EP4683848A1

EP4683848A1 - Driver-assistance system of a motor vehicle, and corresponding method

Info

Publication number: EP4683848A1
Application number: EP24708896.6A
Authority: EP
Inventors: Marco Andreetto; Marco Darin
Original assignee: Centro Ricerche Fiat SCpA
Current assignee: Centro Ricerche Fiat SCpA
Priority date: 2023-03-20
Filing date: 2024-02-28
Publication date: 2026-01-28
Also published as: WO2024194707A1

Abstract

A driver-assistance system of a motor vehicle (2) comprises one or more stationary optical reference markers (9a, 9b, 9c, 9d, 11a, 11b, 12a, 12b) that can be positioned by a user in a manoeuvring area (10), and a 5 motor vehicle (2). The motor vehicle includes an optical sensor (7) such as a video camera, one or more driving systems (3), a user interface (4), and an electronic control unit (5). The optical sensor (7) is configured to detect a scene (8) around the motor vehicle and produce data that represent the scene around the motor vehicle. The electronic control unit (5) is configured 10 to: - receive first data from the optical sensor (7) during a first manual execution of a manoeuvre, and process the first data to determine a set of suggested positions for the markers in the manoeuvring area (10); - indicate to the user, via the user interface (4), the set of suggested 15 positions for the markers; - receive second data produced by the optical sensor (7), and process the second data received via an object-recognition algorithm to detect the presence of the reference markers; - store data indicative of the geometrical configuration of the 20 manoeuvring area (10) based on the detected position of the reference markers in the second data received from the optical sensor during a second manual execution of the manoeuvre, wherein the data indicative of the geometrical configuration of the manoeuvring area (10) constitute a map indicating the areas accessible to the motor vehicle and the areas not 25 accessible to the motor vehicle; and - impart commands to the driving systems (3) of the motor vehicle (2) for controlling the subsequent assisted or automatic repetition of the manoeuvre based on the stored geometrical configuration and on the optical detection of the reference markers.

Description

“Driver-assistance system of a motor vehicle, and corresponding method”

****

TEXT OF THE DESCRIPTION

Field of the invention

The present invention relates in general to driver-assistance systems for motor vehicles. In particular, the invention regards a driver-assistance system that can be used both in a conventional motor vehicle (with assisted driving function) and in a motor vehicle with automated driving function, where the motor vehicle comprises: at least one optical sensor (e.g., a video camera) configured for detecting the scene around the motor vehicle, and an electronic control unit configured for imparting driving commands to one or more driving systems of the motor vehicle in order to make a manoeuvre as a function of the data produced by the optical sensor (e.g., image data produced by the video camera).

Prior art

Known in the art, for example from documents EP 2136275 B1 and EP 3586211 B1 filed in the name of the present applicant, are driverassistance systems for motor vehicles based on viewing systems (e.g., photographic cameras or video cameras), sensor systems (e.g., comprising radar and/or LiDAR sensors), automotive data networks, and V2V (vehicle- to-vehicle) or V2I (vehicle-to-infrastructure) wireless communication systems.

Such driver-assistance systems include data-processing and/or data-fusion algorithms for processing and/or aggregating the data collected by the sensors of the vehicle, for example for processing the image data collected by a viewing system and/or the distance data collected by one or more range sensors, such as ultrasound sensors, LiDAR sensors, and radar sensors. In particular, systems based only on data produced by image processing advantageously prove to be inexpensive in so far as they can be applied to motor vehicles equipped (only) with one or more video cameras (which are rather inexpensive) and without a more complex and sophisticated sensor system, such as radar sensors and/or LiDAR sensors (which are usually rather costly). Document US 2018/0157267 A1 describes a vehicle that comprises a parking-assistance system based on data from a video camera of the vehicle. The vehicle includes an image-capture device and a user display configured to display images received from the image-capture device. The vehicle includes a controller programmed to generate a user prompt to set a home position of the vehicle in response to the vehicle entering a first parked state. The controller is programmed to store a reference image indicative of an area in a vicinity of the vehicle corresponding to the home position of the vehicle. The controller is programmed to collect a current image corresponding to a current position of the vehicle in response to a subsequent approach toward the vicinity of the home position, and compare the current image to the reference image. The controller is programmed to represent on the user display the vicinity, the current position of the vehicle, and the home position of the vehicle.

Also document US 2010/0066515 A1 describes a vehicle that comprises a parking-assistance system based on data from a video camera of the vehicle. An image of a mark is taken by a camera when the vehicle is positioned in the vicinity of a parking space. An image-processing unit extracts a characteristic point from the image of the mark and recognizes two-dimensional coordinates on the image, a positional parameter calculation unit calculates positional parameters including six parameters of the camera with reference to the mark, and a relative position identification unit identifies a relative positional relation between the vehicle and the parking space. A parking-locus calculation unit calculates a parking locus for leading the vehicle to the parking space based on the relative positional relation between the vehicle and the parking space identified by the relative position identification unit, and a guide apparatus outputs drive operation guide information for causing the vehicle to travel along the calculated parking locus.

