Method for evaluating a driver's performance on a specific route

The proposed process for evaluating driving behavior addresses the lack of contextual relevance in existing solutions by using real-time data acquisition and processing to identify specific driving situations and provide adaptive advice, thereby enhancing driving safety and adherence to traffic rules.

FR3155193A1Pending Publication Date: 2025-05-16RENAULT SA
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
FR2023012270
Authority / Receiving Office
FR · FR
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-11-09
Publication Date
2025-05-16

AI Technical Summary

Technical Problem

Existing solutions for evaluating driving behavior lack contextual relevance, failing to provide precise and adaptive advice based on specific driving situations and environments, which can lead to accidents due to relaxed vigilance during regular journeys.

Method used

A process for evaluating driving behavior that involves simultaneous acquisition of external environment data and vehicle state data, processing to identify specific driving situations, determining rule violations, and providing contextualized advice to the driver.

Benefits of technology

This solution enables real-time contextual evaluation of driving behavior, providing relevant and adaptive advice to improve driving safety and adherence to traffic rules without requiring additional vehicle equipment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 00000000_0000_ABST
    Figure 00000000_0000_ABST
Patent Text Reader

Abstract

Method for Evaluating a Driver's Performance on a Predetermined Route. The invention relates to a method for evaluating a driver's performance on a predetermined route, comprising the following steps: simultaneous and synchronized acquisition of data on the vehicle's external environment and its condition during the journey; processing of these data sequences to identify the specific contexts and associated vehicle maneuvers, and to deduce corresponding successive driving situations and the rules applicable to these situations; determination of any rule(s) that may have been violated; association of the data and results of steps b) and c) in separate case files; and communication of said case files to the driver by display, at the end of the journey or subsequently, accompanied by appropriate advice. Figure to be published with the abstract: Fig. 1
Need to check novelty before this filing date? Find Prior Art

Description

Title of the invention: Method for evaluating the driving of a vehicle by a driver on a given route

[0001] The present invention relates to the field of road safety and in particular to autonomous equipment for assisting in improving driving, in particular in correcting risky behavior, in direct and immediate relation to the real driving experiences of a driver, in particular an experienced driver.

[0002] It has been found that drivers with several years of experience tend to neglect safe driving, or even frequently break certain driving rules, and place themselves in accident-prone situations without realizing it.

[0003] The most serious offenses are of course those directly contravening, totally or partially, one or more rules of the highway code. However, there are also less serious offenses, such as failures to follow safe or recommended driving practices depending on the circumstances encountered, which are also dangerous and much more numerous than the first mentioned. By safe or recommended driving, we mean the actions and behavior taught in driving schools when learning the highway code and in accordance with the driving advice given by driving instructors.

[0004] These infractions or breaches are particularly frequent on journeys taken regularly by a driver, during which there are often lapses in terms of vigilance and caution.

[0005] The state of the art provides numerous examples of solutions (on-board or not) for rating (“safety score”) or overall evaluation of a driver’s driving on a journey or for a given driving duration. However, these known solutions do not make it possible to provide the driver with precise and contextualized advice, in relation to a given location or specific circumstances.

[0006] The prior art concerning personalized coaching does not take into account the driving context (visibility, positions of other vehicles, etc.) which nevertheless plays a major role in the way of driving and is necessary for a good evaluation of driving. The advice given then lacks relevance. In addition, this prior art of coaching concerning violations of driving rules or infractions only uses computerized records of infractions committed and noted: it does not automatically determine the rules violated. Furthermore, it does not take into account near violations which may have an accident-prone nature, for example the adaptation of speed to the context and the driving situation.

[0007] Finally, known approaches to evaluating driving on smartphones cannot at all take into consideration the driving context, such as surrounding vehicles, and their evaluation of driving is therefore not relevant, since it lacks any contextual information.

[0008] There is therefore an unmet demand for a solution that allows, automatically and autonomously, to identify violations and near violations of driving rules in particular driving contexts in order to provide relevant and appropriate advice, in relation to a specific driver and journey, aimed at real improvement of driving. In addition, this solution should not require additional hardware resources at the vehicle level and / or be able to be easily adaptable to the equipment level of the "ego" vehicle concerned.

[0009] In order to satisfy this demand, the invention proposes a method for evaluating the driving of a vehicle by a driver on a given route, in relation to the driving rules linked to the highway code and to recommended driving rules, and this in relation to the external context and the circumstances encountered throughout the route, and with determination of successive specific driving situations, this method comprising at least the following steps:

[0010] - a) simultaneous and synchronized acquisition of data on the external environment of the vehicle and on the condition of the latter, and possibly of its driver, throughout the journey,

[0011] - b) processing these data sequences to identify the cir contexts constants and the vehicle maneuvers respectively associated, and deduce from them the corresponding successive particular driving situations and the rules applicable to these situations,

[0012] - c) determination of the rule(s) possibly not respected, this for each of the said particular driving situations encountered,

[0013] - d) association of data and results of steps b) and c) in case files distinct, each corresponding to a particular driving situation selected for which at least one rule has not been respected,

[0014] - e) communication of said case files to the driver by display, at the end of journey or subsequently, where appropriate according to a specific classification, accompanied by appropriate advice, and preferably visual information, in relation to the detailed contexts of the different specific driving situations encountered and retained, for the journey in question.

