METHOD FOR PREDICTING THE SPEED OF A VEHICLE

Abstract

The invention relates to a method (21), implemented in a computer (4) embedded in a vehicle (2), for predicting the speed of the vehicle (2) over a journey between a starting point and an arrival point, the computer (4) being connected to a system (8) for managing static and / or dynamic data relating to road infrastructure and / or road traffic, the computer (4) being configured to calculate an optimal speed profile of the vehicle (2) according to a predefined calculation method, said calculation method minimizing the Hamiltonian of a system of equations modeling the driving of the vehicle (2) and being configured to allow optimization of a criterion taking into account the energy consumption of the vehicle (2) as well as the duration of the journey, the computer (4) storing one or more recordings of past or recent driving phases of the vehicle (2). Abstract figure: Fig. 1.

Description

Title of the invention: METHOD FOR PREDICTING THE SPEED OF A VEHICLE

[0001] The invention relates to a method for predicting the speed of a vehicle. The method is implemented in a computer embedded in the vehicle. The invention also relates to a method, implemented in a computer embedded in the vehicle, for optimizing the vehicle's energy consumption, the method comprising such a sub-method for predicting the vehicle's speed. The invention also relates to a method, implemented in a computer embedded in the vehicle, for controlling the vehicle's speed, the method comprising such a sub-method for optimizing the vehicle's energy consumption. The invention also relates to a computer intended to be embedded in a vehicle, the computer comprising means for implementing the steps of such a method for predicting the vehicle's speed; as well as to a vehicle, in particular a motor vehicle, incorporating such a computer.The vehicle is typically, but not exclusively, an autonomous or semi-autonomous vehicle. The invention also relates to a computer program product comprising program instructions configured to implement the steps of such a vehicle speed prediction method.

[0002] In a motor vehicle, it is known to optimize the energy consumption of the powertrain over a given or planned journey. Such optimization can be carried out on fuel, on electrical energy, on hydrogen consumption (in the case of a vehicle equipped with a fuel cell) or on two or three of these criteria simultaneously.

[0003] As is known, optimization can be performed using the principle known as Pontryagin's Maximum Principle (PMP). This method consists of minimizing the Hamiltonian function (or Hamiltonian) based on the criterion to be optimized, for example the quantity of fuel or hydrogen or the electrical energy consumed, and the description of the system dynamics. The system dynamics are defined based on the state of various vehicle variables (vehicle speed, state of charge of the battery and / or supercapacitors, temperatures, etc.).) and various inputs or instructions (torque instructions to be applied to the vehicle wheels for the internal combustion engine or for the electric machine(s), torque instructions to be applied to the generator set engine(s), and / or power or current instructions for the fuel cell, and / or instructions for heating the catalyst, and / or instructions for controlling the cooling circuit. dissement, etc). The Hamiltonian function is minimized in order to determine the setpoints that allow for the minimum consumption of fuel or hydrogen, and / or electrical energy.

[0004] The Hamiltonian thus determined is then minimized. In other words, the setpoint values ​​for which the Hamiltonian value is lowest are selected and applied to the vehicle. The setpoint values ​​are thus determined in real time according to the current state of the system.

[0005] The application of such a PMP model thus uses an internal model describing the dynamics of the system to be controlled (here, the motor vehicle and its powertrain), via differential equations that provide a representation of the system. Such a PMP model is therefore primarily based on an open-loop optimal control strategy, which calculates the optimal control solution based on the predictions of the internal model, with very little feedback from current measurements on the system.In this context, predicting vehicle speed by a vehicle computer can be useful and is done on a given or predicted driving scenario; this scenario can generally be provided by a navigation system (when the destination is known, given by the driver or guessed by a vehicle assistance system) or by an "electronic information horizon" type system (or "eHorizon" in English, which is classically based on the ADASIS data format standard - from the English "Advanced Driver-Assistance Systems Interface Specifications" - for predictive driver assistance systems, or on any other type of device), giving information on the predicted journey characteristics "ahead of the vehicle", valid for a short, medium or long term time forecast, depending on the available data, the location of the vehicle and the driving conditions.

[0006] When the criterion to be optimized takes into account the energy consumption of the vehicle as well as the duration of the journey, there is a need to be able to have a method for predicting the speed of the vehicle on a journey between a starting point and an arrival point, allowing a temporal prediction of the speed of the vehicle in the short, medium or long term.