However, driver-assistance systems based only on cameras are in general less reliable and may not be sufficiently precise to allow the vehicle to make a manoeuvre (such as parking in a private or in any case controlled area) in an altogether autonomous way.

Object of the invention The object of the present invention is to provide a driver-assistance system of a motor vehicle that makes use of data detected by one or more optical sensors of the vehicle, for example, for autonomous or automatic execution of a manoeuvre in a private area, that will be more precise and reliable as compared to known systems. Such optical sensors may comprise one or more video cameras of the vehicle. In addition or as an alternative, such optical sensors may comprise other sensors sensitive to different wavelengths, such as infrared LiDAR sensors.

Summary of the invention

According to a first aspect, a subject of the invention is a driverassistance system of a motor vehicle. The system comprises one or more stationary optical reference markers (e.g., each constituted by a graphic symbol of pre-set dimensions and/or shape made on a supporting substrate) that can be positioned by a user in a manoeuvring area, and a motor vehicle. The motor vehicle comprises an optical sensor (e.g., a video camera), one or more driving systems, a user interface, and an electronic control unit coupled to the optical sensor, to the driving systems, and to the user interface. The optical sensor is configured to detect a scene around the motor vehicle and produce data (e.g., image data) representing the scene around the motor vehicle. The electronic control unit is configured to:

- receive first data from the optical sensor during a first manual execution of a manoeuvre, and process the first data to determine a set of suggested positions for the stationary optical reference markers in the manoeuvring area;

- indicate to the user, via the user interface, the set of suggested positions for the stationary optical reference markers;

- receive second data from the optical sensor, and process the second data received from the optical sensor via an object-recognition algorithm to detect the presence of the stationary optical reference markers;

- store data indicative of the geometrical configuration of the manoeuvring area based on the detected position of the optical reference markers in the second data received from the optical sensor during a second manual execution of the manoeuvre, where the data indicative of the geometrical configuration of the manoeuvring area constitute a map indicative of the areas accessible to the motor vehicle and the areas not accessible to the motor vehicle; and

- impart commands to the driving systems of the motor vehicle for controlling the subsequent assisted or automatic repetition of the manoeuvre based on the stored geometrical configuration and the optical detection of the stationary optical reference markers.

As will emerge in greater detail from the ensuing description, the fundamental idea underlying the present invention is to provide the user of the vehicle (e.g., at the moment of purchase of the vehicle or even subsequently) with one or more objects that function as optical reference points for the driver-assistance system of the vehicle. The driver-assistance system is programmed for recognizing the aforesaid reference points by processing the data captured by the optical sensor of the vehicle, and for issuing commands to the driving systems of the vehicle to control the path of movement thereof as a function of the above reference points. Advantageously, the objects used as optical references may be positioned by the user himself or herself, as desired, in a given area in which the vehicle has to carry out a recurrent manoeuvre, and the vehicle suggests to the user one or more positions in which to position the markers in order to minimize the number thereof and/or improve the accuracy of localization. For instance, the user can position one or more optical markers in various positions of a private garage to provide the vehicle with a set of easily recognizable and reliable reference points for execution of a parking manoeuvre. In this way, the vehicle is able to carry out an even complex manoeuvre in an autonomous way, albeit basing its operation only on data produced by an optical sensor (such as a video camera), and thus without any need for costly sensors.

According to a preferred embodiment, processing of the first data to determine a set of suggested positions for the reference markers comprises determining a minimum set of suggested positions that is such that, at each moment during execution of the manoeuvre, at least one of the reference markers is located in the field of view of the optical sensor.

According to a preferred embodiment, processing of the first data to determine a set of suggested positions for the reference markers comprises determining a minimum set of suggested positions that is such that, at each moment during execution of the manoeuvre, an uncertainty in the estimate of the position of the motor vehicle made by the electronic control unit is lower than a certain threshold.

According to a preferred embodiment, the minimum set of suggested positions is such that: in a rectilinear stretch during execution of the manoeuvre, no reference marker is located in the field of view of the optical sensor, and/or in a curvilinear stretch during execution of the manoeuvre, at least two reference markers are located in the field of view of the optical sensor.

According to a preferred embodiment, the user interface is configured for indicating to the user the set of suggested positions for the reference markers by displaying on a screen the first data received from the optical sensor during the first manual execution of the manoeuvre and overlaying thereon an indication (e.g., display) of the suggested positions.

According to a preferred embodiment, the user interface is configured for indicating to the user the set of suggested positions for the reference markers by displaying on a screen a map of the manoeuvring area and overlaying thereon an indication (e.g., display) of the suggested positions.

