Method for assessing driving safety levels and providing advice to drivers

Abstract

Method for assessing driving safety and providing advice to the driver. The invention relates to an automatic method for assessing the driving safety level of a subject vehicle by a driver over a given journey or period, in relation to the driving rules linked to the highway code and / or recommended driving rules, taking into account the external context and the situations and circumstances encountered successively throughout the journey, and using the 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 operational steps of collecting the signals or data provided by the external and internal sensors and / or the driver assistance system, and extracting from these signals or data initial and secondary instantaneous information relating to safe driving behavior.to utilize this initial and secondary information by determining deviation values ​​and producing at least one driving score or rating, and to use said score to advise, reward, and / or reward the driver.

Description

Title of the invention: Method for assessing the level of driving safety and advising the driver

[0001] The present invention relates to the field of road safety, and in particular to autonomous on-board equipment for assisting in improving driving, by informing the driver about his driving during a given journey and providing advice relating to the safety aspect at least. The invention more specifically relates to an automatic method for evaluating the level of safety of the driving of a subject vehicle by a driver, and therefore in particular compliance with the driving rules set by local regulations, instantly, over a given journey or period, and advising the driver, as well as a vehicle implementing it.

[0002] Currently, the processes and systems designed to reduce accident risks aim either: - To act just before the accident, such as active emergency braking systems (AEB), - To correct the vehicle's trajectory, such as lane keeping systems, - To inform the driver, such as lane departure warning devices or traffic sign recognition.

[0003] However, these known methods and systems do not allow the driver to be informed globally, nor in a sufficiently precise and detailed manner, about the quality of his driving, for example over one or more journeys, during a given driving period or over a time interval (day, week, month, year), in an attempt to make it safer in the future, by anticipation.

[0004] These known methods and systems also do not make it possible to improve driving behavior in the long term, through a personalized approach based on the use of concrete personal driving experiences and on education, whereas the majority of accidents are linked to human factors and accident-prone driving behavior resulting from poor driving education or from a slackening over time or due to the repetition of the same journeys.

[0005] Various proposals for so-called "safety score" functions have already been proposed, developed from global data such as speed, acceleration and geolocation (GPS position) and displayed on the driver's mobile phone via applications.

[0006] However, this type of system does not allow us to offer a sufficient security score. reliably reliable and based on multiple data and factual information, coming from on-board equipment and providing detailed information on the progress of the journey or driving period, the events that occurred, the circumstances and situations encountered and the behavior of the driver and the vehicle.

[0007] Such scores also do not allow for the delivery of sufficiently relevant safety-oriented driving advice, due to the absence of sufficiently precise contextual data.

[0008] There is therefore an unmet demand for a solution that allows, automatically and autonomously, the analysis of driving behaviors (via a score) and the provision of adapted and detailed advice to improve behaviors (relevant and appropriate advice), in relation to a journey or during a specific period, in an instantaneous and repetitive or cumulative manner (and possibly unique and final). The proposed solution should also aim at the real improvement of driving and the optimization of the use of on-board sensor and / or driver assistance systems (ADAS). In addition, this solution should be able to allow a relevant evaluation, which would be both global and segmented, scalable and sensitive to the different accident-causing causes, and their associated problems.Finally, this solution should also include reward mechanisms for the driver practicing safe and preventive driving, based on an evaluation of the latter and making it possible to encourage safe and recommended driving in the long term.

[0009] To meet this need, the present invention relates to an automatic method for evaluating the level of safety of the driving of a subject vehicle by a driver over a given journey or period, in relation to the driving rules linked to the highway code and / or to recommended driving rules, and this by taking into account the external context and the situations and circumstances encountered successively throughout the journey, and by exploiting the signals or data provided by external and internal sensors and / or by a driving assistance system fitted to the subject vehicle, this method comprising at least the following operating steps:

[0010] -a) collect in real time, and repeatedly throughout the journey or period considered, the signals or data provided by the exterior and interior sensors and / or the driving assistance system,

[0011] -b) draw from these signals or data, for each of the driving situations ren countered during successive iterations, on the one hand, the first instantaneous information relating to safe driving behavior of the driver and the subject vehicle, in accordance with driving rules linked to the highway code or with recommended driving rules and, on the other hand, the second instantaneous information respectively of the same type as said first instantaneous information, but relating to the actual behavior of the driver and the subject vehicle, each type of information being associated with a specific risk factor linked to driving;

[0012] -c) exploiting this first and second information, for example in a manner crossed by determining deviation values, and produce at least one score or driving assessment;

[0013] -d) using said at least one score or said at least one note to advise, value and / or reward the driver, during and / or at the end of the journey or period.

[0014] The invention will be better understood from the following description, which relates to preferred embodiments, given as non-limiting examples, and explained with reference to the appended schematic drawings, in which:

[0015] [Fig-1] represents schematically and in the form of a flowchart, the main steps of the method according to the invention;

[0016] [Fig.2] illustrates as an example a possible architecture for the realization of the method according to the invention and also shows a human-machine interface screen on which information is displayed for the driver at the end of the journey or the driving period concerned;

[0017] [Fig.3] graphically represents a sigmoid function for the determination of the safety distance criterion within the framework of the method according to the invention;

[0018] [Fig.4] is a synoptic representation illustrating a consulting / coaching strategy which can be applied within the scope of the invention.

[0019] The invention relates to an automatic method for evaluating the level of safety of the driving of a subject vehicle by a driver over a given journey or period, in relation to the driving rules linked to the highway code and / or to recommended driving rules, and this by taking into account the external context and the situations and circumstances encountered successively throughout the journey, and by exploiting the signals or data provided by external and internal sensors and / or by a driving assistance system fitted to the subject vehicle.

[0020] According to the invention, and as shown in [Fig.l], this method comprises at least the following operating steps:

[0021] -a) collect in real time, and repeatedly throughout the journey or period considered, the signals or data provided by the exterior and interior sensors and / or the driving assistance system,

[0022] -b) draw from these signals or data, for each of the driving situations ren countered during successive iterations, on the one hand, the first instantaneous information relating to safe driving behavior of the driver and the subject vehicle, in accordance with driving rules linked to the highway code or with recommended driving rules and, on the other hand, the second instantaneous information res respectively of the same type as said first instantaneous information, but relating to the actual behavior of the driver and the subject vehicle, each type of information being associated with a specific risk factor linked to driving;

[0023] -c) exploiting this first and second information, for example in a manner crossed by determining deviation values ​​(MI), and produce at least one score or note (MSi; CMSi; TSj; CTSj; SG) for evaluating the conduct (steps c, d and e in [Fig.l]);

[0024] -d) using said at least one score or said at least one note to advise, value and / or reward the driver, during and / or at the end of the journey or period.

[0025] Thanks to these provisions, the invention makes it possible to propose a complete system which makes it possible to analyze driving behaviors (via a score) and to provide advice to improve behaviors (coach), for example by emphasizing or optimizing the use of driving assistance systems (ADAS).

[0026] The invention is based in particular on a relevant automatic evaluation of multiple parameters linked to the vehicle, the driver and the external environment (exploitation of data / information from various sources: sensors, mapping, external data). This evaluation is both global and segmented (multiple and mutually independent criteria), scalable (possible progressive addition of new criteria) and sensitive to the different accident-causing causes, and their associated problems.

[0027] Advantageously, the operating and production operations of step c) consist at least of:

[0028] -cl) comparing the first and second instantaneous information of the same type with each other, and calculating respective deviation values ​​(Mi), as quantified levels of non-compliance with the aforementioned rules, for each type of information and repeatedly throughout the journey or period considered;

[0029] -c2) determining from said deviation values ​​(Mi) a basic grade or score snapshot (MSi) in relation to a specific mapping and / or detailed reference framework associated with the risk factor considered and a regulatory or recommended driving criterion to be respected, and / or where applicable a cumulative basic score (CMSi) representative of the evolution or cumulative average of said instantaneous score since the start of the journey or period concerned, repeatedly throughout the journey or period considered.