[0007] To this end, several solutions are known, implementing, for example, statistical analysis (based on speed recordings on observed road segments, moderated by traffic conditions, time of day, etc.); analytical modeling (based on certain behavioral models of drivers, vehicle performance, journey characteristics, traffic conditions, etc.); fuzzy logic (modeled from measurements and recordings of recent past driving phases of driver activity on the pedals and gears, common accelerations or decelerations, etc.); artificial intelligence and neural networks (modeled from machine learning phases during repetitive driving tests, simulated or on real roads).

[0008] However, none of these known methods allows for a sufficiently accurate, precise, or stable prediction of the vehicle's speed over the medium or long term. Furthermore, several drawbacks are associated with such prior art methods: - loss of accuracy in unexpected situations or during insufficiently learned phases; - need for a lot of data or complex behavioral models, which can also be specific to each type of driver, each vehicle, each traffic condition, etc.; - requires a long learning process and a period of adaptation before acquiring the required precision; - may not be able to meet future power demand, which is necessary for efficient energy saving control; - cost of the solution if it comes from external providers.

[0009] There is therefore a need for a solution that will at least partially remedy these drawbacks.

[0010] To this end and according to a first aspect, the invention relates to a method, implemented in an on-board computer within a motor vehicle, for predicting the vehicle's speed over a journey between a starting point and an arrival point, the vehicle being further equipped with a vehicle navigation and positioning system connected to the computer, the vehicle's computer being further connected to a system for managing static and / or dynamic data relating to road infrastructure and / or road traffic, said static and / or dynamic data relating to road infrastructure and / or road traffic being provided as input to the computer, the vehicle's route over a predetermined distance being predefined or predicted within the computer, the computer being configured to calculate an optimal speed profile for the vehicle according to a predefined calculation method,said calculation method minimizing the Hamiltonian of a system of equations modeling the driving of the vehicle and being configured to allow optimization of a criterion taking into account the energy consumption of the vehicle as well as the duration of the journey, the computer storing one or more recording(s) of past or recent driving phases of the vehicle, the process comprising the following steps: , - determination, based on one or more recordings of past or recent driving phases of the vehicle, of a driving style of the vehicle's driver, - definition of a series of notable points along the route characterized by a stoppage of the vehicle or a decrease in the vehicle's speed, said series of points re- Markers dividing the route into a series of segments, - for each segment of the journey, calculation of an optimal vehicle speed profile according to said predefined calculation method, the duration of the journey being weighted using a weighting coefficient in the Hamiltonian of said predefined calculation method, - adaptation of said weighting coefficient according to the driving style determined for the driver of the vehicle, said adaptation step providing, for each segment of the journey, an adapted speed profile of the vehicle, - for each segment of the journey, provision of said adapted speed profile as a prediction of the vehicle speed on said segment of the journey.

[0011] The method is implemented for a predefined / given route for the vehicle. Such a route is, for example, predicted by an "electronic information horizon" (or "eHorizon" in English, which is typically based on the ADASIS data format standard – from "Advanced Driver-Assistance Systems Interface Specifications" – for predictive driver assistance systems, or on any other type of device), which is connected to the vehicle's computer. As is known per se, such an "eHorizon" type system can manage both static data relating to road infrastructure (such as, for example, the nature of roads, intersections, applicable speed limits, etc.) and dynamic data (average speed of vehicles on the road, traffic density, dynamic data relating to road infrastructure elements, etc.).Such an "eHorizon" type system is capable of receiving this data, decoding it (via a decoder), reconstructing it (via a data reconstructor), and transmitting it to the vehicle's computer, and can implement vehicle route prediction algorithms using the concept of "most probable path".

[0012] By providing an optimal vehicle speed profile as close as possible to the predictable speed, the method according to the invention makes it possible to obtain a robust prediction of vehicle speed, without requiring preliminary learning phases, complex behavioral models, or additional measurements of "external" or "specific" data related to the vehicle or driver. The method according to the invention also makes it possible to efficiently use an existing optimal speed profile calculation solution to obtain a suitable predictive speed profile, directly incorporating static and dynamic data from an "electronic information horizon" type system, such as speed constraints, road types, traffic, conditions, etc.The method according to the invention also makes it possible to obtain, simultaneously, a prediction of the power demand (with torque or current depending on the powertrain technology), necessary for correct energy analysis, optimization and energy savings. Finally, in the method according to the invention, the step of adapting the weighting coefficient according to the driving style determined for the vehicle driver is quite fast, stable and precise, and is carried out by means of typical "closed loop" feedback corrections (such as PID - Proportional Integral Derivative correction, dichotomy, tendency compensation, etc.), which are capable of keeping up with the slow and reasonable dynamic response time of the usual variations in driving behavior or traffic conditions.

