Method and device for the predictive operation of a motor vehicle

Abstract

Method for the predictive operation of a motor vehicle based on a specification of several driving-relevant control variables (CVs), comprising the following steps: - Providing (S2) a first electronic horizon for a first part of a route to be traveled and a second electronic horizon for a second part of a route to be traveled, wherein the first and the second electronic horizon each have one or more route segments, wherein each of the route segments is assigned one or more route parameters and / or one or more control variable restrictions; - Providing a cost function for several target variables to be optimized, depending on the profiles of the driving-relevant control variables (DC) over the route to be driven; - Performing (S3) a first optimization step for the second electronic horizon, whereby the cost function is optimized under a specified predefined course of at least one of the control variables (SG) in order to obtain a dependency between the multiple target variables as a result of the first optimization step; - Applying the obtained dependency between the target variables to the cost function as a boundary condition at the end of the first electronic horizon; and - Performing (S4) a second optimization step with the applied cost function for the first electronic horizon, wherein the multiple control variables (SG) specify a target speed and a torque distribution, wherein the first optimization step is performed under a defined specification of the target speed profile over the route segments of the second electronic horizon, wherein the target speed profile is determined or provided from a target speed band assigned to the respective route segment, wherein information about the route to be driven, which a driver of the vehicle will drive with a certain probability, is provided by a navigation unit of the motor vehicle.

Description

Technical field

[0001] The invention relates generally to methods for operating a motor vehicle, in particular for specifying several driving-relevant control variables, such as speed, based on a forward-looking consideration of an upcoming driving route. State of the art

[0002] Modern motor vehicles are equipped with automatic cruise control (ACC). Automatic cruise control operates a motor vehicle according to a preset target speed.

[0003] The plan is to combine automatic cruise control with predictive cruise control (PCC), where information about the upcoming route is evaluated and the target speed is variably adjusted within predefined speed bands for each section of the route to achieve an optimization goal, such as fuel savings. This route prediction can be based, for example, on an electronic horizon provided by a navigation system to determine the respective target speed and / or other control parameters. An optimization goal for these control parameters could be, for example, minimizing overall fuel consumption for the entire route and / or the travel time. Especially when considering a

[0004] Due to the high computing capacity required, it is not possible to optimize the torque / power distribution of different drive motors in a hybrid drive system by jointly optimizing all control variables, such as speed, torque distribution, operating state and the like, over longer distances with currently available control units.

[0005] A method for operating a hybrid vehicle capable of operating in a combustion engine and / or electric motor drive mode may be provided. The method may stipulate that, on a route segment suitable for driving in electric motor drive mode, a switch to electric motor drive mode only occurs if a threshold value, representing or correlating with the efficiency advantage achievable on that route segment by switching from combustion engine to electric motor drive mode, is exceeded.Starting from the current position of the hybrid vehicle, the route is proactively scanned for such a route segment, with the threshold being adjusted if such a route segment is detected and driving through the route segment in electric drive mode would likely cause the state of charge of an electrical energy storage device of the hybrid vehicle to drop into a critical state of charge range.

[0006] Furthermore, a method for operating a vehicle is known in which vehicle navigation and traffic patterns associated with a predicted route for the vehicle are monitored, an imminent road load associated with the vehicle navigation and traffic patterns is predicted, and a vehicle propulsion power associated with the predicted imminent road load is estimated. A desired state-of-charge trajectory for the energy storage device is determined, and a state-of-charge trajectory for the energy storage device is predicted that corresponds to the estimated vehicle propulsion power and the desired state-of-charge trajectory for the energy storage device. Furthermore, the operation of the hybrid powertrain system is controlled in response to the predicted state-of-charge trajectory for the energy storage device.

[0007] A method for operating a hybrid drive is known that can be operated either in a first combustion engine operating mode or a second, purely electric motor operating mode. Before the start of a journey for which a route from a starting point to a destination is defined, the entire route is divided into several route segments according to a predefined logic. From these route segments, those segments are selected that are suitable for driving in the second operating mode based on specific route segment criteria. From the selected route segments, those are identified and driven through in the second operating mode where driving in the second operating mode results in a maximum overall fuel consumption advantage compared to driving in the first or any other operating mode.

