Method for preparing road guidance instructions - Patent application

Abstract

The present invention relates to a method for preparing voice guidance instructions for an individual, formulated in the most natural way possible, comprising the steps of: - determining, by a computer, a route to be followed; - acquiring, by image acquisition means, at least one image of the individual's environment; - processing said image in order to detect at least one object in the image and to characterize the object; - preparing voice guidance instructions that provide the user with information to perform an operation in order to follow the route. According to the invention, a step is provided for determining a complexity level of an operation, and in the preparing step, voice guidance instructions are formulated using instructions inferred from the characterization of the object only if the complexity level of the operation is greater than a first threshold.

Description

[Technical Field]

[0001] The present invention relates generally to guidance systems, and particularly (but not exclusively) to guidance systems integrated into motor vehicles.

[0002] The present invention more particularly relates to a method for generating personalized voice guidance instructions, the method comprising: - determining, by a computer, the route to be used; - acquiring at least one image of the individual's environment; - processing the images with a view to detecting in the images at least one object present in the environment and characterizing said object; - determining and transmitting voice guidance instructions informing the individual how to perform maneuvers to follow said route; The present invention relates to a method comprising:

[0003] The invention also relates to a device for generating voice guidance instructions.

[0004] The present invention is preferably applied where the individual is a driver of a motor vehicle. [Background technology]

[0005] Satellite navigation and positioning systems are a particularly effective and well-understood way of determining routes to be taken, especially by drivers of vehicles.

[0006] The advantages over paper maps are considerable, especially since the personal position is updated in real time. Most navigation systems display directions to follow on a map that corresponds to a rough representation of the vehicle's environment.

[0007] Generally, such navigation systems are designed to transmit route guidance instructions to the driver in two different ways: visually via a display screen, and audibly via a speaker provided in the vehicle.

[0008] The visual representation of the maneuver to be performed on the display screen is not always clear because the map distorts the distances and angles between highways, making it difficult for the driver to correctly understand the maneuver to be performed.

[0009] The navigation system's instructions, due to their visual representation and associated audio instructions, are not always consistent, leading to misunderstandings on the part of the driver regarding the operation to be performed, as well as the driver's understanding of the road infrastructure, the configuration of the road infrastructure, and the driving environment.

[0010] It is particularly noted that currently, the voice guidance instructions are generated systematically, like an itinerary, depending on the distance between the vehicle and the next maneuver to be performed.

[0011] Therefore, these instructions are often not well suited to the environment, are repetitive and predictable, and do not meet the driver's expectations.

[0012] Another drawback of known navigation systems is that the instructions are generated piecemeal, meaning that the instructions are unrelated to one another, and the content of each instruction does not take into account the information conveyed in the previous instruction and the information to be conveyed in the upcoming instruction, which depends on the operation that the driver must perform.

[0013] In practice, these drawbacks cause the majority of users to deactivate the voice function of their navigation systems. Summary of the Invention

[0014] In order to remedy the above-mentioned drawbacks of the prior art, the present invention provides a method for generating voice guidance instructions as defined at the beginning, in which provision is made for a step of determining the complexity level of the operation, and in a determination step, the voice guidance instructions are formulated using cues derived from the characterization of the object only if the complexity level of the operation is higher than a first threshold.

[0015] In other words, the invention proposes to generate voice guidance instructions whose sophistication depends on the complexity of the situation in question.

[0016] For example, if a driver is faced with a simple situation such as a crossroad and there are no objects obstructing the driver's view of the crossroad, voice guidance instructions are generated without adding any specific details to the voice guidance instructions.

[0017] In contrast, in more complex situations, for example because an object (such as a truck) obscures part of the intersection or because the intersection itself is more complex, audio guidance instructions are generated along with additional cues that allow the driver to determine where they are. These may be visual cues (e.g., trucks, signs, road infrastructure) or other types of cues, such as contextual cues (e.g., to tell the driver when they need to turn).

[0018] The invention thus proposes to generate voice guidance instructions in a manner as similar as possible to the way a passenger in the passenger seat would do so, thus increasing the ease of understanding of these instructions and adapting their frequency, which makes it possible to reduce the cognitive load on the driver and reduces driver errors and hesitation.

[0019] The following are other advantageous, non-limiting features of the method according to the invention, which features can be implemented individually or in any technically possible combination: - the cue is a visual cue that is either currently visible to the individual or will soon be visible in their vicinity, or a contextual cue that provides the individual with information about when the operation should be performed; - the voice prompts are structured such that at least one action and the cue are inserted into the voice prompts; the complexity level is calculated depending on at least one of the following criteria: the position of the vehicle relative to the zone in which said operation must be performed, the type of road infrastructure present in the zone in which said operation must be performed, the tension felt by the individual, the density of road traffic, the number of objects detected in at least one part of the acquired image, the individual's knowledge of the road infrastructure, the stage in which the vehicle is currently located; - if the road infrastructure is a roundabout, the complexity level is calculated depending on at least one of the following criteria: the angle between the axis of arrival of persons at the roundabout and the axis of exit of persons from the roundabout, the angle between at least two roads leading to the roundabout, the number of roads leading to the roundabout, the difference between the number of entrances to the roundabout and the number of exits from the roundabout, the number of lanes present in the roundabout, and the outer diameter of the roundabout; - if the cue is a visual cue visible to the individual, and the visual cue disappears from the environment visible to the individual between the moment the voice guidance instruction is vocally transmitted and the moment the operation has to be performed, provision is made for a step of generating new voice guidance instructions that do not use the visual cue; - if the cue is a visual cue visible to the individual, and if, between the moment the voice guidance instruction is transmitted vocally and the moment the operation has to be performed, another visual cue appears in the environment visible to the individual, which is indistinguishable from the visual cue used in the voice guidance instruction, provision is made for a step of generating a new voice guidance instruction different from the voice guidance instruction transmitted vocally, - if the complexity level of the operation is higher than a second threshold that is strictly higher than the first threshold, the voice guidance instructions are generated using two cues; - provision is made for determining the cost, related to the additional time or distance, that an error in following said route would cause, and a complexity level is determined depending on said cost; - provision is made to distinguish between at least two separate and consecutive stages preceding the performance of said operation, namely an operation anticipation stage, in which voice guidance instructions may be sent to the individual to indicate how they should position themselves to perform the operation, and an explanation stage, in which voice guidance instructions are sent to the individual to indicate how they should perform the operation, and the voice guidance instructions sent in these stages use distinct cues; - during the anticipation phase, provision is made for verbalizing voice guidance instructions if the road to be used or to be used contains several lanes; - citing signs only if cues used during the anticipatory phase in audio instructions are located outside privately built-up areas; - provision is further made for at least one other stage following the explanation stage, in which voice guidance instructions can be generated, the other stage being a simultaneous accompanying stage to said operation or a subsequent stage to said operation, a reassurance stage allowing the individual to be indicated whether he has performed the operation correctly, or a continuation stage allowing the individual to be informed of the active character of said generation method, - if the voice guidance instructions transmitted during the explanation phase do not refer to any visual indicator and the vehicle slows down, provision is made in the accompanying phase for transmitting voice guidance instructions using cues; - during the reassurance phase, provision is made for transmitting voice guidance instructions using visual indicators that are distinct from one or more visual cues used in the one or more voice guidance instructions transmitted during the preceding phase; During the continuity phase, provision is made to generate different voice guidance instructions depending on whether the radius of curvature of the road used is below or above a threshold value.

[0020] According to one aspect of the invention, the step of determining, by a computer, a route to be taken comprises retrieving from a map database static visual indicators located on the route and attributes associated with the static visual indicators, and determining attributes common to a plurality of static visual indicators. When objects are characterized in the image processing step as having common attributes, audio guidance instructions are formulated in the determination step using cues having the common attributes.

[0021] According to one aspect of the invention, the static visual indicators are road signs, the common attributes comprise a name written on the road sign or a background color of the road sign, and the objects are road signs.

[0022] According to one aspect of the invention, the step of processing the image comprises detecting a plurality of objects, and the generation method further comprises a step of classifying the detected objects, whereby objects characterized as having common attributes are prioritized, and the object selected for the determination step is the prioritized object.