[0030] Advantageously, the operation of producing score(s) or note(s) of step c) also consists of:

[0031] -c3) combine or aggregate the different instantaneous basic scores (MSi), and / or the different cumulative basic scores (CMSi), depending on the nature or the category of the respective corresponding specific risk factors, and deduce instantaneous thematic scores (TSj) and / or cumulative thematic scores (CTSj), at least at the end of the journey or period considered and where appropriate repeatedly throughout said journey or period.

[0032] Advantageously, the operation of producing score(s) or note(s) of step c) also consists of:

[0033] -c4) determine, after completion of the journey or expiration of the period, a score global (SG), used to advise or reward the driver at the end of the journey or period in accordance with step d).

[0034] According to a characteristic of the invention, the method consists of calculating the overall score (SG) by combining or aggregating all the thematic scores (TSj; CTSj) and presenting it to the driver, possibly also with at least one cumulative thematic score (CTSj) and / or a specific cumulative basic score (CMSi), at the end of the journey or period in accordance with step d).

[0035] According to another characteristic of the invention, the method further consists, as a function of the value of the overall score (SG), and possibly of the lowest or worst thematic score (TSj; CTSj), of a combination of the lowest or worst thematic scores (TSj; CTSj) and / or of a combination of the lowest or worst instantaneous basic scores (MSi), in developing or selecting, and then in communicating to the driver, in visual and / or audible form, for example by means of a man / machine interface, suitable advice for improving his driving and / or his use of the driving assistance system fitted to the subject vehicle.

[0036] In order to make it more relevant, it may be provided that the determination of a basic score (MSi, CMSi), instantaneous or cumulative, from a difference value (Mi) between the first and second instantaneous information of the same type, 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 of said vehicle.

[0037] In accordance with a first variant embodiment, the invention may consist of determining the instantaneous or cumulative basic score (MSi, CMSi), corresponding to a deviation value (Mi) calculated between the first and second instantaneous information of the same type, as a result of a thresholding or sigmoid function applied to this deviation value (Mi).

[0038] In accordance with a second variant embodiment, the invention may consist in determining the instantaneous or cumulative basic score (MSi, CMSi), corresponding to a deviation value (Mi) calculated between the first and second instantaneous information of the same type, by means of a function applied to this deviation value (Mi) and to the simultaneous value of at least one other physical quantity qualifying the conduct, such as speed.

[0039] In the context of this second variant, it is preferentially envisaged to determine the instantaneous or cumulative basic score (MSi, CMSi) using a bilinearly interpolated table, the value of the basic score being provided by correlation of the deviation value (Mi), as the first input coordinate, with an instantaneous speed value of the subject vehicle, as the second input coordinate, then by calculating an average.

[0040] The two aforementioned variants can also be implemented in a combined or cumulative manner.

[0041] In order to exclude the consideration of non-significant particular situations or exceptional circumstances, likely to distort the evaluation, the determination of at least one or each basic instantaneous or cumulative score (MSi, CMSi), and / or its consideration in the determination of a thematic or overall score, is suspended when at least one of the following factors or parameters is verified or exceeds an acceptable threshold: subject vehicle stopped or traveling at a speed below a threshold value; absence of vehicle immediately preceding the subject vehicle on the current trajectory of the latter; traffic density too high; type and / or condition of the road unsuitable for normal traffic; extreme weather conditions.

[0042] According to a preferred practical embodiment of the invention, emerging for example from [Fig. 2], each of the instantaneous or cumulative basic scores (MSi; CMSi) is directly linked to a respective criterion of regulatory or recommended driving, which is chosen at least from the criteria 1) of maintaining a safe distance from external targets of interest, 2) of respecting speed limits, 3) of carrying out safe trajectories, 4) of sufficient degree of vigilance of the driver, and 5) of respecting safety instructions, for example in relation to the driver, the passenger(s) or even the vehicle, with each criterion possibly being associated, linked or integrated at least one secondary criterion or sub-criterion each providing a secondary score which influences the value of the basic score concerned, for example by means of an aggregation operation, preferably weighted.

[0043] Possible variations of the different criteria and their associated criteria are discussed below.

[0044] Thus, the invention can provide a first criterion for maintaining a safety distance from external targets of interest, whether mobile or fixed, which takes into account at least the distance and / or the time before potential collision from a primary target formed by the closest vehicle located on the current trajectory of the subject vehicle.

[0045] Advantageously, this criterion of maintaining a safety distance from external targets of interest, moving or fixed, also takes into account the distance and / or time before potential collision, on the one hand, in relation to at least one other moving target close to the subject vehicle or the primary target, in front, at the rear or on the sides, and, on the other hand, in relation to at least one fixed object or part of external infrastructure potentially critical in terms of collision, taking into account the current trajectory of the subject vehicle or the maneuver undertaken by the latter.

[0046] Furthermore, the instantaneous or cumulative basic score (MSi; CMSi) in connection with the criterion of maintaining a safe distance from external targets of interest is determined by taking into account the instantaneous speed of the subject vehicle as a factor influencing the assessment of the calculated deviation value (Mi), the determination of this basic score, or its consideration in the determination of a corresponding thematic score or an overall score, being advantageously suspended when at least one of the following factors or parameters is verified or exceeds an acceptable threshold: subject vehicle stopped or traveling at a speed below a threshold value; absence of a vehicle immediately preceding the subject vehicle on the current trajectory of the latter; excessively high traffic density; type and / or condition of the road unsuitable for normal traffic; extreme weather conditions; degraded driving conditions.

[0047] In relation to a second aspect of this preferred embodiment, the invention may provide an instantaneous or cumulative basic score (MSi; CMSi) in connection with a criterion for compliance with speed limits, which is determined on the basis of deviation values ​​(Mi) calculated between, on the one hand, the measured or displayed value of the instantaneous speed of the subject vehicle, or another quantity relating to the speed of the vehicle and, on the other hand, a verified legal speed value established correlatively from data provided by the detection and interpretation of road signs encountered during the journey and data provided by the road mapping associated with the geolocation of the subject vehicle.

[0048] In relation to a third aspect, the invention may also provide an instantaneous or cumulative basic score (MSi; CMSi) in connection with a criterion for achieving safe trajectories, which is determined, for a given instantaneous speed, at least on the basis of deviation values ​​(Mi) between, on the one hand, measured values ​​of longitudinal and lateral accelerations and / or decelerations of the subject vehicle, and / or values ​​of functions derived from these accelerations and decelerations, and, on the other hand, recommended values ​​of these quantities.

[0049] In relation to a fourth aspect, the invention may also provide an instantaneous or cumulative basic score (MSi; CMSi) linked to a criterion for achieving safe trajectories, which is determined, for a given instantaneous speed, at less on the basis of deviation values ​​(Mi) between, on the one hand, measured values ​​of the yaw rate, roll rate and / or pitch rate of the subject vehicle, and / or values ​​of functions derived from these accelerations and decelerations, and, on the other hand, recommended values ​​of these quantities.

[0050] In the context of this fourth aspect, it may be provided that the determination of the instantaneous or cumulative basic score (MSi; CMSi) in connection with the criterion for achieving safe trajectories also takes into account the indication or not by the driver of a change of trajectory or of a maneuver, for example detected by analysis of the road markings, the degree of anticipation of this indication also influencing the aforementioned basic score.