[0008] Furthermore, a method for managing the energy consumption of a motor vehicle is known in which a journey is divided into segments, each assigned a speed at which the motor vehicle should travel through that segment. A probability is assigned to the speed transitions between each pair of segments. An energy consumption model is applied to these probabilities between the segments to perform optimization.

[0009] DE 10 2008 035 944 A1 describes a hierarchical method for optimizing the driving operation of a motor vehicle, particularly a truck. A higher-level layer segments an entire route based on route parameters and driving conditions and selects an optimization strategy for each segment. A lower-level layer locally optimizes the vehicle's operating points within a short horizon, taking into account boundary conditions from the higher-level layer to achieve economic and environmental objectives.

[0010] US 2009 / 0259355A1 discloses a power management system for hybrid vehicles that uses two-stage Dynamic Programming (DP) to achieve an optimal state of charge (SOC) profile for the battery over an entire trip. At a macro level, a global SOC profile is created for the entire trip to ensure that the battery reaches a defined target state of charge at the end of the trip. At a micro level, the SOC profile and the associated power split ratio are recalculated and adjusted at the end of each trip segment as the vehicle travels the route.

[0011] DE 10 2010 039 653 A1 describes a method for predictively determining the activation point of a range extender in an electric vehicle. This involves estimating the energy expected to be required for the remaining route and taking into account the state of charge of the traction battery. Based on this information and the expected power output of the range extender for each route segment, a decision is made as to when and where the range extender should be activated to ensure optimal operation, considering efficiency, comfort, and legal / environmental constraints. Disclosure of the invention

[0012] According to the invention, a method for the predictive operation of a motor vehicle based on a specification of several driving-relevant control variables according to claim 1, as well as a device, an arrangement, a motor vehicle and a computer program according to the dependent claims are provided.

[0013] Further details are specified in the dependent claims.

[0014] According to a first aspect, a procedure for the predictive operation of a motor vehicle based on a specification of several driving-relevant control variables is provided, which includes the following steps: - Providing a first electronic horizon for a first part of a route to be traveled and a second electronic horizon for a second part of a route to be traveled, wherein the first and second electronic horizons each have one or more route segments, each of which has one or more route parameters and / or one or more control variable restrictions assigned to it; - Providing a cost function for several target variables to be optimized, depending on the profiles of the driving-relevant control variables over the route to be traveled; - Performing a first optimization step for the second electronic horizon, whereby the cost function is optimized under a defined specification of the course of at least one of the control variables in order to obtain a dependency between the several target variables as a result of the first optimization step; - Applying the obtained dependency between the target variables to the cost function as a boundary condition at the end of the first electronic horizon; and - Performing a second optimization step with the applied cost function for the first electronic horizon.

[0015] One idea behind the above method is to optimize several control variables used to operate a motor vehicle, such as target speed and torque distribution in a hybrid drive system, using a two-stage optimization process. This two-stage optimization process is based on considering a first electronic horizon and a second electronic horizon.

[0016] The first electronic horizon designates a first segment (a first part) of the route to be considered for optimization, containing one or more track sections. The second electronic horizon designates a second segment (a second part) of the route to be considered for optimization, containing one or more track sections, and extends from the end of the first segment of the first electronic horizon to the end of the route to be considered for optimization. For each of the track sections, control parameter restrictions, such as minimum and / or maximum values ​​and / or other conditions, as well as one or more route parameters, such as the gradient of the respective track section, are specified for at least one of the several control variables.

