Automatic control method and system for virtually constraining land vehicle in track

Abstract

The invention relates to a method (900; 900 ') for automatically constraining a land vehicle. 1000) for a motor vehicle. In an embodiment of the invention, a method (900; 1000) provides a control unit (30) to carry out the following steps. A set of points of a track in which the vehicle is free to move is defined (900), the track comprising at least one edge defining a boundary of the track that must not be crossed by the vehicle. Vehicle dynamics and location are determined (1001) based on information provided by a plurality of sensors of the vehicle. An intervention area of the track is calculated (1005) based on the current position and dynamics of the vehicle and based on the position from the at least one edge, the area to be avoided extending between the at least one edge of the track towards the current position of the vehicle within the track. It is determined (1007) whether the vehicle is at least partially in the intervention area. In a positive case, at least one actuator is controlled (1013) to modify the vehicle dynamics to perform an exit maneuver to bring the vehicle out of the intervention area. In particular, calculating (1005) the intervention area includes performing the following steps. A first collision area is calculated (10055) based on the current position and dynamics of the vehicle, the position from the at least one side, and the ability of the driver to modify the vehicle dynamics by acting on the at least one actuator. A second collision area is calculated (10057) based on the current position and dynamics of the vehicle, the position from the at least one edge, and the ability to modify the vehicle dynamics of the control unit by acting on the at least one actuator. The first collision region and the second collision region correspond to a track region in which the at least one edge is inevitably reached and / or crossed when the vehicle moving according to the current dynamic state is driven by the driver or the control unit, respectively. Finally, the intervention region is defined (10058) as a region resulting from a difference between the first collision region and the second collision region.

Description

Technical Field

[0001] The present invention relates to the automotive industry. More particularly, the present invention relates to a method for virtually restraining a vehicle on a runway and a related automatic control system. Background Art

[0002] Currently, Electronic Stability Control (ESC) systems help the driver of a vehicle to maintain the trajectory set by him or her, eliminating or at least significantly reducing the risk of accidents.

[0003] Furthermore, automatic or semi-automatic driving systems are known which overlap the driver's control by driving the vehicle instead of the driver according to predetermined criteria, in particular in order to avoid collisions of the vehicle with obstacles on the track.

[0004] For example, US2015 / 0302751 proposes a driver assistance method that intervenes by controlling the braking and steering of a vehicle when an obstacle is identified by onboard sensors. Similarly, US 9637050 describes a driver assistance method that intervenes in steering and braking based on obstacles and information exchanged between vehicles and / or electronic infrastructure around the vehicle's track, i.e., V2V vehicle-vehicle and V2I vehicle-infrastructure communications.

[0005] US10882518 discloses a driver assistance method, which calculates a trajectory to avoid a collision between a vehicle and an obstacle, and forces the vehicle to follow such a trajectory through an automatic braking and steering control system. In addition, the method calculates the trajectory after the avoidance to determine whether there are other obstacles to be avoided in addition to the first obstacle.

[0006] Finally, JP 6654641 proposes an automated driving system for a vehicle which is activated upon detection of a predetermined condition - for example, a risk of collision with an obstacle - and which transfers control to the driver of the vehicle only if the driver's commands substantially correspond to commands generated by the automated driving system.

[0007] In addition, D. Hoehener et al.: “Design of a lane departure driver-assist system under safety specifications”, 2016, IEEE 55th Conference on Decision and Control (CDC), pp. 2468-2474 and US2017 / 183004 describe other autonomous or semi-autonomous driving systems that overlap with driver control.

[0008] Known solutions guarantee greater driving safety, but are not suitable for use in vehicles with essentially performance / entertainment purposes, such as in the case of driving on a racetrack. In fact, these methods and systems tend to overly restrict the driver's maneuverability of the vehicle. This limits the driver's possibilities to improve his or her own driving skills and to learn a sporty and / or safe driving style, and more generally, limits driving pleasure.

[0009] Therefore, there is a need in the art for driver assistance methods and systems that allow a driver to drive a vehicle without interfering with the driver's driving style (eg, a sporty driving style), while ensuring the safety of the driver and the vehicle. Summary of the invention

[0010] The purpose of the present invention is to overcome the shortcomings of the prior art

[0011] In particular, it is an object of the present invention to provide a method and a system that ensure the safety of the driver and the vehicle without interfering with the driver's driving style.

[0012] Specifically, another object of the present invention is to provide a method and related system for identifying vehicle dynamic conditions that cause the vehicle to leave an allowed track (such as a racetrack exit) and quickly correct the vehicle dynamic conditions to avoid the racetrack exit.

[0013] In fact, the Applicant has discovered that if the driver sets an incorrect trajectory, for example by delaying the apex too much, the known ESC systems are not able to ensure that the vehicle remains on the track.

[0014] These and other objects of the invention are achieved by a system incorporating the features of the appended claims, which form an integral part of the present description.

[0015] According to a first aspect, the invention relates to a method for automatically restraining a land vehicle. The vehicle under consideration comprises a control unit, a plurality of sensors configured to measure quantities indicative of the vehicle's dynamics, at least one actuator configured to influence the vehicle's dynamics, the plurality of sensors and the at least one actuator being connected to the control unit.

