Automatic method for evaluating the level of safety of a driver's driving of a vehicle, and vehicle
The method provides continuous, real-time driver safety evaluation using sensor data to address delayed feedback and context-aware scoring, enhancing road safety by immediate driver feedback and advice.
Patent Information
- Application Number
- PCT/EP2026/062598
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2025-05-06
- Filing Date
- 2026-04-30
- Publication Date
- 2026-11-12
AI Technical Summary
Existing driver safety evaluation methods provide delayed and driver-centric feedback, often focusing on vehicle performance rather than real-time driver behavior, and do not account for dynamic driving contexts.
An automatic method that continuously evaluates driver safety during a journey using real-time data from internal and external sensors, applying weighted scoring to provide instant feedback and advice on driving errors.
Enables real-time assessment and immediate feedback on driving errors, promoting safer practices by informing drivers of their actions and suggesting improvements.
Smart Images

Figure EP2026062598_12112026_PF_FP_ABST
Abstract
Description
Description Title of the invention: Automatic method for evaluating the safety level of a driver's ability to operate a vehicle
[0001] The present invention relates to the field of road safety, and in particular to autonomous on-board methods and equipment for improving driving by providing information to the driver about their driving.
[0002] The invention more specifically relates to an automatic method for evaluating the level of safety of a vehicle's operation by a driver and for providing current information to said driver, as well as a vehicle equipped to implement this method.
[0003] Methods for assessing a driver's performance, providing risk assessments based on telematics data, or evaluating specific risks directly related to the driver (fatigue, drowsiness) are already known, and some are already implemented in vehicles in operation, sometimes using remote data processing. These solutions share the common features of generally providing an overall score or recommendation, usually after a delay, typically at the end of a journey or driving period, and / or of not focusing on the driver.
[0004] Furthermore, French patent application No. 2312268 filed by the applicant already discloses an automated method for assessing a driver's safety level while driving a vehicle. This method results in a safety score and provides driving advice. It relies on the use of traffic laws and standard safe driving practices to determine at least one score reflecting driving safety throughout a journey. This score, along with any relevant advice, is provided at the end of a given journey or at the end of a driving period (e.g., a calendar month). Moreover, the score and advice are primarily based on data related to the vehicle (its performance and condition) and focus particularly on its most negative aspects.
[0005] The present invention aims to overcome these limitations of known solutions.
[0006] To this end, its main purpose is an automatic method for evaluating the safety level of a driver's operation of a subject vehicle during a journey or trip, and for providing real-time information to said driver. This method takes into account the internal and external contexts of the vehicle and the situations and circumstances encountered successively during the journey, and uses signals or data provided by external and internal sensors and / or by a driver assistance system fitted to the subject vehicle. This method comprises at least the following sequence of operational steps, executed repeatedly over a predetermined time window, which continuously and continuously explores the progressive temporal unfolding of the journey in a sliding manner:
[0007] a) collect in real time, and repeatedly during the time window, the signals or data provided by the external and internal sensors and / or the driver assistance system, which provide initial instantaneous information relating to the actual behavior and condition of the driver and the vehicle subject, each type of information being associated with a specific risk factor related to driving and being part of a risk category;
[0008] b) to evaluate this first instantaneous information, where appropriate by comparison with second instantaneous information which corresponds either to safe driving behavior of the driver and the vehicle subject, in accordance with driving rules related to the highway code or with recommended driving rules, or to a normal state of vigilance of the driver, and to produce as many measurement or metric values;
[0009] c) to determine from each of these metrics an instantaneous basic rating or score through a matching process, for example with the implementation of a specific mapping for each score;
[0010] d) to continuously produce, for each risk factor and throughout the journey, an average rating or score by a weighted average calculation, a weighted product calculation or by another function of aggregation or merging of the different instantaneous basic ratings or scores determined on the sliding window;
[0011] e) to aggregate or combine these different average notes or scores by risk category to continuously produce thematic scores on the sliding window;
[0012] f) to aggregate or combine these different thematic scores to continuously produce an overall score on the sliding window, which is representative of the overall safety level of driving on the sliding window;
[0013] (g) to provide the driver, continuously and instantaneously, with a visual and / or audible indication relating at least to the overall safety level of his driving since the start of the journey, where appropriate also relating to the overall score on the sliding window or to an aggregate value of said overall score over a configurable time period, and possibly accompanied by occasional, one-off or repeated advice on improving his driving.
