Architecture and method for adaptive tire force prediction for integrated vehicle motion control

Abstract

An adaptive tire force prediction system for use in a motor vehicle includes a control module that executes a program code portion that receives real-time static and dynamic data from motor vehicle sensors, models forces on each tire of the motor vehicle at one or more incremental time steps, estimates actual forces of each tire of the motor vehicle at each of the one or more incremental time steps, adaptively predicts tire forces of each tire of the motor vehicle at each of the one or more incremental time steps, generates one or more control commands for actuators of the motor vehicle, captures a difference between real-time force estimates and nominal force calculations for each tire of the motor vehicle, and applies a compensation parameter to reduce tracking error in the one or more control commands to one or more actuators of the motor vehicle.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to control systems for motor vehicles, and more particularly to systems and methods for modeling automotive tire characteristics. BACKGROUND

[0002] Static and dynamic motor vehicle control systems are increasingly used to manage various static and dynamic motor vehicle performance characteristics. This is particularly true for challenging driving scenarios involving tire slip. In many challenging driving scenarios, control actions such as wheel and / or axle torque should be optimally distributed in order to fully utilize tire load capacity in both longitudinal and lateral directions. Typical tire load capacity management is conducted in on-board computing platforms or controllers and sensors, including inertial measurement units (IMUs), which measure how a motor vehicle moves in space, known as vehicle dynamics. IMUs measure vehicle acceleration on three axes: x-axis (forward / back), y-axis (left / right), and z-axis (up / down). IMUs also measure the rate at which a motor vehicle rotates about the three axes, known as pitch rate (about the y-axis), yaw rate (about the z-axis), and roll rate (about the x-axis). The on-board computing platforms or controllers use the measurement data to estimate forces acting on the vehicle.

[0003] While current tire force prediction systems and methods have achieved their intended purpose, there is a need for a new and improved tire force prediction system and method that can manage control of a motor vehicle during the tire's useful life and under many controlled and uncontrolled tire operating conditions, while maintaining or reducing cost and complexity, improving simplicity, and increasing redundancy and robustness.

[0004] SUMMARY

[0005] According to aspects of the present disclosure, an adaptive tire force prediction system for use in a motor vehicle includes one or more sensors disposed on the motor vehicle, the sensors measuring real-time static and dynamic data of the motor vehicle. The system further includes one or more actuators disposed on the motor vehicle, the actuators altering the static and dynamic behavior of the motor vehicle. The system further includes a control module having a processor, a memory, and an input / output (I / O) port, the control module executing portions of program code stored in the memory. The portions of program code include a first portion of program code receiving, via the I / O port, the real-time static and dynamic data from the one or more sensors, a second portion of program code modeling forces on each tire of the motor vehicle at one or more incremental time steps, and a third portion of program code estimating actual forces on each tire of the motor vehicle at each of the one or more incremental time steps. The portions of program code further include a fourth portion of program code adaptively predicting tire forces on each tire of the motor vehicle at each of the one or more incremental time steps, a fifth portion of program code generating one or more control commands for the one or more actuators of the motor vehicle, and a sixth portion of program code capturing a difference between real-time force estimates and nominal force calculations for each tire of the motor vehicle and applying a compensation parameter to reduce tracking errors in the one or more control commands to the one or more actuators of the motor vehicle.

[0006] In another aspect of the present disclosure, the first portion of program code further receives the real-time static and dynamic data from one or more of an inertial measurement unit (IMU) capable of measuring position, orientation, acceleration, and velocity in at least three dimensions, a wheel speed sensor capable of measuring wheel angular velocity of the motor vehicle, and a throttle position sensor capable of measuring throttle position of the motor vehicle. The first portion of program code also receives the real-time static and dynamic data from an acceleration pedal position sensor capable of measuring acceleration pedal position of the motor vehicle and a tire pressure monitoring sensor capable of measuring tire pressure of the motor vehicle.

[0007] In another aspect of the present disclosure, the real-time static and dynamic data further includes lateral velocity, longitudinal velocity, yaw rate, wheel angular velocity, and longitudinal, lateral, and normal forces on each tire of the motor vehicle.

[0008] In another aspect of the present disclosure, the second portion of program code further includes a piecewise affine model that generates predictions of longitudinal and lateral forces on each tire of the motor vehicle.

[0009] In another aspect of the disclosure, the piecewise affine model further includes a program code portion that calculates a linear approximation of the longitudinal force, lateral force, self-alignment torque, and friction coefficient at the contact patch between the tire and the surface, such that the linear approximation models tire force behavior in the linear region and the nonlinear region at one or more incremental time steps.

[0010] In another aspect of the disclosure, the third program code portion further includes estimating actual forces at each tire of the motor vehicle based on real-time static and dynamic data from one or more sensors using a look-up table.