[0051] In relation to a fifth aspect, the invention may also provide an instantaneous or cumulative basic score (MSi; CMSi) linked to a criterion of sufficient degree of vigilance of the driver, which is determined, in particular depending on the sensors or system(s) equipping the subject vehicle, either on the basis of a state of distraction and fatigue of the driver developed by a driver monitoring system by applying a suitable evaluation method, for example the so-called Karolinska scale, or on the basis of the results of an analysis of the movements of the steering wheel and / or the gaze of the driver, the time elapsed since the start of the journey or uninterrupted driving time being taken into account as a penalizing factor beyond a recommended maximum duration.

[0052] Within the framework of this fifth aspect, it may be provided that the determination of the basic instantaneous or cumulative score (MSi; CMSi) in connection with a criterion of sufficient degree of vigilance of the driver also takes into account compliance or non-compliance with stops, slowdowns and similar behaviors imposed or recommended by the signs, external infrastructures, the configuration of the road network or regulations, as well as the degree of anticipation of the corresponding actions of the driver, for example in terms of reducing speed or signaling maneuvers by the driver.

[0053] Similarly, in connection with this fifth aspect, it may be provided that the determination of the basic instantaneous or cumulative score (MSi; CMSi) in connection with a criterion of sufficient degree of vigilance of the driver also takes into account the state of the driver assessed by means of at least one suitable sensor equipping the vehicle or resulting from information transmitted to the vehicle before the start of the journey or of the assessment period.

[0054] The five aforementioned variants or aspects can also be implemented in a combined or cumulative manner.

[0055] According to a possible favorable characteristic, the instantaneous or cumulative basic score (MSi; CMSi) in relation to a criterion of sufficient degree of vigilance of the driver is taken into account for the determination of at least one other score or rating (MSi; CMSi; TSj; CTSj; SG) for assessing driving. It can therefore have a weighting effect on one, several or all of the other criteria.

[0056] According to a possible practical implementation of the method according to the invention, the latter may consist in providing, during the journey or period, thematic scores (TSj; CTSj), and possibly cumulative basic scores (CMSi), and in providing, at the end of the journey or period, an overall score (SG), illustrating the driving mode on said journey evaluated from a safety point of view, as a result of a function of some or all of the thematic scores (TSj; CTSj), and possibly of certain cumulative basic scores (CMSi), said function being for example chosen from: the calculation of a weighted average of the thematic scores; the selection of the thematic score corresponding to the highest risk; the calculation of the average resulting from the lowest or worst thematic score and the average of the thematic scores; the average of the products or powers of the thematic scores;a transfer or aggregation function parameterized by supervised learning.;

[0057] Preferably, the functions for calculating or determining the instantaneous or cumulative basic scores (MSi; CMSi), the thematic scores (TSj; CTSj) and / or the overall score (SG) are configurable with consideration of the nature and conditions of the journey, the driver's profile, and the type and condition of the subject vehicle, the driver's profile being able, for example, to include his age, the year in which he obtained his current license, the number of years without a reported accident, the average number of points over a given period, the number and type of offenses possibly committed over a given period and / or the occurrence or not of at least one license withdrawal.

[0058] The method is further advantageously characterized, in particular in the context of the aforementioned implementation, in that it consists of providing, during the journey or period, thematic scores (TSj;CTSj), and possibly cumulative basic scores (CMSi), and to provide, at the end of the journey or period, an overall score (SG), and in that it consists of delivering to the driver, at the end of the journey or subsequently, and depending on his overall score (SG) and where appropriate one or more thematic scores, either a congratulatory message without advice when the overall score (SG) is greater than or equal to a predetermined threshold value corresponding to safe driving, or one or preferably several advice messages in the opposite case, these advice messages being advantageously ordered, in accordance with the type of score considered, by increasing degrees of precision, namely, starting from a general or overall advice in relation to the value of the overall score (SG), towards at least one more specific advice in relation to the thematic score(s) (TSj;CTSj) the weakest or worst, and finally towards at least one particular and contextual advice in; relationship with the lowest or worst instantaneous or cumulative basic score(s) (MSi; CMSi).

[0059] In relation to a preparatory parameterization or calibration phase aimed at adapting the evaluation method to local conditions, the invention may consist, prior to the implementation of said method, in adapting the method for determining the deviation values ​​(MI), and / or the production of at least one score or note (MSi; CMSi; TSj; CTSj; SG) for evaluating the driving to the country in which the journey or the driving period to be evaluated is made and to be used for the information relating to the regulatory speed and the driving rules linked to the highway code and / or the recommended driving rules used for determining the deviation values ​​(Mi), the mapping and the local rules of said country.

[0060] The invention will be described below in more detail, but without limitation, using examples of practical implementations of the various characteristics and variants mentioned above.

[0061] The basic principle of the invention is to use the rules of the highway code and the usual rules of safe driving in order to determine a score reflecting the safety of driving throughout a journey and at the end of a journey, or during and at the end of a driving period. To do this, we rely on the data sent back by the sensors on board the vehicle.

[0062] Firstly, as shown in [Fig.l], the on-board system implementing the method according to the invention collects in real time the information coming from different sensors and / or from the driving assistance system of the vehicle, for example the legal speed information via the OSP system (step a).

[0063] This information, each associated with a type of risk linked to driving, is compared with the information from the sensors characterizing the driver's behavior, such as the vehicle speed. This comparison makes it possible to deduce deviation values ​​(MI), hereinafter referred to as "metrics" or metrics, such as the deviation from the legal speed or the deviation from the recommended speed (step b).

[0064] The value of this metric makes it possible to deduce an instantaneous “score” hereinafter called “metric-score” (or MSi-CMSi hereinafter) for each risk considered via a specific mapping (step c) and / or another score representative of the evolution of the instantaneous metric-score since the start of the journey.

[0065] The different metric scores are then combined together by risk category, hereinafter called “topic scores” (or TSj-CTSJ hereinafter), for example the risks linked to overspeeding compared to the legal speed or the recommended speed (step d).

[0066] These topic-scores are then combined to calculate a single overall score (hereinafter SG) for the current journey or period. The overall score and the metrics- scores are presented to the driver and used to define what advice to give to the driver at the end of the journey (step e).

[0067] This approach presents the advantages and benefits which are: - Open and scalable: new metric-score criteria can be gradually added - Independent: each score can be adjusted independently - Responds to differentiated problems of accident causes.

[0068] An example of implementation of this system / method illustrated in the form of architecture, is illustrated in [Fig.2].

[0069] A first criterion that can be considered is that concerning the maintenance of a safety distance between the subject vehicle (EGO) and external targets of interest, fixed or mobile.

[0070] The CIPV (for "Closest In Path Vehicle": closest vehicle in the lane) is the closest vehicle located on the trajectory of the subject vehicle. The front-side radars or the front camera make it possible to measure the distance to the CIPV and, knowing the speed of the vehicle V, makes it possible to calculate the corresponding time by T = InterDistance / V. This time must be of the order of 2 seconds according to French regulations. When there is no vehicle in front, or it is too far away to be detected by the sensor(s), the instantaneous criterion is not calculated.

[0071] The criterion can be extended to the calculation of distances to all the closest targets, for example, the target in the lane (the CIPV), the one preceding the CIPV, the closest beyond the EGO vehicle in the adjacent left, right lane, etc. Each distance from each target, or interdistance, is determined in the same way as before. The most relevant target is chosen by means of a TTC (for "Time to collision") calculation, according to one of the methods known to those skilled in the art.

[0072] We then calculate an instantaneous criterion depending for example on the speed. Indeed, the accident risk of having a short safety distance is all the higher the faster we drive, and therefore it is relevant to have a dependence on the speed of the vehicle.

[0073] This criterion can be obtained for example from an EGO interdistance / speed table, a mathematical function (log, polynomial, etc.), a decision tree, or more generally from an artificial intelligence algorithm (clustering, learning on an expert basis). Overall, this amounts to determining the value of the criterion as a function of the value of the metric and other physical quantities qualifying the driving, such as for example the speed.