[0016] In an embodiment of the invention, the method provides for the control unit to define a set of points of a track within which the vehicle is free to move, said track comprising at least one edge defining a boundary of the track which must not be crossed by the vehicle. Preferably, the control unit periodically determines the vehicle dynamics and the spatial position of the vehicle based on information provided by a plurality of sensors.

[0017] Starting from the position and dynamics of the vehicle and from the position of at least one edge, the control unit calculates the intervention region.

[0018] The control unit then determines whether the vehicle is at least partially within the intervention zone and, in the affirmative, controls at least one actuator to modify the vehicle dynamics to perform an exit maneuver bringing the vehicle outside of said intervention zone.

[0019] The term "dynamics" herein refers to a collection of information related to the movement of a vehicle in space as a function of time. Preferably, but not limiting, the dynamics of a vehicle include the following information: longitudinal velocity, lateral velocity, yaw rate, heading, yaw angle, steering angle, and individual wheel speeds. Optionally, the vehicle dynamics also include information about roll and pitch velocities and suspension extension. It will be apparent to those skilled in the art that the information included in the vehicle dynamics enables the calculation of a plurality of derived information such as corresponding longitudinal, lateral and yaw acceleration values ​​associated with the vehicle.

[0020] In particular, calculating the intervention area comprises performing the following steps. A first collision area based on the current position and dynamics of the vehicle, the position from at least one edge and the ability to modify the vehicle dynamics of the driver by acting on the at least one actuator is calculated, as well as a second collision area based on the current position and dynamics of the vehicle, the position from at least one edge and the ability of the control unit to modify the vehicle dynamics acting on the at least one actuator. The first collision area and the second collision area correspond to track areas within which the at least one edge is inevitably reached and / or crossed when the vehicle moving according to the current dynamics is driven by the driver or the control unit, respectively. Finally, the intervention area is defined as the area resulting from the difference between the first collision area and the second collision area.

[0021] The method guarantees the driver sufficient freedom of maneuver, but at the same time prevents the vehicle from leaving the designated track and thus prevents such an event from causing any damage to the vehicle and / or the driver.

[0022] The method according to an embodiment of the invention completely turns the concept of controlling the stability of a vehicle on its head. In fact, the method eliminates any driver assistance, with the driver freely operating the vehicle commands without any overlap by the stability control system. This allows the possibility to try new driving styles and ensures maximum driving pleasure.

[0023] However, at the same time, continuous monitoring of the vehicle is performed and if it is detected that the dynamic state of the vehicle has reached the intervention area, i.e. the feasibility limit of keeping the vehicle on a trajectory within a specified track (e.g. a runway), a rapid overlap with the driver's commands allows correction of the vehicle dynamics in order to keep the vehicle on track.

[0024] In fact, the calculation of the two collision zones allows to identify in a precise manner the optimal intervention zone which ensures that a collision with the edge of the track is avoided while minimizing intervention in the driving of the driver.

[0025] In other words, the intervention region calculated in this way ensures sufficient spatial extent and time so that the control unit intervenes to correct the vehicle dynamics only when actually necessary.

[0026] It will be obvious to a person skilled in the art that the method can be implemented without the need for external sensors - cameras, radars, lidars. In other words, the method according to an embodiment of the invention can be implemented in a vehicle that provides "standard" on-board instrumentation for the automotive industry. Furthermore, by not requiring the presence of external sensing sensors, the method according to an embodiment of the invention can be implemented on a vehicle without overloading and / or reducing its aerodynamics due to the presence of sensors of this type.

[0027] In one embodiment, the method comprises the step of determining the driver's ability to modify the vehicle dynamics by selecting at least one value indicative of the ability to modify the vehicle dynamics from a set of predetermined values.

[0028] Preferably, the value is based on an analysis of the driving performance of a plurality of sample drivers on the track under consideration.

[0029] Thanks to this solution, it is possible to easily and quickly qualify the driver's ability to modify the dynamics of the vehicle without the need for tests on a runway or simulation tests.

[0030] In an alternative embodiment, the driver's ability to modify the vehicle dynamics is determined by measuring at least one value indicative of the ability to modify the vehicle dynamics while the driver is driving the vehicle, for example on a racetrack or through a simulation.

[0031] In this way, it is possible to accurately assess the driver's actual ability to modify the vehicle dynamics and obtain "tailored" driver assistance.

[0032] In one embodiment, the value indicative of the ability to modify the dynamics of the vehicle comprises at least one of a maximum steering speed, a maximum braking and an acceleration that can be performed by the driver.

[0033] In one embodiment, the steps of calculating the first collision area and the second collision area each comprise calculating a subset of points from which any trajectory settable by the driver or the control unit, respectively, leads to reaching / crossing at least one edge according to a reachability problem based on vehicle dynamics.

[0034] This solution makes it possible to define the collision zone precisely and reliably.

[0035] In one embodiment, the method calculates an intervention region for each side of the track on which the vehicle travels.

[0036] In this way, it is possible to quickly correct the vehicle dynamics at any time during the movement of the vehicle.

[0037] Alternatively, the at least one edge corresponds to an edge of a track that intersects a direction of travel of the vehicle, or an edge of a track that is less than a threshold distance value from the vehicle position.

[0038] Simple and rapid identification of relevant edges with respect to which it is more likely that a corrective intervention in the vehicle dynamics needs to be performed allows limiting the computational load performed by the control unit.