[0014] The invention will be better understood from the following description, which relates to a preferred embodiment, given by way of non-limiting example, and explained with reference to the accompanying schematic drawings, in which:
[0015] [Fig.1] schematically represents the different stages of the process according to the invention;
[0016] [Fig.2] schematically represents the different contributions in terms of basic metrics / scores, then in terms of thematic scores, to form the overall safety score according to the invention.
[0017] As can be schematically shown in Figures 1 and 2, the invention relates to an automatic method for evaluating the level of safety of the driving of a subject vehicle (EGO) by a driver during a journey or trip, and for providing real-time information to said driver, said method taking into account the internal and external contexts of the vehicle and the situations and circumstances encountered successively during the completion of the journey, and by exploiting signals or data provided by external and internal sensors and / or by a driver assistance system equipping the subject vehicle (for example an ADAS system).
[0018] This process includes at least the following sequence of operational steps:
[0019] a) collect in real time, and repeatedly during the time window (W), the signals or data provided by the external and internal sensors and / or the driver assistance system, which provide initial instantaneous information relating to the actual behavior and condition of the driver and the vehicle subject, each type of information being associated with a specific risk factor related to driving and forming part of a risk category;
[0020] b) to evaluate this first instantaneous information, where appropriate by comparison with second instantaneous information which corresponds either to safe driving behavior of the driver and the subject vehicle, in accordance with driving rules related to the highway code or with recommended driving rules, or to a normal state of vigilance of the driver, and to produce as many measurement or metric values (Mji);
[0021] c) to determine from each of these metrics (Mji) an instantaneous basic rating or score (MSji) through a matching process, for example with the implementation of a specific mapping for each score;
[0022] d) to continuously produce, for each risk factor and throughout the journey, an average note or score (MSjW) by a weighted average calculation, a weighted product calculation or by another function of aggregation or merging of the different instantaneous basic notes or scores (MSji) determined on the sliding window (W);
[0023] e) to aggregate or combine these different average notes or scores (MSjW) by risk category to continuously produce thematic scores (TSkW) over the sliding window (W);
[0024] f) to aggregate or combine these different thematic scores (TSkW) to continuously produce an overall score (SGW) on the sliding window (W), which is representative of the overall safety level of driving on the sliding window (W);
[0025] (g) to provide the driver, continuously and instantaneously, with a visual and / or audible indication relating at least to the overall safety level of his driving since the start of the journey, where appropriate also relating to the overall score (SGW) on the sliding window (W) or to an aggregate value of said overall score (SGW) over a configurable time period, and possibly accompanied by occasional, one-off or repeated advice on improving his driving.
[0026] Thanks to the above provisions, the method according to the invention makes it possible to provide at least one instantaneous safety score (SGW), which is calculated continuously over a sliding time window (W), and whose calculation is based on instantaneous metrics (Mji) providing instantaneous basic scores (MSji). This score (SGW) evolves continuously and intuitively according to the driver's behavior and allows them to receive real-time feedback (assessment by a human driver) on their driving behavior from the beginning of the journey. It also makes it possible to provide dynamic advice depending on the instantaneous driving situations and to draw the driver's attention to the driving errors they are making.
[0027] The method according to the invention can thus contribute to improving road safety and driver safety, and to promoting safer driving practices by informing drivers in real time of the driving errors they are making and immediately educating them on areas for improvement. The invention's ability to generate instant scores allows drivers to actively monitor their driving behavior and adjust their driving accordingly, thereby promoting safer driving.
[0028] The invention therefore represents a significant advance in the field of road safety, offering a proactive approach to assess and improve journey safety in real time.
[0029] With the aggregation or merging function or operation of step d), the aim is to aggregate different quantities into a more global quantity, synthesizing the information from these different quantities.
[0030] In accordance with a first advantageous aspect of the invention, arising from [Fig.2], the measurement or metric values (Mji) relate at least to the following criteria: legal or recommended speed, time or distance of following, longitudinal acceleration, transverse acceleration, appropriate and circumstantial use of turn signals, distraction and fatigue / drowsiness.
[0031] Regarding the distraction criterion more specifically, it can be considered in two different forms: short or instantaneous distraction and long distraction.