[0015] 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:

[0016] [Fig.l] schematically represents the successive essential stages a) to e) of the method according to the invention, as well as the various specific operations (numbered 1-12) forming some of these steps;

[0017] [Fig.2] schematically represents the operations allowing the identification of the rules not respected in relation to a specific driving situation;

[0018] [Fig.3] represents a flowchart illustrating the elementary operations executed in relation to a context of intersection with traffic lights and with the rule of adapting the pace to the circumstantial context, and,

[0019] [Fig.4] represents a flowchart illustrating the operations that can be carried out during the last step of the method according to one embodiment of the invention.

[0020] [Fig. 1] illustrates, by way of example, a method for evaluating the driving of a vehicle by a driver on a given route, in relation to the driving rules linked to the highway code and to recommended driving rules, and this in relation to the external context and the circumstances encountered throughout the route, and with determination of successive specific driving situations.

[0021] This method comprises, in accordance with the invention, at least the following steps:

[0022] - a) simultaneous and synchronized acquisition of data on the external environment of the vehicle and on the condition of the latter, and possibly of its driver, throughout the journey,

[0023] - b) processing these data sequences to identify the cir contexts constants and the vehicle maneuvers respectively associated, and deduce from them the corresponding successive particular driving situations and the rules applicable to these situations,

[0024] - c) determination of the rule(s) possibly not respected, this for each of the said particular driving situations encountered,

[0025] - d) association of data and results of steps b) and c) in case files distinct, each corresponding to a particular driving situation selected for which at least one rule has not been respected,

[0026] - e) communication of said case files to the driver by display, at the end of journey or subsequently, where appropriate according to a specific classification, accompanied by appropriate advice, and preferably visual information, in relation to the detailed contexts of the different specific driving situations encountered and retained, for the journey in question.

[0027] Thanks to these provisions, the invention makes it possible to respond satisfactorily to the request expressed above, in particular by achieving contextualization by the synchronized acquisition of data on the external environment, the vehicle and the driver. It is noted that this method is easily adaptable to the level of equipment of the vehicle. But of course, the more multiple and diversified the sources of acquired and processed data are, the more the parameters defining the context and the maneuvers will be complete and precise and the more refined and reliable the determination of the rules broken and the advice given will be.

[0028] In accordance with a first advantageous characteristic of the invention, taking advantage of the equipment of the vehicle in question, it is provided that, in step a), the data on the state of the vehicle are acquired by sensors or detectors specific to the latter and include at least the speed, the longitudinal and lateral accelerations, the angle of the steering wheel and the position and the movement characteristics of at least one pedal, namely the accelerator pedal and / or the brake pedal, data relating to the position and the movement characteristics of at least one organ or part of the driver's body being possibly acquired in an additional manner.

[0029] According to another beneficial characteristic of the invention, the data acquired and processed, in a mutually inter-correlated manner, in relation to each circumstantial context, comprise cartographic data, for example provided by integrated navigation software, and acquired data relating to objects present in the environment of the vehicle and possible particular circumstances linked to the context considered, advantageously acquired by means of sensors or detectors specific to the vehicle, such as for example one or more camera(s) and / or at least one lidar, or a device for communication with other vehicles or objects in the environment external to the vehicle, such as a vehicle-to-everything or V2X antenna.

[0030] According to a preferred embodiment of the invention, and as shown by way of example in [Fig.2], the aforementioned step c) comprises the following consecutive operations:

[0031] - 1) for each of said particular driving situations encountered during of the journey, identification of the successive time sequences constituting it and search for the rule(s) applicable to each of said sequences,

[0032] - 2) verification of compliance with one or more condition(s) associated with the each of the applicable rule(s), a threshold value or a range of values ​​being associated with the or each condition and forming the criterion for selective compliance with the latter,

[0033] - 3) determination of the rule(s) for which at least one of the associated conditions have not been respected, implying non-compliance with the rule(s) concerned.

[0034] In relation to operation 1), it is necessary to discriminate between exceptional situations in which violations of a generally applicable rule are tolerated. In this regard, one can, for example, mention the fact that overtaking on the right is generally prohibited, but in the particular case where the car in front turns left, it is permitted.

[0035] The thresholds mentioned in operation 3) are generally linked to limit values ​​or parameter curves. Thus, in relation to the rule "adaptation of the pace to approaching a traffic light", and in a "green light with reduced visibility" type context, the context-related conditions that can be verified through continuous monitoring (e.g. from 50 m before the light and until the complete exit from the intersection) are as follows:

[0036] - respect of the legal maximum speed

[0037] - progressive deceleration before the fire (value lower than a maximum threshold)

[0038] - preservation of a stopping distance before the fire (possibility of a proven risk).

[0039] According to a practical embodiment variant, it may be beneficially envisaged that each case file includes, in addition to the particular driving situation concerned, also the rule(s) not complied with, the condition(s) not complied with that led to the violation of the associated rule and the data relating to the circumstantial context and maneuver(s) respectively associated, as well as possibly additional information resulting from processing or evaluation of the aforementioned data, such as results of evaluation of the seriousness of the rule(s) not complied with or of the risk incurred. This association, in a single recording, of the different elements of context, situation and maneuver with the rule violated and the condition(s) not complied with makes it possible, on the basis of a single document, to provide relevant advice subsequently, asynchronously, at the end of the journey.

