Control device for a vehicle

Abstract

Control device (30) for a vehicle (10), which control device (30) has at least one input (31) for vehicle data (53, 54) of the vehicle (10), which control device (30) has at least one output (32) for position values ​​(63, 64) for actuators (61, 62, 65, 66) of the vehicle (10), which control device (30) is configured to use a first predetermined vehicle model (MOD1), which first predetermined vehicle model (MOD1) includes the actuators (61, 62, 65, 66) used to influence the yaw moment (M), which control device (30) is designed to calculate, using the first predefined vehicle model (MOD1), as a function of a yaw moment setpoint (Ms) and the current actuated values ​​(63, 64) for the actuators (61, 62, 65, 66), modified actuated values ​​(63, 64) for the actuators (61, 62, 65, 66) by means of predefined calculation rules, which predefined calculation rules are determined by optimization of a performance function (Q) in the first predefined vehicle model (MOD1), which optimization includes a minimization of the deflection of the actuators (61, 62, 65, 66) under the constraint of generating a yaw moment according to the yaw moment setpoint (Ms), and which control device (30) is designed to output the changed control values ​​(63, 64) to the actuators (61, 62, 65, 66) in order to influence the yaw moment (M) of the vehicle (10) caused by the actuators (61, 62, 65, 66) as a function of the yaw moment setpoint (Ms) by means of this feedforward control.

Description

[0001] The invention relates to a control device for a vehicle.

[0002] German patent DE 10 2014 200 299 A1 discloses a method in which the actuators are controlled by a cascade feedforward control system depending on at least one target yaw behavior of the vehicle. Different target substitute models are used in a cascade-like manner, with each cascade adding a target substitute model with an additional actuator.

[0003] German patent application DE 10 2014 203 026 A1 discloses a vehicle dynamics control system in a motor vehicle with at least one electronic control unit and an electric-regenerative drive system. It describes a stored target self-steering behavior, which was empirically determined using a reference vehicle without recuperation.

[0004] DE 10 2014 214 272 A1 discloses a device and a method in which at least one reference yaw moment is determined from a vehicle dynamics target variable. A resulting yaw moment and an available additional yaw moment are determined such that their sum equals the reference yaw moment.

[0005] DE 10 2011 121 454 A1 discloses a motor vehicle control device which has an observer device designed to receive at least one sensor signal from a sensor via a signal input and to determine at least one estimated value for the lateral stiffness of the motor vehicle depending on the at least one sensor signal.

[0006] DE 10 2006 013 788 A1 discloses a vehicle control system with a computer that, depending on boundary conditions, including a resultant target force and a boundary friction circle of each wheel, calculates an integrated controlled variable that is used to control the braking force of each wheel in such a way as to optimize the µ-utilization ratio of the wheel, and a computer that calculates a steering-controlled variable that is used to control only the steering angle of each wheel in order to achieve the resultant target force.

[0007] One of the purposes of the invention is to provide a new control device for a vehicle.

[0008] The problem is solved by the subject matter of claim 1.

[0009] A control device for a vehicle has at least one input for vehicle data, which control device has at least one output for control values ​​for actuators of the vehicle, which control device is configured to use a first predefined vehicle model, which first predefined vehicle model includes the actuators used to influence the yaw moment, which control device is configured to calculate changed control values ​​for the actuators using the first predefined vehicle model as a function of a yaw moment setpoint and the current control values ​​for the actuators according to predefined calculation rules, and which control device is configured to output the changed control values ​​to the actuators in order to influence the yaw moment of the vehicle caused by the actuators as a function of the yaw moment setpoint through this feedforward control.By using a vehicle model that includes the actuators used to influence the yaw moment, optimization can be advantageously achieved, and the control values ​​can be determined directly. Therefore, no cascade-like calculation is required.

[0010] The specified calculation rules were determined by optimizing a performance function in the first given vehicle model. Performing this optimization to obtain the calculation rules results in high-quality feedforward control, thereby reducing the time required for further fine-tuning.

[0011] The optimization involves minimizing the actuator deflection under the constraint of generating yaw moment according to the yaw moment setpoint. It has been shown that minimizing actuator deflection results in more natural vehicle behavior. Users perceive the intervention in the vehicle's state as minimal.

[0012] According to a preferred embodiment, the displacement of the actuators is minimized using the Lagrange multiplier method. The application of the Lagrange multiplier method has proven advantageous for obtaining the calculation rules while taking the constraints into account.