According to a preferred embodiment, the user interface is configured for indicating to the user the set of suggested positions for the reference markers by displaying on a screen an expected path of the manoeuvre and overlaying thereon an indication (e.g., display) of the suggested positions.

According to a preferred embodiment, the user interface is configured for displaying on a screen the second data received from the optical sensor during the second manual execution of the manoeuvre and providing the user with input means for indicating the presence of one or more reference markers in the second data.

According to a preferred embodiment, the input means comprise:

- input means for confirming or rejecting one or more reference markers detected by the electronic control unit via the object-recognition algorithm; and/or

- input means for indicating the presence of one or more reference markers that have not been detected by the electronic control unit via the object-recognition algorithm.

According to another aspect, a subject of the invention is a driver- assistance method of a motor vehicle. The method comprises the steps of:

- detecting, via an optical sensor of the motor vehicle, a scene around the motor vehicle and producing data representative of the scene around the motor vehicle;

- receiving, in an electronic control unit of the motor vehicle, first data produced by the optical sensor during a first manual execution of a manoeuvre;

- processing the first data to determine a set of suggested positions for one or more stationary optical reference markers in a manoeuvring area;

- indicating to a user, via a user interface of the motor vehicle, the set of suggested positions for the reference markers;

- positioning the reference markers in the suggested positions in the manoeuvring area;

- receiving, in the electronic control unit of the motor vehicle, second data produced by the optical sensor of the motor vehicle during a second manual execution of the manoeuvre;

- processing, in the electronic control unit of the motor vehicle, the second data received from the optical sensor via an object-recognition algorithm to detect the presence of the reference markers;

- storing, in a memory area of the electronic control unit of the motor vehicle, data indicating the geometrical configuration of a manoeuvring area based on the detected position of the reference markers in the second data received from the optical sensor during the second manual execution of the manoeuvre, where the data indicating the geometrical configuration of the manoeuvring area constitute a map indicating the areas accessible to the motor vehicle and the areas not accessible to the motor vehicle; and

- imparting commands to the driving systems of the motor vehicle, via the electronic control unit of the motor vehicle, for controlling the subsequent assisted or automatic repetition of the manoeuvre based on the stored geometrical configuration and the optical detection of the reference markers.

Detailed description of the invention

Further characteristics and advantages of the invention will emerge from the ensuing description with reference to the annexed drawings, which are provided purely by way of non-limiting example and in which:

- Figure 1 illustrates a block diagram of an automotive assisted or autonomous driving system;

- Figure 2 is a top plan view that shows a possible scenario of application of the present invention, where a vehicle recognizes the presence of an optical reference object (or optical marker) positioned by a user for localization in the environment and/or mapping of the environment;

- Figure 3 is a top plan view that shows a possible scenario of application of the present invention, where a vehicle recognizes the presence of a number of optical reference objects positioned by a user for localization in the environment and/or mapping of the environment;

- Figure 4 is a block diagram that shows a method of positioning the optical reference objects according to embodiments of the present invention;

- Figure 5 is a block diagram that shows a driver-assistance method for a motor vehicle according to embodiments of the present invention;

- Figures 6A and 6B are a perspective view and a top plan view, respectively, of a possible scenario of application of the present invention, where a vehicle recognizes the presence of a number of optical reference objects positioned by a user for execution of a manoeuvre in assisted or autonomous driving mode; and

- Figures 7A, 7B, and 7C are two perspective views and a top plan view, respectively, of a possible scenario of application of the present invention, where a vehicle recognizes the presence of a number of optical reference objects positioned by a user for execution of a manoeuvre in assisted or autonomous driving mode.

In the figures annexed hereto, corresponding parts are designated by the same reference numbers.

As anticipated, one or more embodiments find application in the field of vehicles with automated or assisted driving function provided with at least one optical sensor, where the path of movement of the vehicle and/or its capacity of localization in the environment are controlled by an electronic control unit of the vehicle as a function of the data produced by the optical sensor.

In the sequel of the present detailed description, reference will be made mainly to a video camera as optical sensor with which the vehicle may be equipped, and to the processing of image data produced by the video camera. However, one or more embodiments of the present invention may be applied to vehicles provided with optical sensors additional to or different from a video camera, such as a LiDAR sensor sensitive to a radiation having a certain wavelength (e.g., infrared radiation).

In this context, Figure 1 illustrates a block diagram of an automotive assisted or autonomous driving system, designated as a whole by the number 1 , designed to get a motor vehicle, designated by the reference number 2, to perform manoeuvres in assisted (e.g., semiautonomous) or autonomous driving mode. As illustrated in Figure 1 , the automotive assisted driving system 1 or autonomous driving system comprises:

- on-board motor-vehicle systems 3 (in particular, driving systems of the vehicle 2) that comprise, for example, a propulsion system, a braking system, a steering system, an infotainment system, and a sensor system designed to detect quantities regarding the motor vehicle 2 such as wheel angle, steering-wheel angle, yaw, longitudinal and lateral acceleration, position, etc.;

- an automotive user interface (HMI - Human-Machine Interface) 4 through which the occupants of the motor vehicle 2 can interact with the automotive assisted driving system 1 ; and

- an electronic control unit 5 (ECU, for instance a controller) operatively connected to the automotive on-board systems 3 and to the automotive user interface 4 through an automotive on-board communication network 6, for example CAN, FlexRay, etc.