[0032] For momentary distraction, it is essential to consider relevant risk weighting factors such as vehicle speed. Indeed, studies have shown that the risk of an accident related to distraction increases significantly with speed. Therefore, a dependence on vehicle speed is relevant for accurately assessing this criterion.
[0033] For example, one method to obtain this criterion could involve using a cumulative time spent in distraction / vehicle speed EGO table, in which data is collected and analyzed for each iteration of the process steps on the sliding window (W). This table allows us to quantify the relationship between vehicle speed and the driver's level of distraction over time.
[0034] Another approach is to use a mathematical function to model this relationship. Logarithmic or polynomial functions can be used to accurately represent the correlation between vehicle speed and driver distraction level.
[0035] A simple example of implementing a function linking the distraction metric to the corresponding instantaneous baseline score is shown below:
[0036] = 0 if time spent in distraction^ 3s M S in S t a nt a ne(Distraction>' = = 100 if time spent distracted < ls
[0037] In the interval ls to 3s, the transition between the values 0 and 100 can be linear, affine, or quadratic.
[0038] Such a relationship of dependence of the instantaneous basic score of distraction with speed, can be extended to all other instantaneous basic scores (MSji).
[0039] Thus, and more generally, the determination of each instantaneous basic score (MSji) from a measurement or metric value (Mji) advantageously takes into account the influence of at least one physical quantity qualifying the driving and behavior of the subject vehicle, such as, for example, its instantaneous speed, and possibly the condition and environment of said vehicle.
[0040] In accordance with a second advantageous aspect of the invention, also apparent from [Fig.2], the risk categories associated with the different thematic scores (TSkW) relate at least to speed (e.g., appropriate / safe speed), safety distances, trajectory safety (e.g., appropriate / safe trajectory), compliance with vertical and horizontal signage, and vigilance.
[0041] The scores for the distraction and fatigue / drowsiness criteria, from for example the DMC, are aggregated or combined to form the thematic alertness score, as also shown in [Fig.2]. This figure also illustrates a two-phase structuring of the process of developing the overall score.
[0042] In order to take into account the problems of momentary malfunctions of sensors or detectors, or more generally of calculation of one or more basic score(s), it may be provided, when an instantaneous basic note or score (MSji) could not be determined one or more times during the sliding window (W), to limit accordingly the contribution of the average note or score (MSjW), which should have included in its calculation this or these missing instantaneous basic note(s) or score(s) (MSji), to the determination of the thematic score (TSkW) concerned and the overall score (SGW).
[0043] In accordance with similar considerations, and in relation to a practical and consistent implementation of the invention, the method may consist of suspending the determination of at least one or each of the instantaneous basic score (MSji), when at least one of the following situations is verified: subject vehicle stopped or moving at a speed below a threshold value; subject vehicle undergoing longitudinal or transverse acceleration below a threshold value; absence of a vehicle immediately preceding the subject vehicle on the latter's current trajectory; excessively high traffic density; type and / or condition of the road unsuitable for normal traffic; extreme weather conditions.
[0044] To determine an average grade or score (MSjW) over a sliding window (W), several methods can be implemented.
[0045] Thus, using for example a weighted arithmetic mean, the different instantaneous basic scores (MSji) of the sliding window (W) are multiplied by predefined weights which correspond to their relative importance, then the sum is divided by the total number of scores (MSji) to obtain the average rating or score (MSjW) considered.
[0046] Alternatively, a weighted product may be used, in which the various individual instantaneous basic scores (MSji) are multiplied together, each product being weighted by its relative importance, and then the result is raised to a power to adjust the impact of each score on the average grade or score (MSjW) under consideration.
[0047] An example of calculating the average score j on the sliding window (MSjW) is given by the formula below:
[0048] In this formula, the variables indicated have the following meanings:
[0049] W: sliding window.
[0050] : iteration (or time step) in the sliding window W for which the instantaneous basic score (MSji) is available (determinable).
[0051] M gj: instantaneous basic score] calculated at time step t. If (j fS^) is not available, then the iteration is not considered in the calculation.
[0052] MS J W : average score j for the sliding window W.
[0053] w j: weights associated with instantaneous basic scores j (positive or possibly negative weights).
[0054] According to an advantageous embodiment of the invention, which can be broken down into several variants, the method consists of implementing the following to perform the aggregation of the different average notes or scores (MSjW) to produce thematic scores (TSkW):
[0055] this involves calculating the geometric mean,
[0056] this involves calculating the arithmetic mean, for example, of the type:
[0057] X C _ WJ T' 1 O ly — .