[0013] Preferably, said predefined calculation method is a calculation method implementing the Pontryagin maximum principle. Such a calculation method is advantageously based on the Pontryagin model, which is self-adaptive and allows real-time implementation for embedded computer systems.

[0014] According to a particular technical feature of the invention, the step of determining the driving style of the vehicle driver is carried out by comparing, in one or more recordings of past or recent driving phases of the vehicle, an actual speed profile of the vehicle to an optimal speed profile of the vehicle calculated using said predefined calculation method. Such a comparison provides an indirect estimation means for determining the driving style of the vehicle driver.

[0015] According to another particular technical feature of the invention, the step of adapting the weighting coefficient according to the driving style determined for the driver of the vehicle is carried out according to a scale of continuous or discretized values.

[0016] Preferably, the step of adapting the weighting coefficient according to the driving style determined for the vehicle driver is carried out via a closed-loop feedback correction. Such a closed-loop feedback correction is able to keep pace with the slow and reasonable dynamic response time to typical variations in driving behavior or traffic conditions.

[0017] According to a second aspect, the invention also relates to a method, implemented in a computer embedded in a vehicle, for optimizing the vehicle's energy consumption over a journey between a starting point and an arrival point, the vehicle being further equipped with a vehicle navigation and positioning system connected to the computer, the vehicle's computer being further connected to a system for managing static and / or dynamic data relating to road infrastructure and / or road traffic, said static and / or dynamic data relating to road infrastructure and / or road traffic being provided as input to the computer, the vehicle's route over a predetermined distance being predefined or predicted within the computer, the computer being configured to calculate an optimal speed profile for the vehicle according to a predefined calculation method, said a calculation method minimizing the Hamiltonian of a system of equations modeling the driving of the vehicle and being configured to allow optimization of a criterion taking into account the energy consumption of the vehicle as well as the duration of the journey, the computer storing one or more record(s) of past or recent driving phases of the vehicle, the method further comprising a sub-method of predicting the speed of the vehicle as described above, and a step of taking into account the predicted speed of the vehicle on each part of the journey, to apply or recommend to the vehicle said predicted speed on said part of the journey.

[0018] According to a third aspect, the invention also relates to a method, implemented in a computer embedded in a vehicle, for controlling the vehicle's speed, the vehicle being further equipped with a vehicle navigation and positioning system and a powertrain, both connected to the computer, the vehicle's computer being further connected to a system for managing static and / or dynamic data relating to road infrastructure and / or road traffic, said static and / or dynamic data relating to road infrastructure and / or road traffic being provided as input to the computer, the vehicle's route over a predetermined distance being predefined or predicted within the computer, the computer being configured to calculate an optimal speed profile for the vehicle according to a predefined calculation method,said calculation method minimizing the Hamiltonian of a system of equations modeling vehicle driving and configured to allow optimization of a criterion taking into account the vehicle's energy consumption and the journey time, the computer storing one or more records of past or recent driving phases of the vehicle, the method further comprising a sub-method for optimizing the vehicle's energy consumption as described above, and a step of transmitting, to the vehicle's powertrain, a speed command established as a function of the predicted vehicle speed taken into account. Such a speed command thus established can then concern an autonomous or semi-autonomous vehicle, or an 'intelligent' cruise control, or even a driver assistance system by displaying this set speed or by visual interface feedback.auditory or haptic feedback directed at the vehicle driver.

[0019] The invention also relates to a computer intended to be embedded in a vehicle, the vehicle's route over a predetermined distance being predefined or predicted within the computer, the computer being configured to calculate an optimal speed profile for the vehicle according to a predefined calculation method, said calculation method minimizing the Hamiltonian of a system of equations modeling the vehicle's driving and being configured to allow optimization of a criterion taking into account the vehicle's energy consumption as well as the duration of a journey, the calculation computer storing one or more recording(s) of past or recent driving phases of the vehicle, the computer including means to implement the steps of the vehicle speed prediction process as described above and / or the vehicle energy consumption optimization process as described above and / or the vehicle speed control process as described above.

