in the assignment of vehicle 350.1 to the destination lane 342 is due to the lateral distance between vehicle 350.1 and the right lane 341 being less than the second lateral distance threshold value Dlat2. Alternatively or in addition, the uncertainty is due to the probability Pass of assigning vehicle 350.1 to the destination lane 342 being in- less than the second Pas2 assignment threshold value.

[0093] As shown in [Fig. 3c], vehicle 350.2 is close to the left lane 343, which is a lane adjacent to the destination lane 342. Uncertainty regarding the assignment of vehicle 350.2 to the destination lane 342 arises because the lateral distance between vehicle 350.2 and the right lane 341 is less than the second lateral distance threshold value Dlat2. Alternatively, or in addition, the uncertainty arises because the probability Pass of assigning vehicle 350.2 to the destination lane 342 is less than the second assignment threshold value Pas2.

[0094] Thus, in the driving situation of [Fig. 3c], vehicle 350.1 or 350.2 is assigned to the destination lane with uncertainty due to its proximity to an adjacent lane. In this case, the AD AS 104 module can apply the fifth rule and allow activation of the SALC function if, in addition, the driver data indicates that the driver is checking the rearview mirror.

[0095] The fifth rule may include the additional condition that the driver turn the steering wheel at least partially towards the destination lane 342 to allow activation of the SALC function.

[0096] Referring again to [Fig.2], following the application of one of the decision-making rules whose conditions are met, the AD AS 104 module authorizes or refuses, at step 204, the activation of the SALC function and the change of direction of vehicle 100.

[0097] In the event of refusal of authorization of the SALC function, for example by application of the first decision-making rule or by application of the second decision-making rule, the AD AS 104 module inhibits the activation of the SALC function at a step 205. In addition, the AD AS 104 module can transmit a message to the driver via the HMI 102, in order to indicate that the activation of the SALC function is refused.

[0098] If authorization is given to activate the SALC function, the AD AS 104 module can display a confirmation request message on the HMI 102, at a step 206, in order to require confirmation from the driver before the SALC function is actually activated.

[0099] If the driver refuses at step 206, the process ends at step 208, without implementing a semi-automated lane change, therefore without executing the SALC function.

[0100] If the driver accepts the procedure in step 206, the AD AS 104 module implements the SALC function in step 207 to assist with lane changes. During step 207, the AD AS 104 module can verify that the driver has their hands on the steering wheel, using data from a sensor located on the steering wheel, and, if the verification is negative, can send a message to the driver instructing them to place their hands on the steering wheel, via the HMI 102. In the event that the driver If you do not place your hands on the steering wheel after a certain period of time has elapsed, the SALC function may be interrupted and the vehicle will remain in the current lane.

[0101] The confirmation request message in step 206 is optional, and, if authorization to activate the SALC function is granted in step 204, the process can proceed directly to step 207 of changing the traffic lane to the destination lane.

[0102] According to some embodiments, step 201 of vehicle detection further includes the identification of a vehicle type, for example from among the following types: car, truck, motorcycle, etc. The threshold values ​​described above may advantageously depend on the vehicle type. Thus, a threshold value is predefined per detected vehicle type for each of the following threshold values: first longitudinal distance threshold value Dmin, second longitudinal distance threshold value Dmin', first lateral distance threshold value Dlatl, second lateral distance threshold value Dlat2, first assignment threshold value Pasl, second assignment threshold value Pas2, first detection threshold value Pdsl, second detection threshold value Pds2.

[0103] In other words, a first threshold value of longitudinal distance Dminl is defined for cars, a first threshold value of longitudinal distance Dmin2 is defined for trucks, a first threshold value of longitudinal distance Dmin3 is defined for motorcycles, etc., the values ​​Dminl, Dmin2 and Dmin3 being different.

[0104] According to some embodiments, the driver data from the DMS 105 system, obtained in step 203, further include an attention indicator and / or a driver drowsiness indicator. The aforementioned threshold values ​​may then depend on the attention indicator and / or the drowsiness indicator. For example, the higher the drowsiness indicator, the higher the threshold values ​​Dmin, Dmin', Dlatl, Dlat2, Pasl, Pas2, Pdsl, and / or Pds2.