[0058] for example, a product of powers of the type: TS^= 100n“1(MS J lv / 100) ,
[0060] or another aggregation method such as T-norms or T-connorms.
[0061] These factors are of particular interest in the context of calculating a security score, and may correspond to one of the following examples:
[0062] -T0(a, b) = if max (a, b) = 1 then min (a, b) otherwise 0 (Weber's drastic product)
[0063] -Tl (a, b) = max (0, a + b - 1) (Lukasiewicz)
[0064] -Tl,5(a, b) = ab / (2 - a - b + ab) (Einstein)
[0065] -T2(a, b) = ab (algebraic or probabilistic)
[0066] -T2.5(a, b) = ab / (a + b - ab) (Harness)
[0067] -T3(a, b) = min (a, b) (Zadeh)
[0068] Furthermore, it is also possible to implement a combination of at least two of the aforementioned aggregation methods, such as:
[0069] TS_momi(a,b) = average(average(a,b), min(a,b)) or
[0070] TS_moma(a,b) = average(average(a,b), max(a,b)),
[0071] In both of these examples, "mean" is either the arithmetic mean or the geometric mean.
[0072] Finally, a generalization of the type:
[0073] Op_l_2_3 = Opl(Op2(a,b), Op3(a,b)),
[0074] in which Opl, Op2 and Op3 designate one of the previously mentioned operators, is also applicable where appropriate.
[0075] As mentioned above, one method that can be used to aggregate mean scores (MSjW) is the geometric mean. This approach takes the nth root of the product of mean scores (MSjW), where n represents the number of instantaneous basic scores (MSji) included in the thematic score (TSkW). The geometric mean is particularly well-suited for accounting for interactions between different metrics while avoiding exaggerating extreme values.
[0076] Another approach is to use a weighted product to aggregate the mean scores (MSjW). In this method, each mean score (MSjW) is weighted by its relative importance in the context of the thematic score (TSkW), and then the weighted mean scores (MSjW) are combined to obtain the thematic score (TSkW).
[0077] Using one of the aggregation methods, the process according to the invention can provide different thematic scores (TSkW) representative of different aspects of safety on a configurable sliding window (of 2s or 3s for example), thus allowing drivers to have a view consistent with the driving actions they are performing.
[0078] In the two examples of calculating such a thematic score (TSkW) mentioned above, namely:
[0079] t i \ c_™ J T Tsk= loonjjws^ / ioo)
[0080] ou [°081 ] k
[0082] In these two formulas, the variables indicated have the following meanings:
[0083] W is the sliding window,
[0084] c W J T are the adjustable fixed weights applied to the average score (MSjW) calculated over the W window, c W J T can be within the interval [0..1],
[0085] TS est L a value of the thematic score on W,
[0086] m k (m to the power of k) is the number of average scores in the thematic score.
[0087] According to another aspect of the invention, the evaluation method may consist of aggregating the thematic scores (TSkW) to produce the overall score (SGW) using a geometric mean calculation, an arithmetic mean calculation or a product of powers.
[0088] In this latter approach, it is the product of thematic scores (TSkW) raised to a power x depending on the theme or criterion considered, which is calculated.
[0089] Using these aggregation methods, the process can provide an overall score (SGW) representative of the overall safety of the journey on a configurable sliding window (W), thus giving drivers an indication of the safety of the actions they are performing, at the level of the instrument panel or on any other screen other than the multimedia screen, on which the journey score can be displayed.
[0090] A specific example of how the overall score is calculated is shown below: M ft SG W = wo j^(rs^ / ioo) L '- £h fc-i
[0091] In this formula:
[0092] W corresponds to the set of time indices describing the sliding window,
[0093] k is the adjustable fixed weight of the thematic score (TSkW) and can, for example belong to the interval [0,1],
[0094] is the overall score,
[0095] TSw is the thematic score (TSkW) on the W window,
[0096] M is the number of thematic scores (TSkW).
[0097] In order to achieve a more pleasant and efficient reading of the score by the driver, the process may consist of smoothing the visual dynamics of overall score (SGW) indication by performing an interpolation from the previous real values of said overall score (SGW), for example by linear or polynomial interpolation.