[0017] Initially, neglecting the first electronic horizon, the first optimization step involves optimizing the manipulated variables based on the manipulated variable constraints and the path parameters of the path segments of the second electronic horizon. This optimization is performed according to an optimization goal, such as reducing fuel or energy consumption, which is defined with respect to one or more target variables, such as fuel consumption. At least one of the manipulated variables to be optimized is fixed for all path segments, i.e., constant or determined according to a predefined profile. The optimization in the first step is therefore only performed for the manipulated variables that are not fixed. This results in a dependency between the one or more target variables, corresponding to the optimization goal for the manipulated variables to be optimized.This results in a transition dependency for the transition between the first and second electronic horizons.

[0018] This transition dependency, which is assumed as a boundary condition, is used to define the optimization problem with respect to the first electronic horizon, which is solved in a second optimization step. For example, the cost function for the control variables to be optimized can be subjected to a transition dependency over the first electronic horizon with respect to at least one target parameter.

[0019] Dividing the optimization problem into two optimization steps allows for a faster execution of the optimization procedure. The first optimization step, with respect to the second electronic horizon, is faster because at least one of the manipulated variables remains constant. The second optimization step, with respect to the first electronic horizon, is also faster because the first electronic horizon can be significantly shorter than the second.

[0020] Dividing the optimization into two steps is particularly advantageous when a variation of one of the control variables, which is disregarded in the optimization of the second electronic horizon, only has short-term effects. Long-term optimization is then performed only for the remaining control variables.

[0021] Furthermore, one or more route parameters can specify the length of the route segment and / or the gradient on the route segment.

[0022] As an alternative to the target speed and torque distribution as control variables in a hybrid drive system, it can be provided that the following multiple control variables are specified: - a drive torque and a torque distribution between drive units of a hybrid drive system, or - a target speed and a partial torque for a drive unit of a hybrid drive system, or - a partial torque for one drive unit and a partial torque for another drive unit of a hybrid drive system.

[0023] In particular, the control variables can also include the selection of an operating state, especially an operating state that indicates whether an internal combustion engine of a hybrid drive system is switched on or not (start / stop operation).

[0024] The torque distribution can also be understood as the power distribution of the partial power outputs of the drive units in a hybrid drive system. Furthermore, a torque distribution of 0 for a hybrid drive system with an internal combustion engine and an electric motor as drive units indicates that the internal combustion engine is off, while a torque distribution greater than 0 defines a hybrid drive mode. In this way, the aforementioned control variables can also be used to model a start / stop operation of the hybrid drive system.

[0025] According to the invention, the multiple control variables specify a target speed and a torque distribution, wherein the first optimization step is carried out under a defined specification of the target speed profile over the track sections of the second electronic horizon, wherein the target speed profile is determined or provided from a target speed band assigned to the respective track section.

[0026] Furthermore, the one or more target variables can specify one or more of the following variables: - fuel consumption; - an electrical energy consumption; - a state of charge of an electrical energy storage device; - an emission of an exhaust pollutant; - aging of one or more components; and - a travel time for traversing the route.

[0027] It may be stipulated that the control variable restrictions specify one or more of the following conditions: - a maximum value of at least one of the control variables; and - a minimum value of at least one of the control variables.

[0028] Furthermore, the part of the route of the first electronic horizon can have a length of between 1 km and 10 km, in particular between 2 km and 3 km.

[0029] According to one embodiment, the first and / or the second optimization step can be carried out using an optimization algorithm designed to take into account the trends of the multiple control variables in the cost function.

[0030] In particular, the optimization algorithm can correspond to a recursive optimization procedure, especially dynamic programming.

[0031] According to another aspect, a device, in particular an optimization unit, is provided for the predictive operation of a motor vehicle based on a specification of several driving-relevant control variables, wherein the device is designed to: - to obtain a first electronic horizon for a first part of a route to be traveled and a second electronic horizon for a second part of a route to be traveled, wherein the first and the second electronic horizon each have one or more route segments, wherein each of the route segments is assigned one or more route parameters and / or one or more control variable restrictions; - to provide a cost function for several target variables to be optimized, depending on the profiles of the driving-relevant control variables over the route to be driven; - to perform a first optimization step for the second electronic horizon, whereby the cost function is optimized under a defined specification of the course of at least one of the control variables in order to obtain a dependency between the several target variables as a result of the first optimization step; - to apply the cost function with the obtained dependency between the target variables as a boundary condition at the end of the first electronic horizon; and - to perform a second optimization step with the applied cost function for the first electronic horizon.