[0023] The present invention relates to a device for generating environmentally tailored route guidance instructions, comprising: - means for acquiring an image of the environment; - Map database and - navigation and geolocation systems; a processing module suitable for receiving and combining elements from the acquisition means, from the map database and from the navigation and geolocation system, and programmed to implement a method as described above; - a voice transmitter suitable for transmitting instructions to individuals; Also provided is a device comprising:

[0024] Of course, the various features, variations and embodiments of the invention may be associated with one another in various combinations, provided that they are not compatible or mutually exclusive with one another.

[0025] The description that follows, with reference to the accompanying drawings, given as non-limiting examples, will make it easier to understand what the invention consists of and how it can be implemented. [Brief explanation of the drawings]

[0026] [Figure 1] 1 is a schematic view of the front part of a motor vehicle equipped with a device for generating route guidance instructions according to the invention; DETAILED DESCRIPTION OF THE INVENTION

[0027] Preferably, the present invention can be used in any motor vehicle (car, truck, motorcycle, etc.).

[0028] In FIG. 1, the cabin of a motor vehicle 1 is shown diagrammatically, equipped with a device for generating route guidance instructions, allowing the invention to be implemented.

[0029] The device comprises means for acquiring images of the vehicle's environment, a guidance unit 5 and an interface 11 suitable for transmitting guidance instructions to the driver of the motor vehicle 1 .

[0030] The image acquisition means comprise, for example, here a camera 10 oriented towards the front of the vehicle and making it possible to acquire images of the environment seen by the driver when he turns his gaze towards the road.

[0031] The interface 11 comprises at least one system for transmitting an audio signal 12. In addition, the interface 11 here comprises a touchscreen 14. As a variant, the interface 11 may be screen-less.

[0032] The audio signal transmission system 12 is designed to transmit audio instructions to an individual. The audio signal transmission system 12 is placed in the vehicle cabin at a distance from the individual such that the individual can clearly hear the communicated audio instructions. For example, it is a matter of a loudspeaker installed in the vehicle for the purpose of allowing, for example, the radio to be listened to, or indeed of a loudspeaker of a portable system such as a telephone or an external navigation device.

[0033] The guidance unit 5 comprises a map database 20, a navigation and geolocation system 22, a computer memory (hereinafter simply referred to as "memory 24"), and a computer (hereinafter simply referred to as "processing module 25").

[0034] The map database 20 comprises two-dimensional and even three-dimensional map data and in particular the location (longitude and latitude) of highways in a given space (for example in Europe). The map database 20 also contains additional information such as the shape of buildings, the location of service stations and of particular points of the environment (for example historical monuments or rivers), the type of road surface (pavement, dirt, tarmac, etc.). The map data can be read by the processing module 25.

[0035] Advantageously, the map database 20 comprises static visual indicators such as road signs and attributes associated with each road sign, such as the background color of the sign and the direction or name written on the sign.

[0036] The navigation and geolocation system 22 is capable of determining the current spatial position of the motor vehicle 1 and determining a route between the current position (or starting position) of the motor vehicle 1 and a desired destination based on data stored in the map database 20. The navigation and geolocation system 22 may be, for example, a Global Positioning System (GPS). The current position and route of the vehicle can be read by the processing module 25.

[0037] The route determined by the navigation and geolocation system 22 is stored in memory 24. Memory 24 also stores parameters, such as thresholds, used in the implementation of the present invention. All elements stored in memory 24 can be read by processing module 25. These elements are described in detail further on in this specification.

[0038] The processing module 25 is programmed to receive and combine various elements from the camera 10, from the map database 20, from the navigation and geolocation system 22, and from the memory 24. The processing module 25 is also programmed to transmit data to the speaker 12 and to the screen 14 in order to communicate guidance instructions to the individual. In practice, this processing module 25 is formed by a microcontroller.

[0039] The device for generating guidance instructions described above allows the method described below to be implemented.

[0040] Step S1: Determining the route The process begins when the motor vehicle 1 is started. The processing module 25 then determines, in a first step, the route that the motor vehicle 1 must take.

[0041] To do this, the current (start) location is determined by the navigation and geolocation system 22. As a variation, the individual may specify a start location different from where they are currently located by manually entering the target start location. For example, the individual may enter the target start location via the screen 14.

[0042] The individual then specifies a desired destination. For example, the desired destination may be entered via screen 14.

[0043] The processing module 25 combines the starting location, the desired destination, and the map data corresponding to the geographic zone of interest. The processing module 25 transmits this information to the navigation and geolocation system 22, which then generates a route.

[0044] The determined route is stored in the memory 24.

[0045] Advantageously, processing module 25 retrieves information on road signs located on the route, and attributes associated with the signs, from map database 20. As will be explained more fully below, processing module 25 is capable of determining whether multiple signs have common attributes in order to identify the common attributes and the signs that comprise the common attributes.

[0046] A route consists of a stretch of road, which is defined, for example, as a section of a highway stretching between two consecutive intersections.

[0047] Each stretch of road may then be characterized by several parameters, such as the number of lanes it has, the direction or directions in which driving is permitted, the type of intersections, and the number of oncoming and offgoing lanes at each intersection.

[0048] The route is, for example, - Road infrastructure (interchanges, roundabouts, crossroads, etc.), - Borough or region, - Important places (pharmacy, bakery, mall, fountain, bus stop), - famous places (museums, monuments, cathedrals, stations, airports, etc.), It passes through many intermediate points, such as:

[0049] At this stage, before the remainder of the method is described, other concepts useful to the understanding of this specification can be explained.

[0050] The first concept is that of an "action," which is a matter of a unit movement to be performed by a vehicle (e.g., "get into the right lane," "follow the red car").

[0051] The concept of "target" represents a zone towards which a motor vehicle must move. Targets can be expressed in terms of elements of the road infrastructure (traffic lights, stop signs, directional signs, highways, etc.).

[0052] The concept of a "maneuver" represents a sequence of one or more actions that allows a "goal" to be realized. Maneuvers are performed to change the position of the motor vehicle relative to the road infrastructure and to achieve the goal. For a given maneuver, multiple pairs of an action and a goal may be linked. For example, a maneuver may be expressed as follows: "Get into the left lane, then turn left onto highway X."

[0053] The concept of "cue" covers any linguistic element that can be used in voice guidance instructions and that is defined based on images acquired by a camera to facilitate understanding of the situation by the vehicle driver when the cue is used.

[0054] The concept of "visual cues" (or "visual indicators") represents static elements (highways, signs, road infrastructure, and more generally, any elements contained in the map database 20) or dynamic elements (vehicles, pedestrians, or more generally, any element detectable by the processing module 25 in the images acquired by the camera 10) that can be used to describe an action to be performed or when an action should be performed. These elements allow the guidance instructions to be enriched either by adding details to the guidance instructions (e.g., "turn right after the traffic light") or by allowing the guidance instructions to be given with reference to this element (e.g., "go past the blue building").

[0055] The concept of "context cue" (or "time cue") describes when a maneuver must be performed. This information indicates the time at which the driver must perform the maneuver. For example, such cues are sometimes expressed as: "now", "immediately", "as soon as possible", "at the roundabout", "at the traffic light".

[0056] The concept of "attribute" refers to a visual cue or target, a characteristic of this visual cue or target that allows the target to be differentiated from other indicators, which can be a matter of color, text, direction, highway type, lane width, road surface type, etc.

[0057] Two categories of attributes, viz. egocentric attributes, in other words attributes related to the position of the motor vehicle 1 (location of the target relative to the position of the vehicle, the position of the various elements surrounding the vehicle, the number of the exit from the roundabout considering the position of the vehicle, its position in the lane, etc.); - allocentric attributes, in other words, absolutely positioned attributes (the location of the visual cue relative to one or more fixed indicators of the environment and therefore not varying with the vehicle's position, e.g., the lane number on a multi-lane highway); is possible.

[0058] Targets are generally egocentric attributes, whereas visual cues are generally allocentric attributes.

[0059] The concept of "familiarity criteria" refers to linguistic elements that allow the human aspect of the formulation of guidance instructions to be strengthened. For example, it can be a matter of formulating "Just stay in your current lane for now", "First, turn towards...", "Then, turn left", etc. This concept also refers to linguistic elements that have a safety function (e.g., "Watch the turn!").

[0060] The concept "geographical indicator" is used to describe the location of a visual cue relative to a target, which is sometimes described as "at the end of," "just ahead of," "between," "fully to the right," "slightly to the right," "opposite," "straight," "in the direction of," etc.