[0011] In another aspect of the disclosure, the fourth program code portion adaptively predicts tire forces at each tire of the motor vehicle at each of the one or more incremental time steps to compensate for effects of tire deformation, tire wear, tire temperature, tire inflation pressure, and friction coefficient of the surface in contact with the tire at the contact patch.

[0012] In another aspect of the disclosure, tire deformation is quantified in terms of longitudinal and lateral slips, including slip angle and slip ratio.

[0013] In another aspect of the disclosure, the slip angle and slip ratio are defined as:

[0014] ,

[0015] wherein the actual tire forces are mathematically defined as:

[0016]

[0017] wherein, represents force calculations for each tire of the motor vehicle, wherein the coefficients , , , , are based on actual tire forces at different slip angles and different normal forces using nonlinear least squares data.

[0018] In another aspect of the disclosure, the tire forces are predicted based on predicted state variables, and the force model for each tire of the motor vehicle in the X and Y directions is defined as:

[0019] ;

[0020] ; and

[0021] wherein the tire forces are used to calculate state variables such as wheel angular velocity according to the following equation: .

[0022] In another aspect of the disclosure, an adaptive force prediction method for use in a motor vehicle includes processing static and dynamic vehicle information by a control module having a processor, a memory, and an I / O port, the control module executing program code portions stored in the memory. The program code portions measure real-time static and dynamic data using one or more sensors placed on the motor vehicle, change static and dynamic behavior of the motor vehicle using one or more actuators placed on the motor vehicle, and receive real-time static and dynamic data from the one or more sensors via the I / O port. The program code portions further model forces at each tire of the motor vehicle at one or more incremental time steps, estimate actual forces at each tire of the motor vehicle at the one or more incremental time steps, and adaptively predict tire forces at each tire of the motor vehicle at each of the one or more incremental time steps. The program code portions further generate one or more control commands for the one or more actuators of the motor vehicle, capture a difference between real-time force estimates and nominal force calculations at each tire of the motor vehicle, and apply a compensation parameter to reduce tracking errors in the one or more control commands to the one or more actuators of the motor vehicle.

[0023] In another aspect of the disclosure, the method further includes program code portions that receive real-time static and dynamic data from one or more of: an inertial measurement unit (IMU) capable of measuring position, orientation, acceleration, and velocity in at least three dimensions, a wheel speed sensor capable of measuring wheel angular velocity of the motor vehicle, a throttle position sensor capable of measuring throttle position of the motor vehicle, an acceleration pedal position sensor capable of measuring acceleration pedal position of the motor vehicle, and a tire pressure monitoring sensor capable of measuring tire pressure of the motor vehicle.

[0024] In another aspect of the disclosure, measuring real-time static and dynamic data further includes measuring lateral velocity, longitudinal velocity, yaw rate, wheel angular velocity, and longitudinal force, lateral force, and normal force on each tire of the motor vehicle.

[0025] In another aspect of the disclosure, the method further includes program code portions that generate predictions of longitudinal force and lateral force on each tire of the motor vehicle using a piecewise affine model.

[0026] In another aspect of the disclosure, generating predictions of longitudinal force and lateral force on each tire of the motor vehicle using a piecewise affine model further includes calculating a linear approximation of longitudinal force, lateral force, self-alignment torque, and coefficient of friction at a contact patch between the tire and a surface, such that the linear approximation models tire force behavior in linear and nonlinear regions at the one or more incremental time steps.

[0027] In another aspect of the disclosure, the method further includes estimating actual forces at each tire of the motor vehicle based on real-time static and dynamic data from one or more sensors using a look-up table.

[0028] In another aspect of the disclosure, the method further includes adaptively predicting tire forces at each tire of the motor vehicle at each step in one or more incremental time steps to compensate for effects of tire deformation, tire wear, tire temperature, tire inflation pressure, and friction coefficient of the surface in contact with the tire at the contact patch.

[0029] In another aspect of the disclosure, the method further includes quantifying tire deformation from longitudinal and lateral slips, including slip angle and slip ratio, where slip angle and slip ratio are defined as:

[0030] ,

[0031] where actual tire forces are mathematically defined as:

[0032]

[0033] where, represents force calculation for each tire of the motor vehicle, where coefficients , , , , are actual tire forces at different slip angles and different normal forces based on using nonlinear least square data.

[0034] In another aspect of the disclosure, the method further includes predicting tire forces based on predicted state variables, and force model for each tire of the motor vehicle in X and Y directions is defined as:

[0035] ;

[0036] ; and

[0037] where tire forces are used to calculate state variables such as wheel angular velocity from the following equation: .