The invention described herein may be applied in the case where the vehicle 2 is configured to carry out in assisted or autonomous driving mode one or more recurrent and low-speed complex manoeuvres, as described in the document EP 3586211 B1 cited previously. In such cases, the electronic control unit 5 of the vehicle 2 is configured for: identifying what are the recurrent and low-speed complex manoeuvres; localizing the vehicle 2 within the environment in which the aforesaid recurrent manoeuvres are performed; and repeating such manoeuvres in assisted or autonomous driving mode. In particular, localization of the vehicle 2 in the environment (which may be a private or in any case controlled area) may be carried out using algorithms of a SLAM (Simultaneous Localization And Mapping) type, which are in themselves known and which enable the vehicle to create and store a virtual representation of the environment, i.e. , a sort of map of an area that may possibly not be covered by digital road maps. As is known in the art, SLAM algorithms seek significant reference points in the surrounding environment that are to be used as landmarks for localization of the vehicle 2 within the map. Typically, such reference points are objects or elements that can easily be distinguished from the background of the image.

Figure 2 is a schematic representation of the concept underlying the present invention and is a top plan view of a motor vehicle 2 equipped with a video camera 7 (in this case a frontal camera, even though it does not necessarily have to be a frontal camera), which is configured for detecting the scene around the motor vehicle 2 in the cone of vision 8 (e.g., the scene ahead of the vehicle) and producing image data that represent the aforesaid scene, the image data being sent to the electronic control unit 5, for example via the vehicle communication network 6. In the example of Figure 2, the vehicle 2 is in a private or in any case controlled area 10, where a user (e.g., the driver or owner of the motor vehicle) has positioned an optical reference object or marker 9 (here represented in a front view, as detected by the video camera 7, and not in top plan view) that can be recognized by the vehicle 2 via an image-processing algorithm applied to the images captured by the video camera 7. As illustrated in Figure 3, in the same manoeuvring area 10 there could be positioned by the user a plurality of markers 9a, 9b, 9c, 9d that can be recognized (possibly, being distinguished from one another) by the vehicle 2 (which also in Figure 3 are represented in a front view, as they would be detected by the video camera 7, and not in top plan view). The electronic control unit 5 receives the image data from the video camera 7 of the vehicle 2 and processes these data via an objectrecognition algorithm to detect the presence of the markers 9a, 9b, 9c, 9d. During at least one first manual execution of a manoeuvre (i.e. , an execution in which the driving commands are imparted by the driver) in the area 10, the electronic control unit 5 stores a set of data that are indicative (or descriptive) of the geometrical configuration of the manoeuvering area 10 on the basis of the detected position of the markers 9a, 9b, 9c, 9d in the image data. Consequently, the vehicle 2 is able to store a map indicating the areas accessible to the motor vehicle and the areas not accessible to the motor vehicle in so far as it is able to associate the driving commands for execution of the manoeuvre to the detected positions of the markers 9a, 9b, 9c, 9d, as will emerge more clearly from the further ensuing examples. Once the data that describe the geometrical conformation of the manoeuvring area 10 have been stored, the electronic control unit 5 can hence impart commands to the driving systems of the motor vehicle 2 to control the subsequent automatic repetition of the manoeuvre (e.g., parking in a private garage) based on the stored map and the optical detection of the markers 9a, 9b, 9c, 9d. In other words, the reference points or markers 9a, 9b, 9c, 9d positioned by the user in the area of interest to the user provide stable references that are easy to detect for localization and execution of manoeuvres in assisted or autonomous driving mode, without the need to provide in the vehicle 2 a set of costly sensors or a complex and costly processing and/or data-fusion platform.

As illustrated in the block diagram of Figure 4, the method 40 for positioning the markers 9a, 9b, 9c, 9d is based on a co-operative interaction between the user and the vehicle.

In particular, in a step 402, during a first manual execution of the manoeuvre, the video camera 7 captures one or more images of the environment in which the manoeuvre is performed (e.g., a private garage).