[0039] In one embodiment, the step of defining a set of points of a track within which the vehicle is free to move comprises acquiring geographical data relating to a geographical area comprising said track, identifying geographical coordinates of said track based on the acquired geographical data.

[0040] Thus, the geographical coordinates of a track are defined as points corresponding to the track, while the remaining geographical coordinates are defined as points not belonging to the track. Finally, the at least one edge is defined as a subset of points of the track that are adjacent to each other and to at least one point not belonging to the track.

[0041] This definition of the track that can be traveled by the vehicle is particularly easy to implement and requires a closed storage space. Therefore, the track can be calculated by an electronic device containing resources and / or can be quickly transmitted via a wired or wireless connection.

[0042] In one embodiment, it is possible to add to the set of points of the track the coordinates of at least an additional area adjacent to a part of the set of points of the track, said additional area being suitable for the movement of the vehicle.

[0043] Thanks to this operation, it is possible to define the track within which the movement of the vehicle is contained, based on the actual configuration of the geographical area containing the track, the type of vehicle and / or the driving style of the driver.

[0044] In one embodiment, controlling at least one actuator to modify vehicle dynamics includes calculating an exit trajectory from the intervention area based on the vehicle's current position and dynamics and the vehicle's distance from at least one edge, and performing the exit maneuver by controlling the at least one actuator to follow the exit trajectory.

[0045] Preferably, the exit trajectory is calculated as the solution to the following relation:

[0046]

[0047] Make

[0048]

[0049] d ost (x(t))>d min ,

[0050] Where t refers to time, T is the prediction period, and f(x,u) represents the dynamics of the vehicle d ost (x) is the distance between the vehicle and at least one edge, d min is a distance threshold allowed between the vehicle and the at least one edge, x represents the position of the vehicle in space, and u represents a dynamic variable controllable by the at least one actuator.

[0051] Exit trajectories are calculated in a simple and fast manner, which allows timely modification of vehicle dynamics to ensure that track exits are avoided.

[0052] Different aspects of the invention relate to a vehicle comprising a control unit, a plurality of sensors configured to measure quantities indicative of vehicle dynamics, at least one actuator configured to influence the vehicle dynamics. The plurality of sensors and the at least one actuator are connected to the control unit.

[0053] Furthermore, the control unit is configured to perform the method according to any one of the above embodiments.

[0054] A vehicle having the above characteristics can obtain the same advantages as described above.

[0055] Further features and advantages of the present invention will become more apparent from the description of the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] The present invention will be described below with reference to some examples, which are provided for the purpose of explanation and not limitation and are shown in the accompanying drawings. These drawings show different aspects and embodiments of the present invention, and where appropriate, reference numerals showing structures, components, materials and / or similar elements in different drawings are indicated by similar reference numerals.

[0057] Figure 1 is a block diagram of a system according to an embodiment of the present invention;

[0058] Figure 2 is a qualitative representation of a trajectory diagram that can be traveled by a vehicle according to an embodiment of the present invention;

[0059] Figure 3 is a flow chart of a mapping method according to an embodiment of the present invention;

[0060] Figure 4 is a flowchart of a method for restraining a vehicle according to an embodiment of the present invention;

[0061] Figure 5 is a logic block diagram of a software application executed by an onboard control unit according to an embodiment of the present invention;

[0062] Figure 6 is included in Figure 4 The flowchart of Figure 5 A flowchart of a process for calculating an intervention area implemented by a module of a software application;

[0063] Figure 7 is through Figure 6 A qualitative representation of the intervention area calculated by the process, and

[0064] Figure 8 It is used to calculate Figure 6 Qualitative representation of the physical quantities used in the process of the intervention area. DETAILED DESCRIPTION

[0065] Although the present invention is susceptible to various modifications and alternative constructions, some preferred embodiments are shown in the drawings and described in detail below. It must be understood in any case that there is no intention to limit the present invention to the specific embodiments shown, but on the contrary, the present invention is intended to cover all modifications, substitutions and equivalent structures within the scope of the present invention defined by the claims.

[0066] The use of "such as", "etc", "or", unless otherwise stated, refers to non-exclusive alternatives without limitation. The use of "including", unless otherwise stated, means "including, but not limited to",.

[0067] refer to Figure 1 , an automatic control system for virtually restraining a vehicle on a runway according to an embodiment of the present invention (hereinafter referred to as "system 1") is included in a vehicle V, which is a car in the example considered.

[0068] The system 1 comprises a plurality of actuator groups, denoted as a whole by the reference numeral 10 , known per se and, for the sake of brevity, not described in detail here. In the non-limiting example considered herein, the system 1 comprises:

[0069] - a steering or active steering actuator group 11, and

[0070] -Brake pump pressure control assembly 12 (or master cylinder).

[0071] Preferably, although not in a limiting way, the following actuator groups are included:

[0072] - an actuator group for driving torque 13, and

[0073] A braking torque control unit 14 acting independently on each wheel of the vehicle v.

[0074] Furthermore, the system 1 comprises a plurality of sensors, generally indicated by the reference numeral 20 , and in particular:

[0075] - a positioning sensor 21 of the vehicle V, for example a satellite positioning system GNSS (short for Global Navigation Satellite System) or a positioning system based on a camera or positioning algorithms;

[0076] - an inertial platform 22 , preferably with a centre of gravity having six degrees of freedom, for example comprising three accelerometers and three gyroscopes;

[0077] - at least one speed sensor 23 coupled to one wheel of the vehicle; preferably a speed sensor 23 for each wheel of the vehicle V, and

[0078] - Steering angle sensor 24.