[0038] In another aspect of the disclosure, a tire force prediction method for a motor vehicle includes processing static and dynamic vehicle information by a control module having a processor, a memory, and an I / O port, the control module executing program code portions stored in the memory. The program code portions measure real-time static and dynamic data using one or more sensors placed on the motor vehicle, the real-time static and dynamic data including: changing static and dynamic behavior of the motor vehicle using one or more actuators placed on the motor vehicle, and receiving real-time static and dynamic data from one or more of the following devices via the I / O port: an inertial measurement unit (IMU) capable of measuring position, orientation, acceleration, and velocity in at least three dimensions, a wheel speed sensor capable of measuring wheel angular velocity of the motor vehicle, a throttle position sensor capable of measuring throttle position of the motor vehicle, an acceleration pedal position sensor capable of measuring acceleration pedal position of the motor vehicle, and a tire pressure monitoring sensor capable of measuring tire pressure of the motor vehicle. The method further includes generating a prediction of longitudinal and lateral forces on each tire of the motor vehicle to model forces on each tire of the motor vehicle at one or more incremental time steps using a piecewise affine model that calculates a linear approximation of longitudinal force, lateral force, self-alignment torque, and friction coefficient at a contact patch between the tire and a surface, such that the linear approximation models tire force behavior in linear and nonlinear regions at the one or more incremental time steps. The method further includes estimating actual forces at each tire of the motor vehicle at the one or more incremental time steps using a look-up table that estimates actual forces at each tire of the motor vehicle based on the real-time static and dynamic data from the one or more sensors. The method further includes adaptively predicting tire forces at each tire of the motor vehicle at each of the one or more incremental time steps to compensate for effects of tire deformation, tire wear, tire temperature, tire inflation pressure, and friction coefficient of a surface in contact with the tire at the contact patch. The method further includes generating one or more control commands for one or more actuators of the motor vehicle, capturing a difference between real-time force estimates at each tire of the motor vehicle and nominal force calculations, and applying a compensation parameter to reduce tracking error in the one or more control commands to the one or more actuators of the motor vehicle. The compensation parameter includes quantifying tire deformation as a function of longitudinal and lateral slip including slip angle and slip ratio, defined as:

[0039] ,

[0040] wherein actual tire forces are mathematically defined as:

[0041]

[0042] wherein, representing the force calculation for each tire of the motor vehicle, wherein the coefficients , , , , are predicted based on actual tire forces at different slip angles and different normal forces using a non-linear least square data, and based on predicted state variables, and the force model on each of the tires of the motor vehicle is defined in X and Y directions as:

[0043] ;

[0044] .

[0045] The tire forces are used to calculate state variables such as wheel angular velocities according to the following equations: .

[0046] Further areas of applicability will become apparent from the description provided herein. It should be understood that the description and specific examples are intended for purposes of illustration only and are not intended to limit the scope of the present disclosure. BRIEF DESCRIPTION OF DRAWINGS

[0047] The drawings described herein are for illustrative purposes only and are not intended to limit the scope of the present disclosure in any way.

[0048] Figure 1 is a schematic representation of a motor vehicle having an adaptive tire force prediction system for integrated motion control in the motor vehicle in accordance with one aspect of the present disclosure;

[0049] Figure 2A is a block diagram of an adaptive tire force prediction system for integrated motion control in the motor vehicle of Figure 1 in accordance with another aspect of the present disclosure;

[0050] Figure 2B is a block diagram of a portion of an adaptive tire force prediction system for integrated motion control in the motor vehicle of Figure 2A in accordance with another aspect of the present disclosure, depicting real-time calculation of forces;

[0051] Figure 3 is a partial functional block diagram of a portion of an adaptive tire force prediction system for integrated motion control in the motor vehicle in accordance with one aspect of the present disclosure; and

[0052] Figure 4 is a flow chart depicting an adaptive tire force prediction method for integrated motion control in the motor vehicle in accordance with one aspect of the present disclosure. DETAILED DESCRIPTION

[0053] The following description is merely exemplary in nature and is not intended to limit the present disclosure, application, or uses.

[0054] Referring to Figures 1-2B An adaptive tire force prediction system 10 for integrated vehicle motion control of a motor vehicle 12 is shown. The system 10 includes a motor vehicle 12 and one or more controllers 14. The motor vehicle 12 is shown as a car, but it should be understood that the motor vehicle 12 can be a van, a bus, a tractor trailer, a semi-trailer, an SUV, a truck, a trike, a motorcycle, an airplane, an amphibious vehicle, or any other such vehicle that is in contact with the ground without departing from the scope or intent of the present disclosure. The motor vehicle 12 includes one or more wheels with tires 18 and a drivetrain 20. The drivetrain can include various components such as an internal combustion engine (ICE) 22 and / or an electric motor 24; and a transmission 26 capable of transmitting power generated by the internal combustion engine (ICE) 22 and / or the electric motor 24 to the wheels 27 and ultimately to the tires 18 attached to the wheels. In one example, the motor vehicle 12 can include an ICE 22 acting on a rear axle 28 of the motor vehicle 12 and one or more electric motors 24 acting on a front axle 30 of the motor vehicle 12. However, it should be understood that the motor vehicle 12 can use one or more ICEs 22 and / or one or more electric motors 24 configured in other ways without departing from the scope or intent of the present disclosure. For example, the motor vehicle 12 can have an ICE 22 acting on the front axle 30 only, while one or more electric motors 24 act on the rear axle 28 only. In another example, an ICE 22 can act on both the front and rear axles 30, 28, and electric motors can act on both the front and rear axles 30, 28.