[0020] The invention also relates to a vehicle, in particular a motor vehicle, in which the vehicle carries a computer as described above and a vehicle navigation and positioning system connected to the computer, the computer being able to be connected to a static and / or dynamic data management system relating to road infrastructure and / or road traffic.

[0021] The invention also relates to a computer program product, downloadable from a communication network and / or recorded on a computer-readable medium and / or executable by a processor, the computer program product comprising program instructions, said program instructions being configured to implement the steps of the vehicle speed prediction method as described above and / or the vehicle energy consumption optimization method as described above and / or the vehicle speed control method as described above when said instructions are executed on a computer as described above.

[0022] The following are non-limiting examples of embodiments of the present invention, with reference to the accompanying figures in which: - [Fig. 1] schematically illustrates a vehicle according to the invention, the vehicle being equipped with an on-board computer; and - [Fig.2] is a flowchart representing a method for controlling the speed of a vehicle, including a sub-method for predicting the speed of the vehicle over a journey between a starting point and an arrival point, implemented by the computer of [Fig.1], according to the present invention.

[0023] With reference to [Fig. 2], the present invention relates to a method, implemented in a computer 4 embedded in a vehicle 2 (visible in [Fig. 1]), for controlling the speed of the vehicle 2. The vehicle 2 is typically a motor vehicle. The term "motor vehicle" means a road vehicle powered by at least one internal combustion engine, or at least one electric motor, or at least one gas turbine, such as, for example, a car, a van, a truck, etc., or even a hybrid motor vehicle.

[0024] The computer 4 is, for example, part of a data processing unit storing a computer application or program capable of cooperating with the computer 4 (the data processing unit and the computer application or program not being (not shown in [Fig. 1] for clarity). Alternatively, the application or computer program is stored directly in the control unit 4. The control unit 4 is connected to the vehicle's powertrain (not shown) and to a system 8 for managing static and / or dynamic data relating to road infrastructure and / or road traffic. More specifically, the control unit 4 is, for example, connected to the static and / or dynamic data management system 8 via wireless communication means 10 embedded in the vehicle 2. The wireless communication means 10 consist, for example, of a data transmitter / receiver coupled to an electronic communication card of the SIM card type (Subscriber Identity Module). The static and / or dynamic data management system 8 is, for example, configured according to a cloud architecture.Management system 8, for example, is an "electronic information horizon" type system (or "eHorizon" in English, which is classically based on the ADASIS data format standard – from "Advanced Driver-Assistance Systems Interface Specifications" – for predictive driver assistance systems, or on any other type of device). As is known, such an "eHorizon" type system allows for the management of both static data relating to road infrastructure (such as, for example, the nature of roads, intersections, applicable regulatory speed limits, etc.) and dynamic data (average speed of vehicles on the road, traffic density, dynamic data relating to road infrastructure elements 6, etc.).Such an "eHorizon" type system is capable of receiving this data, decoding it (via a decoder), reconstructing it (via a data reconstructor), and transmitting it to computer 4, and implements vehicle route prediction algorithms 2 using the concept of "most probable path".

[0025] The computer 4 is configured to calculate an optimal speed profile for vehicle 2 according to a predefined calculation method. As will be detailed later, this predefined calculation method (which is, for example, pre-recorded in the data processing unit comprising the computer 4) minimizes the Hamiltonian of a system of equations modeling the vehicle's driving and is configured to allow optimization of a criterion taking into account the energy consumption of vehicle 2 as well as the duration of the vehicle 2's journey between a starting point and an arrival point. This predefined calculation method is, for example, a calculation method implementing the Pontryagin maximum principle.Alternatively, the calculation method may be any other calculation method implementing optimization by road segments, considering different final speed constraints for each road segment, free a priori or forced a posteriori to the speed limit values ​​provided by the road infrastructure, and allowing optimization. of a criterion taking into account the energy consumption of vehicle 2 as well as the duration of the journey of vehicle 2 between the starting point and the arrival point.

[0026] The calculator 4 stores one or more recording(s) of past or recent driving phases of vehicle 2.

[0027] In addition to the computer 4 and the wireless communication means 10, the vehicle 2 also includes a vehicle navigation and positioning system 12 connected to the computer 4. The navigation and positioning system 12 is, for example, a GPS (Global Positioning System). The navigation map implemented within the navigation and positioning system 12 is instrumented. Preferably, the vehicle 2 also includes a display device 14 (such as a screen, for example), connected to the computer 4. According to a particular embodiment of the invention, the vehicle 2 also includes a cruise control system (not shown) equipped with means for calculating a set speed and connected to the vehicle 2's powertrain on the one hand, and to the computer 4 on the other.