[0032] According to another aspect, an arrangement is provided comprising the above optimization unit, a navigation unit for providing the first electronic horizon and the second electronic horizon, and a propulsion system.

[0033] Furthermore, the arrangement can be equipped with a hybrid drive system with multiple drive units.

[0034] According to another aspect, a computer program is provided which is set up to carry out all the steps of the above procedure. Brief description of the drawings

[0035] The embodiments are explained in more detail below with reference to the accompanying drawings. These show: Fig. 1. A schematic representation of a control system for operating a motor vehicle; and Fig. 2. A flowchart illustrating a procedure for the predictive operation of a motor vehicle. Description of embodiments

[0036] Fig. Figure 1 shows a control system 1 with an optimization unit 2, which performs an optimization procedure for specifying control variables SG for operating a motor vehicle based on a forecast of the route to be traveled. The optimization unit 2 receives information from a navigation unit 3 about a planned route, which has one or more segments.

[0037] The route is divided into electronic horizons. A first electronic horizon denotes a sub-route (a first part) of the route to be considered for optimization, containing one or more track segments. A second electronic horizon denotes a sub-route (a second part) of the route to be considered for optimization, containing one or more track segments, and extends from the end of the first electronic horizon to the end of the entire route.

[0038] Each section of the route is defined by one or more control variable restrictions, such as minimum and / or maximum values ​​of the considered control variables SG and / or other conditions for the control variables SG, as well as one or more route parameters. The route parameters can, for example, specify the length of the section in question, the gradient of the section, for example, in the form of information about the geographical elevation at the beginning and end of the section.The control variable restrictions can, for example, in the case of a control variable "target speed", specify a speed limit due to geographical or local conditions, such as curves or legal speed limits (entering a town), and / or a speed band that determines a preferred minimum and preferred maximum speed on the relevant section of the route.

[0039] Optimization unit 2 provides manipulated variables SG to a drive control unit 5 as specifications for operating the vehicle's drive system 4. The manipulated variables SG correspond to the manipulated variables to be optimized by optimization unit 2 and are provided in real time for the operation of the drive system 4.

[0040] In the present embodiment, the drive system 4 relates to a hybrid drive system, which is formed by the drive control unit 5, an electric drive 6, and an internal combustion engine 7 as drive units. The drive control unit 5 serves to control the electric drive 6 and the internal combustion engine 7 in a suitably appropriate manner. The electric drive 6 is supplied with electrical energy from an electrical energy storage device 9, and the internal combustion engine 7 is supplied with chemical energy. The electric drive 6 can also be operated recuperatively, i.e., mechanical energy from the vehicle or the internal combustion engine 7 can be converted into electrical energy for storage in the electrical energy storage device 9.

[0041] The hybrid drive system described here is only an example; however, an axial hybrid drive system or other drive systems, such as a hydraulic hybrid and the like, can also be used.

[0042] The manipulated variables SG can relate to the degree of control, such as the partial torque provided by each of the individual drive units 6, 7, or a specification for the distribution of the partial torques or partial powers to provide a total drive torque or total power. The manipulated variables SG, or the resulting partial torques to be controlled, can be converted into corresponding physical control variables S1, S2 in the drive control unit 5. For example, the physical control variables S1, S2 are the magnitude of the phase currents for the electric drive 6 and, for example, a throttle valve position, an ignition angle, injection timing and fuel injection quantities, and the like, for the internal combustion engine 7 (gasoline engine), so that the drive units 6, 7 are controlled in a known manner to provide the respective partial torque.