[0055] In several aspects, drivetrain 20 includes one or more in-plane actuators 32. In-plane actuators 32 can include all-wheel drive (AWD) systems, including electronically controlled or electric AWD (eAWD) 34 systems; and limited slip differentials (LSD), including electronically controlled or electric LSD (eLSD) 36 systems. In-plane actuators 32 including eAWD 34 and eLSD 36 systems can generate and / or modify forces in the X and Y directions at tires 18 acting onto road contact patch 38 within certain load capacities. eAWD 34 systems can transfer torque from the front of motor vehicle 12 to the rear and / or from one side of motor vehicle 12 to the other. Likewise, eLSD 36 can transfer torque from one side of motor vehicle 12 to the other. In some examples, eAWD 34 and / or eLSD 36 can directly vary or manage torque transfer from ICE 22 and / or electric motor 24 and / or eAWD 34 and eLSD 36 can act on braking system 40 to adjust the amount of torque transferred to each of tires 18 of motor vehicle 12.

[0056] In further examples, motor vehicle 12 can include means for varying the normal force of each of tires 18 of motor vehicle 12 via one or more out-of-plane actuators 42, such as active aerodynamic actuators 44 and / or active suspension actuators 46. Active aerodynamic actuators 44 can actively or passively vary the aerodynamic profile of motor vehicle via one or more active aerodynamic elements 48, such as airfoils, spoilers, fans or other suction devices, actively managed Venturis, etc. Active suspension actuators 46, such as active dampers 50, etc. In several aspects, active dampers 50 can be magneto-rheological dampers or other such electrically, hydraulically, or pneumatically adjustable dampers without departing from the scope or intent of the present disclosure. For simplicity, in the following description, ICE 22, electric motor 24, eAWD 34, eLSD 36, braking system 40, aerodynamic control systems (including active aerodynamic elements 48, active dampers 46), etc. will be more broadly referred to as actuators 52.

[0057] The terms "forward," "rearward," "inward," "inwardly," "outward," "outwardly," "upward," and "downward" are terms used with respect to the orientation of the motor vehicle 12 as shown in the drawings of the present application. "Forward" refers to a direction toward the front of the motor vehicle 12, and "rearward" refers to a direction toward the rear of the motor vehicle 12. "Left" refers to a direction toward the left side of the motor vehicle 12 with respect to the front of the motor vehicle 12. Likewise, "right" refers to a direction toward the right side of the motor vehicle 12 with respect to the front of the motor vehicle 12. "Inward" and "inwardly" refer to a direction toward the interior of the motor vehicle 12, while "outward" and "outwardly" refer to a direction toward the exterior of the motor vehicle 12, "downward" refers to a direction toward the bottom of the motor vehicle 12, and "upward" refers to a direction toward the top of the motor vehicle 12. Furthermore, the terms "top," "over," "bottom," "side," and "above" are terms used with respect to the orientation of the actuator 52 and the motor vehicle 12 more generally shown in the drawings of the present application. Thus, while the orientation of the actuator 52 or the motor vehicle 12 can vary with particular uses, these terms remain applicable to the orientation of the system 10 and the motor vehicle 12 components shown in the drawings.

[0058] The controller 14 is a non-generic electronic control device having a preprogrammed digital computer or processor 54, a non-transitory computer readable medium or memory 56 for storing data (e.g., control logic, software applications, instructions, computer code, data, lookup tables, etc.), and input / output (I / O) ports 58. The computer readable medium or memory 56 includes any type of media capable of being accessed by a computer, such as read only memory (ROM), random access memory (RAM), hard drives, compact discs (CDs), digital video discs (DVDs), or any other type of memory. The "non-transitory" computer readable memory 56 does not include wired, wireless, optical, or other communication links transferring transitory electromagnetic signals or other signals. The non-transitory computer readable memory 56 includes media where the data can be permanently stored and media where the data can be stored and overwritten, such as rewritable optical discs or erasable memory devices. Computer code includes any type of program code, including source code, object code, and executable code. The processor 54 is configured to execute the code or instructions. The motor vehicle 12 can have a controller 14 that includes a dedicated Wi-Fi controller or engine control module, transmission control module, body control module, infotainment control module, etc. The I / O ports 58 can be configured to communicate via wired communication, wirelessly through Wi-Fi protocols under IEEE 802.1 lx, or using similar means without departing from the scope or intent of the present disclosure.