In a step 404, the electronic control unit 5 analyzes the images captured by the video camera in relation to the manoeuvre executed and determines one or more positions identified as “optimal” for positioning of the markers 9a, 9b, 9c, 9d. The electronic control unit 5 implements an algorithm to determine the suggested positions for the markers 9a, 9b, 9c, 9d. The criteria used by the vehicle for proposing positioning of the markers 9a, 9b, 9c, 9d may have the aim of reducing (e.g., minimizing) the number of markers that the user has to position, in any case maintaining a sufficient support to the localization algorithm implemented by the vehicle 2. In particular, one of the following two criteria (or a combination thereof), which have the aim of minimizing or eliminating the odometry drift, may be used:

- given the estimated path that the vehicle has to cover to carry out the manoeuvre (learnt during the first manual execution by the driver), the electronic control unit 5 determines what is the positioning of the markers that reduces (minimizes) the number of markers at the same guaranteeing that the vehicle always has at least one marker in sight;

- given the estimated path that the vehicle has to cover, the electronic control unit 5 determines the number and position of the markers necessary for the uncertainty of estimation of localization to remain below a warning threshold; for example, the vehicle will suggest positioning a larger number of markers in the proximity of a bend in the road (in so far as a greater uncertainty accumulates), whereas it could tolerate the absence of markers in the sight of the vehicle for short rectilinear stretches in so far as it calculates as manageable (i.e. , lower than the threshold) the increase in uncertainty caused by the odometry drift.

In a step 406, the optimal positions for the markers 9a, 9b, 9c, 9d are suggested to the user via the interface 4. For instance, there may appear on a display of the interface 4 one (or more) of the images captured by the video camera 7 during the first execution of the manoeuvre, superimposed on which is the indication of the position (or positions) in which the vehicle suggests positioning of one or more markers (e.g., showing the aforesaid position in the image with a coloured area or an appropriate graphic symbol). In addition or as an alternative, there may appear on the display of the interface 4 a map of the environment generated by the vehicle, superimposed on which is the indication of the position (or positions) in which the vehicle suggests positioning of one or more markers. Once again in addition or as an alternative, there may appear on the display of the interface 4 an estimate of the path that the vehicle has to cover to carry out the manoeuvre, superimposed on which is the indication of one or more points of the path in which the vehicle suggests positioning of one or more markers.

In a step 408, the user positions the markers 9a, 9b, 9c, 9d in the positions suggested by the vehicle, for the purpose of a subsequent recognition of such markers via the video camera 7 and the electronic control unit 5 during a subsequent execution of the manoeuvre in assisted or autonomous driving mode.

As represented in the block diagram of Figure 5, the method 50 for execution of the manoeuvre in assisted or autonomous driving mode may be based on a co-operative interaction between the user and the vehicle.

In particular, in a step 502, during assisted or autonomous execution of the manoeuvre, the video camera 7 captures images of the surrounding environment, and the electronic control unit 5 applies object-recognition processing to these images to identify the markers 9a, 9b, 9c, 9d previously placed by the user (according to the indications supplied by the vehicle itself in a previous set-up stage).

In a step 504, the user interface 4 of the vehicle 2 provides the occupant of the vehicle with the possibility of “helping” the vehicle in recognizing the presence of the markers, above all in the case where the vehicle is unable to identify them (e.g., on account of poor ambient light). For instance, the image captured in real time by the video camera 7 may appear on the display of the interface 4, and the user can indicate in this image the presence of one of the markers 9a, 9b, 9c, 9d. For instance, the display shows the image in real time, and indicates in this image: the markers identified with a fair degree of certainty (i.e. , with a confidence level higher than a given threshold) by the vehicle so that no action is required to the user; and/or the markers identified by the vehicle with a certain degree of uncertainty (i.e., with a confidence level lower than the threshold), for which the user can act on the interface 4 to confirm/reject the marker identified. In certain cases, the interface 4 can show the image in real time without the indication of any marker (whether “certain” or “uncertain”), and the user can indicate the presence of a marker in the image even though this has not been recognized in any way by the vehicle 2.

In a step 506, the electronic control unit 5 determines the set of operations to carry out to perform the desired manoeuvre (e.g., localization of the vehicle and planning of the path, also known as “localization and planning”), based on recognition of the markers made in step 504.

In a step 508, the driving systems 3 of the vehicle 2 are controlled by the electronic control unit 5 to carry out the operations determined in step 506.

Advantageously, the optical reference objects 9a, 9b, 9c, 9d may be particularly inexpensive. For instance, they may each be constituted by a graphic symbol provided on a substrate. In the simplest case, the graphic symbols may be printed on a sheet of paper or plastic material, possibly adhesive on the back, in such a way that the user can easily affix them in the positions of interest.