[0105] It is therefore permissible to take into account a physiological state of the driver, in particular a state of attention / fatigue of the driver, and the safety associated with the control of the SALC function is thus improved.

[0106] According to some embodiments, the condition "the driver consulted the rearview mirror" used in the five decision-making rules detailed above can be more restrictive and may require consulting the rearview mirror for a period exceeding a predefined threshold duration. Alternatively or in addition, the condition "the driver consulted the rearview mirror" can be replaced by the condition that the driver looked at the road ahead of the vehicle before consulting the rearview mirror.

[0107] According to some embodiments, the driver's gaze position is evaluated by the AD AS 104 module based on driver data received in step 203, before Implementation of step 204. For example, a two-dimensional or three-dimensional point corresponding to an intersection of the driver's eye axes can be used as the gaze position. The gaze position thus evaluated is compared with the position of at least one vehicle detected and assigned to a traffic lane following steps 201 and 202. The threshold values ​​Dlatl, Dlat2, Pasl, Pas2, Pdsl, and / or Pds2 used in step 204 for the detected vehicle can then depend on a distance between the driver's gaze position and the position of the detected vehicle. Thus, the smaller this distance, the lower the threshold values ​​Dlatl, Dlat2, Pasl, Pas2, Pdsl, and / or Pds2 applied to this vehicle in step 204 can be.

[0108] Figure 4 shows the structure of a 400 control module according to a rea ​​mode implementation of the invention.

[0109] The control module 400 can be integrated into, or can be, the AD AS 104 module described above, suitable for implementing the steps of the process according to the invention described with reference to [Fig.2].

[0110] Alternatively, the control module 400 can be integrated into, or can be, the ECU 101 control device described above, when the method according to the invention is implemented in the ECU 101 and not in the AD AS 104 module. In this case, the ECU 101 authorizes or denies the activation of the SALC function which is implemented by the AD AS 104 module.

[0111] The control module 400 includes a processor 401 configured to communicate unidirectionally or bidirectionally, via one or more buses or via a direct wired connection, with a memory 402 such as a Random Access Memory (RAM), a Read Order Memory (ROM), or any other type of memory (Flash, EEPROM, etc.). Alternatively, the memory 402 comprises several memories of the aforementioned types.

[0112] The memory 402 is capable of storing, permanently or temporarily, at least some of the data used and / or resulting from the implementation of the steps of the process according to the invention illustrated with reference to [Fig.2].

[0113] In particular, memory 402 may be able to store the threshold values ​​described above as well as at least one of the decision-making rules described above, preferably several of the decision-making rules described above.

[0114] The processor 401 is capable of executing instructions, stored in memory 402, for the implementation of the steps of the process according to the invention, described with reference to [Fig. 2]. Alternatively, the processor 401 can be replaced by a microcontroller designed and configured to perform the steps of the process according to the invention, described with reference to [Fig. 2].

[0115] The control module 400 includes a first interface 403 capable of receiving a activation command for a lane change assistance function from a current lane to a destination lane, from ECU 101 or from interface 106.

[0116] The control module 400 further includes a second interface 404 capable of receiving data from the first sensor 103.1, from which the control module 400 can detect one or more vehicles located behind the vehicle 100.

[0117] The control module 400 includes a third interface 405, capable of receiving data from the second sensor 103.2, from which the control module 400 is able to identify the traffic lanes, and then assign a traffic lane to each detected vehicle.

[0118] The control module 400 further includes a fourth interface 406 capable of receiving driver data from the DMS 105 system.

[0119] The control module 400 includes a fifth interface 407 capable of communicating with the SALC function of the AD AS 104 module. When the control module 400 is integrated into the AD AS 104 module, it does not include a fourth interface 406, since it can itself implement the SALC function.