[0098] To provide personalized and contextual guidance to improve safety during the current journey, the system may also provide specific real-time advice, either visual or audible, when at least one average score (MSjW) reaches a critical level, particularly when this occurs repeatedly across multiple, possibly successive, windows during the journey. This advice is preferably directly linked to and specific to the average score that has reached a critical level, and the advice can be displayed simultaneously on several screens or screen areas depending on the criticality of the situation (critical level of a metric or thematic score). This advice effectively complements the existing alerts on the vehicle's dashboard, enhancing overall driving safety without disturbing or distracting the driver, and can be deactivated if necessary.
[0099] In relation to an efficient and ergonomic implementation, the process may consist of displaying the overall score (SGW) on the dashboard or a screen other than a multimedia screen, in the form of a dynamic gauge which varies throughout the journey and which presents a symbolic color palette, possibly simultaneously with a numerical value for example out of 10 or out of 100, allowing its short-term variation to be displayed.
[0100] In order to avoid tiring the driver or diminishing their attention and receptiveness to alerts and advice, by excessively frequent and unselective signage, the process may consist of arbitrating between different possible advice based on the accident impact or accident risk associated with each of these pieces of advice and / or reducing, by filtering, possibly progressively, the rate of occurrence of advice based on the value of the current overall score (SGW), in particular when it is greater than at least a threshold value.
[0101] According to a practical variant of an advantageous implementation of the invention, it may be provided that a regularly repeated time window (W) can be used as a sliding window with an overlap rate between two successive windows of between 0% and 99%, advantageously between 40% and 90%, preferably between 50% and 80%.
[0102] Favorably, and in accordance with a feature of the invention, the sliding window (W) has a duration between 1.5 s and 4.5 s, preferably between 2 s and 3 s, inclusive.
[0103] In accordance with another favorable feature of the invention, steps a) to f), mentioned above, are repeated between two and twenty times during the sliding window (W), preferably between four and ten times, with iterations spaced at a constant time step.
[0104] For complete safety information for the driver, it may also be provided to continuously display, in addition to the overall safety level of driving since the start of the journey, and possibly the overall score (SGW), the thematic scores (TSkW) since the start of the journey, and possibly those produced on the sliding window (W), for example in the form of as many respective graduated gauges.
[0105] Advantageously, these two types of score displays are refreshed every 1 to 20 seconds, preferably every 2 to 5 seconds, and are done close to each other, advantageously on the same screen.
[0106] Each tip can be displayed for a configurable duration (e.g., 5 seconds). Furthermore, the interval between consecutive displays (e.g., 2 seconds) is also configurable, ensuring sufficient visibility without unduly distracting the driver. Additionally, each tip can be deactivated if necessary.
[0107] The table below provides an example of advice that can be displayed or broadcast audibly when critical situations are detected. Regulatory alert - Advice displayed Vigilance Distraction Stay focused on the road Keep your eyes on the road driving Fatigue and drowsiness: remember to take a break. Keep your eyes on the road. Microsleep and so mmeil: Fatigue alert - take a break. Sudden illness Emergency stop - resume NA nose control Appropriate speed limit (NA) Pay attention to speed limits Follow distance NA Keep your distance from the vehicle in front Safe trajectory / Cl NA Don't forget your turn signal. Safe trajectory / Ac NA Slow down before the turn
[0108] The invention also relates to a motor vehicle equipped with external and internal sensors and / or a driving assistance system, and possibly including an on-board computer and radio communication means, said system or said computer being configured for the implementation of a program carrying out the evaluation process described above.
[0109] This vehicle is therefore equipped with an on-board virtual driving safety monitor, allowing for safer and more responsible driving.
[0110] Of course, the invention is not limited to the embodiment described and shown in the accompanying drawings. Modifications remain possible, particularly with regard to the composition of the various elements or by substitution of technical equivalents, without departing from the scope of protection of the invention.