[0059] The controller 14 further includes one or more applications 60. An application 60 is a software program configured to perform a particular function or set of functions. The applications 60 can include one or more computer programs, software components, sets of instructions, procedures, functions, objects, classes, instances, related data, or portions thereof, adapted for implementation in suitable computer readable program code. The applications 60 can be stored in the memory 56 or in additional or separate memory. Examples of applications 60 include audio or video streaming services, games, browsers, social media, etc. In other examples, the applications 60 are used to manage body control system functions, suspension control system functions, aerodynamic control system functions, or similar functions in the example motor vehicle 12.

[0060] In several aspects, to manage the performance of the tires 18, the system 10 utilizes one or more applications 60 to model the tires 18. In one example, the applications 60 include an offline adaptation algorithm 61 and an online adaptation algorithm 63. The offline adaptation algorithm 61 models parameters of the tires 18 and compensates for slip effects of the tires 18 and actual tire 18 grip capabilities. Conversely, the online adaptation algorithm 63 models parameters of the tires 18 and compensates for wear of the tires 18, pressure of the tires 18, temperature of the tires 18, friction coefficients of the road surface or contact patch 38, etc. In several aspects, the applications 60 can be a model predictive control (MPC) algorithm or similar algorithm, or other known techniques for modeling and predicting behavior of the motor vehicle in short term ranges. In some examples, the short term prediction range is 10-15 sample times, or about 100-150 milliseconds, but the precise sample times and / or times defining the prediction range can vary from the above values without departing from the scope or intent of the present disclosure. In the MPC algorithm 62, the system 10 primarily operates in a feedback control model to adjust real-time constraints, optimize capabilities of the tires 18 and the actuators 52, and thereby maintain stability, yaw rate, lateral velocity, etc. of the motor vehicle 12. That is, the controller 14 combines the offline adaptation algorithm 61, the online adaptation algorithm 63, and motor vehicle state information in a hybrid tire force calculation 65, which can be used to generate a real-time optimization 67 that maximizes capabilities of the tires 18 and the actuators 52 to maintain performance of the motor vehicle 12 under various driving conditions.

[0061] The controller 14 receives data from various sensors 64 equipped on the motor vehicle 12 and obtains vehicle state information. The sensors 64 can measure and record a wide variety of motor vehicle 12 data. In several examples, the sensors 64 can include an inertial measurement unit (IMU) 66, a suspension control unit, such as a semi-active damping suspension (SADS) 68, a global positioning system (GPS) sensor 70, a wheel speed sensor 72, a throttle position sensor 74, an accelerator pedal position sensor 76, a brake pedal position sensor 78, a steering position sensor 80, a tire pressure monitoring sensor 82, an aerodynamic element position sensor 84.

[0062] The piecewise affine tire model 86 defines another application program 60 stored in the memory 56. The piecewise affine tire model 86 obtains the longitudinal and lateral forces at each tire 18 at each time step of the prediction horizon. The piecewise affine tire model 86 is adaptive, accounting for the effects of normal force variations and tire deformations in real time. In one example, as the motor vehicle 12 traverses a road or off-road surface, the tires 18 can encounter various surface disturbances present on the road or off-road surface. The surface disturbances can be potholes, road discontinuities, road cambers and / or shoulders, rocks, mounds of dirt, water, asphalt, or any other variety of surface friction coefficient and / or shape variations. When the tires 18 encounter a disturbance, the tires 18 can deviate from an intended path of travel on the surface in one or more of the X, Y, and Z directions due to the elasticity of each tire 18, as well as the suspension system components to which the wheels and tires 18 are connected. Also, the elasticity of the tires 18 can cause them to deform from a circular shape when encountering a surface disturbance. The actual forces on each tire 18 are obtained at the beginning of each prediction horizon through online estimation. A predicted tire 18 force is then calculated using an adaptive tire model that relies on known nominal tire parameters and potential compensations in order to match the real-time estimated forces under different driving conditions.

[0063] In several aspects, the piecewise affine tire model 86 is a linear approximation of a so-called “magic formula” (MF) tire model. The MF tire model is suitable for a variety of tire 18 types, structures, and operating conditions. In the MF tire model, each tire 18 is characterized by a plurality of coefficients for each force related to tire 18 performance. In some examples, the plurality of coefficients relate to contact patch, lateral and longitudinal forces, self-alignment torque, etc. These coefficients are used as a best fit between the experimentally determined performance data of the tire 18 and the MF model. These coefficients can then be used to generate equations showing how much force is generated on the tire 18 for a particular vertical load, as well as camber, slip angle

[0064] ​

[0065]

[0066]

[0067]

[0068]

[0069]

[0070]

[0071]

[0072]

[0073]

[0074] where, for the combined slip in longitudinal and lateral directions, the parameters can be calculated as:

[0075]

[0076]

[0077]

[0078]

[0079]

[0080]

[0081] However, when the combined slip in longitudinal and lateral directions is not used, Likewise, when the cornering slip is not used in the above equations, 1.