According to an advantageous characteristic, the graphic symbols of the markers 9a, 9b, 9c, 9d may have a pre-set shape and/or size. The data on the shape and size of the graphic symbols are stored in the electronic control unit 5 of the vehicle 2 in such a way that the electronic control unit 5 can process the images detected by the video camera 7 and use these data to improve the precision of localization in the manoeuvring area 10. For instance, the electronic control unit 5 can apply a given algorithm that determines the distance between the motor vehicle 2 and a given optical marker 9 as a function of the size of the marker as displayed in the image supplied by the video camera 7 and of the known information on its actual size (the smaller the size of the marker 9 in the image captured, the greater the distance of the vehicle 2 from that marker). In addition or as an alternative, the electronic control unit 5 can apply a given algorithm that determines the orientation of the motor vehicle 2 with respect to a given optical marker 9 as a function of the shape of the marker as displayed in the image supplied by the video camera 7 and of the known information on its actual shape. For instance, if the markers 9 are square, as a result of the perspective, the marker displayed in the image of the video camera 7 will be all the more “deformed” in the form of a trapezium, the greater the angle of deviation between the direction perpendicular to the marker 9 and the direction of the optical axis of the video camera 7.

In various embodiments, the graphic symbol represented on each marker 9 may be used to encode optically-detectable information. For instance, as illustrated in the figures attached hereto, each graphic symbol may comprise visual elements with high contrast, such as a QR code or a bar code, optionally containing an identifier code of each graphic symbol (i.e., each marker). The electronic control unit 5 may consequently be configured for decoding the optically-detectable information carried by the graphic symbols 9 and imparting commands to the driving systems 3 of the motor vehicle 2 as a function of the decoded information.

Operation of a driver-assistance system according to the present invention will emerge more clearly from the examples illustrated in the Figures 6A, 6B, 7A, 7B, and 7C that follow. Figures 6A and 6B represent a possible scenario of use of the driving system according to the invention, where a user of the vehicle 2 has affixed a first marker 11 a on a wall alongside the entrance of his or her own garage, and a second marker 11 b on an end wall of the garage (Figure 6A: perspective view as captured via the video camera 7; Figure 6B: top plan view, except for the markers 11 a, 11 b, represented in front view). During a first set-up stage, the user executes the manoeuvre manually, and the vehicle determines the optimal positions of the markers, which the user will position accordingly. During manual execution of the parking manoeuvre of vehicle 2 in the garage, the assisted-driving system easily detects the presence of the markers 11a, 11 b via the video camera 7 and stores a correlation between the detected position of the markers 11 a, 11 b and the driving commands made by the driver for execution of the manoeuvre. During a subsequent autonomous execution of the parking manoeuvre, the motor vehicle 2 can repeat the same sequence of driving commands as a function of optical detection of the markers 11 a, 11 b. In addition or as an alternative, the markers 11a, 11 b can encode information content that can be decoded by the vehicle 2. For instance, marker 11 a can transmit information that says “keep to the left” and marker 11 b can transmit information that says “stop here” in such a way that the electronic control unit 5 can impart driving commands to the driving systems 3 of the vehicle 2 also as a function of the information content of the markers 11 a, 11 b. Figure 6B exemplifies the map that the vehicle 2 is able to store after the first manual execution of the manoeuvre in the area 10, the map comprising information on the position of the markers 11 a, 11 b.

Figures 7A, 7B, and 7C represent another possible scenario of use of the driving system according to the invention, where a user of the vehicle 2 has affixed a first marker 12a on an end wall of a manoeuvring lane of a private garage space in which the particular garage of the user is located and a second marker 12b on an end wall of the garage itself (Figures 7A and 7B: perspective views as captured via the video camera 7; Figure 7C: top plan view, except for the markers 12a, 12b, which are represented in front view). Also in this case, during a first set-up stage, the user executes the manoeuvre manually, and the vehicle determines the optimal positions of the markers, which the user will position accordingly. During a first manual execution of the parking manoeuvre of vehicle 2 in its garage, the assisted- driving system detects the presence of the markers 12a, 12b via the video camera 7 and stores a correlation between the detected position of the markers 12a, 12b and the driving commands made by the driver for execution of the manoeuvre. During a subsequent autonomous execution of the parking manoeuvre, the motor vehicle 2 can repeat the same sequence of driving commands as a function of the optical detection of the markers 12a, 12b. In addition or as an alternative, the markers 12a, 12b may encode information content that can be decoded by the vehicle 2. For instance, marker 12a may transmit information that says “turn right here”, and marker 12b may transmit information that says “stop here” in such a way that the electronic control unit 5 can impart driving commands to the driving systems 3 of the vehicle 2 also as a function of the information content of markers 12a, 12b. Figure 7C exemplifies the map that the vehicle 2 is able to store, after the first manual execution of the manoeuvre in the area 10, the map comprising information on the position of the markers 12a, 12b.

In various embodiments, it is envisaged that the information content encoded by each graphic symbol appearing on each marker (9, 11 , or 12) will basically correspond to an identifier code of the marker itself. In this case, it may be envisaged that the user of the vehicle 2 can configure the response of the vehicle 2 to recognition of a certain marker.

The invention described herein consequently proves advantageous in so far as it enables implementation of a mapping function based on markers (reference points) that can be readily recognized by the algorithm of the vehicle 2, and makes it possible to close the control loop for execution of manoeuvres in assisted or automatic driving mode on the basis of the markers (and possibly their information content) instead of on the basis of scan-matching techniques, thus requiring a minimal computational capacity from the electronic control unit 5 and possibly without any need to resort to complex and costly sensors.