[0074] A simple example of implementation of this function linking the Interdistance metric to the corresponding score metric is the threshold function:

[0075]

[0076]

[0077]

[0078]

[0079] x [ = 0 if Interdistance < 1.Os Interdistance ) = ], , 7 l - 100 if Inter distance 1.0s The thresholding nevertheless gives values ​​0 or 100 for values ​​very close to the metric like 0.99 and 1.01s for which the interpretation is essentially the same from the driver's point of view (only 0.02s of tracking time separates these 2 values). To solve this problem, it is possible to use a sigmoid function such as: distal = 7Æ7) [Fig.3] is a visual illustration of such a function. It can be noted that this calculation is independent of the EGO speed. However, it may be preferable to also have a speed dependency according to a 2-input correspondence table where the intermediate values ​​are bilinearly interpolated. For example, using the following table ("EGO Speed" means EGO vehicle speed / "Follow up time" means "follow-up time"): EGO Speed(km / h) Follow up time 5 30 40 50 60 70 80 90 100 110 120 130 140 0.2 100 -10 -50 -90 -13 0 -16 0 -20 0 -24 0 -28 0 -31 0 -35 0 -39 0 -430 0.45 10 -30 -60 -90 -12 0 -15 0 -18 0 -21 0 -24 0 -28 0 -31 0 -340 0.6 30 0 -30 -50 -80 -10 0 -13 0 -15 0 -18 0 -20 0 -23 0 -250 0.9 60 50 40 30 10 0 -10 -30 -40 -50 -60 -80 1.5 100 1.6 100 100 100 100 100 100 100 100 100 100 100 100

[0080] In order not to disturb its value due to contribution during, in particular, the stopping phase, this instantaneous metric-score may preferably not be evaluated below a low speed threshold (of the order of 30 km / h), nor in the absence of a preceding vehicle.

[0081] Then, an average is determined either by means of a 1st order filter of type l / (l+s*t), or simply by means of a summative time average. This average then corresponds to the metric-score of Safety Distance (or "Safe Distance” - see [Fig.2]).

[0082] This interdistance criterion is well suited to characterizing the safety of the distance from the preceding vehicle when the two vehicles are in a situation of established and lasting following, such as driving in a line of vehicles or following a slower vehicle without the possibility of overtaking.

[0083] In certain situations, such as: the insertion of a vehicle in front of the subject vehicle, the change of lane of the EGO vehicle, the approach by the preceding vehicle of a slower vehicle, the CIPV changes, a CIPV appears when there was none, or there are simultaneously several vehicles in relation to which the distance is a safety criterion.

[0084] It is therefore advantageous to introduce the notion of additional targets, identified for example as follows: the CIPV is noted “target” 0, target 1 is the vehicle located in front of target 0, in the same traffic lane, target 2 is the vehicle located in front of the subject vehicle in the adjacent lane, target 3 is the vehicle located in front of the subject vehicle or EGO, in the other adjacent lane.

[0085] For example, when the vehicle (target 1) located in front of the CIPV (target 0) slows down, the anticipation by the subject vehicle of the slowing down of target 1 is a safety factor. Another example: when a vehicle enters the lane of the subject vehicle, the distance from the vehicle that is entering (target 2 or 3 becoming target 0) and the distance from the vehicle located in front (target 0 which becomes target 1) are important for safety during the entire maneuver. Finally, when the subject vehicle changes lanes, the distance from the vehicle located in front of it before the lane change (target 0 becoming target 2 or 3) and the distance from the vehicle located in front of it in the destination lane (target 2 or 3 becoming target 0) are important.

[0086] In this way, we introduce into the calculation of the interdistance criterion the taking into account of several targets, either by considering the shortest distance in relation to the targets of interest during the maneuver, or by calculating distance criteria in relation to the different targets and combining them.

[0087] On the other hand, during these transient phases, the subject vehicle partly “undergoes” the inter-vehicle distance, and moreover the driver acts in such a way as to establish a new inter-vehicle distance. The inter-distance criterion, even when extended to take into account several targets, is not always well adapted or not sufficient to characterize the safety of the interaction between the vehicles in these situations. Either because it would tend to generate unjustified penalties when the cause of the temporary distance reduction is the action of another driver, or because it does not completely characterize the safety of the strategy applied by the driver to carry out a maneuver.

[0088] To overcome this, these situations are detected by suitable logic, and all or part of the interdistance calculation can be suspended for a period depending on the speed of the subject vehicle, that of the target vehicle, the traffic density, the type of road or other.

[0089] Furthermore, this criterion is then replaced or supplemented by a “time before collision” criterion with the CIPV and the other targets which represents, at any instant, the evaluation of a metric corresponding to the time it would take for the subject vehicle to collide with each of these targets if their relative speed or their relative acceleration remained constant (typically corresponding to an absence of reaction from the subject vehicle).

[0090] More generally, this collision time criterion is relevant for characterizing driving safety in situations that are much more diverse than vehicle following or interactions with other vehicles traveling in the same direction on a multi-lane road. It can in particular be applied to interactions with other road users or even with infrastructure elements such as the following: - Approaching a slower vehicle: even when the following time is sufficient, an approach that is too fast means that late and hard braking will be necessary, or a late and very dynamic lane change: lack of anticipation. - Approaching a stopped vehicle. - Approaching a bicycle or pedestrian, before overtaking. - Approaching a fixed obstacle (island, barrier, wall, etc.). - Vehicle, motorcycle, cyclist, pedestrian or other vulnerable user (scooter, etc.) crossing the path of the subject vehicle, in front of it. - Change of trajectory of the subject vehicle (e.g. intersection) causing it to cross the trajectory of a vehicle, pedestrian, cyclist, or other vulnerable user. - Circulation or encroachment of the ego vehicle on the adjacent lane intended for oncoming traffic (road with non-physically separated traffic directions), to overtake, avoid an obstacle, or other: a vehicle arriving in the oncoming direction is a potentially critical object. - Change of lane of the subject vehicle, while a vehicle is approaching from behind on the road towards which the subject vehicle is heading: there is a risk of collision with this vehicle.

[0091] The calculation of the “time before collision” criterion is therefore carried out from a “time before collision with a critical object” metric determined by analysis of a potential collision risk with the CIPV but also with any relevant vehicle target, vulnerable user, obstacle. In the case where there are several potentially critical, the criterion can be calculated relative to the most critical object (lowest TTC, for example) or by a combination of criteria relating to the different objects. In the case of an approach to the critical vehicle from behind, the sensors used to detect it are side radars, ultrasonic sensors, or rear cameras.

[0092] In certain cases other than vehicle tracking, even if the time before collision is not critical (sufficiently high) or if there is no direct risk of collision (no intersection between the planned trajectories), safe driving nevertheless requires maintaining a certain safety distance, longitudinal or lateral, with traffic participants or fixed objects. Example: lateral distance from a bicycle or pedestrian that one is overtaking, lateral distance from a central concrete reservation, etc. We therefore introduce a 3rd criterion characterizing the “distance with a critical object”, constructed from a metric “distance from a potentially critical target”. When there are several potentially critical objects, the criterion is calculated by considering the most critical or by a combination of criteria relating to an object.

[0093] As for the interdistance criterion, the evaluation of these 2 criteria “time before collision” and “distance from a critical object” from the corresponding metrics is carried out by a function depending on the value of the metric and other physical quantities qualifying the driving, such as for example the speed, and according to the category and the characteristics of the situation among those described above and any others not detailed.

[0094] Like the interdistance criterion, these two additional time criteria are not calculated under certain conditions for which they would not be relevant, in particular below a certain speed which depends on the situation.