[0040] In order to provide the driver with the most relevant advice possible, in particular in connection with a constraint of limited availability and / or limited attention span on the part of the driver, the invention may provide that, in step e), the classification of the case files, and their order of communication to the driver, results from a prioritization operation based on one or more criteria such as the degree of seriousness of the rule(s) not respected, the number of repetitions of the infringement of the same rule during the journey, or even the recurrence of the same infringement on several distinct journeys, identical or different, in relation to similar or different circumstantial contexts.

[0041] In practice, in step e), each case is preferably communicated to the driver by display according to a prioritization classification on an interactive audio-visual interface device present in the vehicle or portable, for example in the form of a reconstructed and annotated aerial image, being accompanied by recommendations and / or advice to be applied to enable him to comply with the conditions of the rule(s) infringed in the case in question. Such a visual display of the situation makes it possible to provide relevant and understandable advice, which may possibly also be provided simultaneously, at least in part, in the form of an oral message.

[0042] The prioritization of cases can, for example, follow a hierarchy depending on all first, in decreasing order, of the seriousness of the rules broken (for example classified according to the penalties resulting from their non-compliance), then of the level of risk resulting from possible bad practices of the driver (without a rule actually being broken).

[0043] In order not to, for example, repeat the same advice several times or, on the contrary, specifically insist on a frequently recurring infraction or dangerous behavior, the determination of the recommendations and / or advice to be transmitted to the driver can be carried out, in each case, by implementing a strategy of interaction with the driver and personalization in relation to the latter, based on a list of cases established by means of classification by prioritization and taking into account a history of the recommendations and / or advice already given to the same driver.

[0044] In accordance with a possible additional alternative exploitation of the information and analysis results of steps a) to d), and as shown by the optional operations numbered 11) and 12) in [Fig.l], the method according to the invention may further consist of carrying out statistical and comparative analyses and providing a driving score for the journey considered, and displaying it on an interactive audio-visual interface device present in the vehicle or portable belonging to the driver.

[0045] Several alternatives can be envisaged for the concrete realization of the steps of the method, this depending in particular on the computer processing resources available in the vehicle (hardware and software) authorizing or not immediate processing of the data, distributed over the course of the journey and the creation of case files as and when, or else a simple recording of the data during the journey and a grouped processing of these at the end of the journey, by means specific to the vehicle (on-board computer) or by remote means with which the vehicle communicates at the end of the journey.

[0046] Thus, and in accordance with a first concrete variant of execution of the method, it is provided that at least steps a) to d) are carried out as the vehicle travels the route, by electronic means and software embedded in the latter, preferably combined with an interactive audio-visual interface device.

[0047] In accordance with a second concrete variant of execution of the method, it is provided that at least steps d) and e), and possibly also step c), are carried out at the end of the journey, either by electronic means and software on board the vehicle, preferably combined with an interactive audio-visual interface device, or by remote electronic means and software with which said vehicle can communicate.

[0048] Finally, within the framework of the two aforementioned variant embodiments, and in accordance with a advantageous implementation of the invention, it may be provided that at least the operations of step b), preferably at least the operations of steps b) and c), are carried out by means of at least one time series classification algorithm, for example in the form of at least one artificial intelligence software having been subjected to adequate prior learning, configured to process the time sequences of acquired data and deduce therefrom the corresponding successive particular driving situations and the rules applicable to these situations, and possibly verify compliance with the conditions associated with said rules.

[0049] In particular, the classifier known as "ROCKET" can be used for these operations. However, algorithms of the "decision tree" or "random forest" type, or that known as "XGBoost" can also be implemented.

[0050] Certain aspects of the invention are illustrated below in a practical manner in the form of tables and on the basis of examples illustrating a possible concrete implementation, at least partial and based on the rules of the highway code, of the invention.

[0051] Thus, tables 1 to 4 below indicate, as concrete, non-limiting and non-exhaustive examples, for different rules of the French Highway Code, the common use cases that may be encountered by a driver, the actions, behavior and maneuvers expected from said driver, in accordance with regulatory / recommended conduct, and the evaluation methods and thresholds of the conditions that must be verified for the corresponding rule to be respected. These common use cases are considered in relation to so-called normal conditions (weather / road condition).