[0013] According to a preferred embodiment, the control device has tuning maps assigned to at least some of the actuators, and it is configured to control the assigned actuator depending on the respective changed setpoint and assigned tuning map. This division into basic calculation and downstream tuning maps enables good overall vehicle tuning and leads to a significant reduction in the effort required for vehicle tuning.

[0014] According to a preferred embodiment, the control values ​​for the actuators include at least one of the control values ​​from the first group consisting of: - Setting value for a rear-wheel steering device; - Setpoint for a roll moment distribution device; - Setpoint for a longitudinal drive force distribution device in a vehicle with two-axle drive; - Setpoint for a transverse drive force distribution device; - Setpoint for a vehicle dynamics control system.

[0015] According to a preferred embodiment, the control device has at least two control values ​​from the first group, preferably at least three control values ​​from the first group, more preferably at least four control values ​​from the first group, and more preferably all five control values ​​from the first group. The specified actuators allow for a very comprehensive influence on the driving behavior and are therefore advantageous.

[0016] According to a preferred embodiment, the first specified vehicle model is selected from a second group consisting of: - linear single-track model, - non-linear single-track model, - linear two-track model and - non-linear two-track model.

[0017] The linear single-track model is relatively easy to use and requires less computing power than the other models. The non-linear models and the two-track model are more accurate, especially under extreme conditions, and therefore enable better driving performance.

[0018] According to a preferred embodiment, the control values ​​are calculated independently of one another, so the order in which they are calculated is irrelevant. This is possible because the calculations take place within the predefined vehicle model and the control value calculations are not dependent on the results of the other control value calculations. This independence also allows the calculations to be performed in parallel, for example, using multi-core processors. This results in very fast control.

[0019] According to a preferred embodiment, the control device is configured to determine the yaw moment setpoint from at least one of the following calculations from the third group consisting of: - Determination of the yaw moment setpoint as a function of the vehicle data in the first predefined vehicle model, and - Determining the yaw moment target value as a function of the vehicle data in a second predefined vehicle model in order to achieve the behavior of the second vehicle model.

[0020] These are two preferred methods for determining the yaw moment setpoint.

[0021] According to a preferred embodiment, a vehicle has a control device and actuators for influencing the yaw moment. Such a vehicle can advantageously utilize the control device.

[0022] According to a preferred embodiment, the control device is designed as a central control device.

[0023] Further details and advantageous embodiments of the invention will become apparent from the exemplary embodiments described below and illustrated in the drawings, which are in no way to be understood as limiting the invention, as well as from the dependent claims. It is understood that the features mentioned above and those to be explained below can be used not only in the combinations specified, but also in other combinations or individually, without departing from the scope of the present invention. The drawings show: Fig. 1 a vehicle and an associated single-track model, Fig. 2 in a schematic representation the vehicle with the control device, and Fig. 3. A quality criterion for minimization.

[0024] Fig. Figure 1 shows a vehicle 10. The model shown is a representation of the single-track model, which can be used to describe and calculate vehicle behavior, especially lateral dynamics. In the single-track model, both wheels of an axle are conceptually combined into a single central wheel. The center of gravity of vehicle 10 is marked with reference symbol 25, and the front wheel 21 has a distance I. F from the center of gravity 25. In the same way, the rear wheel 22 has a distance I R from the center of gravity 25. The total length between the front axle (front wheel) 21 and the rear axle (rear wheel) 22 is denoted as I.

[0025] In the exemplary embodiment, the vehicle 10 has both front-wheel steering and rear-wheel steering. The front-wheel steering angle is denoted by δF, and the rear-wheel steering angle by δR. The yaw angle ψ indicates the rotation of the vehicle about the vehicle-fixed vertical yaw axis 26 at the center of gravity 25. The yaw moment M is the torque about the yaw axis 26. Correspondingly, the yaw rate ψ' gives a measure of the change in the yaw angle ψ with time, and the yaw acceleration ψ'' gives the time derivative of the yaw rate ψ'. The yaw moment M corresponds to the product of the yaw moment of inertia I and the yaw acceleration ψ''.

[0026] The angle of drift β between the direction of movement of the vehicle's center of gravity 25 and the vehicle's longitudinal axis 28 is also shown.