[0061] Step S2: Acquiring the video When the vehicle starts, the camera 10 captures images of the vehicle's environment and transmits the images to the processing module 25. All or only some of the images captured by the camera 10 may be processed by the processing module 25 at frequencies higher than 1 Hertz.

[0062] The processing module 25 is programmed to process these images to generate voice guidance instructions that communicate to the driver, in particular, which actions the driver will soon be required to perform. The processing module 25 is programmed to generate at least a portion of these instructions that take into account the environment visible to the driver.

[0063] As the vehicle follows the route defined by the driver, the processing module 25 tracks the vehicle's progress through the navigation and geolocation system 22. This allows the processing module 25 to predict before each intersection whether the vehicle will have to perform a maneuver, and if so, which maneuver the vehicle will have to perform.

[0064] Each maneuver that the vehicle must perform is then broken down by the processing module 25 into five successive stages, in each of which voice guidance instructions are generated and then transmitted, if necessary.

[0065] The first stage is the anticipation stage, which allows the driver to be told how to position his vehicle on the road to tackle the next event (intersection, roundabout, etc.) in the best possible way.

[0066] The second stage is an explanation stage that allows the driver to be informed of the operation he must perform so that he can understand the operation before he has to perform it.

[0067] The third stage is an accompanying stage dedicated to assisting the driver during maneuvering to increase his confidence.

[0068] The fourth stage is a reassurance stage that allows the driver to be given confirmation that he or she has performed the correct maneuver.

[0069] The fifth stage is a reassurance stage, which occurs when two operations are separated from each other by a distance or time greater than a predefined threshold, and allows the driver to be indicated that the processing module 25 is still active (after a prolonged absence of guidance instructions).

[0070] Guidance instructions are thus generated, in some cases, during each of these five stages.

[0071] These guidance instructions are generally generated sequentially in a number of successive steps that consist of detecting visible static and moving objects in the images, identifying roads and routes in each captured image, locating the detected objects on a map obtained from map database 20, relating the detected static objects to known static objects in map database 20, and applying statistical and / or sequential reasoning to the objects for the purpose of generating audio guidance instructions.

[0072] In this document, first a detailed description of how guidance instructions are generated is given. In the second part, details are given on how these guidance instructions can be enhanced using environmental objects detected in the acquired images. Finally, in the third part, instructions are given on how exactly these instructions are generated at the stage the vehicle is currently in (anticipation, explanation, accompaniment, etc.).

[0073] Step S3: Detecting objects The step of detecting moving and static objects in the images acquired by the camera 10, and the attributes that characterize each of these objects, is performed by the processing module 25 using existing artificial intelligence, machine learning technologies.

[0074] It should be noted that an object is defined here as an element of the environment that is potentially or actually visible to the driver of the vehicle.

[0075] The list of attributes depends on the type of object detected.

[0076] Thus, as an illustrative example, if the object is a "four-wheeled motor vehicle", the attributes would be as follows: - Category (cars, light trucks, medium trucks, heavy trucks, buses, etc.), - the main color of the car body, - the direction in which the motor vehicle is being driven; - the part seen by the driver (front, rear, left, right), - Status (parked, stopped, moving), - the orientation of the object relative to the vehicle, - the percentage of possible views, is.

[0077] A confidence index expressed in percentages is given to the detected object and to each attribute. The closer this confidence index value is to 100%, the more easily the object and attribute can be used. Attributes or objects below a certain threshold are no longer taken into account. The greater the number of attributes of a given object, the more easily this object can be recognized in each iteration of scene analysis (in other words, in each processed image).

[0078] Step S4: Positioning the object To identify one or more objects to be used in formulating voice guidance instructions, the processing module 25 matches the objects detected in the processed images with data stored in the map database 20, which allows the relationship of each object to the environment and to the driver's perception to be understood.

[0079] To do this, the processing module 25 uses data stored in the map database 20, in particular the objects referenced in the map database 20 (points of interest, road signs, traffic lights, bus stops, etc.) and their descriptions (name, type, color, etc.).

[0080] The processing module 25 then identifies roads and routes in the processed images.

[0081] Processing module 25 then locates each detected object on the map by matching referenced objects with the detected object. Processing module 25 also locates detected, unreferenced objects (vehicles, pedestrians, etc.) on the map.

[0082] Step S6: Classifying At this stage, the processing module 25 classifies all detected objects.

[0083] The first classification technique consists of using existing artificial intelligence, machine learning technologies.

[0084] This technique may consist of training an artificial intelligence to identify at least 10 detected objects on its own, and then classify these 10 objects in order of importance, focusing on the right zone of the image if the vehicle has to turn right, the center zone if the vehicle has to go straight, and the left zone if the vehicle has to turn left.

[0085] The second classification technique consists of performing statistical calculations to assign a weight to each detected object.

[0086] The weights are listed below: - the persistence of the object in the scene, i.e., in each image (the longer an object remains visible, the more its weight increases, and if its presence alternates between visible and invisible, the more its weight decreases); - The area occupied by objects in the scene and the percentage of objects that can be seen (trucks are more visible than cars) - object salience (the more different the color of an object is from the rest of the environment, the greater the weight of the object; for example, a red vehicle is more visible than a grey one); - uniqueness of the object (the use of just one blue vehicle is less ambiguous than the use of a red vehicle among other red or orange vehicles), - the position of the object relative to the operation to be performed (the closer the object is to the operation to be performed, the greater the weight of the object); - the stage the vehicle is currently in (depending on the stage, some classes of objects are more useful in preparing guidance than others, as will be detailed further on in this specification); can be calculated according to a criterion selected from

[0087] The third technique involves strategies based on the continuity of visual cues along the route.

[0088] An object is prioritized for use in generating instructions in step S7, described below, if the processing module 25 identifies an attribute of the object as an attribute common to multiple objects or a cue that has already been used in generating a previous guidance instruction for the current route and / or that can potentially be used to generate a subsequent guidance instruction on the same route.

[0089] Advantageously, the common attributes are identified by the processing module 25 when it determines the route S1.

[0090] For example, when determining route S1, processing module 25 identifies a number of road signs that have the name "Long Island" written on them. The common attribute that is identified is the name "Long Island" written on the road signs. In a video acquisition step S2, camera 10 acquires an image comprising the signs. In an object detection step S3, processing module 25 identifies signs in the image and detects whether the signs are characterized by the previously identified common attribute. The attribute may be detected by decoding the name written on the sign in the image or by obtaining a match via geolocation with a sign contained in map database 20, which is associated with the attribute also contained in map database 20. In a classification step S5, processing module 25 prioritizes the signs to ensure that they are used in generating the next guidance instructions in step S7.

[0091] Thus, if signs with the same name written on them are found at multiple consecutive intersections along a route, the guidance instructions relating to these intersections will privilege the use of a given visual cue, ensuring consistency and ensuring a continuity of visual cues that will facilitate user guidance while reducing the cognitive load placed on the user.

[0092] For example, on a route with an intersection followed by a first roundabout and a road that requires no deviation from it, the following successive guidance instructions may be generated: "At the roundabout, turn right towards Long Island," "Continue straight towards the Long Island sign," and "Stay in your current lane and continue towards Long Island."

[0093] Step S7: Generating instructions Guidance instructions are then generated, possibly using cues that are appropriate to the context and the operation to be performed and that depend on the attributes of the selected object.

[0094] One purpose is to describe at what moment and in what context the object should be used to generate voice guidance instructions to best help the driver find their way.

[0095] The context depends in particular on the complexity of the operation to be performed, which here is determined by the following criteria: - the position of the vehicle relative to the maneuver (if a lane change is required, the situation is likely to be more complex); - the configuration of the road infrastructure (the more lanes an intersection has, the more complex it is considered to be); - Driving context (driver tension, i.e. stress, busy roads, etc.), - the presence of a large number of objects in the driver's field of view, - Driver knowledge of road infrastructure, - the stage the vehicle is currently in, and - statistical weights of the various detected visual indicators, is estimated according to at least one of, and preferably according to each of, them.

[0096] Preferably, the visual indicator is used in the voice guidance instructions only if the visual indicator is found to be necessary.