Claims
Demands
1. An automatic method for evaluating the safety level of a subject vehicle's (EGO) operation by a driver during a journey or trip, and for providing real-time information to said driver, said method taking into account the vehicle's interior and exterior contexts and the situations and circumstances encountered successively during the journey, and using signals or data provided by exterior and interior sensors and / or by a driver assistance system fitted to the subject vehicle, said method comprising at least the following sequence of operational steps, executed repeatedly over a predetermined time window (W), which continuously and smoothly explores the progressive temporal unfolding of the journey: a) collecting in real time, and several times during the time window (W), the signals or data provided by the exterior and interior sensors and / or the driver assistance system,which provide initial, instantaneous information relating to the actual behavior and condition of the driver and the vehicle in question, each type of information being associated with a specific driving-related risk factor and belonging to a risk category; b) to evaluate this first instantaneous information, where appropriate by comparison with second instantaneous information which corresponds either to safe driving behavior of the driver and the subject vehicle, in accordance with driving rules related to the highway code or with recommended driving rules, or to a normal state of vigilance of the driver, and to produce as many measurement or metric values (Mji); c) to determine from each of these metrics (Mji) an instantaneous basic rating or score (MSji) through a matching process, for example with the implementation of a specific mapping for each score; d) to continuously produce, for each risk factor and throughout the journey, an average note or score (MSjW) by a weighted average calculation, a weighted product calculation or by another function of aggregation or merging of the different instantaneous basic notes or scores (MSji) determined on the sliding window (W); e) to aggregate or combine these different average notes or scores (MSjW) by risk category to continuously produce thematic scores (TSkW) on the sliding window (W); f) to aggregate or combine these different thematic scores (TSkW) to continuously produce an overall score (SGW) on the sliding window (W), which is representative of the overall level of driving safety on the sliding window (W); (g) to provide the driver, continuously and instantaneously, with a visual and / or audible indication relating at least to the overall safety level of his driving since the start of the journey, where appropriate also relating to the overall score (SGW) on the sliding window (W) or to an aggregate value of said overall score (SGW) over a configurable time period, and possibly accompanied by occasional, one-off or repeated advice on improving his driving.
2. A method according to claim 1, characterized in that the measurement or metric values (Mji) relate at least to the following criteria: legal or recommended speed, following time or distance, longitudinal acceleration, transverse acceleration, appropriate and contextual use of turn signals, distraction and fatigue / drowsiness, and in that the risk categories associated with the different thematic scores (TSkW) relate at least to speed, safety distances, trajectory safety, compliance with vertical and horizontal signage, and vigilance.
3. A method according to claim 1 or 2, characterized in that the determination of each instantaneous basic score (MSji) from a measurement or metric value (Mji) takes into account the influence of at least one physical quantity qualifying the driving and behavior of the subject vehicle, such as, for example, its instantaneous speed, and possibly the state and environment of said vehicle.
4. A method according to any one of claims 1 to 3, characterized in that it consists, when an instantaneous basic note or score (MSji) could not be determined one or more times during the sliding window (W), in limiting accordingly the contribution of the average note or score (MSjW), which should have included in its calculation this or these instantaneous basic note(s) or score(s) (s) (MSji) missing(s), to the determination of the thematic score (TSkW) concerned and the overall score (SGW).
5. A method according to any one of the preceding claims, characterized in that it consists, in order to carry out the aggregation of the different average notes or scores (MSjW) to produce thematic scores (TSkW), of implementing a geometric mean calculation, an arithmetic mean calculation, a power product calculation or an aggregation method of the T-norm or T-conorm type, or a combination of at least two of the aforementioned aggregation methods.
6. A method according to any one of the preceding claims, characterized in that it consists of providing specific real-time advice, visual or audible, when at least one average rating or score (MSjW) reaches a level deemed critical, in particular repeatedly over several windows, possibly successive, during the journey.
7. A method according to any one of the preceding claims, characterized in that it consists of displaying the overall score (SGW) on the dashboard or a screen other than a multimedia screen, in the form of a dynamic gauge which varies throughout the journey and which presents a symbolic color palette, optionally simultaneously with a numerical value out of 10 or out of 100, allowing its short-term variation to be displayed.
8. A method according to any one of the preceding claims, characterized in that it consists of arbitrating between different possible recommendations based on the accident impact or the accident risk associated with each of these recommendations and / or reducing, by filtering, possibly progressively, the rate of occurrence of recommendations based on the current overall score (SGW) value, in particular when it is greater than at least a threshold value.
9. A method according to any one of the preceding claims, characterized in that it consists of repeating steps a) to f) between two and twenty times during the sliding window (W), preferably between four and ten times, with iterations spaced at a constant time interval.
10. Motor vehicle equipped with external and internal sensors and / or a driver assistance system, and optionally including an on-board computer and radio communication means, said system or computer being configured to implement a program carrying out the method according to any one of the preceding claims.