[0082] To calculate the derivative of the longitudinal force of each tire 18 in an analytical way and to reduce the calculation time, effort and resource utilization, piecewise affine linear approximations are used:

[0083]

[0084] where the tire parameters are found such that the linear approximation provides an accurate tire force behavior in both linear and nonlinear regions.

[0085] The deformation of the tires 18 can be quantified in terms of longitudinal and lateral slip 87. To calculate the longitudinal and lateral slip 87, the first longitudinal and lateral velocity coordinates associated with the body 88 of the motor vehicle 12 are calculated according to the following equations.

[0086]

[0087] where, and define the velocity of the front left tire 18 of the motor vehicle 12 in the X and Y directions, respectively. Likewise, and define the velocity of the front right tire 18 of the motor vehicle 12 in the X and Y directions, respectively. Likewise, and and and define the velocity of the rear left and right tires 18 of the motor vehicle 12 in the X and Y directions, respectively. A rotation matrix can be used to convert the velocities to wheel coordinates:

[0088]

[0089] The slip angle and slip ratio can then be calculated as:

[0090]

[0091] Since the side slip angle is only a function of the lateral and longitudinal velocities of the motor vehicle 12 at each tire 18 and depends only on the state of the body 88 and not the state of the wheels and tires 18 themselves, it is unlikely that the side slip angle will change drastically over a short time horizon. Therefore, for the purposes of certain calculations, it can be assumed that the side slip angle is constant over the length of the short time horizon.

[0092] The piecewise affine tire model 86 further includes a tire curve fitting process 90. The tire curve fitting process 90 is a computer executable program code portion or algorithm that matches the forces calculated at each tire 18 and from the MF tire model to the simplified piecewise affine tire model 86 for different slip angles and normal forces. The result of the above analysis is a number of coefficients that describe the various forces on each tire 18. More specifically, the coefficients that are obtained are based on the actual forces at each tire 18 for different slip angles and normal forces using a non-linear least squares data fitting method, which is mathematically defined as follows:

[0093] ​​​​

[0094] where represent the MF force calculations. Once the coefficients of different slip angles and normal forces , , , , are obtained, a look-up table 92 can be designed to provide coefficients of a range of slip angles and normal forces , , , , for use in the piecewise affine tire model 86.

[0095] To calculate the nominal forces on each tire 18, the system 10 estimates and / or predicts the state variables of the motor vehicle 12. Then, the longitudinal and lateral slip 87 information of the tires 18 of the motor vehicle 12 is calculated. More specifically, the slip rates and slip angles are calculated based on the available estimates at the beginning of the prediction horizon. Then, a look-up table containing the tire 18 curve information is used to calculate the forces on each tire 18 and predict the state variables of the body 88 and the wheels 27 in the piecewise affine prediction model 86. The predicted state variables are used in the context of the sampling time to calculate updated predicted slip rates and slip angles , and then, new forces on each tire 18 are calculated. The system 10 continuously and recursively updates the estimates and / or predictions of the state variables throughout the driving of the motor vehicle 12. Thus, the system 10 calculates the longitudinal and lateral forces on the tires 18 of the motor vehicle 12 in real-time.

[0096] Turning now to Figure 3 and with continued reference to Figures 1-2B , to provide robust tire 18 force predictions and achieve optimal control performance under different road and environmental conditions, including controlled and uncontrolled, the compensator algorithm 100 captures the difference between the real-time force estimates and the nominal tire 18 force calculations. Broadly stated, the formulaic representation of the compensator algorithm 100 can be presented as follows:

[0097]

[0098]

[0099] where,

[0100]

[0101]

[0102] In fact, and is a compensation parameter that matches real-time force estimates to model calculations of the tires 18 in order to be robust to road surface, tire 18 temperature and pressure variations, etc. Real-time estimates are provided to the compensator algorithm 100 by the controller 14. More specifically, in some aspects, the controller 14 is a vehicle dynamics controller (VDC) 102 or control module. The VDC 102 provides lateral and longitudinal force estimates on each tire 18 to the compensator algorithm 100. The compensator algorithm 100 also receives state estimates from an extended Kalman filter (EKF) 104. In several aspects, the EKF 104 is an optimal estimation algorithm that estimates the state of the system 10 from indirect and / or uncertain measurements. The compensation calculator 106 determines compensation parameters and from the outputs of the EKF 104 and the VDC 102 and are then input into a linear time-varying (LTV) MPC 108 that matches real-time force estimates to model calculations of the tires 18 over a prediction horizon.