[0043] The drive control unit 5 can include an automatic speed control unit 51, which determines the drive torque to be applied based on a setpoint speed, which is provided, for example, as a first manipulated variable SG. The speed control unit 51 performs a control operation known per se, which outputs the drive torque to be applied as the controlled variable.

[0044] Using a torque distribution specified as a second control variable SG, which corresponds to an indication of what proportion of the drive torque to be supplied is provided by each of the drive units, the drive torque to be supplied can then be divided into the partial torques for the electric drive 6 and the combustion engine 7. As an alternative control variable SG corresponding to the above variant, which is provided by the optimization unit 2 as optimized control variable SG, the following can also be used: - Drive torque and torque distribution, or - the target speed and partial torque for one of the drive units 6, 7, or - the partial torque for the combustion engine 7 and the partial torque for the electric drive 6 are considered as control variables SG.

[0045] The following describes an embodiment based on the target speed and the torque distribution as control variables SG.

[0046] The speed control unit 51 determines the total torque to be applied from the specified target speed. This total torque is then divided into a partial torque for the electric drive 6 and a partial torque for the combustion engine 7, based on the torque distribution. The partial torques are converted into respective physical control variables S1 and S2 and made available to the respective drive units 6 and 7 for control purposes.

[0047] Optimization Unit 2 performs a predictive optimization procedure using Optimization Block 8. Optimization Block 8 provides an algorithm for performing an optimization procedure based on a given cost function (the objective function for the optimization) to determine an optimized control variable profile SG, i.e., a profile for the target speed and a torque distribution for the electronic horizon under consideration. By calling the optimization algorithm in Optimization Block 8, Optimization Unit 2 executes the predictive operation procedure for the vehicle.

[0048] In Fig. Figure 2 is a flowchart illustrating the procedure for the predictive operation of a motor vehicle, using the control variables to be optimized, "target speed" and "torque distribution," as an example.

[0049] In step S1, for example, navigation unit 3 transmits information about the route that the driver is likely to travel. This route can be determined, for instance, by using automatic route guidance (route navigation) to a destination specified by the driver. Based on the destination input, navigation unit 3 determines a preferred route and presents it to the driver for selection. After confirmation by the driver, the driver is guided along the selected route by directional instructions, which they typically follow.

[0050] Navigation unit 3 provides optimization unit 2 with information about the selected route in the form of control variable constraints and route parameters for individual route segments. The route parameters can be specified, for example, as the length of the respective route segment and / or the elevation difference to be overcome on that segment, e.g., by specifying the gradient. Control variable constraints can include a minimum and / or maximum value of the control variable SG, e.g., a minimum and / or maximum value of the target speed, which must be assumed based on route characteristics or legal requirements. Specifications regarding torque distribution can also be provided.

[0051] Furthermore, an average speed can be specified for each section of the route or determined from the provided minimum and maximum target speeds.

[0052] In step S2, a first electronic horizon and a second electronic horizon are determined from the information about the route. The first electronic horizon covers a relatively short segment of the total route, extending from the vehicle's current position to a position on the route at an average distance of, for example, 1 to 10 km, e.g., at 2 km. The second electronic horizon covers the remainder of the route, i.e., from the end of the first electronic horizon to, for example, the end of the entire likely route. The end of the second electronic horizon can also be assumed to be the end of a segment of the total route if, for example, the route is very long or contains a greater number of segments than a predefined maximum.

[0053] In step S3, the behavior of a manipulated variable SG or a subset of the manipulated variables SG is optimized for the second electronic horizon based on a predefined cost function of a first optimization problem. The cost function describes the influence of the manipulated variables SG on several target variables. These target variables can, for example, be information about the consumption of electrical energy to operate the electric drive 6 and / or the travel time for the distance traveled and / or the consumption of chemical energy to operate the combustion engine 7.

[0054] The optimization goal could be, for example, minimizing fuel consumption and / or minimizing travel time.