[0028] In order to collect the information necessary for managing the calculation of the optimized speed profile, the data processing unit which is embedded in the computer 4 is connected to various devices of the vehicle 2.

[0029] The data processing unit is, for example, connected to a speed measuring device, a path determination device, and (optionally) an event detection device (such devices are not shown in the figures for clarity). It is understood that, in another embodiment, the data processing unit can be connected to other devices or equipment (not shown), such as, for example, devices related to a heat engine or a fuel cell, or devices related to an electric machine.

[0030] Each device is characterized by at least one state variable, allowing the operating state of the device to be described.

[0031] Vehicle 2 further comprises means for measuring the state(s) described by the state variable(s) and means for measuring the criterion to be optimized (such measuring means are not shown in [Fig. 1] for clarity). Preferably, vehicle 2 also comprises means for measuring disturbances and / or setpoints applied to vehicle 2 (such as, for example, torque and / or speed setpoints of vehicle 2).

[0032] The criterion to be optimized using the predefined calculation method implemented in the computer 4 takes into account the energy consumption of vehicle 2 as well as the duration of the journey of vehicle 2 between the starting point and the arrival point. The criterion to be optimized is represented by a criterion equation g(u,q). The criterion equation g(u,q) is a function of at least the instantaneous setpoint values ​​u and the variables state q. Preferably, the criterion equation is also a function of the disturbances and / or the setpoints w applied to vehicle 2, and is then written g (u,q,w).

[0033] The system is represented by a system of state equations f(u,q) modeling the dynamics of vehicle 2. The state equations f(u,q) are functions at least of the instantaneous setpoint values ​​u and the state variables q. Preferably, the state equations are also functions of the disturbances and / or setpoint points w applied to vehicle 2, and are then written as f(u,q,w).

[0034] The computer 4 is capable of receiving the measurement of each state variable value relating to each device. In addition, the computer 4 is capable of controlling each device to which it is connected, by issuing a setpoint, according to the value(s) of the variable(s) relating to that device.

[0035] The computer 4 is also configured to implement the principle of the PMP method, in other words, Pontryagin's Maximum Principle method, by determining the Hamiltonian function H(x, u*, X) from the different setpoint values ​​of an applicable setpoint domain. The notation u*, which represents the optimal control, is introduced here. Adjoint states X (also called "adjoint parameters," "Lagrange parameters," "adjoint vectors," or "co-state vectors") are also introduced. These adjoint states are associated with the state equations that represent the conditions of the dynamic behavior of the physical system and will allow for the complete solution of the optimization problem.

[0036] The criterion to be optimized taking into account the energy consumption of vehicle 2 as well as the duration of the journey of vehicle 2 between the starting point and the arrival point, the Hamiltonian function H (u,q,X,w) is then expressed as: H ( u, q, X, w ) = [ g ( u, q , w ) ] + + ÂT. [ f ( u, q, w ) ] with XT the transpose of the adjoint vector X, and / 7 > 0 ; 7 being a weighting coefficient allowing to weight the duration of the journey.

[0037] The calculator 4 includes a processor capable of implementing a set of instructions to perform these functions.

[0038] As illustrated in [Fig.2], the vehicle 2 speed control method includes a sub-method 20, implemented in the computer 4 embedded in the vehicle 2, for optimizing the energy consumption of the vehicle 2. This sub-method 20 itself includes a sub-method 21, implemented in the computer 4 embedded in the vehicle 2, for predicting the speed of the vehicle 2 over a journey between the starting point and the arrival point.

[0039] Initially, static and / or dynamic data relating to road infrastructure and / or road traffic and managed by the management system 8 are provided as input to the computer 4 via the management system 8. In addition, the route of vehicle 2 over a predetermined distance is predefined (for example via a user input) or predicted (for example via the management system 8 which transmits the information to the computer 4) within the computer 4. The computer 4 executes, for example, the application or computer program for the implementation of the process.

[0040] The sub-method 21 includes an initial step 22 in which the computer 4 determines a driving style for the driver of vehicle 2, based on one or more recordings of past or recent driving phases of vehicle 2. Preferably, step 22 of determining a driving style for the driver of vehicle 2 is performed by comparing, in said one or more recordings of past or recent driving phases of vehicle 2, an actual speed profile of vehicle 2 with an optimal speed profile of vehicle 2 calculated using the predefined calculation method. The driving styles determined for the driver of vehicle 2 can, for example, be divided into: an "economy" mode, a "sport" mode, and a "normal" mode.