[0079] Finally, the system 1 comprises a processing unit or control unit 30 configured to drive the actuator group 10 and to acquire the measurements provided by the sensors 20. In the example considered, the control unit 30 comprises one or more computers 31 - such as processors, microprocessors, microcontrollers, ASICs, FPGAs, DSPs, etc. - a storage module 32 provided with one or more non-volatile and volatile storage elements - and auxiliary modules, such as a power supply module 33 and a communication module 34, configured, for example, to exchange data via a CAN / LIN bus and / or via a wireless communication channel - such as Bluetooth, Wi-Fi, ZigBee, etc.

[0080] In an embodiment of the present invention, the control unit 30 stores information on the tracks that can be driven, hereinafter referred to as tracks. In the example considered, the control unit 30 stores Figure 2Detailed map M of a circuit or track P schematically depicted in FIG. 1 , in which the vehicle V can travel. The map M defines the track P by a collection of points corresponding to geographical coordinates, preferably provided by the same positioning system used by the positioning sensor 21 .

[0081] In an embodiment of the present invention, the map M basically comprises two sets of points:

[0082] a. the set of track points pT where vehicles are allowed to appear, and

[0083] b. A set pN of non-track points where vehicles are not allowed to appear.

[0084] Typically, the points of the set of track points pT correspond to the surface of the runway P on which the vehicle V is expected to travel. Optionally, the set of track points pT may include points pA that are part of additional surfaces, such as escape routes, secondary paths of the runway P and / or areas adjacent to the runway P on which the vehicle V can travel without being damaged. Otherwise, the non-track points pN include all the remaining points of the map M

[0085] In an embodiment of the present invention, a set pM of edge points is defined, which is a subset pT of track points adjacent to at least one non-track point pN.

[0086] In one embodiment, a map M of a runway P is created by a mapping method 900 described below, where Figure 3 Preferably, the method 900 is performed by an electronic device 50 (such as Figure 1 ) is executed, for example, by a general purpose computer provided with a software application 90 that is appropriately configured to execute the method 900.

[0087] Initially, geographic data for an area including runway P is acquired (step 901). The geographic data includes one or more of a set of geographic coordinates defined by one or more GNSS, GIS geographic information system data, cartographic data, aerial / satellite images, etc. Such geographic data is acquired by one or more corresponding computer systems (not shown, for example, a geographic map repository including GPS coordinates).

[0088] The track defining runway P is thus identified (step 903 ). This operation may be performed manually and / or by a software application configured to identify the track of runway P by analyzing one or more satellite images of the area including runway P.

[0089] Therefore, the geographical coordinates corresponding to the identified runway P are added to the set of track points pT (step 905).

[0090] Optionally, any additional surfaces are identified manually and / or with the aid of a software application, whose points pA are added to the set of track points pT (step 907). For example, the software application is configured to identify possible additional surfaces based on an analysis of one or more satellite images of the area including the runway P, and the addition of the points pA associated with these additional surfaces is then manually confirmed by a human operator.

[0091] The remaining coordinates are added to the set pN of non-track points (step 911).

[0092] Subsequently, edge points pM are identified (step 913). Advantageously, the edge points are further divided into a pair of edge subsets pM1 and pM2, each edge subset comprising coordinates adjacent to each other. In other words, each subset of edge points pM1 and pM2 comprises points defining one of the two edges of the runway p.

[0093] Finally, the map M, including the set of track points pT, the subsets of edge points pM1 and pM2 and the set of non-track points pN, is made available for transmission to the control units 30 of one or more vehicles V (step 915).

[0094] The map M is then transmitted to the control unit 30 to which the electronic device 50 is connected via a wired or wireless connection.

[0095] In an embodiment of the invention, the control unit 30 executes a method 1000 of automatically restraining a vehicle V, the purpose of which is to protect the driver from dangerous situations while restricting the driver's freedom of maneuver as little as possible.

[0096] Figure 4 is a flow chart of a method 1000, which provides for performing the following steps, preferably periodically, for example, with a sampling period comprised between about 5 and 10 milliseconds:

[0097] - determining (step 1001) the dynamics of the vehicle V, wherein the term "dynamics" refers to the set of information related to the movement of the vehicle in space as a function of time, including but not necessarily limited to: longitudinal velocity, lateral velocity, yaw rate, yaw angle, steering angle and individual wheel speeds, and optionally also roll and pitch rates and suspension extensions;

[0098] - determining (step 1003 ) the position of the vehicle V within the runway P included in the map M, this position being obtained by means of the positioning sensors 21 and / or by a dead reckoning operation based on values ​​provided by the inertial platform 22 or by means of a positioning algorithm, the position of the vehicle V corresponding, in the example considered, to the position of the centre of mass of the vehicle V;

[0099] - based on the position of the vehicle V, in particular the points relative to the edge subsets pM1 and pM2, the dynamics of the vehicle V and the type of control that the control unit 30 can exercise on the vehicle V, calculate (step 1005) at least one intervention zone Zi included in the runway P, i.e. included in the set pT of track points;