[0052] [Tables 1] Rule Use case Assessment methods Adjusting your speed at traffic lights or intersections -Traffic lights Articles R 412 -29-30-31-32-33 The traffic lights regulating the flow of vehicles are green, yellow or red. Yellow and red traffic lights may be flashing. Green light = I pass with caution Article R 412-33 Green traffic lights authorize the passage of vehicles, subject to... Green light general case Check the rear-view mirrors Check and adjust the speed (slowing down, stabilizing, accelerating) Check intersections on the left and right Crossing at the traffic light. Or even adapting the speed in relation to the vehicle in front, with preparation Green light with reduced visibility (all cases combined) Check the rear-view mirrors Check and adjust the speed (slowing down,stabilization) Awareness and understanding of a potential risk Place your foot in front of the brake or even braking to be able to stop Green light turns right Check right interior exterior mirrors Check the blind spot Activate the right indicator Position yourself (stay in the right lane) Adapt speed Check intersections Check the pedestrian crossing Change direction Stop if pedestrian Green light turns left Check interior rearview mirror and general green light Monitoring at 50 meters in built-up areas and 100 meters outside built-up areas Mirror control Speed ​​threshold,set to authorized speed (no more than 50 km / h if limited to 50) always true Change to 50 possible if good visibility and no observable risk Intersection controls Message to anyone accelerating beyond the authorized speed until leaving the intersection Green light reduced visibility Monitoring at 50 meters in built-up areas and 100 meters outside built-up areas Mirror control Speed ​​threshold, set to authorized speed (no more than 50 km / h if limited to 50) always true Slowing Speed ​​allowing stopping in the event of a proven risk Thresholds defined according to the reduction in visibility and the uncertainty of the situation but also the weather conditions... Intersection controls Example bus = poor visibility on PP and on the right + high pedestrian risk (case on si- , left exterior Check the blind spot Activate the left indicator Move into the lane or along the road axis Check intersections Advance to the center of the intersection Give way in front (if vehicle) Check if pedestrians are involved Turn left Stop if pedestrian • If the risk is proven: brake and stop • If the risk is not proven: accelerate and cross Threshold case of hidden pedestrian risk: Take into account the pedestrian risk in the distances K 50 meters = 50 km / h At 20 meters = 40 km / h At 10 meters = 30 Km / h At 5 meters = 20 km / h Foot in front of the brake Threshold = poor visibility at intersection only Intersection control At 50 meters = 50 km / h At 20 meters = 40 km / h Green light turns right Monitoring at 50 meters in built-up areas and 100 meters outside built-up areas Mirror control Speed ​​threshold,set to authorized speed (no more than 50 km / h if limited to 50) always true Slowdown Thresholds defined according to the angle of the right turn, visibility, presence of pedestrians, weather... Left intersection checks Right blind spot check Pedestrian crossing and surroundings check on the right Look at corner exit Green light turns left Monitoring at 50 meters in built-up areas and 100 meters outside built-up areas Mirror check Speed ​​threshold, set to speed, authorized (no more than 50 km / h if limited to 50) always true Slowing Thresholds defined according to the angle of the left turn, visibility, the presence of crossing vehicles, the presence of pedestrians, the weather... Right intersection checks Left blind spot checks Pedestrian crossing and surroundings checks on the left Turn exit look

[0053]

[0054] [Tables2] Rule Use case Assessment methods Article R 412-3 Every driver must stop in front of a fixed yellow traffic light, except in the case where, when said light comes on, the driver can no longer stop his vehicle in sufficient safety conditions Orange traffic light Orange light = stop Checking mirrors Checking and adjusting speed Taking my foot off the accelerator Progressive braking Stopping before the light Exception 1: If the distance / speed ratio makes stopping impossible or dangerous Checking mirrors Checking and adjusting speed Awareness that stopping is impossible Maintaining speed or accelerating if necessary Passing before the light turns red Exception 2: If the vehicle is following me too closely,with risk of collision in case of sudden braking Rearview mirror check Speed ​​check Spotting a very close vehicle Awareness: braking and stopping dangerous in relation to the vehicle behind Decision to pass to avoid the risk of rear-end collision Maintaining speed or accelerating Orange light Stop before the light if pedestrian crossing before the pedestrian crossing if bicycle box before or before markings Rearview mirror check Exception 1 (too short to stop): If I am: 25 meters or more = stop if speed allowed in town 20 meters = stop if speed less than or equal to 40 10 meters = stop if speed less than or equal to 30 Rearview mirror checks Gradual slowing down Exception 2 (a vehicle close, threshold less than 2 seconds): Same thresholds but justified decision to pass because of the presence of a vehicle close Rearview mirror checks Acceleration to pass, Passing before the light turns red Exception due to weather or road conditions, with distances that may be different for exceptions 1 and 2...

[0055]

[0056] [Tables3] Rule Use cases Assessment methods Article R 412-3: Every driver must stop completely in front of a red traffic light, whether fixed or flashing. The stop is made: 1° When a stop line is marked, respecting the limit of this line; 2° When a stop line is not marked, respecting the limit of a line located before the pedestrian crossing if it precedes the light and, in other cases, directly above the traffic light.Red light Mirror check Checking and adapting speed Gradual slowing Stop before the light Alternative behavior (eco-driving) Mirror check (person behind me) Checking speed Adapting speed to avoid stopping Intersection checks Going through a green light Mirror Gradual slowing Stopping directly above the light or behind vehicles in front of me At 50 meters = 50 km / h At 20 meters = 30 km / h At 10 meters = 20 km / h At 5 meters = 10 km / h Stopping at a height Alternative behavior Mirrors Lifting the foot Slowing down (without being able to set a threshold) Resuming speed and accelerating possible No thresholds (eco-driving).

[0057]