[0055] The first optimization step is performed with a reduction in degrees of freedom; that is, a fixed, predetermined path is assumed for one or more of the manipulated variables SG over the path of the second electronic horizon. In other words, the optimization is carried out with respect to a reduced number of manipulated variables SG, while the path of one or more of the manipulated variables SG over the path of the second electronic horizon is fixed.

[0056] For example, the manipulated variable "target speed" can be assumed to have a profile over the path of the second electronic horizon that corresponds to the provided or determined average speed; that is, the first optimization step is performed based on a fixed, predefined speed profile. Thus, the optimization of the first optimization step is performed only based on one or more of the remaining, non-predefined manipulated variables SG.

[0057] As a result of the first optimization step, which is carried out in optimization block 8 based on specifications for the target variables at the end of the second electronic horizon, the dependency between the several target variables at the beginning of the second electronic horizon—i.e., a dependency of the minimum (optimized) fuel consumption on the state of charge of a traction battery at the end of the second electronic horizon—is available in the form of an equation or a characteristic curve. This dependency contains the optimization of the target variables, namely fuel consumption or electrical energy consumption or the state of charge of the electrical energy storage device 9 of the traction battery for traversing the route of the second electronic horizon.

[0058] The state represented by the determined dependency also applies to the end of the first electronic horizon, since the second electronic horizon directly adjoins the first. There, a transition dependency is defined as a boundary condition for the determined dependency. This transition dependency can then be used to define a second optimization problem for a subsequent optimization step with respect to the path of the first electronic horizon.

[0059] The optimization problem underlying the second optimization step is based on the cost function, to which the transition dependency determined in the first optimization step is added, or rather, to which the transition dependency determined in the first optimization step is applied. Specifically, the integrated fuel consumption across the first electronic horizon is to be supplemented with the minimized fuel consumption, which depends on the state of charge of the traction battery at the end of the first electronic horizon. The resulting cost function forms the basis for the optimization of the second optimization step and is accordingly calculated in step S4 of optimization block 8 by a suitable optimization algorithm.

[0060] This allows the optimization of the control variables SG over the entire journey to be divided into two optimization steps, which are easier to solve compared to the original problem. In the embodiment described above, the target variables can be fuel consumption, journey time, and the state of charge of the traction battery. In addition, driving comfort, emissions, and component aging effects can also be considered as target variables and should be minimized accordingly.

[0061] In optimization block 8, optimization can be performed using various optimization algorithms. One preferred optimization algorithm is dynamic programming. Dynamic programming is a backward-computed, recursive optimization procedure that, starting from the specifications for the target variables at the end of the second electronic horizon, determines a control variable profile (SG) with the lowest costs according to the cost function.

[0062] In the above procedure, the torque distribution is planned as one of the control variables SG over the distance of the second (long-term) electronic horizon and the target speed according to the first (short-term) electronic horizon.

[0063] During the optimization process, further boundary conditions for the target variables can be considered, such as boundary conditions relating to ride comfort, emissions, and component aging effects. In particular, a predetermined state of charge can be specified for the traction battery at the end of the second electronic horizon, for example, a state of charge of 0%, 10%, 50%, or a full (100%) charge of the traction battery.

[0064] The optimization procedure described above, steps S1 to S4, is repeated cyclically, particularly after each passage through a section of the route. The determined profiles of the several control variables SG, i.e., the target speed and the torque distribution, are provided to the drive control unit 5 in the time-defined sequence for controlling the drive units 6 and 7.