[0095] These three instantaneous safety distance criteria (or “Safe Distance”) are then individually averaged via a low-pass filter or a time average, to constitute “metric scores” or safety distance metric scores. They are then aggregated together to constitute a single “topic score” or topical safety distance score, via an aggregation method which will be detailed later.

[0096] Another criterion that can be taken into account is the safety speed criterion (or “Safe speed - see [Fig.2]”.

[0097] The front camera can detect speed signs. It is associated with a map which provides, thanks to the GPS position, the regulatory speed at said position. Consolidated regulatory speed information Robust legal speed (VLR) is produced from both the information from the map and the detection of the signs by the camera.

[0098] The use of a panel detection camera makes it possible to take into account cases that a map cannot provide: regulated speed limit, work zone, etc.

[0099] Other information to determine specific or variable speed limits may be used (I2V communication).

[0100] Obviously, there is an obligation to respect the speed limits and the corresponding functional block of the system determines whether the speed V of the EGO vehicle is greater than the speed VLR.

[0101] When V > VLR (speeding), an instantaneous criterion is calculated depending on the difference between the vehicle speed and the legal speed, or the difference between the vehicle speed and the speed recommended by another driving assistance system, or other physical quantities qualifying the driving. For example: , „ (0 if Overspeed > 20 km / h MSspeed (Overspeed) = ], „. (100 if Overspeed s 2 km / h

[0102] As before, this instantaneous criterion value can also be given in the form of a table from which the value is deduced by bilinear interpolation or by another interpolation method between the different operating points mentioned in the table, as for the other metrics, for example as below (“EGO Speed” means EGO vehicle speed / “Overspeed” means “overspeed”): Ego speed (km / h) Overspeed 5 10 20 30 40 50 60 70 80 90 100 110 120 130 150 0 100 100 2 100 100 100 100 100 100 3 100 100 100 100 100 100 100 100 100 20 100 75 50 25 0 -2 -5 -7 -10 -10 -10 -10 -10 30 0 40 -1 -4 -11 -18 -23 -29 -35 -40 -46 -50 -50 50 -10 60 100 -20 -33 -45 -55 -65 -70 -85 -80 -10 0

[0103] For very large speed differences, this instantaneous speed criterion will be very low for example.

[0104] And minimal speed deviations, particularly on motorways, may be penalized less in terms of the score than significant deviations, particularly in urban situations where the risk of accidents is very high, which a speed-dependent table allows for precise adjustment. Other information also allowing specific or variable speed limits to be determined can be used (I2V communication).

[0105] The aforementioned criterion can possibly be extended to the notion of recommended / adapted speed with several possibilities:

[0106] 1-taking into account contextual information

[0107] 2-taking into account the speed usually practiced (for example by 85% of drivers), the type of road, the radius of curvature, the slope, etc.

[0108] 3-taking into account the recommended speed of the REM type

[0109] 4-taking into account work areas

[0110] Then, the average of the instantaneous speed score can be determined by either a 1st order filter of type l / (l+s*t) or simply a summative time average. This average then corresponds to the safe speed score (or “Safe Speed”).

[0111] Yet another criterion, which can be used within the framework of the invention, is that associated with the creation of safe trajectories (or “Safe trajectory” criterion).

[0112] This criterion of safe trajectory(ies) aims to provide a safety indicator on how to take the trajectories. It may include sub-criteria relating to changes in traffic lanes and accelerations of the EGO vehicle (see [Fig.2]).

[0113] The front camera detects the road markings, their nature (continuous, discontinuous, no markings) and detects whether a line has been crossed. Obviously, in accordance with the rules, crossing continuous lines is not permitted and crossing discontinuous lines is permitted provided the indicator is activated.

[0114] On the other hand, it is preferable to adopt cautious driving with moderate accelerations both longitudinally and laterally, to guarantee the maintenance of grip on the ground and therefore the safety of the vehicle, but also of other users.

[0115] Advantageously, an instantaneous criterion of safe trajectories depending on the level of acceleration (longitudinal and lateral), or other criteria, is first calculated.

[0116]

[0117]

[0118]

[0119]

[0120]

[0121]

[0122] qualifying the driving provided by another driving assistance system, such as the trajectory stability criterion provided by the electronic stability control system (ESC / ESP) by means of a correspondence table giving the values ​​of the criterion as a function of the vehicle speed. Indeed, the risk of accidental accidents linked to the analysis of accelerations is higher the faster you drive, and therefore it is relevant to have a dependence on the vehicle speed, which a speed-dependent table allows you to adjust precisely. Having a higher lateral acceleration in a roundabout can be normal and does not correspond to risk-taking, which a speed-dependent table allows you to take into account. An example of calculating the “longitudinal acceleration” metric score for a vehicle speed equal to 50 km / h could correspond to: MS?ra jeta ir e (Accel Longi) = { 0 if Accel Longi. > 5 m / s2 (100 if Accel Longi < 3 m / s2 Another example in table form (bilinearly interpolated for intermediate values) is given below: Longitudinal acceleration MetricS (m / s2! 0 30 40 50 60 70 80 90 100 110 120 130 140 3 100 100 100 100 100 100 100 100 100 100 100 100 100 4 100' 60 50 40 40 30 30 20 20 10 10 10 0 5 100 10 0 -10 -20 -30 -40 -50 -60 -70 -80 -80 -90 7 100 10 -10 -20 -30 -40 -50 -60 -70 -80 -90 -100 -100 9 100 -10 -30 -50' -80 -70 -90 -100 -110 -120 -130 -140 -150 An example of calculating the “lateral acceleration” metric score for a vehicle speed equal to 50 km / h could correspond to: Kl SrCaj ectoiï'e (Accel Lat) [ 0 if Aece / Lat > 5 m / s2 1.100 if Accel Lat < 3 m / s2 Again, another example in table form (bilinearly interpolated for intermediate values) is given below: Ego speed (km / h) Transverse acceleration Metric3 {m / s2i 0 30 40 50 60 70 80 90 100 110 120 130 140 3 100 100 100 100 100 100 100 100 100 100 100 100 100 Ï100 4 100 90 80 70 60 40 20 0 -30 -50 -80 -110 -150 <7 100 0 -30 -70 -100 -130 -170 -200 -230 -270 -300 -330 -370 7 100 0 -40 -70 -110 -140 -170 -210 -240 -280 -310' -350 -380 9 ■100 -10 -40 -80 -110 -150 -180 -220 -250 -290 -320 -360 -390 Then, an average is determined by either a 1st order filter of type l / (l+s*t) or simply a summative time average: sum (trajectory score secure instantaneous *Te) / time (Te = sampling period). This average then corresponds to the secure trajectory score.

[0123] It is possible to improve this score and also take into account irregular overtaking situations which correspond to driving errors, such as overtaking without indicating, for example.

[0124] The front camera is capable of giving the position relative to the road markings and of verifying that there is indeed another user in front (overtaking condition); in the case where there is overtaking detected by the position at the marking and the detected presence of a vehicle in front, and the indicator has not been activated, a penalty on the instantaneous score is applied, which will have the effect of reducing the safe trajectory score in these circumstances.

[0125] The penalty on the safe trajectory score may depend separately or in aggregate on the anticipation time of the indicator before crossing a line when the indicator has been activated on the correct side and, when the indicator has not been activated on the correct side, on the intrusion distance into the adjacent lane.