[0097] If the situation is simple, in other words, if the complexity level is below a first complexity threshold, it is preferable not to use a visual indicator so as not to increase the cognitive load on the driver, such that saying "Turn left at this intersection" is preferable to saying "Turn left after the bus stop" because the latter instruction requires the driver to locate the bus stop and the intersection rather than just the intersection itself.

[0098] In contrast, if the complexity level is higher than this first complexity threshold, instructions are generated using time or visual cues (or "indicators") that improve understanding of the operation to be performed. This is called "enhanced voice guidance instructions."

[0099] This enhancement can be performed in different ways depending on different parameters.

[0100] A first way to enhance voice guidance instructions consists of incorporating the notion of time in the instructions to facilitate understanding of the actions to be performed. This can be illustrated by several examples using contextual cues.

[0101] In a first example, if the road is congested or an object is blocking the maneuver or obscuring the goal of the maneuver, the processing module 25 is programmed to integrate a contextual cue, such as "as soon as possible," into the middle of the guidance instructions so that the driver feels comfortable repositioning their vehicle at the best moment.

[0102] In a second example, if the operation can be performed and must be immediate, the processing module 25 is programmed to incorporate another context cue, such as "now" or "immediately," to encourage the driver to reposition his vehicle without delay.

[0103] In a third example, if a maneuver is possible but not immediately required, the processing module 25 is programmed to incorporate a context cue such as "as soon as possible" so that the driver feels comfortable repositioning their vehicle at the best possible moment.

[0104] Each guidance instruction is then formed by combining at least one action, one contextual cue and one attribute.

[0105] In the case of a guidance instruction linked to another guidance instruction, the processing module 25 is programmed to replace the context cue "as soon as possible" with "next" or "later."

[0106] In this case, the complete guidance instruction is formed by combining the action and the egocentric attributes, contextual cues and other attributes linked to the action.

[0107] For example, the complete navigation instructions may be expressed in the form "Turn right, then keep left", or "Turn right, then keep left", or "Turn right, then follow the direction of Long Island".

[0108] A second way to enhance the voice guidance instructions is to employ a moving indicator to facilitate understanding of the action to be performed. This approach is used when the object detected in the acquired image with the highest weight is a moving indicator.

[0109] The processing module 25 successfully - Detects various types of movement indicators (cars, trucks, buses, vans, bicycles, motorcycles), - positioning the movement indicators on the lanes of the map obtained from the map database 20; - detecting attributes of these movement indicators that distinguish them from each other; - identified unique movement indicators that are capable of being seen by a driver without ambiguity by characterizing the unique movement indicators by at least a unique movement indicator type, a unique movement indicator color, and a unique movement indicator location; and - Identifying that the unique motion indicator is stable (i.e., that the unique motion indicator appears in different successively processed images); It can be considered that.

[0110] In this case, the processing module 25 is programmed to use this unique movement indicator to help the driver correctly position his or her vehicle.

[0111] A complete navigation instruction is then formed by combining the actions, geographic indicators, visual cues and attributes.

[0112] For example, the guidance instruction may be expressed in the form "Stay in the same lane as the red car."

[0113] A third way to enhance the voice guidance instructions consists of employing static indicators, such as elements of road infrastructure, to facilitate understanding of the action to be performed. This way is used when the object detected in the acquired image with the highest weight is a static indicator.

[0114] To do this, the processing module 25 normally - Detected various types of static indicators (bridges, traffic lights, etc.), - associating the detected static indicators with objects stored in the map database 20; - detecting the attributes of these static indicators so as to distinguish them from one another; - identified unique static indicators that are capable of being seen by a driver without ambiguity by characterizing the unique static indicators by at least a unique static indicator type, a unique static indicator color, and a unique static indicator location; and - Identified that the indicator is stable, It can be considered that.

[0115] In this case, the processing module 25 is programmed to use this unique static indicator to help the driver correctly position his or her vehicle.

[0116] A complete navigation instruction is then formed by combining the actions, goals, geographic indicators and visual cues.

[0117] For example, the complete directions may be expressed in the form "Take the exit just after the bridge."

[0118] A fourth way to enhance the voice guidance instructions is to employ directional signs to facilitate understanding of the action to be performed. This approach is used when the object detected in the acquired image with the highest weight is a sign.

[0119] To do this, the processing module 25 normally - Detected various types of signs and sign attributes (traffic lights, stop signs, etc.), - associating the detected signs with the signs contained in the map database 20, in particular with the aim of determining the attributes of each sign (color, text, pointed direction, etc.); - This determines the attributes of each sign that distinguish them, - Identified a unique sign that can be seen by the driver without ambiguity (e.g., by the color of the unique sign, the text written on the unique sign, etc.); - identified that the sign is stable (in particular, that the text or displayed shape does not change over a short period of time); It can be considered that.

[0120] In this case, the processing module 25 is programmed to use this unique landmark and attributes of the unique landmark to help the driver correctly position his or her vehicle.

[0121] A complete guidance instruction is then formed by combining the actions, geographic indicators, egocentric attributes and objectives.

[0122] For example, the complete guidance instruction may be expressed in the form "Stay in the left lane towards the green sign."

[0123] A fifth way to enhance the voice guidance instructions consists in employing important places (restaurants, pharmacies, etc.) to facilitate understanding of the actions to be performed. This way is used when the object detected in the acquired image with the highest weight is an important place.

[0124] To do this, the processing module 25 normally - Detected various types of static indicators that form important locations, - associating the detected static indicators with indicators contained in the map database 20; - detecting the attributes of these indicators that distinguish them from each other; - Identifying the unique indicators that can be best seen by the driver without ambiguity, taking into account the indicator attributes (type, color and position of the indicator); and - Identified that the indicator is stable, It can be considered that.

[0125] In this case, the processing module 25 is programmed to use this significant location and attributes of the significant location to help the driver correctly position his vehicle.

[0126] A complete guidance instruction is then constructed by combining actions, egocentric attributes, geographic indicators, and visual cues.

[0127] For example, the directions may be expressed in the form "Turn right just past the restaurant."

[0128] A sixth way to enhance the voice guidance instructions consists in employing prominent locations (monuments, cathedrals, etc., known important buildings whose function will not change over the next few years) to facilitate understanding of the actions to be performed. This approach is used when the object detected in the acquired image with the highest weight is a prominent location.

[0129] In this case, to help the driver correctly position his vehicle, the processing module 25 is programmed to use only the name of this well-known place and not add any attributes.

[0130] A complete guidance instruction is then constructed by combining actions, egocentric attributes, geographic indicators, and visual cues.

[0131] For example, the complete directions may be expressed in the form "Turn right just past the Empire State Building."

[0132] A seventh way to enhance the voice guidance instructions is to employ two important and / or prominent locations to facilitate understanding of the action to be performed. This way is used when the object detected in the acquired image with the highest weight is a prominent or important location and when one of the other detected objects is also a prominent or important location.

[0133] To do this, the processing module 25 normally - Detected various types of static indicators that form important or prominent places, - associating the detected static indicators with indicators contained in the map database 20; - detecting the attributes of these indicators that distinguish them from each other; - identified that the target of the operation is located between two detected static indicators that can be seen by the driver without ambiguity, - Identifying that these indicators are stable, It can be considered that.

[0134] In this case, the processing module 25 is programmed to use these prominent or important locations and attributes of the prominent or important locations to help the driver correctly position his or her vehicle.

[0135] A complete guidance instruction is then formed by combining the action, the egocentric attribute, the geographic indicator, the first visual cue and the second visual cue.

[0136] For example, the directions may be expressed in the form "Turn right between the church and the pharmacy."

[0137] An eighth way to enhance voice guidance instructions consists of using road characteristics to facilitate understanding of the action to be taken.

[0138] In particular, the map database 20 stores information regarding highway characteristics that may be used in the context of the present invention.

[0139] The processing module 25 may then be considered to have successfully detected the attributes that distinguish the highway to be used from other highways visible to the driver.

[0140] In this case, the processing module 25 is programmed to use this attribute to help the driver correctly position his vehicle.

[0141] The complete guidance instruction is then formed by combining the actions, egocentric attributes and goals, as well as the goals' attributes.

[0142] For example, the complete navigation instructions may be expressed in the form "Turn right onto a side street" or "Turn right onto a minor street" or "Turn right onto a paved road."