[0058] [Tables4] Rule Use case Assessment methods R415-1: Any driver approaching a road intersection must check that the roadway they are about to cross is clear, drive at a speed that is all the more moderate as visibility conditions are poorer and, if necessary, announce their approach. Article R415-6 At certain intersections indicated by a so-called stop sign, all drivers must stop at the edge of the roadway they are approaching.He must then give way to vehicles traveling on the other road(s) and only enter them after ensuring that he can do so without Intersections Intersection with priority of passage Outside built-up areas (speed limited to 80 or 90 km / h) and without Mirror control Intersection control No thresholds In built-up areas (speed 50 or vehicle on the right or left: Good visibility Mirror control My visibility is good Intersection control No vehicle I am passing Note: slowing down is not necessary if the checks are early With priority of passage and with 30) Mirror control Intersection control No thresholds With crossing vehicle in and outside built-up areas) Mirror control Intersection control Slowing down (lifting foot, braking?) No threshold Note: checking that sufficient slowing is achieved to avoid the collision trajectory is complex: 1.Because I may have seen the other driver slow down 2. I cannot ask a driver to position himself in the event that each time, the other driver would not give me the right of way. 3. I can also sometimes accelerate to pass before someone arrives... Low visibility (in or out a vehicle could cross my path: Good visibility Mirror check My visibility is good Intersection control A vehicle could cross my path I take my foot off I put my foot in front of the brake I look for the driver's gaze or a clue that allows me to think that he might not stop The vehicle is going to stop I pass With right of way and low visibility Mirror check Search for information intersection built-up area) Mirror Intersection control Slowing down (taking my foot off, . danger. I must see that I do not see braking) no threshold Article 415-7. A Lift your foot Stop certain inter- Foot in front brake Rearview mirror sections indicated Speed ​​adjustment Control intersection by a signage called "give way" Stop until the passage stops, all the driver must Lift your foot At 50 meters = 50km / h give way Slow down gradually At 20 meters = 30 km / h to vehicles moving on the other Take information At 5 meters = 10 km / h or other roads and only enter after crossing on the right and at Thresholds outside built-up areas after ensuring that he can do so without a vehicle At 50 meters = 60 km / h Mirror check At 30 meters: 40 km / h danger. Information collection A 20 meters = 30 km / h Article 515-5.Lift the foot At 10 meters = 20 Km / h When two drivers approach an intersection by different roads, I pass without stopping Give way or "priority at intersection by With vehicle right" No vehicle At 5 meters = 10 km / h I pass without stopping Give way or "priority at intersection by different roads, Check my mirrors Without vehicle the driver Taking information Check mirrors coming from the left Lift the foot Checks intersection Vehicle likely to cross my Slow down according to the road profile (visibility, readability, angle taken...) the other driver Progressive slow down of the intersection and the Any driver Case 1: I stop and give way to the direction taken...) getting ready to pass No threshold and no obligation to leave a road Case 2: I arrange to pass slow down prescribed on his left must after the vehicle With vehicle keep to the left...Case 3: I accelerate to pass Mirror checks He must yield before Intersection checks Changes of direction Case 1: slow down until vehicles coming in the opposite direction leave the road Turns to stop (see threshold defined above on "priority" for stop) Mirror checks (inside Case 2: slow down without stopping. preparing to leave outside left) Blind spot Left indicator Placement at the center of the road or on a marked lane) Slowing down Intersection control Give way to the vehicle in front (if present) Turn left Turn right Leave the “priority” road Interior mirror checks,right exterior Blind spot Right turn signal Placement on the right or on marked lane) Slowing down Intersection control Turn right Watch out for pedestrians (I go after) Case 3: maintain speed or accelerate within authorized limits (I go before without obstructing) Turn left Mirror control Intersection controls Case 1: slow down to a stop (see threshold defined above) Case 2: slow down without stopping Turn right Mirror control Intersection controls Slow down without stopping (no threshold) Average distance thresholds (beware of exceptions) Outside built-up areas at 80 km / h Changes to the right or left Mirror controls (front flashing) Flashing between (100 m and 40 meters) Intersection controls (as soon as possible, depending on the layout of the area) In built-up areas at 50 km / h Change to the right or left Mirror controls front flashing Flashing between 50 meters and 25 meters See special case of roundabout intersection ,Intersection checks (as soon as possible, depending on the layout of the area) Article 414-4. Before overtaking, every driver must ensure that it is safe to do so. He may only attempt to overtake a vehicle if: He has the possibility of returning to the normal flow of traffic without obstructing it; The relative speed of the two vehicles will allow overtaking to be carried out in a sufficiently short time. He is not himself about to be overtaken. He must, in addition, warn the user he wishes to overtake of his intention.To overtake, he must move over Passing, moving out of the way To the left Check the mirrors Make sure the lane is clear That the signs allow it That the power reserve is sufficient Indicator Blind spot check Move out of his lane by moving far enough overtaking Check if overtaking is complete (check your mirrors) Turn your right indicator Move back into the right lane To the right Identify that overtaking on the right is authorized (vehicle overtaken turns left, traffic in uninterrupted lanes, direction of the two vehicles different...) Mirror checks Blind spot Right turn signal if necessary Overtaking on the right Changes of direction Right Mirrors Right blind spot Right turn signal Overtaking on the left Check if the time is right (before changing direction, overtaking, stopping) Overtaking on the right Mirror checks Blind spot check Right turn signal if necessary 1. Mirror check 2. Turn signal 3. Blind spot check 4. Change of speed 5. Overtaking 6. Blind spot 7. Reclining . enough to avoid the risk of hitting the user he wants to overtake. In any case, he must not approach it laterally to within one meter in built-up areas and one and a half meters outside built-up areas if it is an animal-drawn vehicle, a two- or three-wheeled vehicle, a pedestrian, a horse rider or an animal.Change of direction Left Rearview mirrors Left blind spot Left indicator Change of direction Roundabout Rearview mirrors Indicator depending on the lane when entering Driving on the appropriate lane Indicator to the right before exiting Stopping / parking Taking a parking space Rearview mirrors Blind spot Indicator Stopping or parking Leaving a parking space Rearview mirrors Blind spot Indicator Leaving the parking space Do not overtake on the right Situations not covered by the permitted cases I observe a vehicle in the central (or left) lane My speed is higher than theirs I am not allowed to overtake them on the right I adjust my speed to theirs and stay behind them 1. Two-lane roads with a vehicle on the left (overtaking on the left) Overtaking is done on the left. II.- As an exception to this rule, any driver must overtake on the right: 1° A vehicle whose driver has signalled that he is about to change direction to the left; 2° A vehicle travelling on a railway line using the roadway when the gap between this vehicle and the . right prohibited) edge of the road is sufficient