[0041] The sub-process 21 includes a subsequent step 24 in which the computer 4 defines a series of notable points along the predefined path of vehicle 2, such notable points being characterized by a stop of vehicle 2 or a decrease in vehicle 2's speed. The series of notable points divides the path of vehicle 2 into a series of segments. Such a series of notable points associated with a path allows the path to be divided into a series of segments. The notable points of the path preferably characterize known locations along the path where vehicle 2 is likely to stop or slow down significantly (for example, below 30 km / h or more than 50% of its speed). By way of non-limiting examples, these notable points could be, for instance, a roundabout, a traffic circle, a speed bump, a change in the speed limit, etc., for predictable static data.

[0042] The sub-process 21 includes a subsequent step 26 in which the computer 4 calculates an optimal speed profile for the vehicle 2 according to the predefined calculation method. The travel time between the starting point and the arrival point of the vehicle 2 is weighted using the weighting coefficient in the Hamiltonian H(u,q,X,w) of the predefined calculation method.

[0043] The sub-process 21 includes a subsequent step 28 in which the computer 4 adapts the weighting coefficient according to the driving style determined for the driver of vehicle 2. For each segment of the journey between the starting point and the arrival point of vehicle 2, the adaptation step 28 provides an adapted speed profile for vehicle 2. The impact of the coefficient is explained as follows: - when the value of the weighting coefficient is low, it means that the travel time is of little importance, and therefore the minimization of the Hamiltonian is more focused on the energy consumption of vehicle 2, with a lower average speed for vehicle 2 - this can typically correspond to an "economy" mode determined for vehicle 2; - when the value of the weighting coefficient is high, it means that the duration of the journey is of great importance, and therefore the minimization of the Hamiltonian will tend to minimize this duration of the journey, with a higher average speed for vehicle 2 - this can typically correspond to a "sport" mode determined for vehicle 2; - when the value of the weighting coefficient V is intermediate between "low" and "high", this leads to a compromise between the energy consumption of vehicle 2 and the duration of the journey - this can typically correspond to a "normal" mode determined for vehicle 2.

[0044] According to one embodiment of the invention, step 28, which adapts the weighting coefficient according to the driving style determined for the driver of vehicle 2, is carried out using a scale of continuous values. According to another embodiment of the invention, step 28, which adapts the weighting coefficient, is carried out using a scale of discretized values ​​(such a scale being, for example, the scale described above: weighting coefficient value 7 "low", "medium", or "high").Preferably, step 28, which adapts the weighting coefficient according to the driving style determined for the driver of vehicle 2, is carried out via a closed-loop feedback correction (typically implemented by a Proportional-Integral-Derivative controller, a bisection controller, a trend-compensating controller, or any other feedback-type computational method known to those skilled in the art that provides a correction value to be applied at each instant to reduce the error between estimation and measurement). Step 28, which adapts the weighting coefficient, thus allows the actual speed profile of vehicle 2 to "match" a posteriori the optimal speed profile considered a priori as a prediction of this speed of vehicle 2 calculated using the predefined calculation method.

[0045] The sub-process 21 includes a subsequent step 30 in which the computer 4 provides, for each portion of the journey between the starting point and the arrival point, the adapted speed profile during step 28. This adapted speed profile is provided as a prediction of the speed of the vehicle 2 on this portion of the journey.

[0046] Once vehicle 2 has passed the end of the current portion of the journey, sub-process 21 loops back to step 22 of determining a driving style of the driver of vehicle 2, and so on until vehicle 2 reaches the destination point of the journey.

[0047] The sub-process 20 for optimizing the energy consumption of vehicle 2 includes a subsequent step 32 during which the computer 4 takes into account the predicted speed of vehicle 2 on each portion of the journey during step 30, to apply or recommend to vehicle 2 this predicted speed on the portion of the journey in question.