As anticipated at the beginning of the present detailed description, the driver-assistance system described herein may advantageously be applied also to an optical sensor other than a video camera, or may be used in addition to other more complex sensor systems installed on board the motor vehicle 2, implementing a data-fusion algorithm on the data produced by a number of sensor systems of the motor vehicle 2.

For instance, the motor vehicle 2 may be equipped with a LiDAR sensor sensitive to electromagnetic radiation having a given wavelength (e.g., in the infrared, IR), and the optical markers could be made of a material such as to reflect the electromagnetic radiation to which the LiDAR sensor is sensitive. In this way, the optical markers could function as optical positioning references also for the LiDAR sensor.

Consequently, in one or more embodiments there could be present also other sensors in addition to the video camera 7 of the motor vehicle 2, or the video camera 7 could even be optional. In fact, the use of more complex scan-matching techniques can be facilitated by the optical markers both in the case of direct use (i.e. , when the optical markers can be directly detected by a LiDAR sensor) and in the case of indirect use (i.e., when the video camera detects the optical markers and communicates to the LiDAR sensor the position where these optical markers are located).

Moreover, also in the case where the optical markers can be detected by the video camera 7 alone of the motor vehicle 2, the information on the presence of the optical markers may in any case facilitate the tasks attributed to the other sensors, for example the creation of an occupancygrid map via range sensors.

Of course, without prejudice to the principle of the invention, the details of construction and the embodiments may vary widely with respect to what has been described and illustrated herein purely by way of example, without thereby departing from the scope of the present invention, as defined in the annexed claims.

Claims

1. A driver-assistance system of a motor vehicle (2), comprising:

- one or more stationary optical reference markers (9a, 9b, 9c, 9d, 11 a, 11 b, 12a, 12b) that can be positioned by a user in a manoeuvring area (10); and

- a motor vehicle (2) comprising an optical sensor (7), one or more driving systems (3) of the motor vehicle, a user interface (4), and an electronic control unit (5) coupled to said optical sensor (7), to said one or more driving systems (3) of the motor vehicle and to said user interface (4); wherein said optical sensor (7) is configured to detect a scene (8) around the motor vehicle and produce data that represent the scene around the motor vehicle, and wherein said electronic control unit (5) is configured to:

- receive (402) first data from said optical sensor (7) during a first manual execution of a manoeuvre, and process (404) said first data to determine a set of suggested positions for said one or more stationary optical markers (9a, 9b, 9c, 9d, 11 a, 11 b, 12a, 12b) in said manoeuvring area (10);

- indicate (406) to the user, via said user interface (4), said set of suggested positions for said one or more stationary optical markers (9a, 9b, 9c, 9d, 11a, 11 b, 12a, 12b);

- receive (502) second data from said optical sensor (7) during a second manual execution of said manoeuvre, and process (502) said second data via an object-recognition algorithm to detect the presence of said one or more stationary optical markers (9a, 9b, 9c, 9d, 11a, 11 b, 12a, 12b);

- store data indicative of the geometrical configuration of said manoeuvring area (10) based on the detected position of said one or more stationary optical markers (9a, 9b, 9c, 9d, 11 a, 11 b, 12a, 12b) in said second data received from said optical sensor (7), wherein said data indicative of the geometrical configuration of the manoeuvring area (10) constitute a map indicating the areas accessible to the motor vehicle and the areas not accessible to the motor vehicle; and

- impart commands to said one or more driving systems (3) of the motor vehicle (2) for controlling the subsequent assisted or automatic repetition of said manoeuvre based on said stored geometrical configuration and on the optical detection of said one or more stationary optical markers (9a, 9b, 9c, 9d, 11 a, 11 b, 12a, 12b).

2. The driver-assistance system of a motor vehicle (2) according to claim 1 , wherein the step of processing (404) said first data to determine a set of suggested positions for said one or more stationary optical markers (9a, 9b, 9c, 9d, 11a, 11 b, 12a, 12b) comprises determining a minimum set of suggested positions such that, at any moment during execution of said manoeuvre, at least one of said stationary optical markers (9a, 9b, 9c, 9d, 11 a, 11 b, 12a, 12b) is located in a field of view of said optical sensor (7).

3. The driver-assistance system of a motor vehicle (2) according to claim 1 or claim 2, wherein the step of processing (404) said first data to determine a set of suggested positions for said one or more stationary optical markers (9a, 9b, 9c, 9d, 11 a, 11 b, 12a, 12b) comprises determining a minimum set of suggested positions such that, at any moment during execution of said manoeuvre, an uncertainty in the estimate of the position of said motor vehicle (2) made by said electronic control unit (5) is lower than a certain threshold.