[0100] - determining (decision step 1007 ) whether the vehicle V at least partially enters the intervention zone Zi;

[0101] If the vehicle has not entered the intervention zone Zi (exit branch of step 1007 ), the method provides for not executing (step 1009 ) any intervention Zi, leaving the control of the vehicle V completely to the driver;

[0102] - if, on the other hand, the vehicle has entered the intervention zone Zi (exit branch Y of step 1007 ), the method 1000 provides for calculating a trajectory for exiting from a correction zone that keeps the vehicle V inside the runway, ie included in the zone defined by the set of track points pT (step 1011 );

[0103] - controlling (step 1013) the actuators 10 of the vehicle V to modify the dynamics of the vehicle V so as to perform an exit maneuver following the calculated exit trajectory, and

[0104] - After performing the exit maneuver, control of the vehicle V is returned (step 1015) to the driver.

[0105] Preferably, method 1000 is repeated at the exit of step 1007 or step 1011 as long as vehicle V is moving on runway p.

[0106] In one embodiment of the present invention, the control unit 30 stores and executes a software product 100 implementing the method 1000 .

[0107] In the example considered, the software 100 executed by the control unit 30 implements three main logical modules - e.g. Figure 5 As shown in the block diagram: an inevitable collision set calculation module or module ICS101 (acronym for inevitable collision set), a reference generation module or module RG 102 (acronym for reference generation), and a vehicle dynamics control module or module VDC 103 (acronym for vehicle dynamics control).

[0108] As described in more detail below, the module ICS 101 is configured to define, preferably in real time, one or more intervention zones Zi (steps 1001-1005 of the method 1000). The module RG 102 is configured to calculate an exit trajectory from the intervention zone Zi (steps 1007-1011 of the method 1000). Finally, the module VDC 103 is configured to operate the actuator 10 so that the vehicle V follows the exit trajectory calculated by the module RG 102, and to return control of the vehicle to the driver (steps 1013 and 1015 of the method 1000).

[0109] Specifically, the intervention area Zi definition providing module ICS101 executes Figure 6 The following sub-steps are shown in the flowchart.

[0110] Preliminarily, for example in a system initialization step, the module ICS101 acquires (step 10051) the following data related to the vehicle V (for example, stored in the memory module 32):

[0111] - a set of characteristics of the vehicle V, such as but not limited to the type of tires installed, the nominal friction of the vehicle V, the wheelbase of the vehicle V and the mass of the vehicle V;

[0112] - Control configuration - i.e. which actuators 10 can be autonomously controlled by the control unit 30, for example selected from:

[0113] a) Active braking only,

[0114] b) Active braking and active steering,

[0115] c) active steering and torque vectoring – i.e. the ability to apply a desired yaw moment (e.g. via differential torque), or

[0116] d) Active steering only;

[0117] - the ability of the control unit 30 to modify the dynamics of the vehicle V - for example, the response time required for the control unit 30 to control one or more of the actuators 10 (e.g., to perform braking and / or steering) to modify the dynamic characteristics of the vehicle V, the maximum steering speed (measured in rad / s), the maximum braking or acceleration capability (i.e., deceleration measured in m / s), the ability to apply additional yaw moments through torque vectoring or differential braking (measured in Nm), etc., and

[0118] -The ability to modify the dynamics of the driver's vehicle V in response to external stimuli (e.g., the presence of an obstacle or the setting of a curve) - for example, the response time required for the driver to provide commands to the vehicle (e.g., to perform braking and / or steering), the maximum steering speed (measured in rad / s), the maximum braking or acceleration capacity (i.e., deceleration measured in m / s), etc.

[0119] The characteristics of the vehicle V, its control configuration and the ability to modify the dynamics of the control unit 30 are known data, for example provided to the control unit 30 by the manufacturer / assembler of the vehicle V via the device 50 in a manner similar to that described above with respect to FIG. 24. Otherwise, the ability to modify the driver's dynamics is selected from a set of values ​​- preferably adjustable based on the driver's experience level. In one embodiment, the value is based on an analysis of the driving performance of a plurality of sample drivers on the track under consideration - for example, for each driver, the maximum steering speed and the maximum deceleration (or acceleration) set by the driver via the actuators under his or her control are calculated; these two measurements are then weighted to obtain a value indicative of the driver's level of experience in driving the vehicle and included in the above set of values.

[0120] Alternatively, the ability to modify driver dynamics is determined empirically, through real or simulated driving tests, during a preliminary calibration step of the system 1 .

[0121] Based on the data collected by the vehicle V and the measurements provided by the sensor 20, the module ICS101 is configured to determine, preferably in parallel, the intervention area for each subset pM1 and pM2 of the edge points of the runway P, also referred to hereinafter as edges pM1 and pM2 for simplicity. For simplicity, the steps for calculating the intervention area are described below only with respect to one edge (edge ​​pM1), but the same process is performed mutatis mutandis for the other edge pM2 of the runway P.

[0122] The module ICS101 calculates (step 10053) a first collision zone ICS1 of the runway P with reference to the edge pM1 (i.e. the set of track points pT, called the unavoidable collision set). In particular, when the vehicle V enters the first collision zone ICS1, the driver, by operating the actuators only under his or her control and taking into account the above-mentioned boundaries, cannot perform maneuvers such as preventing the vehicle V from reaching the edge pM1 and exiting the runway p.