[0059]

[0060] In addition to Tables 1 to 4 above, Table 5 below indicates, as concrete, non-limiting and non-exhaustive examples, various specific cases which differ from the previous common use cases by increased complexity in their evaluation and / or by a significantly lower frequency of occurrence.

[0061] [Tableauxô] Specific case (description) Evaluation method Green light Being caught in a flow of vehicles approaching a green light, when my speed, more particularly the slowing down, is due to the vehicles in front of me If the approach is different from the use case, then no evaluation Is my behavior a choice or imposed? Use case? Green light Traffic blocked in the intersection and risks blocking the intersection further, then stop Stop directly above the light Wait for free passage and green light Green light Priority vehicle on the right or left Stop directly above the light Wait for free passage and green light Green light Traffic officer asking me to stop despite the green light Stop directly above the light Wait for the officer's signal Green light Green light in the shape of an arrow, then follow the direction without giving way even when turning left Failure to give way to cars Give way to pedestrians on the left Green light The light in front has a red cross indicating thetraffic light status Failure to yield to oncoming vehicles Yield to pedestrians Light off or malfunction or flashing orange in the middle I respect the intersection signs Return to intersection use cases Construction site light Rules identical to the general traffic light General traffic light use case according to color Flashing orange light at the bottom Replaces the green light but signals an intersection with specific risk and stopping in front of other vehicles (special caution) Slower than at a green light alone Orange light alone Orange light alone = danger Rearview mirror control Slowing down Flashing red light only Tram, level crossing = absolute stop Mirror check Stop before the light (so that it is visible) or at the road markings Intersection Being caught in a flow of vehicles approaching an intersection, when my speed, more particularly the slowing down, is due to the vehicles in front of me Intersection With sign speed limit lower than that generally prescribed Observing the prescribed speed Intersection Intersection without modification of the main lane (go straight ahead,follow the road) Cases of exemption for the indicator Without modification for the rest Roundabout Direction straight ahead Cases of exemption for indicator Intersection Indicator disengaged by the vehicle Cases of exemption Exit from intersection Indicator maintained after change of direction Reminder that it may be necessary to remove the indicator (see duration) Intersection Indicator on the right with risk of misinterpretation (example stop on the right after intersection, two intersections) Advice on use of the indicator Overtaking Uninterrupted traffic. The fact that the right lane is faster does not constitute an overtaking situation. Cases of exemption from the rule prohibiting overtaking on the right Overtaking on the right If the overtaken vehicle turns left on a two-way road with positioning at the center and possibility of Cases of exemption from the prohibition of overtaking Cases of exemption from the , overtake indicator Overtaking on the right If the vehicle turns left into a reserved lane on a one-way road and overtaking while remaining in the right-hand lane Exemption from the overtaking ban Exemption from the indicator Overtaking on the left Overtaking ban (road markings or sign) Reminder of the overtaking ban

[0062]

[0063] Finally, in the context of an example of practical application of the invention, the three tables 6 to 8 below illustrate three common rules of conduct to be respected, namely:

[0064] - adaptation of speed when approaching a traffic light or an intersection (straight ahead or with change of direction),

[0065] - activation of the indicators before a change of direction,

[0066] - ban on overtaking on the right (on motorways),

[0067] the acquired data sequence, the maneuver performed and the corresponding driving situation (for the first rule, also indication of certain circumstances).

[0068] [Tableauxô] Infrastructure Traffic Lights Data sequence Traffic light status / intersection signaling Maneuver Visibility Situation - Start: 50 meters before a traffic light - End: exit from the intersection - Traffic light position Green Go straight Good SI Bad S2 Go right / S3 Go left No vehicle present in front S4 Vehicle present in front S5 - Start: 50 meters before a traffic light - End: traffic light position Orange Go straight / S6 Go right / S7 Go left / S8 Red Go straight / S9 Go right / S10 Go left / SU Intersections - Start: 50 meters before the intersection entrance - End: exit from the intersection - Position of the intersection Priority road Go straight Good without vehicles present (right or left) S12 Good with vehicles present (right or left) S13 Bad S14 Go right / S15 Go left No vehicle present in front S16 Vehicle present in front S17 - Start: 50 meters before the intersection entrance - End: position of the intersection entrance Give way / No vehicles S18 Vehicles present S19 Stop Go straight / S20 Go right / S21 Go left / S22

[0069]

[0070] [Tables?] Data sequence Maneuver Situation Disengagement (Fixed obstacle) Start and end of Maneuver From the left S23 Overtaking Start and end of Maneuver From the left S24 From the right S25 Change of direction (Intersection or doubt) Start and end of Maneuver Right / Left / Roundabout S26 Stopping / parking Start and end of Maneuver Taking a parking space S27 Exiting a parking space S28

[0071]

[0072] [Tables8] Data sequence Maneuver Situation Overtaking on the right Start and end of Maneuver With lane change S25 Without lane change S29

[0073]

[0074] In addition to tables 1 to 3, 5 and 6, [Fig. 3] illustrates in the form of a flowchart, for the rule relating to the adaptation of the speed at the level of the traffic lights (here rule 1), a possible sequence of the different steps a) to c) of the method according to the invention. It can be noted that this sequence is easily transposable to other situations and rules, by appropriately adapting the operations included in the two vertical brackets.