[0143] A ninth way of enhancing voice guidance instructions is used when enhanced guidance instructions are recommended in the anticipation or explanation phase, but visual indicators are not available at that time, whereas in the accompanying phase, visual indicators are available.

[0144] In this case, the processing module 25 is programmed to use this visual indicator to generate enhanced voice guidance instructions at the earliest opportunity.

[0145] This visual indicator can be used before the driver is able to see it (in the explanation phase, or even in the anticipation phase) or from the moment the driver is able to see it. In particular, it is possible to envisage that the processing module 25 commands the repetition of voice guidance instructions, which are supplemented using this visual cue.

[0146] A tenth way of enhancing voice guidance instructions is used depending on the complexity of the operation to be performed and on the detected level of hesitation.

[0147] This provides for associating multiple types of reinforcement when the complexity of the maneuver is determined to be extremely high (higher than a second predetermined complexity threshold) or when the driver's level of hesitation exceeds a predetermined threshold.

[0148] The driver's level of hesitation can be assessed in various known ways, for example, depending on the route taken (whether the route taken deviates from what was expected), depending on the sway of the steering wheel, depending on the way the steering wheel is held, depending on the driver's position, etc.

[0149] A first example of association consists of employing reinforcement using the notion of time, and any other type of reinforcement, in the voice guidance instructions.

[0150] A second example of association consists of employing reinforcement using important or prominent locations and reinforcement using characteristics of the road to be utilized.

[0151] In this example, a complete guiding instruction is then formed by combining the action, the egocentric attribute, the geographic indicator, the visual cue, the goal, and the goal's attributes. This complete guiding instruction may then be expressed in the form, "Just past the Empire State Building, turn right and take the narrow side street."

[0152] A third example of association consists of employing reinforcement using directional signs and reinforcement using important or prominent locations.

[0153] A fourth example of association consists of employing two reinforcements that use important or prominent locations.

[0154] In this example, a complete navigation instruction is then formed by combining the action, the egocentric attribute, the visual cue, the geographic indicator, and another visual cue. This complete navigation instruction may be expressed in the form "Turn right past the Empire State Building between the church and the pharmacy."

[0155] A fourth example of association consists of employing reinforcement using moving indicators and reinforcement using fixed indicators or elements of road infrastructure.

[0156] In this example, a complete guiding instruction is then formed by combining the action, the egocentric attribute, the first visual cue, the other visual cue, and the attribute of the cue. This complete guiding instruction may be expressed in the form "Turn left past the traffic light like a red car."

[0157] It was taught above that in low complexity situations, voice guidance instructions are not reinforced.

[0158] It was also taught that the level of complexity of an operation depends on several variables.

[0159] Preferably, provision can be made for the complexity level to also depend on another parameter, in other words the cost in terms of time or distance that an error by the driver (for example if the driver takes the wrong exit) results.

[0160] Thus, if an error is made by the driver during operation, and the processing module 25 predicts that this error will add a lot of time to the remaining journey, the processing module 25 is programmed to increase the value of the complexity level, which leads to intensified voice guidance instructions.

[0161] It will therefore be appreciated that in this particular case provision is also made for using more frequently and even systematically combinations of reinforcements, such as those detailed above.

[0162] Specifically, the processing module 25 must provide the driver with assurance of the success of a maneuver if not performing the maneuver would significantly affect the estimated time of arrival at the destination, which is especially true on highways, roads with few entrances and exits, or one-way streets.

[0163] In practice, the processing module 25 is programmed to calculate the additional time (or additional distance to the destination) required to reach the destination if the driver takes a wrong turn during a maneuver to be performed, so that when a predetermined threshold (expressed in terms of time or distance to reach the destination or percentage of time remaining) is exceeded, the processing module 25 generates stronger voice guidance instructions than would be generated if the calculated time were below this threshold.

[0164] Preferably, the processing module 25 may also use a firmer intonation than usual when verbalizing the voice guidance instructions so that the driver understands the importance of performing the correct maneuver.

[0165] The cost of not performing the operation (and therefore the choice for generating more enhanced voice guidance instructions) may be calculated depending on other parameters (type of route, type of road, etc.), if clearly identified, e.g. - The vehicle is not correctly positioned to perform the maneuver; - the vehicle is correctly positioned for the maneuver to be performed, but the appropriate direction indicator lamp is not activated, - the driver did not act as suggested before the end of the anticipation phase or before the end of the explanation phase; This may be increased if necessary.

[0166] In these cases, the processing module 25 is programmed to repeat the voice guidance instructions, in some cases, with emphasis added to the voice guidance instructions, for example, "Now, enter the ... lane," or with another enhancement added to the voice guidance instructions that allows the action to be performed to be emphasized, for example, "Now, enter the ... lane ... in the direction of ...."

[0167] As the vehicle moves forward, the context may change, making the guidance instructions ambiguous at best and useless at worst.

[0168] This leads to two possible cases.

[0169] The first case is when an element comes between the vehicle and the object selected for generating guidance instructions, blocking the driver's view of this object.

[0170] The second case occurs when another object similar to the object selected for generating guidance instructions appears in the driver's field of view, which can occur, for example, when a bus leaves a bus stop and reveals a second bus stop.

[0171] These two situations result in increased stress levels and cognitive load for the driver, and introduce the risk of errors or delayed execution of maneuvers.

[0172] In these two cases, the processing module 25 is then programmed to generate new instructions that fit the new context.

[0173] As mentioned above, the voice guidance instructions to be verbalized vary depending on the stage the vehicle is currently in. Accordingly, how the guidance instructions are formulated in each of these stages will now be described.

[0174] Anticipation stage It will be recalled that an anticipation phase takes place before the maneuver to be performed. The anticipation phase begins and ends at a location away from the zone in which the maneuver will be performed. The anticipation phase allows the driver to position the vehicle as best as possible and / or prepare to execute this maneuver (especially when the target of the maneuver becomes visible only at the last moment).

[0175] During such an anticipatory phase, enhanced voice guidance instructions may be recommended, but the use of famous or important locations is avoided.

[0176] The processing module 25 is programmed to verbalize what are called advance voice guidance instructions when the stretch of road being used includes multiple lanes.

[0177] The manner in which this advance guidance indication is reinforced may vary depending on the situation encountered.

[0178] Typically (in other words, in situations other than those considered below), reinforcement is carried out in the following manner:

[0179] When the vehicle is on a multi-lane highway and positioned in the lane most compatible with performing the next maneuver, the processing module 25 is programmed to generate advance voice guidance instructions to reassure the driver.

[0180] Voice guidance instructions are then formed by combining the actions, familiarity criteria, geographic indicators and egocentric attributes.

[0181] For example, the voice guidance instruction may be expressed in the form "Stay in your current lane."

[0182] If the vehicle is located on a multi-lane road and is positioned in a lane that is compatible with performing the next maneuver (typically there is no solid white line between the lane in which the vehicle is located and the point of the next maneuver), but a lane change is required, the processing module 25 is programmed to generate advance voice guidance instructions that guide the driver to quickly change lanes.

[0183] Voice guidance instructions are then formed by combining the actions, the geographic indicators and the contextual cues.

[0184] For example, the voice guidance instruction may be expressed in the form "Enter lane x now (or as soon as possible)."

[0185] If the vehicle must enter a multi-lane road and the vehicle is located in an approach lane of limited length, the processing module 25 is programmed to generate advance voice guidance instructions to guide the driver to change lanes.

[0186] A voice guidance instruction is then formed by combining at least one action and one geographic indicator.

[0187] For example, the voice guidance instruction may be expressed in the form of "Stay in your lane."

[0188] If the vehicle must enter a multi-lane road and the vehicle is located on an unlimited length approach lane (in other words, a lane that merges with a lane on the road) that is most compatible with performing the next maneuver, the processing module 25 is programmed to generate advance voice guidance instructions guiding the driver to stay in this lane.

[0189] A voice guidance instruction is then formed by combining at least one action, one geographic indicator and one egocentric attribute.

[0190] For example, the guidance instruction may be expressed in the form "stay in your current lane."

[0191] If the vehicle must enter a multi-lane road and the vehicle is located on an unlimited length approach lane that is not the most compatible lane for the next maneuver to be performed, the processing module 25 is programmed to count the number of lane changes required to reach the most compatible lane and then generate advance voice guidance instructions to guide the driver to change lanes.