[0126] An example of calculating the “turn signal anticipation” metric score for a vehicle speed equal to 70 km / h is given by: MSt,ïïjectory (Age flashing') = 0 if Flashing Age = 0 s 100 if Age flashing < 3.5 s

[0127] Another example of determining the metric can exploit a timed duration and be based on a bilinearly interpolated table as below (“EGO Speed” means EGO vehicle speed / “Timer value” means “stopwatch value”): Ego Speed ​​Timer value Metric (s) 70 110 0 0 0 0.5 1 50 1.5 2 2.5 50 3 3.5 100 4 5 100 6

[0128] A fourth criterion that can be taken into consideration within the framework of the invention may correspond to a criterion of degree of vigilance of the driver (or “Safe Vigilance” Criterion - see [Fig.2]).

[0129] This criterion makes it possible to establish the level of risk associated with the driver's degree of vigilance.

[0130] For vehicles equipped with an interior camera, a detection of the driver's state of distraction and fatigue is developed according to the Karolinska scale by the DMS system (driver monitoring system). For vehicles that are not equipped with one, an estimate can also be made from an analysis of steering wheel movements using a technology known as DAA (driver attention alert).

[0131] In the case where the vehicle is equipped with an interior camera, the direction of gaze can also be used to determine whether the driver is attentive to possible obstacles in the vicinity.

[0132] Knowing the time elapsed T since the start of driving, the sufficient vigilance score is developed taking into account T and the instantaneous score, in order to penalize the overall score SG all the more the longer the driver drives (a break every 2 hours is recommended).

[0133] It is possible to improve this score and take into account other criteria relating to the analysis of driving, such as for example: - An anticipation criterion to reduce speed when approaching an intersection - Check compliance with STOP signs, red lights, etc. ... which are detectable by the front camera

[0134] And in the event of non-compliance with these criteria (non-limiting), calculate a score which is deducted from the sufficient vigilance score developed on the basis of driver monitoring.

[0135] In the above, when several sub-criteria or secondary criteria are used to calculate a criterion, their sub-scores will be aggregated by means of an aggregation operator of the average, product, minimum, other extended T-Norm, etc. type chosen according to the level of compensation / compromise sought between the different sub-criteria. This aggregation results in the score of the criterion.

[0136] The aggregation of the criteria scores can be done instantly, or over a journey, over a period, over a duration (one month, one year), etc.

[0137] The determination / calculation process within the framework of the method according to the invention ends with the consolidation of the final or overall score.

[0138] From each of the aforementioned individual scores, an overall SG score is calculated (step e), which reflects the driving style on the journey concerned or over the given period (from a safety point of view).

[0139] To do this, several calculation methods can be considered: - Weighted average => the weightings allow to give more weight to one or the other of the scores in order to link the weight of a theme or topic to its relative importance: - Objectively in accidentology, - Subjectively, with regard to customer experience: the score according to the invention is above all an educational tool For example, greater weight can be given to the “Safe Speed” criterion because speed represents 1 / 3 of fatal accidents on roads in Europe (Advantage = simple to implement and easy to understand) - Minimum topic scores: the overall SG score becomes the score of the topic or theme presenting the greatest risk - The average of the average of the topic scores and the minimum among the topic scores. This method of calculation has the advantage of taking into account all the topic scores in the overall SG score, but by weighting more heavily the one which represents the most risk - Any other linear or non-linear combination of metric-scores and / or topic-scores - Use of the availability of metrics to modify their contribution in the calculation of the overall SG score (For example, a calibration table giving the weight according to the availability of a metric for the calculation of weighted averages) - Products of the scores, also with possible weightings, or another aggregation function chosen according to the level of compensation / compromise sought between the topic scores (Minimum, other extended T-Norm, etc.). The aggregation function can vary according to the subgroups of topic scores to which the aggregation is applied (Advantage = a bad sub-score will have a greater impact on the overall score than an average: it will therefore be more severe) - Use of different maps per country, particularly if the accident rate is different from one country to another - Transfer function by supervised learning (accident database) - The calibrations associated with the different methods can be dependent on the driver profile (for example the weights associated with each metric-score and topic-score can be a function of the driver's age, the number years of license, number of years without an accident, etc.)

[0140] Below are examples of formulas for calculating the cumulative metric scores (or cumulative scores CMSi), the topic scores (or cumulative thematic scores CTSj) and the overall score SG.

[0141] Calculation of the cumulative metric score: f y- Metric Score j = MSi, = J 1 Card(n}

[0142]

[0143]

[0144]

[0145]

[0146]

[0147]

[0148]

[0149] With : MSji: instantaneous metric score] calculated at time i. If MSji is not available, the iteration concerned is not taken into account in the calculation. MSjT: score of metric j since the start of path T. n: number of times during the journey where the score of metric j is available. Card(n): Number of time intervals, since the start of the journey, during which a score of metric j is available. MSjT is not available if all MSji are unavailable. Calculation of the cumulative topic score: MSl x c_w? TS* = --— y™ r j: weights assigned to metric scores.

[0150] TSkT: topic score k since the start of the journey T

[0151] If MSjT is not available, the relevant metric score is not taken into account in the calculation.

[0152]

[0153]

[0154]

[0155]

[0156] TSkT is not available if all MSjT are unavailable, mk: number of metric scores of the topic score k If a metric score is available then TS = MS Calculation of the overall score SG:

[0157] With: f. k , the weight of the topic score TSk

[0158] SGT: overall score since the start of journey T

[0159] TSkT: Topic score j since the start of journey T

[0160] o: topic score number

[0161] SGT is not defined if all MSjTs are unavailable.

[0162] As indicated above for the instantaneous speed criterion, the possible values ​​of each instantaneous criterion can possibly be given by a table from from which the value of the criterion concerned is deduced by bilinear interpolation or by another method of interpolation between the different operating points mentioned in the said table.

[0163] The various data collected and scores and metrics calculated can be transmitted to an external server allowing them to be retransmitted to the driver who can consult them on his smartphone (application) or on a site dedicated to analysis. This can be statistical data but also more precise data on driving faults observed on the route so that the driver can analyze them himself and identify what he could improve.

[0164] This functionality can also be used for driving training organizations such as driving schools, accompanied driving, etc.

[0165] This functionality can also be useful for business-type fleet operators wishing to strengthen their CSR policy, subject to compliance with the GDPR.

[0166] Furthermore, the various data collected and scores and metrics calculated can also be transmitted to a third party with the driver's agreement, such as their insurer, in order to obtain an offer optimized for their use.

[0167] Alternatively, the score history can be offered to the driver, to enable him to assess his progress day after day.

[0168] Advice or “coaching” of the driver, in relation to the driver safety score, can be done on different levels linked to the process of calculating the metric scores, the topic scores and then the overall score.

[0169] In fact, the driver is advised or coached in a global manner according to his overall score via a label (very short sentence) which can congratulate him by telling him that he has very safe driving if he has a good score (overall score >= 90 for example), or inform him about his driving style according to his score. In addition to the congratulations label, different labels can be provided which correspond to different score ranges (for example four different ones).

[0170] The driver is also coached on a more precise level based on his worst topic score. This coaching is carried out using a very short introduction to the real advice. This introduction to the advice is advantageously displayed near the label to congratulate the driver by encouraging him to continue to behave well if he has a good score (overall score >= 90), or to advise him to improve his lowest topic score. There is therefore an introduction to the real advice for each topic.

[0171] The driver is also coached on an even more precise level based not only on his worst topic score, but also on his worst metric score. Indeed, the coaching is then carried out after a 2-step identification: identification of the worst topic score, then identification of the worst metric of the worst topic score (except in the case of a single metric in a topic). This type of coaching is the most detailed and in-depth, it is visible to the driver from the moment he chooses to go to the advice page to be well coached following his last journey or his last period (during the mission / finalized, at the end of the mission) or following all of his journeys made during the whole month.

[0172] To amplify the driver's awareness of the importance of this type of coaching and to avoid falling into the repetitiveness of advice, an advice logic can be put in place. This logic is based on three levels for example: - A first level of basic advice which serves to warn the driver on the worst corresponding criterion based on the highway code. - A second level of advice which serves to help the driver by directing them towards the use of driving assistance functions. - A third level of advice which allows the driver to become aware of the seriousness of having the lowest score on the corresponding criterion from an accident point of view.