[0048] The method for controlling the speed of vehicle 2 includes a final step 34 in which the control unit 4 transmits a speed command to the vehicle 2's powertrain. This command is based on the vehicle 2's predicted speed as determined in step 30. When vehicle 2 is equipped with cruise control, the speed command is transmitted to the cruise control by the control unit 4. The cruise control then calculates a target speed based on this speed command. The speed of vehicle 2 is then regulated according to this target speed (which may vary depending on the calculated speed command). Alternatively, vehicle 2 may be an autonomous or semi-autonomous vehicle, in which case the speed command may be transmitted directly by the control unit 4 to the vehicle 2's powertrain.Alternatively, the energy consumption optimization process 20 for vehicle 2 can be used to recommend to the driver the optimal speed profile for vehicle 2 taken into account by the computer 4 (in this case, there is no longer a final step 36). This recommendation is made, for example, via a display of information on the display device 14, such information taking, for example, the form of a colored (dynamic) graphic recommendation area, within which the driver is prompted to position a needle representing the speed of vehicle 2 (the position of the needle being controlled by the accelerator pedal), or a haptic pedal (responding under the foot), or an auditory interface.

[0049] The vehicle 2 speed control method then loops back to step 22 of determining a driving style of the driver of vehicle 2, in order to be able to follow the evolutions of the route predictions given over time by the vehicle 2 navigation and positioning system 12.

[0050] The vehicle speed prediction method according to the invention makes it possible to obtain a robust prediction of the vehicle speed, without requiring preliminary learning phases, complex behavioral models, or additional measurements of "external" or "specific" data related to the vehicle or the driver. The method according to the invention also makes it possible to efficiently use an existing optimal speed profile calculation solution to obtain a suitable predictive speed profile, directly incorporating static and dynamic data from an "electronic information horizon" type system, such as speed constraints, road types, traffic, conditions, etc. The method according to the invention also makes it possible to simultaneously obtain a prediction of the Power demand (with torque or current depending on the powertrain technology) is necessary for correct energy analysis, optimization, and energy savings. Finally, in the method according to the invention, the step of adapting the weighting coefficient according to the driving style determined for the vehicle driver is quite fast, stable, and precise, and is carried out by means of typical "closed-loop" feedback corrections, which are capable of keeping up with the slow and reasonable dynamic response time to usual variations in driving behavior or traffic conditions.

Claims

Demands

1. A method (21), implemented in a computer (4) embedded in a vehicle (2), for predicting the speed of the vehicle (2) over a journey between a starting point and an arrival point, the vehicle (2) further being equipped with a vehicle (2) navigation and positioning system (12) connected to the computer (4), the computer (4) of the vehicle (2) further being connected to a system (8) for managing static and / or dynamic data relating to road infrastructure and / or road traffic, said static and / or dynamic data relating to road infrastructure and / or road traffic being provided as input to the computer (4), the route of the vehicle (2) over a predetermined distance being predefined or predicted within the computer (4), the computer (4) being configured to calculate an optimal speed profile of the vehicle (2) according to a predefined calculation method,said calculation method minimizing the Hamiltonian of a system of equations modeling the driving of the vehicle (2) and being configured to allow optimization of a criterion taking into account the energy consumption of the vehicle (2) as well as the duration of the journey, the computer (4) storing one or more recording(s) of past or recent driving phases of the vehicle (2), characterized in that the process (21) comprises the following steps:, - determination (22), based on one or more recordings) of past or recent driving phases of the vehicle (2), of a driving style of the driver of the vehicle (2), - definition (24) of a series of notable points on the route characterized by a stoppage of the vehicle (2) or a decrease in the speed of the vehicle (2), said series of notable points dividing the route into a series of portions, - for each segment of the journey, calculation (26) of an optimal speed profile of the vehicle (2) according to said predefined calculation method, the duration of the journey being weighted using a weighting coefficient in the Hamiltonian of said predefined calculation method, - adaptation (28) of said weighting coefficient according to the driving style determined for the driver of the vehicle (2), said adaptation step (28) providing, for each portion of the journey, an adapted speed profile of the vehicle (2), - for each portion of the journey, provision (30) of said adapted speed profile as a prediction of the speed of the vehicle (2) on said portion of the journey.

2. Method (21) according to claim 1, characterized in that said predefined calculation method is a calculation method implementing the Pontryagin maximum principle.

3. Method (21) according to claim 1 or 2, characterized in that the step (22) of determining a driving style of the driver of the vehicle (2) is carried out by comparing, in said one or more recordings) of past or recent driving phases of the vehicle (2), an actual speed profile of the vehicle (2) to an optimal speed profile of the vehicle (2) calculated using said predefined calculation method.

4. Method (21) according to any one of claims 1 to 3, characterized in that the step (28) of adapting the weighting coefficient according to the driving style determined for the driver of the vehicle (2) is carried out according to a scale of continuous or discretized values.