[0192] A voice guidance instruction is then formed by combining at least one action and an attribute.

[0193] For example, the voice guidance instruction may be expressed in the form "Enter the nth lane."

[0194] In this situation, the way in which it is expressed which lane the vehicle should move into can be determined as follows.

[0195] If there are no restrictions on which lane to use among the lanes of the highway, the processing module 25 is programmed to generate voice guidance instructions directing the driver to simply enter.

[0196] A complete guidance instruction is then formed by combining at least one action and one goal.

[0197] For example, the guidance instructions may be expressed in the form "Turn into..." The name of the road or the type of road may then be specified.

[0198] If the highway has only two lanes of traffic, the processing module 25 is programmed to generate voice guidance instructions indicating whether the vehicle should place itself in the "right" lane or the "left" lane.

[0199] A guidance instruction is then formed by combining at least one action, one geographic indicator and one egocentric attribute.

[0200] If the highway has exactly three lanes of traffic, the processing module 25 is programmed to generate voice guidance instructions indicating whether the vehicle should place itself in the "right" lane, the "left" lane, or the "center" lane.

[0201] A guidance instruction is then formed by combining at least one action, one geographic indicator and one egocentric attribute.

[0202] If the highway has exactly four lanes of traffic and only one of them is compatible with the maneuver to be performed, the processing module 25 is programmed to generate voice guidance instructions indicating whether the vehicle should place itself in the "rightmost" lane, the "leftmost" lane, the "second lane," or the "third lane."

[0203] The guidance instructions are then formed by combining at least one action, one geographic indicator and one allocentric attribute.

[0204] If the highway has exactly four lanes of traffic and more than one of them is compatible with the maneuver to be performed, the processing module 25 is programmed to generate voice guidance instructions indicating whether the vehicle should place itself in the "rightmost" lane, the "leftmost" lane, or the "center" lane.

[0205] The guidance instructions are then formed by combining at least one action, one geographic indicator and one allocentric attribute.

[0206] If the highway has an odd number of traffic lanes greater than four and only the center lane is compatible with the maneuver to be performed, the processing module 25 is programmed to generate a voice guidance instruction indicating that the vehicle should position itself in the "center" lane.

[0207] The guidance instructions are then formed by combining at least one action, one geographic indicator and one allocentric attribute.

[0208] If a highway has multiple lanes and at least one of them is closed to vehicles, the processing module must integrate this into the total number of lanes.

[0209] When the vehicle is located outside a built-up area, enhanced voice guidance instructions in the anticipation phase may primarily point out to the driver why he or she is positioning himself or herself in that way (in other words, the purpose of positioning) by explaining the next operation to be performed.

[0210] The objective is thereby indicated in terms of distance between the vehicle and the maneuver, in terms of the type of maneuver to be performed (highway exit, interchange, roundabout, etc.), and in terms of road infrastructure visible to the driver (directional signs, etc.).

[0211] On highways, the processing module 25 must use visual indicators, which are preferably static (of the directional sign type), or alternatively moving indicators (in the absence of such signs), or alternatively elements of the road infrastructure (in the absence of signs and moving indicators).

[0212] The following example illustrates this rule:

[0213] When the processing module 25 detects a sign and is able to read the inscription on the sign, the processing module 25 is programmed to generate advance voice guidance instructions that guide the driver to follow the direction indicated on the sign.

[0214] A complete guidance instruction is then formed by combining at least one action, one geographic indicator, one egocentric attribute, one objective, one geographic indicator, one visual cue and one attribute.

[0215] For example, the voice guidance instructions may be expressed in the form "Get into the left lane and prepare to take the exit towards...".

[0216] If the processing module 25 detects a sign and the processing module 25 is not able to read the inscription on the sign but the map database 20 contains this information, the processing module 25 operates in the same manner as described above.

[0217] If the processing module 25 detects a sign among other signs whose color is unique, the processing module 25 is programmed to generate voice guidance instructions that guide the driver to follow the direction of the sign of the identified color.

[0218] The voice guidance instruction is then formed by combining at least one action, one geographic indicator, one egocentric attribute, one objective and one attribute.

[0219] For example, the guidance instruction may be expressed in the form "Get ready to get into the left lane and follow the blue signs." If the writing on the sign is readable or known, this instruction may be supplemented by pointing to the direction indicated on the sign.

[0220] When the vehicle is located in a built-up area, signs are more often obscured or covered by vegetation. As a result, the processing module 25 selects the most appropriate visual cues. - movement indicator, - Road infrastructure elements, - static indicators other than signs, and - important or prominent places, The indicators must be distinguished among other indicators, including:

[0221] The following four examples illustrate this situation.

[0222] In a first example, processing module 25 generates the following guidance instruction: "Get in the same lane as the red car." In this example, the guidance instruction is formed by combining at least one action, one geographic indicator, one visual cue, and one attribute.

[0223] In a second example, the processing module 25 generates the following guidance instruction: "Get into the left lane heading in the direction of the green sign." In this example, the guidance instruction is formed by combining at least one action, one geographic indicator, one egocentric attribute, one geographic indicator, one visual cue, and one attribute.

[0224] In a third example, the processing module 25 generates the following guidance instruction: "Get into the left lane, then take the third exit at the roundabout." In this example, the guidance instruction is formed by combining at least one action, one geographic indicator, one visual cue, one connecting component, another visual cue, an action, and an allocentric attribute.

[0225] In a fourth example, processing module 25 generates the following guidance instruction: "Get into the right lane and prepare to turn right at the cross street." In this example, the guidance instruction is formed by combining at least one action, one geographic indicator, one visual cue, one objective, and another visual cue.

[0226] Operation explanation stage The second stage is the explanatory stage. In this second stage, audio guidance instructions, called explanatory audio guidance instructions, characterize the actions that must be performed to reach the goal.

[0227] In this instruction, the processing module 25 conveys the information the driver needs to determine where they are and to locate the target relative to the driver's environment (particularly relative to detected visual indicators).

[0228] When the vehicle reaches an intersection, the processing module 25 calculates the angle of each road starting from the intersection relative to the road the vehicle is on, this angle being expressed in degrees, for example, between -180° and 180°.

[0229] The guidance instructions can then be enhanced with geographic attributes tailored to the intersection, which can be "hard right" or "ahead" (or "straight ahead").

[0230] The manner in which guidance instructions are reinforced may vary depending on the type of intersection.

[0231] In the case of a T-junction, when the vehicle passes through the vertical line of the T and arrives at a road that forms the horizontal line of the T, the processing module 25 is programmed to employ the word "intersection" as the case may be.

[0232] If the angle between the vertical and horizontal lines of the T is substantially perpendicular (eg, within 10, 20, or 30 degrees), the processing module 25 employs the verb "bend" or "utilize."

[0233] If the angle between the vertical and horizontal lines of the T is not substantially perpendicular, two cases must be considered.

[0234] The first case is when the vehicle must deviate at an intersection by an angle substantially less than 90 degrees relative to the vehicle's initial trajectory. In this case, the processing module 25 supplements the verb used with the expression "slightly." The processing module 25 also prefers the use of the verb "use" over the verb "turn."

[0235] The second case is when the vehicle must deviate at an intersection by an angle substantially greater than 90 degrees relative to the vehicle's initial trajectory. In this case, processing module 25 supplements the verb used with the expression "completely." Processing module 25 also prefers the use of the verb "turn" over the verb "take advantage."

[0236] A "fork" type intersection is an intersection consisting of exactly two exit lanes located from and in the extension of the incoming lane.

[0237] The way in which voice guidance instructions are formulated at such intersections then depends on the configuration of the locale in question.

[0238] If two exit lanes are to the right of the entry lane and the vehicle must use the far right lane, the instruction is formulated using a phrase such as "use the far right lane." In contrast, if the vehicle must use another exit lane, the guidance instruction needs to be strengthened to indicate to the driver which exit lane they must use.

[0239] Conversely, if two exit lanes are to the left of the entry lane and the vehicle must use the leftmost lane, the instruction is formulated using a phrase such as "use the leftmost lane." In contrast, if the vehicle must use another exit lane, the guidance instruction needs to be strengthened to indicate to the driver which exit lane they must use.

[0240] When two exit lanes are on either side of the entry lane, instructions are formulated using phrases such as "keep left lane" or "keep right lane."