[0027] The speed at which the front wheel 21 moves is given by v F denoted, and the speed of the rear wheel 22 is given by v Rdenoted. The angle between the direction of travel of the front wheel 21 and the actual speed v. F is called the front slip angle α F denoted. Similarly, the rear slip angle α denotes R the angle between the direction of travel of the rear wheel 22 and the direction of speed of the rear wheel 22.

[0028] The centripetal force acting on the center of gravity 25 in the single-track model is given by m · a y The lateral force acting on the front wheel 21 is shown. y,F denoted and the lateral force acting on the rear wheel 22 is denoted by F y,R designated.

[0029] The aim of the vehicle's control device is to influence the vehicle 10 with the help of actuators, which include, for example, an active chassis, in such a way as to generate a desired yaw moment.

[0030] The desired yaw moment is preferably specified as the yaw moment setpoint Ms.

[0031] Fig. Figure 2 shows a schematic representation of the vehicle 10 with the control device 30. The control device 30 has an input 31 and an output 32. Vehicle data 53 and 54 are supplied to the control device 30 via input 31. Vehicle data 53 originates, for example, from a device 51, which is a steering wheel position sensor. Vehicle data 54 originates, for example, from a device 52, which measures the current speed of the vehicle or the lateral acceleration. Vehicle data such as lateral acceleration is also referred to as vehicle status.

[0032] The driver or an autonomous driving device, as well as the vehicle 10 itself, provide vehicle data for the vehicle 10, in particular one or more from the group consisting of: - Front wheel steering angle δF or steering wheel position - Speed ​​v - Longitudinal acceleration a x - Lateral acceleration a y - Longitudinal slippage s

[0033] The vehicle data is transmitted via input 31 to a sequence of steps A, B, C, D, and from step D, control values ​​are output via output 32, for example to actuators 61, 62, 65 and 66. These are, for example, - Rear wheel steering device 61; - Roll moment distribution device 62; - Longitudinal drive force distribution device 65; - Vehicle dynamics control 66; - Transverse drive force distribution device.

[0034] The longitudinal drive force distribution device 65 is also referred to as the longitudinal drive torque distribution device, and it can variably distribute the drive torque in the longitudinal direction in a four-wheel drive vehicle 10 and thereby influence the behavior of the vehicle 10.

[0035] The lateral drive force distribution device is also referred to as a lateral drive torque distribution device, and it can variably distribute the drive torque in the lateral direction, thereby influencing the behavior of the vehicle 10. The adjustment of the distribution is achieved, for example, by a locking differential or an electronic transverse locking device.

[0036] The control device 30 works with a first vehicle model 92, which is designated MOD1, and optionally with a second vehicle model 93, which is designated MOD2.

[0037] In step A, the vehicle data 53, 54 of vehicle 10 are recorded via at least one input 31.

[0038] In step B, a yaw moment target value Ms for vehicle 10 is calculated based on vehicle data 53 and 54. The yaw moment target value Ms can be calculated from vehicle data 53 and 54 using the first vehicle model MOD1 to achieve a desired driving behavior for vehicle 10. Alternatively, the yaw moment target value Ms can be calculated based on the vehicle data and the second vehicle model MOD2 to achieve the behavior of vehicle 10 according to the second vehicle model MOD2. This makes it possible, for example, to adapt the driving behavior of vehicle 10 to that of another vehicle that is sportier or has different characteristics.

[0039] In step C, the yaw moment generation is distributed among the individual actuators 61, 62, 65, 66 depending on the yaw moment setpoint Ms by generating control values ​​for the actuators. This is described in more detail below. Preferably, the control device 30 also includes control variable limits, for example, a maximum steering angle of the rear wheel steering.

[0040] In step D, the setpoints for the individual actuators 61, 62, 65, 66 are optionally influenced by applying tuning characteristic maps AKF to the actuators 61, 62, 65, 66.

[0041] The control values ​​that may have been influenced in step D are then output via at least one output 32.

[0042] As a result, the systems of vehicle 10 are controlled by the method shown in such a way that its driving behavior exhibits the desired yaw moment within the limits of the systems' capabilities.

[0043] A suitable tire model must also be selected for the single-track model, as the interaction between the tires and the road significantly influences the vehicle's movement. Ideally, the tire model for the single-track controller should consider at least the following parameters: - Longitudinal slippage, - Hideaways, - Wheel load.