[0120] The control module 400 may also include a sixth interface 408 capable of communicating with the HMI 102, for the delivery of messages in particular to the driver of the vehicle 100.

[0121] The present invention is not limited to the embodiments described above by way of example; it extends to other variants.

Claims

Claims

1. Method for controlling a lane change assistance function of a vehicle (100), called an ego-vehicle, the method comprising: - receiving (200) a command to activate a lane change assistance function, for a change from a current lane (301; 321; 341) to a destination lane (302; 322; 342); - detecting (201), on the basis of data from at least one sensor (103.1; 103.2), at least one vehicle (310.1; 310.2; 330; 350.1; 350.2) in a rearward position relative to said ego-vehicle, and assigning (202) said at least one detected vehicle to a lane; - receiving (203) driver data from a driver monitoring system (105); - decision (205) to authorize or not the activation of the lane change assistance function, depending on the driver data and the assignment of the detected vehicle to the lane.

2. The method of claim 1, wherein the driver data includes an indication of a consultation of a rearview mirror by the driver, and wherein the decision (205) whether or not to authorize the lane change command depends on the indication of consultation of the rearview mirror by the driver.

3. The method of claim 2, wherein the decision (205) whether or not to allow activation of the lane change assistance function comprises comparing the driver data and the assignment of the detected vehicle to the lane, with conditions of at least one predefined decision-making rule, wherein at least one predefined decision-making rule comprises a condition dependent on the driver's indication of rearview mirror consultation.

4. Method according to claim 3, wherein said at least one vehicle (310.1; 310.2; 330; 350.1; 350.2) is detected with a given detection probability, wherein at least one predefined decision-making rule comprises a condition relating to a comparison between the detection probability of the detected vehicle and a detection threshold value.

5. A method according to claim 3 or 4, wherein said at least one vehicle (310.1; 310.2; 330; 350.1; 350.2) detected is assigned to the traffic lane (301-303; 321; 322; 341-343) with a given assignment probability, wherein at least one predefined decision-making rule comprises a condition relating to a comparison between the assignment probability of the detected vehicle and an assignment threshold value.

6. Method according to one of claims 3 to 5, wherein at least one predefined decision-making rule comprises a first condition relating to a first comparison between a distance (323) separating the ego-vehicle (100) from said at least one vehicle (310.1; 310.2; 330; 350.1; 350.2) detected and assigned to the destination lane (302; 322, 342), and a first longitudinal distance threshold value (324) and comprises a second condition relating to a second comparison of said distance with a second longitudinal distance threshold value (325), the second longitudinal distance threshold value being less than the first longitudinal distance threshold value.

7. The method of claim 6, wherein the decision-making rule comprising the first and second conditions, authorizes the activation of the lane change assistance function if the rearview mirror consultation indication indicates that the driver is consulting the rearview mirror and if said distance (323) is between the second longitudinal distance threshold value (325) and the first longitudinal distance threshold value (324).

8. A method according to claim 5 and according to claim 6 or 7, further comprising identifying a vehicle type for said at least one detected vehicle (310.1; 310.2; 330; 350.1; 350.2), wherein the vehicle detection threshold value, the vehicle assignment threshold value, the first longitudinal distance threshold value (324) and / or the second longitudinal distance threshold value (325) is determined based on the vehicle type identified for said at least one detected vehicle.

9. Computer program comprising instructions for implementing the method according to one of the preceding claims, when these instructions are executed by a processor (401).

10. Control module (400) of a lane change assistance function of a vehicle (100), called ego-vehicle, comprising: - an interface (403) capable of receiving a command to activate a lane change assistance function, for a change from a current lane to a destination lane; - a processor (401) capable of detecting, on the basis of data from at least one sensor (103.1; 103.2), at least one vehicle in a rear position relative to said ego-vehicle, and of assigning said at least one detected vehicle to a traffic lane; - another interface (406) capable of receiving driver data from a driver monitoring system (105); wherein the processor is further capable of deciding whether or not to authorize the activation of the lane change assistance function, based on the driver data and the assignment of said at least one detected vehicle to the lane.