[0241] At roundabouts, the voice guidance instructions are defined in particular according to the number of exits from the roundabout.

[0242] A roundabout comprises a ring road and roads that connect to this central road, which roads will be referred to below as "adjacent roads". A roundabout has entrances and exits, which may be located on the same adjacent road.

[0243] The complexity of a rotary is determined by the following criteria: - the angle between the road on which the vehicle arrives and the road on which the vehicle must exit, - the angle between the exit to be used and the nearest exit, - number of adjacent roads, - the difference between the number of entrances and the number of exits, - the number of lanes on the ring road, - outer diameter of the rotary, Note that this depends on

[0244] A roundabout is considered simple if it has the same number of entrances as exits, distributed in pairs on adjacent roads, if the number of adjacent roads is less than or equal to four, if these adjacent roads are oriented in clearly different directions, and if priority on the roundabout is given to vehicles operating on the ring road. In other cases, the roundabout is considered complex.

[0245] A first example of a simple roundabout is a roundabout whose adjacent roads are four in number and angularly separated by approximately 90 degrees. Reference is made to an "orthogonal roundabout." In this example, processing module 25 is programmed to verbalize voice guidance instructions such as, "At the roundabout, exit right (or exit left, or go straight)," as the case may be.

[0246] A second example of a simple roundabout is a roundabout whose adjacent roads are three or four in number, at least one of which is angularly separated in pairs by an angle substantially different from 90 degrees. In this example, processing module 25 is programmed to verbalize a voice guidance instruction, possibly augmented by the number of the exit lane to be taken. The instruction may be expressed in the form, "At the roundabout, take the first exit on the right."

[0247] A third example of a simple roundabout is a small diameter roundabout, in which case the processing module 25 considers the roundabout to be a simple crossroads (see below).

[0248] Maneuvering voice instructions, which allow complex roundabouts to be navigated, are constantly being enhanced.

[0249] Various examples of reinforcement of instructions can be given.

[0250] In the first example, the roundabout does not have four pairs of orthogonal exits.

[0251] Thus, the first case to be considered is when the exit is perpendicular (within 10 degrees) to the incoming lane through which the vehicle arrives and corresponds to the first or second exit. In this case, the processing module 25 is programmed to preferentially verbalize a voice guidance instruction reinforced by the number of the exit to be taken. This instruction may be expressed in the form: "At the roundabout, take the first exit on the right."

[0252] A second case to be considered is when the exit is not perpendicular to the incoming lane through which the vehicle arrives, but corresponds to the first or second exit. In this case, the processing module 25 is programmed to obligatory verbalize a voice guidance instruction reinforced by the number of the exit to be taken. This instruction may be expressed in the form: "At the roundabout, take the first exit on the right."

[0253] A third case to consider is when the exit corresponds to the third exit or a subsequent exit. In this case, the processing module 25 is programmed to verbalize an allocentric attribute-enhanced voice guidance instruction. This instruction may be expressed in the form, "Take the exit on the left before the restaurant."

[0254] A second example of a complex roundabout is one with a different number of entrances and exits.

[0255] It is particularly possible to consider the case where there is one entrance more than there are exits, and this entrance is located on a neighboring one-way road that is different from the lane through which the vehicle arrives.

[0256] In this example, the processing module 25 applies the rules defined in the preceding example, while occasionally refraining from indicating the exit to be taken by its number.

[0257] Thus, if the exit to be taken is located after a one-way entrance, the processing module 25 is programmed to verbalize an allocentric attribute-enhanced voice guidance instruction and not point to the exit number, possibly expressed in the form "Take the next exit on the right after..."

[0258] In contrast, if the exit to be used is located before a one-way entrance, the processing module 25 is programmed to verbalize the voice guidance instructions in the same manner as in the first example given above of a complex roundabout.

[0259] A third example of a complex roundabout is one in which the exits are very close to each other, either in distance or angle. In particular, a roundabout is considered complex when the angle between the two exits is less than or equal to 30°.

[0260] In this example, the processing module 25 applies the same rules as those applied in the first example given above of a complex roundabout, while mandatorily enhancing the voice guidance instructions.

[0261] A fourth example of a complex roundabout is one in which vehicles arriving at the roundabout have priority over those already driving around the roundabout (the Place de l'Etoile roundabout in Paris, France is the best-known example of this type of roundabout).

[0262] In this example, the processing module 25 applies the same rules as those applied in the first example given above of a complex roundabout, while obligatorily reinforcing the voice guidance instructions with warnings such as "Caution! Vehicles coming from the right have priority at the roundabout."

[0263] At crossroads with at least four lanes leading from them, the processing module 25 applies the same rules as applied in the various roundabout examples given above, substituting the term "roundabout" with "crossroad" or "intersection."

[0264] The processing module 25 particularly counts vehicle exits from lanes, and in so doing may count all lanes, even those that are one-way or closed.

[0265] If the exit lane to be taken is offset relative to other exit lanes, the processing module 25 is programmed to add a command to the instructions such as "Proceed through the cross street and take the second exit on the left in the direction of..."

[0266] The accompanying stage of the operation The accompanying stage essentially indicates to the driver the exact moment when he or she can begin the actual maneuver, i.e., change of direction. During this stage, the processing module 25 sends what are called accompanying voice guidance instructions only if it determines that crossing an intersection is complicated.

[0267] In this situation, whether and how the voice guidance instructions are intensified depends on the type of intersection.

[0268] Thus, when multiple T-junctions follow one another on a roadway and there are no nearby visual indicators (infrastructure, signs, prominent locations, etc.) to distinguish the multiple T-junctions, the processing module 25 is programmed to verbalize accompanying voice guidance instructions, which may include contextual cues and be formulated as follows: "Turn right now..."

[0269] At fork type intersections, no accompanying voice guidance instructions are transmitted.

[0270] At roundabouts, if the diameter of the roundabout is smaller than a predetermined threshold and the exits are not pairwise orthogonal, the processing module 25 is programmed to generate accompanying voice guidance instructions during the maneuver, either by repeating instructions already transmitted during the maneuver explanation phase, or by using new distinctive or prominent locations or elements of road infrastructure that were not previously visible. The distinctive or prominent locations or elements of road infrastructure used must, of course, be visible, unique, and immediately identifiable by the driver.

[0271] This instruction may in some cases be formulated as "Exit the roundabout just before Hotel X now."

[0272] If the processing module 25 detects that the driver has gone around the roundabout, the processing module 25 may be programmed to generate accompanying voice guidance instructions that repeat the voice guidance instructions transmitted during the explanation phase.

[0273] If the voice guidance instructions transmitted during the explanation phase did not refer to any visual indicators because the indicators were not clearly visible at the time and the vehicle's speed was significantly reduced (either due to driver hesitation or traffic congestion), the processing module 25 is programmed to generate accompanying voice guidance instructions using distinctive or prominent locations or elements of road infrastructure that have become visible.

[0274] The drop here must be greater than a predetermined threshold.

[0275] Reassurance Stage Once the maneuver is performed, the processing module 25 assesses the need to reassure the driver about the quality of the maneuver he has just performed, depending on the complexity of the maneuver and any driver hesitation.

[0276] To do this, if processing module 25 detects hesitation during or when exiting a maneuver, processing module 25 is programmed to generate an enhanced voice guidance prompt, the decision to generate this prompt may also depend on other factors, such as the complexity of the maneuver performed, the imminent presence of another maneuver to be performed, etc.

[0277] This instruction is preferably reinforced using a visual cue located near the end of the operation or visible only when the operation is completed, which cue is preferably separate from any visual cues previously used.

[0278] Continuity Stage When no operation is required for a distance or time greater than a threshold (e.g., more than one minute or more than one kilometer), the processing module 25 preferably indicates to the driver that the processing module 25 is still active.

[0279] The voice guidance instructions may then be formulated differently depending on the configuration of the highway used and on the environment visible to the driver.

[0280] Consider the case where at least one intersection is detected between the vehicle's position and the next maneuver to be performed, no road change is planned, and no lane change is possible. In this situation, the processing module 25 is programmed to generate voice guidance instructions that depend on the curvature of the road.

[0281] Thus, when the curvature radius of the road is greater than a predetermined threshold, a voice guidance instruction is formed by combining at least one action, one geographic indicator, one egocentric attribute, one purpose, and one geographic indicator. The voice guidance instruction may be expressed in the form of "Go through the intersection and go straight...".