[0044] In the single-track model, further dependencies on the yaw rate Ψ' and the sideslip angle β can be added, for example via the front slip angle α. F and the rear slip angle α R , which may exhibit corresponding dependencies. Such dependencies transform the single-track model into a nonlinear single-track model.

[0045] Fig. Figure 3 shows a quality criterion Q, which can be used to determine the calculation rules for the setpoint of the actuators.

[0046] The vector u contains the modified normalized control values ​​u cur + u c , where u cur The current actuator position before the change and the setpoint change are both defined. Normalization is performed with respect to the setpoint range u. I max - u I min of the actuator with index I. For certain actuators, the zero position corresponds to a target value u. I tar = 0, for example, the target value u I tar For rear-wheel steering, the target value is 0°. For other actuators, such as the longitudinal distribution of drive torque in a four-wheel drive system, the target value may not be zero, and in such a case, the normalized setpoint for the actuator with the index I is defined as follows: (ulcur−uic−ultar) / (uImax−ulmin)

[0047] The size of the vector u depends on the number of actors to be considered. The scalar product of the transposed vector uT The vector u is used to form the setpoint changes. The setpoint changes uc are normalized; thus, for the actuator with index I, they result from the setpoint change u. I C , divided by the difference u I max - u I min of the maximum setpoint u I max and the minimum setpoint u I min .

[0048] In the right-hand part of the performance criterion Q, Lagrange multipliers are used as a constraint, stipulating that the yaw moment Ms generated by the actuators must be generated by the changes in the control input. Matrix B is called the effectiveness matrix, and it describes the effect of each change in the control input on the yaw moment.

[0049] In order to obtain the equations required for calculating the control values ​​for the individual actuators, an optimization of the quality criterion Q is carried out by minimizing it while taking the Lagrange multipliers into account.

[0050] The term u T The optimization process, u, leads to a minimization of actuator deflections, and the Lagrange multipliers ensure compliance with the constraint that the required yaw moment Ms in the vehicle model is achieved through the control parameter changes. Minimizing the actuator deflections / changes of the individual actuators results in vehicle behavior 10 that is perceived by the driver as natural. The driver does not perceive an ideal control device or controller, and the vehicle drives, for example, like a perfect passive reference vehicle.

[0051] The effectiveness matrix B defines, for each of the actuators with the assigned index I, the effect of the control value change on the yaw moment M. This can be achieved, for example, by a specification BI=∂Mz / ∂ul(uImax−ulmin)

[0052] Here, M Z the yaw moment, and ∂M z / ∂u I denotes the gradient of the yaw moment as a function of the setpoint of actuator I.

[0053] The optimization process yields formulas for the control values ​​of the individual actuators, depending on the vehicle data. These control values ​​can be calculated directly and independently of each other. Therefore, as with a cascade approach, it is not necessary to iteratively calculate a first control value in a first step and then, in a second step, calculate a second control value based on the first. The ability to calculate the individual control values ​​independently also enables parallel and thus faster calculations. Compared to an iterative calculation, the direct calculation method described here is less computationally intensive and can therefore be performed faster or with a lower computational load.

[0054] The optimization of the quality criterion Q shown makes it easy to consider additional factors. For each additional factor, its influence must be defined in the effectiveness matrix B, and then the optimization can be performed.

[0055] Some of the control values ​​calculated in the single-track model can be used directly in vehicle 10, for example by using the rear wheel steering angle δR for both rear wheels. Other control values, such as the control value for roll moment distribution, are converted in a further step into control values ​​for the stabilizers or for the active suspension or active dampers.

[0056] A special feature of the method is that the control values ​​are preferably not determined by the control device 30 through a control system, but are derived from the current state of the vehicle using the single-track models. This is also referred to as feedforward control and leads to good reproducibility of the vehicle behavior.

[0057] Today's vehicles are equipped with increasingly powerful actuators, which in turn offer new degrees of freedom. For example, a very strong torque vectoring method can be used in the wheel hub motors of electric vehicles. The behavior of the vehicle can thus be significantly influenced, shifting it towards the characteristics of another vehicle, such as a sportier vehicle or one with more dynamic stability. This results in increasing configuration options for customer vehicles.

[0058] Since all the different actuators can be taken into account in the control device 30, the control device 30 can be designed as a central control device 30.

[0059] The calculations shown allow for the modular consideration of additional actuators and vehicle data, and the intervention of the control device can be adapted.