Claims

[1] Method for the predictive operation of a motor vehicle based on a specification of several driving-related control variables (CV), comprising the following steps: - Providing (S2) a first electronic horizon for a first part of a route to be traveled and a second electronic horizon for a second part of a route to be traveled, wherein the first and the second electronic horizon each have one or more route segments, wherein each of the route segments is assigned one or more route parameters and / or one or more control variable restrictions; - Providing a cost function for several target variables to be optimized, depending on the profiles of the driving-relevant control variables (DC) over the route to be driven; - Performing (S3) a first optimization step for the second electronic horizon, whereby the cost function is optimized under a specified predefined course of at least one of the control variables (SG) in order to obtain a dependency between the multiple target variables as a result of the first optimization step; - Applying the obtained dependency between the target variables to the cost function as a boundary condition at the end of the first electronic horizon; and - Performing (S4) a second optimization step with the applied cost function for the first electronic horizon, wherein the multiple control variables (SG) specify a target speed and a torque distribution, wherein the first optimization step is performed under a defined specification of the target speed profile over the route segments of the second electronic horizon, wherein the target speed profile is determined or provided from a target speed band assigned to the respective route segment, wherein information about the route to be driven, which a driver of the vehicle will drive with a certain probability, is provided by a navigation unit of the motor vehicle. [2] Method according to claim 1, wherein one or more route parameters specify a length of the route segment and / or a gradient on the route segment. [3] Method according to claim 1 or 2, wherein the multiple control variables (CV) are specified: - a target speed and a torque distribution, or - a drive torque and a torque distribution between drive units (6, 7) of a hybrid drive system (4), or - a target speed and a partial torque for a drive unit (6, 7) of a hybrid drive system (4), or - a partial torque for a drive unit (6, 7) and a partial torque for another drive unit (6, 7) of a hybrid drive system (4). [4] Method according to any one of claims 1 to 3, wherein the one or more target variables specify one or more of the following variables: - fuel consumption; - an electrical energy consumption; - a state of charge of an electrical energy storage device (9); - an emission of an exhaust pollutant; - aging of one or more components; and - a travel time for traversing the route. [5] Method according to any one of claims 1 to 4, wherein the control variable restrictions specify one or more of the following conditions: - a maximum value of the manipulated variable (SG); and - a minimum value of the manipulated variable (SG). [6] Method according to any one of claims 1 to 5, wherein the part of the travel distance of the first electronic horizon has a length of between 1 km and 10 km. [7] Method according to any one of claims 1 to 6, wherein the first and / or the second optimization step is carried out using an optimization algorithm designed to take into account the trends of the multiple control variables (CV) in the cost function. [8] Method according to claim 7, wherein the optimization algorithm corresponds to a recursive optimization method. [9] Computer program configured to perform all steps of a method according to any one of claims 1 to 8. [10] Machine-readable storage medium on which a computer program according to claim 9 is stored. [11] Device, in particular optimization unit (2), for predictive operation of a motor vehicle based on a specification of several driving-relevant control variables (SG), wherein the device is designed to use the computer program according to claim 9 and / or the machine-readable storage medium according to claim 10: - to obtain a first electronic horizon for a first part of a route to be traveled and a second electronic horizon for a second part of a route to be traveled, wherein the first and the second electronic horizon each have one or more route segments, wherein each of the route segments is assigned one or more route parameters and / or one or more control variable restrictions; - to provide a cost function for several target variables to be optimized as a function of the curves of the driving-relevant control variables (SC) over the route to be driven; - to perform a first optimization step for the second electronic horizon, whereby the cost function is optimized under a defined specification of the course of at least one of the control variables (CV) in order to obtain a dependency between the several target variables as a result of the first optimization step; - to apply the cost function with the obtained dependency between the target variables as a boundary condition at the end of the first electronic horizon; and - to perform a second optimization step with the applied cost function for the first electronic horizon, wherein the multiple control variables (CV) specify a target speed and a torque distribution, wherein the first optimization step is performed under a defined specification of the target speed profile over the route segments of the second electronic horizon, wherein the target speed profile is determined or provided from a target speed band assigned to the respective route segment, wherein information about the route to be driven, which a driver of the vehicle will travel with a certain probability, is provided by a navigation unit of the motor vehicle. [12] Arrangement encompassing: - an optimization unit (2) according to claim 11; - a navigation unit (3) for providing the first electronic horizon and the second electronic horizon; and - a drive system (4). [13] Arrangement according to claim 12 comprising a hybrid drive system with multiple drive units (6, 7) as the drive system (4).

Images