[0282] Conversely, if the road curvature radius is less than this threshold, the voice guidance instruction is formed by combining at least one action, one geographic indicator, one egocentric attribute, one objective, one geographic indicator, and another egocentric attribute. The voice guidance instruction may be expressed in the form of "Go through the next intersection and follow the main lane...".

[0283] Also consider the case where at least one intersection is detected between the vehicle's position and the next maneuver to be performed, no road change is planned, and no lane change is desirable in view of the next maneuver to be performed. In this situation, processing module 25 is programmed to generate voice guidance instructions that also depend on the curvature of the road.

[0284] Thereby, when the curvature radius of the road is greater than a predetermined threshold, a voice guidance instruction is formed by combining at least one action, one geographic indicator, one egocentric attribute, one objective, and one geographic indicator concatenated with another action, a familiarity criterion, another geographic indicator, and another egocentric attribute. The voice guidance instruction may be expressed in the form of "Go straight through the intersection and stay in your current lane...".

[0285] In contrast, if the road curvature radius is less than this threshold, the voice guidance instruction is formed by combining at least one action, one geographic indicator, one attribute, one purpose, and one geographic indicator concatenated with another action, a familiarity criterion, another geographic indicator, and another egocentric attribute. The voice guidance instruction may be expressed in the form of "Follow the main lane through the next intersection and stay in your current lane."

[0286] Finally, also consider the case where only one intersection is detected between the vehicle's position and the next maneuver to be performed, and no road change is planned, but a lane change is required in consideration of the next maneuver to be performed. In this situation, the processing module 25 is programmed to generate voice guidance instructions that also depend on the curvature of the road.

[0287] Thereby, when the curvature radius of the road is greater than a predetermined threshold, a voice guidance instruction is formed by combining at least one action, one geographic indicator, one attribute, one purpose, and one geographic indicator linked with another action, familiarity criterion, and goal. The voice guidance instruction may be expressed in the form of "Go straight through the intersection, then enter lane X."

[0288] Conversely, if the curvature radius of the road is less than this threshold, the voice guidance instruction is formed by combining at least one action, one geographic indicator, one attribute, one objective, and one geographic indicator linked with another action, familiarity criterion, and objective. The voice guidance instruction may be expressed in the form of "Follow the main lane at the next intersection, then move into lane X."

[0289] When the distance between the vehicle's position and the next maneuver to be performed (or the time required to reach the maneuver's zone) is greater than a predetermined threshold (e.g., 2 km), the processing module 25 is programmed to integrate the notion of distance into the voice guidance instructions, which may be expressed in the form "keep straight for x km...".

[0290] In the so-called continuity phase, the voice guidance instructions may also mention elements of the road infrastructure (roundabouts, bridges, tunnels, crossroads) or distinctive or prominent places (train stations, gas stations, neon signs) that become visible (and are unique) when approaching the maneuver to be performed. Here again, two cases can be considered:

[0291] If the radius of curvature of the road is greater than a predetermined threshold, the voice guidance instructions may be expressed in the form "Continue straight to the gas station."

[0292] Conversely, if the road curvature radius is less than this threshold, the voice guidance instructions may be expressed in the form "Follow the main road to the next bridge."

[0293] If the same operation can be performed in multiple locations, the voice guidance instructions may be expressed in the form of "keep straight until the next instruction" or "keep straight until the end of the street" or "keep straight until you reach the tall building...".

Claims

1. 1. A method for generating personalized audio guidance instructions, comprising: - determining, by a computer (25), the route to be taken; - acquiring, by image acquisition means (10), at least one image of the environment of said individual; - processing said image in order to detect at least one object in said image and attributes characterizing said object; - determining and transmitting voice guidance instructions informing said individual how to perform maneuvers to follow said route; Equipped with a step of determining a complexity level of the operation is provided; In the determining step, the voice guidance instructions are formulated using cues formed from the attributes of the object only if the complexity level of the operation is higher than a first threshold; A method characterized by:

2. The complexity level is determined based on the following criteria: - the position of the vehicle relative to the zone in which said operation must be carried out; the type of road infrastructure present in said zone in which said operation must be carried out; - tension felt by said individual, - road traffic density, the number of objects detected in at least one portion of the acquired image, - personal knowledge of said road infrastructure; - the stage at which the vehicle is currently located, and If said road infrastructure is a roundabout, the following criteria must be met: - the angle between the person's direction of entry into the roundabout and the person's direction of exit from the roundabout; - the angle between at least two roads leading to said roundabout, - the number of roads leading to said roundabout; - the difference between the number of entrances to the roundabout and the number of exits from the roundabout, - the number of lanes present in the roundabout; and - the outer diameter of the rotary, is calculated according to at least one of The method of claim 1 .

3. 3. The generation method of claim 1 or 2, wherein if the cue is a visual cue visible to the individual, and the visual cue disappears from the environment visible to the individual between the moment the voice guidance instruction is transmitted vocally and the moment the operation must be performed, a step of generating new voice guidance instructions that do not use the visual cue is performed.

4. A generation method as described in any one of claims 1 to 3, wherein if the cue is a visual cue visible to the individual, and if another visual cue indistinguishable from the visual cue used in the voice guidance instruction appears in the environment visible to the individual between the moment the voice guidance instruction is transmitted vocally and the moment the operation must be performed, a step is performed of generating new voice guidance instructions different from the voice guidance instructions transmitted vocally.

5. 5. The method of claim 1, wherein if the complexity level of the operation is higher than a second threshold that is strictly higher than the first threshold, the voice guidance instructions are generated using two cues.

6. 6. A method according to claim 1, wherein a cost relating to additional time or distance caused by an error in following the route is determined, and the complexity level is determined in response to said cost.

7. At least two separate and consecutive stages are distinguished prior to the execution of the operation: an operation anticipation stage, in which voice guidance instructions can be sent to the individual to indicate how they should position themselves to perform the operation, and an explanation stage, in which voice guidance instructions are sent to the individual to indicate how they should perform the operation, and the voice guidance instructions sent in these stages use distinct cues, and there is further made at least one other stage following the explanation stage, in which voice guidance instructions can be generated, and which other stage comprises: - a concomitant step simultaneously with said operation, or a reassurance stage after said operation, allowing said person to indicate whether they have correctly performed said operation, or a continuity step that enables the individual to be informed of the active characters of the production method; That is, A method for producing a liquid according to any one of claims 1 to 6.

8. The method of claim 7, wherein the voice guidance instructions transmitted during the explanation stage do not refer to any visual indicators, and if the vehicle slows down, voice guidance instructions using cues are transmitted in the accompanying stage.

9. 9. The method of claim 7 or 8, wherein in the reassurance stage, voice guidance instructions are transmitted using visual indicators that are distinct from one or more visual cues used in one or more voice guidance instructions transmitted during the preceding stage.

10. 10. A method according to any one of claims 7 to 9, wherein in the continuity stage different voice guidance instructions are generated depending on whether the radius of curvature of the road used is below or above a threshold value.

11. 11. The method of claim 1, wherein the step of determining, by a computer (25), the route to be taken comprises retrieving static visual indicators located on the route and attributes associated with the static visual indicators from a map database (20), and wherein attributes common to a plurality of static visual indicators are determined, and when the object is characterized as having the common attribute in the step of processing the image, the voice guidance instructions are formulated in the determination step using cues having the common attribute.

12. The method of claim 11 , wherein the static visual indicator is a road sign, the common attribute comprises a name written on a road sign or a background color of the road sign, and the object is a road sign.

13. 13. The method of claim 11 or 12, wherein the step of processing the image comprises detecting a plurality of objects, and the method further comprises a step of classifying the detected objects, wherein objects characterized as having the common attribute are prioritized, and the object selected for the determination step is the prioritized object.

14. 1. A device for generating environmentally tailored route guidance instructions, comprising: means (10) for acquiring an image of said environment; - a map database (20), a navigation and geolocation system (22), a processing module (25) suitable for receiving and combining elements from said acquisition means (10), from said map database (20) and from said navigation and geolocation system (22); a voice transmitter (11) suitable for transmitting instructions to the individual; Equipped with 14. The processing module (25) is programmed to implement a method according to any one of claims 1 to 13. device.

Images