4. The driver-assistance system of a motor vehicle (2) according to claim 3, wherein said minimum set of suggested positions is such that:

- at a rectilinear stretch of said manoeuvre, no stationary optical marker (9a, 9b, 9c, 9d, 11 a, 11 b, 12a, 12b) is located in a field of view of said optical sensor (7); and/or

- at a curvilinear stretch of said manoeuvre, at least two of said stationary optical markers (9a, 9b, 9c, 9d, 11a, 11 b, 12a, 12b) are located in the field of view of said optical sensor (7).

5. The driver-assistance system of a motor vehicle (2) according to any of the previous claims, wherein said user interface (4) is configured to indicate (406) to the user said set of suggested positions for said one or more stationary optical markers (9a, 9b, 9c, 9d, 11a, 11 b, 12a, 12b) by displaying on a screen said first data received from said optical sensor (7) during the first manual execution of said manoeuvre and overlaying thereon an indication of said suggested positions.

6. The driver-assistance system of a motor vehicle (2) according to any of the previous claims, wherein said user interface (4) is configured to indicate (406) to the user said set of suggested positions for said one or more stationary optical markers (9a, 9b, 9c, 9d, 11a, 11 b, 12a, 12b) by displaying on a screen a map of said manoeuvring area (10) and overlaying thereon an indication of said suggested positions.

7. The driver-assistance system of a motor vehicle (2) according to any of the previous claims, wherein said user interface (4) is configured to indicate (406) to the user said set of suggested positions for said one or more stationary optical markers (9a, 9b, 9c, 9d, 11a, 11 b, 12a, 12b) by displaying on a screen an expected path of said manoeuvre and overlaying thereon an indication of said suggested positions.

8. The driver-assistance system of a motor vehicle (2) according to any of the previous claims, wherein said user interface (4) is configured to:

- display (504) on a screen said second data received from said optical sensor (7) during the second manual execution of said manoeuvre; and

- provide the user with input means for indicating the presence of said one or more stationary optical markers (9a, 9b, 9c, 9d, 11a, 11 b, 12a, 12b) in said second data.

9. The driver-assistance system of a motor vehicle (2) according to claim 8, wherein said input means comprise:

- input means for confirming or rejecting one or more stationary optical markers (9a, 9b, 9c, 9d, 11a, 11 b, 12a, 12b) detected by the electronic control unit (5) via said object-recognition algorithm; and/or

- input means for indicating the presence of one or more stationary optical markers (9a, 9b, 9c, 9d, 11 a, 11 b, 12a, 12b) that have not been detected by the electronic control unit (5) via said object-recognition algorithm.

10. A driver-assistance method for a motor vehicle (2), the method comprising:

- detecting, via an optical sensor (7) of the motor vehicle (2), a scene (8) around the motor vehicle (2) and producing data that represent the scene around the motor vehicle;

- receiving (402), in an electronic control unit (5) of the motor vehicle (2), first data produced by said optical sensor (7) during a first manual execution of a manoeuvre;

- processing (404) said first data to determine a set of suggested positions for one or more stationary optical markers (9a, 9b, 9c, 9d, 11 a, 11 b, 12a, 12b) in a manoeuvring area (10);

- indicating (406) to a user, via a user interface (4) of the motor vehicle, said set of suggested positions for said one or more stationary optical markers (9a, 9b, 9c, 9d, 11 a, 11 b, 12a, 12b);

- positioning (408) said one or more stationary optical markers (9a, 9b, 9c, 9d, 11 a, 11 b, 12a, 12b) in said suggested positions in said manoeuvring area (10);

- receiving (502), in said electronic control unit (5) of the motor vehicle (2), second data produced by said optical sensor (7) of the motor vehicle (2) during a second manual execution of said manoeuvre;

- processing (502), in said electronic control unit (5) of the motor vehicle (2), said second data via an object-recognition algorithm to detect the presence of said one or more stationary optical markers (9a, 9b, 9c, 9d, 11 a, 11 b, 12a, 12b);

- storing, in a memory area of said electronic control unit (5) of the motor vehicle (2), data indicative of the geometrical configuration of said manoeuvring area (10) based on the detected position of said one or more stationary optical markers (9a, 9b, 9c, 9d, 11 a, 11 b, 12a, 12b) in said second data received from said optical sensor (7) during the second manual execution of said manoeuvre, wherein said data indicative of the geometrical configuration of the manoeuvring area (10) constitute a map indicating the areas accessible to the motor vehicle and the areas not accessible to the motor vehicle; and

- imparting commands to one or more driving systems (3) of the motor vehicle (2), by said electronic control unit (5) of the motor vehicle (2), for controlling the subsequent assisted or automatic repetition of said manoeuvre based on said stored geometrical configuration and on the optical detection of said one or more stationary optical markers (9a, 9b, 9c, 9d, 11 a, 11 b, 12a, 12b).