[0103] The forces on each tire 18 can then be predicted from the predicted state variables and the tire model, as shown in the following modified MF equation:

[0104]

[0105]

[0106] Likewise, the forces on the tires 18 can also be used to calculate state variables, such as wheel angular velocity:

[0107]

[0108] Furthermore, state variables of the vehicle body 88 can be calculated in a similar manner.

[0109] The compensation parameters and can be estimated in several different ways. In one example, a least squares method (LSM) is used to estimate the compensation parameters in real-time to capture the difference between real-time force estimates and force calculations of the nominal tires 18. However, it should be understood that the LSM can be replaced by any optimization method that is capable of finding the compensation parameters with the least possible error without departing from the scope or intent of the present disclosure. In one example, given data {( , ),... ( , )}, the error related to is defined as Then, the goal is to find and the value of the error is minimized. To achieve this goal, the and are computed , and . Then, by LSM estimation, and the analytical solution of the minimum possible value of

[0110]

[0111] Referring now to the Figure 4 and with continued reference to Figures 1-3 an adaptive tire force prediction method 200 for integrated vehicle motion control is shown in flowchart form. The method 200 begins at block 202 where one or more sensors 64 equipped on the motor vehicle 12 continuously measure vehicle state information in real-time, including static and dynamic motor vehicle 12 state data. At block 204, the controller 14 executes a first portion of program code to receive the vehicle state information from the sensors.

[0112] At block 206, the controller 14 executes a second portion of program code that uses a piecewise affine model 86 to determine the forces at each tire 18 of the motor vehicle 12 at one or more incremental time steps. More specifically, the piecewise affine model 86 is a linear approximation of a so-called "magic formula" (MF) based tire model. In the MF tire model, each tire 18 is characterized by a plurality of coefficients for each force related to the tire 18 performance. In some examples, the plurality of coefficients relate to contact patch, lateral and longitudinal forces, self-alignment torque, etc. These coefficients are used as a best fit between experimentally determined performance data of the tire 18 and the MF model. These coefficients can then be used to generate equations showing how much force is generated on the tire 18 for a particular vertical load, as well as camber, slip angle and other parameters.

[0113] In several aspects, the piecewise affine linear approximation is used as:

[0114]

[0115] where the tire parameters are found such that the linear approximation provides accurate tire force behavior in both linear and nonlinear regions.

[0116] At block 208, the controller executes a third portion of program code that estimates the actual forces on each tire 18 of the motor vehicle 12 at each of one or more incremental time steps. More specifically, the system 10 estimates and / or predicts state variables of the motor vehicle 12. Then, longitudinal and lateral slip 87 information for the tires 18 of the motor vehicle 12 is calculated. Slip rate and slip angle are calculated based on the available estimates at the beginning of the prediction horizon. Then, a look-up table containing tire 18 curve information is used to calculate the forces on each tire 18 and predict the state variables of the vehicle body 88 and the vehicle wheels 27 in a piecewise affine prediction model 86. The predicted state variables are used in the context of the sampling time to calculate updated predicted slip ratios and slip angles Then, the new forces on each tire 18 are calculated. The system 10 continuously and recursively updates the estimates and / or predictions of the state variables throughout the driving of the motor vehicle 12. Thus, the system 10 calculates the longitudinal and lateral forces on the tires 18 of the motor vehicle 12 in real-time.

[0117] At block 210, the controller 14 executes a fourth portion of program code that adaptively predicts the forces at each tire 18 of the motor vehicle 12 at each of one or more incremental time steps. More specifically, the fourth portion of program code is adapted to account for tire 18 parameters that can change over the life of the given tire 18. For example, the fourth portion of program code compensates for the effects of tire 18 wear, tire 18 temperature, tire 18 inflation pressure, the coefficient of friction of the surface in contact with the tire 18, and the like.

[0118] At block 212, the controller 14 executes a fifth portion of program code that generates one or more control commands for one or more actuators 52 of the motor vehicle 12. The control commands can include torque requests, steering inputs, or the like. At block 214, the controller 14 executes a sixth portion of program code that accounts for the difference between the real-time force estimates and the nominal force calculations for each tire 18 of the motor vehicle 12. The sixth portion of program code also applies compensation parameters to reduce tracking errors in the one or more control commands to one or more actuators 52 of the motor vehicle 12 in order to maximize the adhesion between the tires 18 and the road surface at the contact patches 38 for the motor vehicle 12 in any given dynamic or complex driving situation.

[0119] At block 216, the method ends and returns to block 202, where the method 200 is iteratively, continuously, and / or recursively run again as the motor vehicle 12 is driven.