[0123] Furthermore, the module ICS101 calculates (step 10055) a second zone ICS2 of the runway p. In particular, when the vehicle V enters the first collision zone ICS1, no maneuvers can be performed by the control unit 30 acting with all actuators under its control to prevent the vehicle V from exiting the runway p.

[0124] like Figure 7 As schematically shown in FIG. 1 , collision regions ICS1 and ICS2 extend from edge pM1 of runway P and extend towards the interior of runway P. In particular, first collision region ICS1 has a wider surface than second collision region ICS2 because first collision region ICS1 has a greater ability to modify the dynamics of control unit 30 than the driver has the ability to modify the dynamics of vehicle V.

[0125] In one embodiment, the points included in the collision areas ICS1 and ICS2 are calculated based on the dynamics of the vehicle V, which is defined according to the following equation:

[0126]

[0127] Among them - such as Figure 8 As shown in the figure, β is the drift angle, that is, the angle between the velocity vector v of the vehicle and its longitudinal axis; C r and C f is the so-called "cornering stiffness" or the speed at which the rear wheels r and front wheel t f The lateral slip and the lateral force Fy applied to the wheels r and Fy f V is the modulus of the forward speed of the vehicle V; m is the mass of the vehicle V; δ is the wheel t r and t f Steering angle at I f and I r are the distances between the center of mass cM of the vehicle V and the front and rear axles of the vehicle V, respectively, J is the moment of inertia of the vehicle, r is the yaw rate of the vehicle, and MY is the additional yaw moment caused by differential braking. In addition, refers to the turning speed, and is the acceleration command provided by the driver.

[0128] In particular, the points included in the collision zones ICS1 and ICS2 are calculated as points for which all possible control sequences executable by the driver for ICS1 and all possible control sequences executable by the control unit 30 for ICS2 result in reaching / crossing the edge pM1 starting from the current dynamics of the vehicle V, based on the reachability theory defined in the following literature: Kolmanovsky and, for example, Gilbert, “Theory and computation of disturbance invariant sets for discrete-time linear systems”, Mathematical Problems in Engineering, Vol. 4, No. 4, pp. 317–367, 1998; I. Mitchell, “Comparing forward & backward reachability as tools for safety analysis”. Springer, 2007, pp. 428-443; and T. Fraichard and H. Asama, "Inevitable Collision States-A step towards safer robots?" Advanced Robotics, Vol. 18, No. 10, pp. 1001-1024, 2004, which are incorporated herein by reference.

[0129] In the example considered, it is not possible to obtain a closed form of the trajectory in state space - that is, the variables that appear in the form of derivatives in equations 1-7 - that leads to reaching or crossing the edge pM1 starting from relations 1-7. Therefore, module ICS101 is configured to solve the following digital system using relations 1-7:

[0130]

[0131] where _xD835__xDC94__xD835__xDFCE_ includes the initial point of the trajectory that brings the vehicle V to a point on or beyond the edge pM1 (i.e., the vehicle V reaches the set pN of non-track points) - included in the set pT of track points, is the set of collision states, is a set of possible control actions that can be performed by the driver for the first collision zone ICS1 or by the control unit 30 for the second collision zone ICS2, and is included in the group One of the control actions in .

[0132] The set of maximum reachable points - or maximum reachable set - starting from the initial point _xD835__xDC94__xD835__xDFCE_ where the vehicle v is located The definition is as follows:

[0133] (9)

[0135] And the set of reachable points that do not conflict with the edge - or the conflict-aware reachable set - The restrictions are as follows:

[0136]

[0137] where τ refers to the time variable contained between 0 and t

[0138] The collision areas ICS1 and ICS2 are therefore defined as:

[0139]

[0140] Advantageously, the definition of the collision zones ICS1 and ICS2 is performed by the module ICS101 by solving a recursive formula, discretizing in time.

[0141] According to an embodiment of the invention, a pseudo code of the calculation of collision areas ICS1 and ICS2 performed by module ICS101 is reported below.

[0142] enter:

[0143] - state space (S) - corresponds to the set of orbital points pT,

[0144] - prohibited geographical area (K) - corresponds to the set pN of non-orbital points,

[0145] - the step size of the time discretization △t,

[0146] a set U of possible corrective actions that can be performed by the driver for the first collision zone ICS1 or by the control unit 30 for the second collision zone ICS2

[0147] Pseudo algorithm:

[0148] initialization:

[0149]

[0150] k=0,

[0151] repeat:

[0152] k←k+1,

[0153]

[0154] X f ←X s ~{Effects of all control actions},

[0155]

[0156] until:

[0157]

[0158] return:

[0159]

[0160] In particular, using the steps

[0161] X f ←X s

[0162] The states that can be avoided using available control actions are from the group X f According to Pontryagin, this operation consists of differences. Specifically, consider as possible control actions The function of the trajectory modification of the function determines whether it is possible to identify from a specific point avoiding the group calculated in the previous iteration A vector of points from the set computed in the current iteration Excluded.

[0163] In addition, using the steps

[0164]

[0165] The free movement is inverse - that is, the evolution of the vehicle V does not take into account the input, i.e., the command provided by the driver or system 1 - to expand the collision zone. In other words, the free movement is integrated backwards in time rather than forwards. This is equivalent to starting from the endpoint (e.g., the point of edge pM1) and tracing the trajectory of the vehicle backwards until it reaches its starting point.