5. Method (21) according to any one of claims 1 to 4, characterized in that the step (28) of adapting the weighting coefficient according to the driving style determined for the driver of the vehicle (2) is carried out via a closed-loop feedback correction.

6. A method (20), implemented in a computer (4) embedded in a vehicle (2), for optimizing the energy consumption of the vehicle (2) over a journey between a starting point and an arrival point, the vehicle (2) further being equipped with a vehicle navigation and positioning system (12) connected to the computer (4), the computer (4) of the vehicle (2) further being connected to a system (8) for managing static and / or dynamic data relating to road infrastructure and / or road traffic, said static and / or dynamic data relating to road infrastructure and / or road traffic being provided as input to the computer (4), the route of the vehicle (2) over a predetermined distance being predefined or predicted within the computer (4), the computer (4) being configured to calculate an optimal speed profile of the vehicle (2) according to a predefined calculation method, said method of

7.

8. a calculation minimizing the Hamiltonian of a system of equations modeling the driving of the vehicle (2) and being configured to allow optimization of a criterion taking into account the energy consumption of the vehicle (2) as well as the duration of the journey, the computer (4) storing one or more recording(s) of past or recent driving phases of the vehicle (2), characterized in that the method further comprises a sub-method (21) of predicting the speed of the vehicle (2) according to any one of claims 1 to 5, and a step (32) of taking into account the speed of the vehicle (2) predicted on each portion of the journey, to apply or recommend to the vehicle said predicted speed on said portion of the journey. A method, implemented in a computer (4) embedded in a vehicle (2), for controlling the speed of the vehicle (2), the vehicle (2) further being equipped with a vehicle (2) navigation and positioning system (12) and a powertrain, both connected to the computer (4), the computer (4) of the vehicle (2) further being connected to a system (8) for managing static and / or dynamic data relating to road infrastructure and / or road traffic, said static and / or dynamic data relating to road infrastructure and / or road traffic being provided as input to the computer (4), the route of the vehicle (2) over a predetermined distance being predefined or predicted within the computer (4), the computer (4) being configured to calculate an optimal speed profile of the vehicle (2) according to a predefined calculation method,said calculation method minimizing the Hamiltonian of a system of equations modeling the driving of the vehicle (2) and being configured to allow optimization of a criterion taking into account the energy consumption of the vehicle (2) as well as the duration of the journey, the computer (4) storing one or more recording(s) of past or recent driving phases of the vehicle (2), characterized in that the method further comprises a sub-method (21) for optimizing the energy consumption of the vehicle according to claim 6, and a transmission step (34), to the powertrain of the vehicle (2), of a speed command established as a function of the predicted speed of the vehicle (2) taken into account. Computer (4) intended to be installed in a vehicle (2), the route of the vehicle (2) over a predetermined distance being predefined or predicted within the computer (4), the computer (4) being configured to calculate an optimal speed profile of the vehicle (2) according to a predefined calculation method, said calculation method minimizing the Hamiltonian of a system of equations modeling the driving of the vehicle (2) and being configured to allow optimization of a criterion taking into account the energy consumption of the vehicle (2) as well as the duration of a journey, the computer (4) storing one or more recording(s) of past or recent driving phases of the vehicle (2), characterized in that the computer (4) includes means for implementing the steps of the method (21) for predicting the speed of the vehicle (2) according to any one of claims 1 to 5 and / or of the method (20) for optimizing the energy consumption of the vehicle (2) according to claim 6 and / or of the method for controlling the speed of the vehicle (2) according to claim 7.

9. Vehicle (2), in particular automobile, characterized in that the vehicle (2) carries a computer (4) according to claim 8 and a navigation and positioning system (12) for the vehicle (2) connected to the computer (4), the computer (4) being capable of being connected to a system (8) for managing static and / or dynamic data relating to road infrastructure and / or road traffic.

10. Product computer program, downloadable from a communication network and / or stored on a computer-readable medium and / or executable by a processor, characterized in that it comprises program instructions, said program instructions being configured to implement the steps of the method (21) for predicting the speed of the vehicle (2) according to any one of claims 1 to 5 and / or the method (20) for optimizing the energy consumption of the vehicle (2) according to claim 6 and / or the method for controlling the speed of the vehicle (2) according to claim 7 when said instructions are executed on a computer (4) according to claim 8.

Images