[0075] [Fig.4] illustrates a possible sequence of steps d) and e) of the method of the invention in the continuity of the example of [Fig.3], or in the context of another situation encountered and rule to be verified.

[0076] The invention also relates to a motor vehicle comprising, in addition to an interactive audio-visual interface device or an on-board computer combined with such an interactive audio-visual interface device, also i) navigation software, ii) sensors or detectors for understanding the external environment around the vehicle, such as for example one or more camera(s) and / or at least one lidar, iii) sensors or detectors for understanding the state of the vehicle and its operating parameters, namely at least the speed, longitudinal and lateral accelerations, the steering wheel angle and the position and movement characteristics of the accelerator pedal and / or the brake pedal, iiii) possibly at least one sensor or detector providing data relating to the position and movement characteristics of at least one organ or part of the driver's body, and iiiiii) possibly a device for communication with other vehicles or objects in the environment, such as a vehicle-to-everything or V2X antenna.

[0077] In accordance in particular with the first concrete variant of execution of the method mentioned above and according to a first embodiment, this vehicle also comprises specific software means used by the on-board computer present, preferably with at least one artificial intelligence software having been subjected to adequate prior learning and associated with at least one database, these software means being configured for carrying out the steps of the method described previously, and the data provided by these sensors and / or detectors, and where appropriate by this communication device, being processed by the on-board computer of the vehicle, with recording of the results of the evaluation and contextual visualization of the latter, as well as possible advice, on an interactive display means, for example the display screen of said on-board computer.

[0078] In accordance in particular with the second concrete variant of execution of the method mentioned above and according to a second embodiment, this vehicle also comprises at least recording and transmission means, where appropriate associated with an on-board computer, which are configured to record the data provided by these sensors and / or detectors, and where appropriate by this communication device, and to transmit them to a remote processing unit, which implements specific software means, preferably with at least one artificial intelligence software having been subjected to adequate prior learning and associated with at least one database.These different means and this unit are configured to carry out the steps of the method mentioned above and the contextual visualization of the results of the evaluation, accompanied by possible advice, is carried out on the interactive audio-visual interface device, for example the screen of an on-board computer of the vehicle concerned.

[0079] Of course, those skilled in the art will understand that the present invention has been described above in relation to the French highway code, but that it can be applied in its essential characteristics with any highway code. Furthermore, 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 constitution of the various elements or by substitution of technical equivalents, without departing from the scope of protection of the invention.

Claims

Claims

1. Method for evaluating the driving of a vehicle by a driver on a given route, in relation to the driving rules linked to the highway code and to recommended driving rules, and this in relation to the external context and the circumstances encountered throughout the route, and with determination of successive specific driving situations, this method comprising at least the following steps: - a) simultaneous and synchronized acquisition of data on the vehicle's external environment and on the state of the latter, and possibly of its driver, throughout the journey, - b) processing these data sequences to identify the circumstantial contexts and the vehicle maneuvers respectively associated with them, and to deduce corresponding successive particular driving situations and the rules applicable to these situations, - c) determination of the rule(s) possibly not respected, for each of the said particular driving situations encountered, - d) association of the data and results of steps b) and c) in separate case files, each corresponding to a particular driving situation selected for which at least one rule has not been respected, - e) communication of said case files to the driver by display, at the end of the journey or subsequently, where appropriate according to a specific classification, accompanied by appropriate advice, and preferably visual information, in relation to the detailed contexts of the different specific driving situations encountered and retained, for the journey in question.

2. Method according to claim 1, characterized in that, in step a), the data on the state of the vehicle are acquired by sensors or detectors specific to the latter and comprise at least the speed, the longitudinal and lateral accelerations, the angle of the steering wheel and the position and the movement characteristics of at least one pedal, namely the accelerator pedal and / or the brake pedal, data relating to the position and the movement characteristics of at least one organ or part of the driver's body being possibly acquired in addition.

3. Method according to claim 1 or 2, characterized in that the data acquired and processed, in a mutually inter-correlated manner, in relation to each circumstantial context, comprise cartographic data, for example provided by integrated navigation software, and acquired data relating to objects present in the environment of the vehicle and possible particular circumstances linked to the context considered, advantageously acquired by means of sensors or detectors specific to the vehicle, such as for example one or more camera(s) and / or at least one lidar, or a device for communication with other vehicles or objects in the environment external to the vehicle, such as a vehicle-to-everything or V2X antenna.

4. Method according to any one of claims 1 to 3, characterized in that step c) comprises the following consecutive operations: - for each of said particular driving situations encountered during the journey, identification of the successive time sequences constituting it and search for the rule(s) applicable to each of said sequences, - verification of compliance with one or more condition(s) associated with the or each of the applicable rule(s), a threshold value or a range of values being associated with the or each condition and forming the criterion for selective compliance thereof, - determination of the rule(s) for which at least one of the associated conditions has not been complied with, implying non-compliance with the rule(s) concerned.