[0173] This is a rotating operating logic that allows the driver to see the three levels, each level of advice is displayed alone, depending on the number of consecutive occurrences of the same worst criterion. Each time the occurrence of the worst criterion is incremented, the level of advice changes from Level 1 to Level 3.

[0174] For example, the levels of advice on the speed criterion can concretely be presented as follows:

[0175] Level 1: "Remember to leave a sufficient safety distance. Try to keep an interval of at least 2 seconds between your vehicle and the vehicle in front."

[0176] Level 2: "Assistance is available on your vehicle to help you maintain safe distances: Following distance information or, depending on your vehicle, adaptive cruise control with the following distance set by default."

[0177] Level 3: "Driving too close to other vehicles is one of the main causes of major accidents in Europe."

[0178] The driver can also go to each topic or theme (by clicking on it) to access a window defining the associated topic where he can find detailed explanations of the topic in question.

[0179] It is planned, in a later stage of the safety score, that the coaching will be increasingly adapted so as to target more directly the most serious incivilities which occur during the trip.

[0180] This coaching can depend on the profile of the scores, to define more suitable advice. Several methods are possible, such as for example the choice of advice based on a table of score values, from a decision tree testing the values ​​of each score (metric score, topic score or global score), a score clustering algorithm, a neural network having learned the advice to provide according to a learning base from experts, etc.

[0181] Advice can be given in real time or only at the end of a journey.

[0182] The invention also relates to a motor vehicle equipped with exterior and interior sensors and / or a driving assistance system, and possibly comprising an on-board computer and radio communications means, said system or said computer being configured for the implementation of a program carrying out the method described above.

[0183] Advantageously, this vehicle is equipped with wireless communication means allowing connectivity of the driving assistance system or the on-board computer with a driver's communication terminal, allowing the latter to consult the recommendations and advice again, to follow the evolution of his overall score and to learn about any rewards awarded in the event of good driving and / or to compare himself with other drivers.

[0184] Of course, the invention is not limited to the embodiments described and shown in the attached drawings. Modifications remain possible, in particular from the point of view of the constitution of the various elements or by substitution of technical equivalents, without departing from the scope of protection of the invention.

Claims

Claims

1. Automatic method for evaluating the level of safety of the driving of a subject vehicle by a driver over a given journey or period, in relation to the driving rules linked to the highway code and / or to recommended driving rules, and this by taking into account the external context and the situations and circumstances encountered successively throughout the journey, and by exploiting the signals or data provided by external and internal sensors and / or by a driving assistance system fitted to the subject vehicle, this method comprising at least the following operating steps: -a) collecting in real time, and repeatedly throughout the journey or period considered, the signals or data provided by the external and internal sensors and / or the driving assistance system, -b) deriving from these signals or data, for each of the driving situations encountered during the successive iterations, on the one hand, first instantaneous information relating to safe driving behavior of the driver and the subject vehicle, in accordance with driving rules linked to the highway code or with recommended driving rules and, on the other hand, second instantaneous information respectively of the same type as said first instantaneous information, but relating to the actual behavior of the driver and the subject vehicle, each type of information being associated with a specific risk factor linked to driving; -c) exploiting this first and second information, for example in a cross-referenced manner by determining deviation values (MI), and producing at least one score or rating (MSi; CMSi; TSj; CTSj; SG) for evaluating the driving; -d) use said at least one score or said at least one rating to advise, value and / or reward the driver, during and / or at the end of the journey or period.

2. Method according to claim 1, characterized in that the exploitation and production operations of step c) consist at least in: -cl) comparing the first and second instantaneous information of the same type with each other, and calculating respective deviation values (Mi), as quantified levels of non-compliance with the aforementioned rules, this for each type of information and repeatedly throughout the journey or period considered; -c2) determining from said deviation values (Mi) an instantaneous basic score or rating (MSi) in relation to a specific detailed mapping and / or reference system associated with the risk factor considered and a regulatory or recommended driving criterion to be respected, and / or where appropriate a cumulative basic score (CMSi) representative of the evolution or cumulative average of said instantaneous score since the start of the journey or period concerned, this repeatedly throughout the journey or period considered, and in that the operation of producing score(s) or rating(s) of step c) also consists of -c4) determining, after completion of the journey or expiration of the period, an overall score (SG), used to advise or reward the driver at the end of the journey or period in accordance with step d).

3. Method according to claim 2, characterized in that the operation of producing score(s) or note(s) of step c) also consists of: -c3) combining or aggregating together the different instantaneous basic scores (MSi), and / or the different cumulative basic scores (CMSi), depending on the nature or category of the respective corresponding specific risk factors, and deducing therefrom instantaneous (TSj) and / or cumulative (CTSj) thematic scores, at least at the end of the journey or period considered and where appropriate repeatedly throughout said journey or period.

4. Method according to claims 2 and 3, characterized in that it consists of calculating the overall score (SG) by combining or aggregating all the thematic scores (TSj; CTSj) and presenting it to the driver, possibly also with at least one cumulative thematic score (CTSj) and / or a specific cumulative basic score (CMSi), at the end of the journey or period in accordance with step d).

5. Method according to claim 4 or claims 2 and 3, characterized in that it further consists, as a function of the value of the overall score (SG), and possibly of the lowest or poor thematic score (TSj; CTSj), of a combination of the lowest or poor thematic scores (TSj; CTSj) and / or of a combination of the lowest or poor instantaneous basic scores (MSi), in developing or selecting, and then in communicating to the driver, in visual and / or audible form, for example by means of a man / machine interface, suitable advice for improving his driving and / or his use lization of the driving assistance system fitted to the subject vehicle.

6. Method according to any one of claims 1 to 5, characterized in that the determination of a basic score (MSi, CMSi), instantaneous or cumulative, from a difference value (Mi) between the first and second instantaneous information of the same type, 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 of said vehicle.

7. Method according to any one of claims 1 to 6, characterized in that it consists in determining the instantaneous or cumulative basic score (MSi, CMSi), corresponding to a deviation value (Mi) calculated between the first and second instantaneous information of the same type, as a result of a thresholding or sigmoid function applied to this deviation value (Mi) and / or by means of a function applied to this deviation value (Mi) and to the simultaneous value of at least one other physical quantity qualifying the driving, such as for example the speed, where appropriate using a bilinearly interpolated table, the value of the basic score then being provided by correlating the deviation value (Mi), as a first input coordinate, with an instantaneous speed value of the subject vehicle, as a second input coordinate, then by calculating an average.

8. Method according to any one of claims 1 to 7, characterized in that the determination of at least one or each basic instantaneous or cumulative score (MSi, CMSi), and / or its consideration in the determination of a thematic or global score, is suspended when at least one of the following factors or parameters is verified or exceeds an acceptable threshold: subject vehicle stopped or traveling at a speed below a threshold value; absence of vehicle immediately preceding the subject vehicle on the current trajectory of the latter; traffic density too high; type and / or condition of the road unsuitable for normal traffic; extreme weather conditions.

9. Method according to any one of claims 1 to 8, characterized in that each of the instantaneous or cumulative basic scores (MSi; CMSi) is directly linked to a respective criterion of regulatory or recommended driving, which is chosen at least from the criteria 1) of maintaining a safety distance from external targets of interest, 2) of respecting speed limits, 3) of rea-

10.