[0166] The algorithm is repeated until the collision area no longer grows between two consecutive iterations, i.e.

[0167] Finally, the overall collision area ICS can be defined as the group The union of the prohibited geographic area K and the intersection with the state space S, that is,

[0168] The intervention area Zi is calculated (step 10057) as a subset of the points of the set pT of trajectory points included in the difference between the first collision area ICS1 and the second collision area ICS2.

[0169] In other words, the collision zones ICS1 and ICS2 are similar to a dynamic "widening" of a physical boundary that is not crossed, i.e., the edge of the runway P. The more the vehicle V "points" toward the physical boundary - that is, the more the direction of advance, and more generally, the dynamics bring the vehicle V toward one of the edges of the runway P - the more this boundary must be widened, or similarly, "approached" the vehicle V, thereby creating a virtual buffer zone that allows correcting the dynamics of the vehicle V and avoiding a runway exit.

[0170] The intervention area Zi thus calculated is transmitted (step 10059 ) to the module RG 102 .

[0171] When module RG 102 determines that vehicle V, in particular the position of its center of mass, is within the intervention zone Zi calculated by module ICS101 (at step 1007 ), the position and dynamics of vehicle V are used as starting points to generate an exit trajectory allowing to keep the edges of vehicle V within runway p.

[0172] In particular, it is essential to ensure the continuity of the control variables when calculating the exit trajectory (performed in step 1011 of method 1000). In a preferred embodiment, module RG 102 is configured to calculate the formulated exit trajectory in differential mode (or velocity form) starting from the problem.

[0173] In the example discussed, the exit trajectory is calculated as the solution to the following relation:

[0174]

[0175] So that:

[0176]

[0177] d ost (x(t))>d min , (14)

[0178] Where t refers to time, T is the prediction period, and f(x,u) represents the dynamics of the vehicle d ost (x) is the distance between the vehicle V and the edge of the runway P from which the collision zones ICS1 and ICS2 extend—usually corresponding to the nearest edge—and d minis the distance threshold allowed between the vehicle and the edge of the runway - for example greater than or equal to half the width of the vehicle V, x represents the position of the vehicle in space, and u comprises the variables controlled by the control unit 30 that are allowed by the control configuration in use.

[0179] It can be observed that the prediction period T is limited by the computing power of the control unit 30 and that, in general, the longer the prediction period T, the better the calculated trajectory. However, the relationship (12) is solved at each sampling instant in the "receding horizon" logic and therefore the prediction period T does not necessarily include the time required for the vehicle to reach an edge (for example an obstacle or the limit of the runway at a bend). Studies carried out by the Applicant have made it possible to determine that a prediction period of the order of between 1 and 3 seconds is sufficient to provide sufficiently accurate predictions.

[0180] The trajectory calculated by the module RG 102 is transmitted to the module VDC 103 to modify the dynamics of the vehicle V (at step 1013 of the method 1000 ) so as to follow this trajectory. In particular, the module VDC 103 operates essentially as a closed-loop control system whose task is to actuate the actuators 10 so as to follow the trajectory calculated by the module RG 102 .

[0181] The module VDC 103 also enables the return of control of the vehicle V to the driver (performed at step 1015 of the method 1000). In the example considered, the module VDC 103 is configured to return control to the driver only when the vehicle has exited the first collision zone ICS1 (i.e. the zone of maximum amplitude), and the command to the actuator 10 being calculated by the module VDC 103 - i.e. the control variable - substantially corresponds to the command applied by the driver through the vehicle V by means of the steering wheel, pedals and / or other input elements. In particular, the driver's command comprises one or more of the acceleration V applied to the vehicle (determined based on the position of the accelerator pedal due to the throttle-by-wire technology), the braking torque (determined by measuring the pressure in the brake pump), the steering torque (i.e. a measure of the torque applied by the driver on the steering wheel). In detail, when the values ​​of the applied acceleration and / or braking torque correspond to the values ​​applied by module VDC 103, and / or when the steering torque is below a predetermined threshold (indicative of the fact that the driver is holding the steering wheel without trying to counteract the action of module VDC 103), the driver is considered to be ready to resume control of the vehicle V, and module VDC 103 transfers control to the driver, i.e. stops sending commands to the actuators.

[0182] However, it is obvious that the above examples are not to be interpreted as limiting and that the invention thus conceived is susceptible of numerous modifications and variations.

[0183] For example, the map may include obstacles present on the runway at the set of non-track points. To this end, in one embodiment, the system is configured to detect and / or receive information about the state of the runway and / or the presence of obstacles / vehicles within the runway and to update the map accordingly. For example, the map may be updated as soon as the state of the runway and / or the presence of obstacles / vehicles changes, or the map may be updated periodically. In this case, the system will determine the collision zone taking into account the obstacles so that if the driver reaches a position where he or she cannot avoid the obstacle by acting only on the actuators under his or her control, the trajectory correction is intervened.

[0184] In an alternative embodiment, the creation of the map may be performed directly by the control unit of the vehicle and verified / modified by the operator or driver himself through the vehicle user interface. In other words, the vehicle may define a map, and in particular the allowed tracks within which the vehicle must move, by uploading information previously processed by an external device (e.g., the map itself) or by directly processing geographic data, images and / or user input in order to obtain such a map.