[0120] The adaptive tire force prediction system 10 and method 200 for integrated vehicle motion control of the present disclosure provide several advantages. These advantages include providing the motor vehicle 12 driver or operator with the maximum possible performance, stability, handling, maneuverability, steerability of the motor vehicle 12 under a variety of conditions, including inclement weather, tire 18 deformation, tire 18 wear, tire 18 temperature variations, tire 18 inflation levels, etc. Furthermore, the system 10 and method 200 can operate on the motor vehicle 12 in complex driving scenarios, including performance driving situations where the driver can attempt power slides, drifts, etc., and the system 10 and method 200 will generate the appropriate forces at the tire 18 / pavement or contact patch 38 while providing maximum tire 18 / pavement or contact patch 38 adhesion in driving scenarios where maximum grip is required. These advantages can all be obtained using the system 10 and method 200 described herein while maintaining or reducing cost and complexity, reducing calibration efforts, improving simplicity, and increasing redundancy and robustness.

[0121] The description of the present disclosure is merely exemplary in nature and variations that do not depart from the spirit and scope of the present disclosure are intended to be within the scope of the present disclosure. Such variations are not to be regarded as a departure from the spirit and scope of the present disclosure.

Claims

1. An adaptive tire force prediction system for use in a motor vehicle, the system comprising: one or more sensors placed on the motor vehicle that measure real-time static and dynamic data about the motor vehicle; one or more actuators placed on the motor vehicle that change the static and dynamic behavior of the motor vehicle; a control module having a processor, a memory, and an input / output (I / O) port, the control module executing program code portions stored in the memory, the program code portions comprising: a first program code portion that receives real-time static and dynamic data from the one or more sensors via the I / O port; a second program code portion that models forces at each tire of the motor vehicle at one or more incremental time steps; a third program code portion that estimates actual forces at each tire of the motor vehicle at each of the one or more incremental time steps; a fourth program code portion that adaptively predicts tire forces at each tire of the motor vehicle at each of the one or more incremental time steps; a fifth program code portion that generates one or more control commands for the one or more actuators of the motor vehicle; and a sixth program code portion that captures a difference between real-time force estimates and nominal force calculations at each tire of the motor vehicle and applies a compensation parameter to reduce tracking errors in the one or more control commands to the one or more actuators of the motor vehicle; wherein the second program code portion further comprises: a piecewise affine model that produces predictions of longitudinal and lateral forces on each tire of the motor vehicle; wherein the piecewise affine model further comprises: a program code portion that calculates a linear approximation of longitudinal forces, lateral forces, self-alignment torques, and a friction coefficient at a contact patch between the tire and a surface, such that the linear approximation models tire force behavior in linear and nonlinear regions at one or more incremental time steps; wherein the fourth program code portion adaptively predicts tire forces at each tire of the motor vehicle at each of the one or more incremental time steps to compensate for effects of tire deformation, tire wear, tire temperature, tire inflation pressure, and a friction coefficient of a surface in contact with the tire at a contact patch; wherein the tire deformation is quantified in terms of longitudinal and lateral slips, including slip angle and slip ratio; wherein slip angle and slip ratio are defined as: ； wherein, is the slip angle, is the slip ratio, is the speed of the tire in the Y direction, is the speed of the tire in the X direction; wherein actual tire forces are mathematically defined as: ； wherein y i represents the force calculation for each tire of the motor vehicle, wherein the coefficients c1, c2, c3, c4, c5 are based on actual tire forces at different slip angles and different normal forces using a non-linear least square data, is the tire normal force, is a piecewise affine function.

2. The adaptive tire force prediction system for use in a motor vehicle of claim 1, wherein, the first program code portion further comprises: receiving real-time static and dynamic data from one or more of: an inertial measurement unit (IMU) capable of measuring position, orientation, acceleration, and velocity in at least three dimensions; a wheel speed sensor capable of measuring wheel angular velocity of the motor vehicle; a throttle position sensor capable of measuring throttle position of the motor vehicle; an acceleration pedal position sensor capable of measuring acceleration pedal position of the motor vehicle; and a steering wheel angle sensor capable of measuring steering wheel angle of the motor vehicle. A tire pressure monitoring sensor capable of measuring tire pressure of the motor vehicle.

3. The adaptive tire force prediction system for use in a motor vehicle of claim 2, wherein, The real-time static and dynamic data further comprises: a lateral velocity; a longitudinal velocity; a yaw rate; a wheel angular velocity; and a longitudinal force, a lateral force, and a normal force of each tire of the motor vehicle.

4. The adaptive tire force prediction system for use in a motor vehicle of claim 1, wherein, The third program code portion further comprises estimating actual forces at each tire of the motor vehicle based on the real-time static and dynamic data from the one or more sensors using a look-up table.

Images