11. 4) sufficient degree of vigilance of the driver, and 5) compliance with safety instructions, for example in relation to the driver, the passenger(s) or the vehicle, with each criterion possibly being associated, linked or integrated with at least one secondary criterion or sub-criterion each providing a secondary score which influences the value of the basic score concerned, for example by means of an aggregation operation, preferably weighted. Method according to any one of claims 1 to 9, characterized in that it consists in providing a criterion for maintaining a safety distance from external targets of interest, mobile or fixed, which takes into account at least the distance and / or the time before potential collision with respect to a primary target formed by the closest vehicle located on the current trajectory of the subject vehicle, this criterion for maintaining a safety distance from external targets of interest, mobile or fixed, possibly also taking into account the distance and / or the time before potential collision, on the one hand, with respect to at least one other mobile target close to the subject vehicle or the primary target, in front, at the rear or on the sides, and, on the other hand, with respect to at least one fixed object or part of external infrastructure potentially critical in terms of collision,this taking into account the current trajectory of the subject vehicle or the maneuver undertaken by the latter. Method according to claim 10, characterized in that the instantaneous or cumulative basic score (MSi; CMSi) in relation to the criterion of maintaining a safe distance from external targets of interest is determined by taking into account the instantaneous speed of the subject vehicle as a factor influencing the assessment of the calculated deviation value (Mi), the determination of this basic score, or its consideration in the determination of a corresponding thematic score or an overall score, being advantageously suspended when at least one of the following factors or parameters is verified or exceeds an acceptable threshold: subject vehicle stopped or traveling at a speed below a threshold value; absence of a vehicle immediately preceding the subject vehicle on the current trajectory of the latter; traffic density too high; type and / or condition of the road unsuitable for normal traffic; extreme weather conditions;degraded driving conditions.;

12. Method according to any one of claims 1 to 11, characterized in that it consists of providing: - an instantaneous or cumulative basic score (MSi; CMSi) linked to a criterion for compliance with speed limits, which is determined on the basis of deviation values (Mi) calculated between, on the one hand, the measured or displayed value of the instantaneous speed of the subject vehicle, or another quantity relating to the speed of the vehicle and, on the other hand, a verified legal speed value established correlatively from data provided by the detection and interpretation of road signs encountered during the journey and data provided by the road mapping associated with the geolocation of the subject vehicle; - an instantaneous or cumulative basic score (MSi; CMSi) linked to a criterion for achieving safe trajectories, which is determined, for a given instantaneous speed, at least on the basis of deviation values (Mi) between, on the one hand, measured values of longitudinal and lateral accelerations and / or decelerations of the subject vehicle, and / or values of functions derived from these accelerations and decelerations, and, on the other hand, recommended values of these quantities; - an instantaneous or cumulative basic score (MSi; CMSi) linked to a criterion for achieving safe trajectories, which is determined, for a given instantaneous speed, at least on the basis of deviation values (Mi) between, on the one hand, measured values of the yaw rate, the roll rate and / or the pitch rate of the subject vehicle, and / or the values of functions derived from these accelerations and decelerations, and, on the other hand, the recommended values of these quantities; and / or - an instantaneous or cumulative basic score (MSi;CMSi) in connection with a criterion of sufficient degree of vigilance of the driver, which is determined, in particular depending on the sensors or system(s) fitted to the subject vehicle, either on the basis of a state of distraction and fatigue of the driver developed by a driver monitoring system by applying a suitable assessment method, for example the so-called Karolinska scale, or on the basis of the results of an analysis of the movements of the steering wheel and / or the driver's gaze, the time elapsed since the start of the journey or uninterrupted driving time being taken into account as a penalizing factor beyond a maximum recommended duration.;

13. Method according to claim 12, characterized in that the determination of the instantaneous or cumulative basic score (MSi; CMSi) in relation to the The criterion for achieving safe trajectories also takes into account whether or not the driver indicates a change of trajectory or a maneuver, for example detected by analysis of the road markings, the degree of anticipation of this indication also influencing the aforementioned basic score.

14. Method according to claim 12 or 13, characterized in that the determination of the basic instantaneous or cumulative score (MSi; CMSi) in relation to a criterion of sufficient degree of vigilance of the driver also takes into account compliance or non-compliance with stops, slowdowns and similar behaviors imposed or recommended by signage, external infrastructure, the configuration of the road network or regulations, as well as the degree of anticipation of the corresponding actions of the driver, for example in terms of reducing speed or signaling maneuvers by the driver.

15. Method according to any one of claims 12 to 14, characterized in that the determination of the instantaneous or cumulative basic score (MSi; CMSi) in relation to a criterion of sufficient degree of vigilance of the driver also takes into account the state of the driver assessed by means of at least one suitable sensor fitted to the vehicle or resulting from information transmitted to the vehicle before the start of the journey or the assessment period.

16. Method according to any one of claims 12 to 15, characterized in that the instantaneous or cumulative basic score (MSi; CMSi) linked to a criterion of sufficient degree of vigilance of the driver is taken into account for the determination of at least one other score or note (MSi; CMSi; TSj; CTSj; SG) for evaluating the driving.

17. Method according to any one of claims 1 to 16, characterized in that it consists in providing, during the journey or period, thematic scores (TSj; CTSj), and possibly cumulative basic scores (CMSi), and in providing, at the end of the journey or period, an overall score (SG), illustrating the driving mode on said journey evaluated from a safety point of view, as a result of a function of some or all of the thematic scores (TSj; CTSj), and possibly of certain cumulative basic scores (CMSi), said function being for example chosen from: the calculation of a weighted average of the thematic scores; the selection of the thematic score corresponding to the highest risk; the calculation of the average resulting from the lowest or worst thematic score and the average of the thematic scores matics; the average of the products or powers of the thematic scores; a transfer or aggregation function parameterized by supervised learning, in that it consists of delivering to the driver, at the end of the journey or subsequently, and depending on his overall score (SG) and where appropriate one or more thematic scores, either a congratulatory message without advice when the overall score (SG) is greater than or equal to a predetermined threshold value corresponding to safe driving, or one or preferably several advice messages in the opposite case, these advice messages being advantageously ordered, in accordance with the type of score considered, by increasing degrees of precision, namely, starting from a general or global advice in relation to the value of the overall score (SG), towards at least one more specific advice in relation to the thematic score(s) (TSj;CTSj) the lowest or worst, and finally towards at least one particular and contextual advice in relation to the lowest or worst instantaneous or cumulative basic score(s) (MSi; CMSi), and; in that it consists, prior to its implementation, in adapting the method of determining the deviation values (MI), and / or the production of at least one score or note (MSi; CMSi; TSj; CTSj; SG) of driving assessment in the country in which the journey or driving period to be assessed is made and to be used for information relating to the regulatory speed and the driving rules linked to the highway code and / or the recommended driving rules used for the determination of the deviation values (Mi), the mapping and the local rules of said country.

18. Method according to claim 17, characterized in that the functions for calculating or determining the instantaneous or cumulative basic scores (MSi; CMSi), the thematic scores (TSj; CTSj) and / or the overall score (SG) are configurable with consideration of the nature and conditions of the journey, the driver's profile, and the type and condition of the subject vehicle, the driver's profile being able, for example, to include his age, the year in which he obtained his current license, the number of years without a reported accident, the average number of points over a given period, the number and type of offenses possibly committed over a given period and / or the occurrence or not of at least one license withdrawal.

19. Motor vehicle equipped with exterior and interior sensors and / or a driving assistance system, and possibly comprising an on-board computer and radio communications means, said system or said computer being configured for the implementation of a program carrying out the method according to any one of claims 1 to 18.

20. Motor vehicle according to claim 19, characterized in that it is equipped with wireless communication means allowing connectivity of the driving assistance system or the on-board computer with a driver's communication terminal, allowing the latter to consult the recommendations and advice again, to follow the evolution of his overall score and to take note of any rewards awarded in the event of good driving and / or to compare himself with other drivers.

Images