[0185] It will be apparent to those skilled in the art that one or more steps of the above method may be performed in parallel with each other, or in an order different from that described above. Similarly, one or more optional steps may be added or deleted from one or more of the above processes.

[0186] Naturally, all the details may be replaced by other technically equivalent elements.

[0187] In summary, the materials, shapes and possible sizes used in the above-mentioned devices, apparatuses and terminals can be determined according to specific implementation requirements without departing from the protection scope of the attached claims.

Claims

1. A method (900; 1000) for automatically restraining a land vehicle, the land vehicle comprising a control unit (30), a plurality of sensors (20) configured to measure quantities indicative of vehicle dynamics, at least one actuator (10) configured to influence the vehicle dynamics, the plurality of sensors (20) and the at least one actuator (10) being connected to the control unit (30), wherein the method (900; 1000) comprises the control unit (30) performing the following steps: a set of points defining (900) a track within which the vehicle is free to move, the track comprising at least one edge defining a boundary of the track, the boundary having to be not crossed by the vehicle, determining (1001) vehicle dynamics based on information provided by the plurality of sensors, determining (1003) the spatial position of the vehicle based on information provided by the plurality of sensors, calculating (1005) an intervention area of ​​the track based on the position and the dynamics of the vehicle and the position from the at least one edge, the intervention area being contained within the track, determining (1007) whether the vehicle is at least partially within the intervention area, and in the affirmative, controlling (1013) the at least one actuator to modify the vehicle dynamics to perform an exit maneuver that takes the vehicle outside the intervention area, Wherein calculating (1005) the intervention area comprises: calculating (10055) a first collision region based on the vehicle's current position and dynamics, the position from the at least one edge, and the driver's ability to modify the vehicle's dynamics by acting on the at least one actuator, calculating (10057) a second collision area based on the current position and dynamics of the vehicle, the position from the at least one edge and the ability of the control unit to modify the vehicle dynamics acting on the at least one actuator, wherein the first collision area and the second collision area correspond to a track area within which the at least one edge is inevitably reached and / or crossed when the vehicle moving according to the current dynamics is driven by the driver or the control unit, respectively, and The intervention area is defined (10058) as the area resulting from the difference between the first collision area and the second collision area.

2. The method (900; 1000) according to claim 1, further comprising the step of determining (10051) the driver's ability to modify the vehicle dynamics by: selecting at least one value indicative of said ability to modify vehicle dynamics from a set of predetermined values, or While the driver is driving the vehicle, at least one value indicative of the ability to modify the dynamics of the vehicle is measured.

3. The method (900; 1000) according to claim 2, wherein: The value indicative of the ability to modify the dynamics of the vehicle includes at least one of: capable of being performed by the driver Maximum turning speed, and Maximum braking and maximum acceleration.

4. The method (900; 1000) according to any one of the preceding claims, wherein: The steps (10055, 10057) of calculating the first collision area and the second collision area each comprise calculating a subset of points from which any trajectory set by the driver or the control unit, respectively, leads to reaching / crossing at least one edge according to an accessibility problem based on the vehicle dynamics.

5. The method (900; 1000) according to any one of the preceding claims, wherein: The at least one edge corresponds to an edge of a track that intersects with the vehicle's advancing direction, or an edge of a track that is less than a threshold distance value from the vehicle's position.

6. The method (900; 1000) according to any one of the preceding claims, wherein: The step of defining (900) a set of points of a track within which the vehicle is free to move comprises: Acquiring (901) geographic data associated with a geographic area including the track; identifying (903) geographic coordinates of the track based on the acquired geographic data; defining (905) the geographic coordinates as points of the track; defining (911) the remaining geographic coordinates as points that do not belong to the track, and The at least one edge is identified (913) as a subset of the points of the track that are adjacent to each other and to at least one point that is not part of the track.

7. The method (900; 1000) according to claim 8, wherein: It also comprises the step of adding to said set of points of said track the coordinates of at least an additional area adjacent to a portion of said set of points of said track, said additional area being suitable for travel of said vehicle.

8. The method (900; 1000) according to any one of the preceding claims, wherein: The step of controlling (1013) the at least one actuator for modifying the dynamics of the vehicle comprises: calculating an exit trajectory from the intervention area based on the current position and the dynamics of the vehicle and the distance of the vehicle from the at least one edge, and The exit maneuver is performed by controlling the at least one actuator to follow the exit trajectory.

9. The method (900; 1000) according to claim 8, wherein: The exit trajectory is calculated as the solution to the following relation: Make d ost (x(t))>d min , Where t refers to time, T is the prediction period, and f(x,u) represents the dynamics of the vehicle. ost (x) is the distance between the vehicle and the at least one edge, d min is a distance threshold allowed between the vehicle and the at least one edge, x represents the position of the vehicle in space, and u represents at least one dynamic variable controllable by the at least one actuator.

10. A vehicle (V) comprising a control unit (30), a plurality of sensors (20) configured to measure quantities indicative of vehicle dynamics, at least one actuator (10) configured to influence the dynamics of the vehicle, the plurality of sensors (20) and the at least one actuator (10) being connected to the control unit (30), and Wherein the control unit (30) is configured to perform the method (900; 1000) according to any of the preceding claims.

Images