5. Method according to any one of claims 1 to 4, characterized in that each case file comprises, in addition to the particular driving situation concerned, also the rule(s) not respected, the condition(s) not respected which led to the infringement of the associated rule and the data relating to the circumstantial context and maneuver(s) respectively associated, as well as possibly additional information resulting from processing or evaluation of the aforementioned data, such as results of evaluation of the seriousness of the rule(s) not respected or of the risk incurred.

6. Method according to any one of claims 1 to 5, characterized in that, in step e), the classification of the case files, and their order of communication to the driver, results from a prioritization operation in depending on one or more criteria such as the degree of seriousness of the rule(s) not respected, the number of times the same rule is violated during the journey, or the recurrence of the same violation on several separate journeys, identical or different, in relation to similar or circumstantial contexts

7. no. Method according to any one of claims 1 to 6, characterized in that, in step e), each case is communicated to the driver by display according to a prioritization classification on an interactive audio-visual interface device present in the vehicle or portable, for example in the form of a reconstructed and annotated aerial image, being accompanied by recommendations and / or advice to be applied to enable him to respect the conditions of the rule(s) infringed in the case considered.

8. Method according to claim 7, characterized in that the determination of the recommendations and / or advice to be transmitted to the driver is carried out, in each case, by implementing a strategy of interaction with the driver and personalization in relation to the latter, based on a list of cases established by means of classification by prioritization and taking into account a history of the recommendations and / or advice already given to the same driver.

9. Method according to any one of claims 1 to 8, characterized in that it consists of carrying out statistical and comparative analyses and providing a driving score for the journey considered, and displaying it on an interactive audio-visual interface device present in the vehicle or portable belonging to the driver.

10. Method according to any one of claims 1 to 9, characterized in that at least steps a) to d) are carried out as the vehicle travels the route, by electronic means and software embedded in the latter, preferably combined with an interactive audio-visual interface device.

11. Method according to any one of claims 1 to 9, characterized in that at least steps d) and e), and possibly also step c), are carried out at the end of the journey, either by electronic means and software on board the vehicle, preferably combined with an interactive audio-visual interface device, or by remote electronic means and software with which said vehicle can communicate.

12. Method according to any one of claims 1 to 11, characterized in that at least the operations of step b), preferably at least the operations of steps b) and c), are carried out by means of at least one time series classification algorithm, for example in the form of at least one artificial intelligence software having been subjected to adequate prior learning, configured to process the time sequences of acquired data and deduce therefrom the corresponding successive particular driving situations and the rules applicable to these situations, and possibly verify compliance with the conditions associated with said rules.

13. Motor vehicle comprising, in addition to an on-board computer combined with an interactive audio-visual interface device, i) navigation software, ii) sensors or detectors for understanding the external environment around the vehicle, such as for example one or more cameras and / or at least one lidar, iii) sensors or detectors for understanding the state of the vehicle and its operating parameters, namely at least the speed, the longitudinal and lateral accelerations, the angle of the steering wheel and the position and movement characteristics of the accelerator pedal and / or the brake pedal, iiii) optionally at least one sensor or detector providing data relating to the position and movement characteristics of at least one organ or part of the driver's body, and mu) optionally a device for communication with other vehicles or objects in the environment, such as a vehicle-to-everything or V2X antenna,vehicle characterized in that it also comprises specific software means used by this on-board computer, preferably with at least one artificial intelligence software having been subjected to adequate prior learning and associated with at least one database, these software means being configured for carrying out the steps of the method according to any one of claims 1 to 12, and the data provided by these sensors and / or detectors, and where appropriate by this communication device, being processed by the on-board computer of the vehicle, with recording of the results of the evaluation and contextual visualization of the latter, as well as possible advice, on an interactive display means, for example the display screen of said on-board computer.,

14. Motor vehicle comprising, in addition to an interactive audiovisual interface device, i) navigation software, ii) sensors or de- detectors for understanding the external environment around the vehicle, such as for example one or more cameras and / or at least one lidar, iii) sensors or detectors for understanding the state of the vehicle and its operating parameters, namely at least the speed, the longitudinal and lateral accelerations, the angle of the steering wheel and the position and movement characteristics of the accelerator pedal and / or the brake pedal, iiii) possibly at least one sensor or detector providing data relating to the position and movement characteristics of at least one organ or part of the driver's body, and iiiiii) possibly a device for communication with other vehicles or objects in the environment, such as a vehicle-to-everything or V2X antenna, vehicle characterized in that it also comprises at least recording and transmission means, where appropriate associated with an on-board computer,which are configured to record the data provided by these sensors and / or detectors, and where appropriate by this communication device, and to transmit them to a remote processing unit, which implements specific software means, preferably with at least one artificial intelligence software having been subjected to adequate prior learning and associated with at least one database, these different means and this unit being configured for carrying out the steps of the method according to any one of claims 1 to 12 and the contextual visualization of the results of the evaluation, accompanied by possible advice, being carried out on the interactive audio-visual interface device, for example the screen of an on-board computer of the vehicle concerned.,

Citation Information

Patent Citations

  • Contextual scoring of automobile drivers

    US20170088142A1

  • Real-time driver observation and progress monitoring

    US20170132951A1

  • Detection of driving actions that mitigate risk

    US20200166897A1

  • Driving risk computing device and methods

    US20210370955A1