[0060] The in step D of Fig. The two tuning characteristic curves mentioned above, AKF, enable subsequent fine-tuning of the actuators. For example, a setting in step C of Fig.The determined setpoint can be increased or decreased under a given condition by the tuning map AKF assigned to the actuator in order to change the behavior of the vehicle 10. Such fine-tuning can be used to equip the vehicle with driving characteristics typical for the respective model series, and fine-tuning by a downstream tuning map AKF can lead to a desired result much faster than adjusting the equations determined by optimization to determine the setpoints.

[0061] Naturally, numerous variations and modifications are possible within the scope of the present invention.

[0062] In the exemplary embodiment, a linear or non-linear single-track model was used, depending on the selected tire model. Alternatively, the use of a linear or non-linear dual-track model or another model is possible.

Claims

[1] Control device (30) for a vehicle (10), which control device (30) has at least one input (31) for vehicle data (53, 54) of the vehicle (10), which control device (30) has at least one output (32) for position values ​​(63, 64) for actuators (61, 62, 65, 66) of the vehicle (10), which control device (30) is configured to use a first predetermined vehicle model (MOD1), which first predetermined vehicle model (MOD1) includes the actuators (61, 62, 65, 66) used to influence the yaw moment (M), which control device (30) is designed to calculate, using the first predefined vehicle model (MOD1), as a function of a yaw moment setpoint (Ms) and the current actuated values ​​(63, 64) for the actuators (61, 62, 65, 66), modified actuated values ​​(63, 64) for the actuators (61, 62, 65, 66) by means of predefined calculation rules, which predefined calculation rules are determined by optimization of a performance function (Q) in the first predefined vehicle model (MOD1), which optimization includes a minimization of the deflection of the actuators (61, 62, 65, 66) under the constraint of generating a yaw moment according to the yaw moment setpoint (Ms), and which control device (30) is designed to output the changed control values ​​(63, 64) to the actuators (61, 62, 65, 66) in order to influence the yaw moment (M) of the vehicle (10) caused by the actuators (61, 62, 65, 66) as a function of the yaw moment setpoint (Ms) by means of this feedforward control. [2] Control device according to claim 1, in which the minimization of the deflection of the actuators (61, 62, 65, 66) is carried out by the method of Lagrange multipliers. [3] Control device according to one of the preceding claims, which has tuning characteristic maps (AKF) assigned to at least some of the actuators (61, 62, 65, 66), and which is designed to control the assigned actuator (61, 62, 65, 66) depending on the respective changed setpoint (63, 64) and assigned tuning characteristic map (AKF). [4] Control device (30) according to one of the preceding claims, wherein the actuating values ​​(63, 64) for the actuators (61, 62, 65, 66) comprise at least one of the actuating values ​​(63, 64) from the first group consisting of: - Setting value for a rear wheel steering device (61); - Setpoint for a roll moment distribution device (62); - Setpoint for a longitudinal drive force distribution device (65) in a vehicle with two-axle drive; - Setpoint for a transverse drive force distribution device; - Control value for a vehicle dynamics control system (66). [5] Control device (30) according to claim 4, which has at least two control values ​​(63, 64) from the first group, preferably at least three control values ​​(63, 64) from the first group, more preferably at least four control values ​​(63, 64) from the first group, and more preferably all five control values ​​(63, 64) from the first group. [6] Control device (30) according to one of the preceding claims, wherein the first predetermined vehicle model (MOD1) is selected from a second group consisting of: - linear single-track model, - non-linear single-track model, - linear two-track model and - non-linear two-track model. [7] Control device (30) according to one of the preceding claims, in which the calculation of the actuating values ​​(63, 64) is carried out independently of each other, so that the order of the calculation of the actuating values ​​(63, 64) is irrelevant. [8] Control device (30) according to one of the preceding claims, which is configured to determine the yaw moment setpoint (Ms) from at least one of the following calculations from the third group consisting of: - Determination of the yaw moment setpoint (Ms) as a function of the vehicle data in the first predefined vehicle model (MOD1), and - Determination of the yaw moment setpoint (Ms) as a function of the vehicle data in a second predefined vehicle model (MOD2) in order to achieve the behavior of the second vehicle model (MOD2). [9] Vehicle with a control device (30) according to one of the preceding claims, which has actuators (61, 62, 65, 66) for influencing the yaw moment (M).

Images