Gain optimization and selection for vehicle systems

Abstract

Gain optimization and selection for vehicle systems are provided. A vehicle system for selecting a gain set for controlling a vehicle includes a plurality of sensors configured to detect one or more vehicle characteristics and a control module. The control module is configured to: optimize a plurality of gain sets for a plurality of different driving conditions, each gain set comprising one or more values representing a relationship between an input of the control module and an output of the control module; determining a current driving condition associated with the vehicle based on the one or more vehicle characteristics; selecting a gain set of the plurality of optimized gain sets based on a current driving condition associated with the vehicle; and generating a control signal for controlling at least one operation of the vehicle based on the one or more values of the selected gain set. Other example vehicle systems and control methods are also disclosed.

Description

Technical Field

[0001] The information provided in this section is for the purpose of presenting the general background of this disclosure. To the extent described in this section, the work of the currently named inventors, and aspects of the description that may otherwise not conform to the prior art at the time of filing, are neither explicitly nor implicitly considered to be prior art of this disclosure.

[0002] This disclosure relates to gain optimization and selection for vehicle systems. Background Technology

[0003] Vehicles include control systems for controlling various vehicle operations, such as controlling motion and / or trajectory. For example, some vehicles (e.g., autonomous vehicles, semi-autonomous vehicles, etc.) may rely on sensor feedback to plan and / or control vehicle movement, such as acceleration, deceleration, turning, lane changing, etc. In some cases, vehicles may utilize a set of system gains for this type of vehicle control. For example, a vehicle control system may be provided with a set of system gains used for speed estimation across a variety of driving conditions. Summary of the Invention

[0004] A vehicle system for selecting a set of gains for controlling a vehicle includes: a plurality of sensors configured to detect one or more vehicle characteristics, and a control module communicating with the plurality of sensors. The control module is configured to: optimize a plurality of gain sets for a plurality of different driving conditions, each gain set including one or more values ​​representing a relationship between inputs and outputs of the control module; determine current driving conditions associated with the vehicle based on one or more vehicle characteristics; select a gain set of the plurality of optimized gain sets based on the current driving conditions associated with the vehicle; and generate control signals based on one or more values ​​of the selected gain set for controlling at least one operation of the vehicle.

[0005] Among other features, the control module is configured to: receive a dataset associated with multiple driving conditions; and, based on the received dataset, optimize the cost function using an optimization algorithm based on the performance of estimators for multiple gain sets for multiple driving conditions.

[0006] Among other features, the dataset includes lateral vehicle acceleration values ​​and wheel slip ratio values ​​for each driving condition.

[0007] Among other features, the control module is configured to determine the optimal values ​​of multiple gain sets based on a cost function equation with correlation coefficients and root mean square error (RMSE) components.

[0008] Among other features, the control module is configured to detect singularity in a portion of the dataset and prevent the optimization algorithm from exploiting this portion of the optimization space when determining the optimal value of multiple gain sets based on the cost function equation.

[0009] Among other features, the control module includes a first model and a second model, and the control module is configured to optimize a first subset of multiple gain sets for the first model and a second subset of multiple gain sets for the second model, respectively.

[0010] Among the other features, the first model is a dynamic model, and the second model is a kinematic model.

[0011] Among other features, the control module is configured to freeze a first subset of multiple gain sets and optimize a second subset of multiple gain sets for the second model while freezing the first subset.

[0012] Among other features, the control module is configured to: determine a first vehicle control state using a first model based on a first subset of multiple gain sets; determine a second vehicle control state using a second model based on the first vehicle control state and a second subset of multiple gain sets; and generate a control signal based on the second vehicle control state.

[0013] Among the other features, the second vehicle control state is the lateral speed estimation.

[0014] Among the other features, the current driving condition is the first current driving condition, and the gain set of multiple optimized gain sets is the first gain set.

[0015] Among other features, the control module is configured to: determine a second current driving condition associated with the vehicle based on one or more vehicle characteristics; select a second gain set of multiple optimized gain sets based on the determined second driving condition associated with the vehicle; blend a first gain set into the second gain set to provide a smooth transition; and generate a control signal based on one or more values ​​of the second gain set for controlling at least one operation of the vehicle.

[0016] Among the other features, one or more vehicle characteristics include the vehicle's lateral acceleration value and the slip ratio value of each wheel of the vehicle.

[0017] Among the other characteristics, driving conditions include one or more of the following: conditions of decreased surface friction coefficient, conditions of increased lateral acceleration, and normal driving conditions.

[0018] A vehicle system for selecting a set of gains for controlling a vehicle includes: a plurality of sensors configured to detect one or more vehicle characteristics, and a control module communicating with the plurality of sensors. The control module is configured to: receive a plurality of optimized gain sets for a plurality of different driving conditions, each gain set including one or more values ​​representing a relationship between inputs and outputs of the control module; determine current driving conditions associated with the vehicle based on one or more vehicle characteristics; select a gain set of the plurality of optimized gain sets based on the current driving conditions associated with the vehicle; and generate a control signal based on one or more values ​​of the selected gain set for controlling at least one operation of the vehicle.

[0019] Among other features, the control module is a first control module inside the vehicle, and the vehicle system also includes a second control module outside the vehicle.

[0020] Among other features, the second control module is configured to optimize multiple gain sets for multiple different driving conditions and transmit the multiple optimized gain sets to the first control module.

[0021] Among other features, the control module includes a first model and a second model, and the gain set includes a first subset for the first model and a second subset for the second model.

[0022] Among other features, the control module is configured to: determine a first vehicle control state using a first model based on a first subset of the gain set; determine a second vehicle control state using a second model based on the first vehicle control state and a second subset of the gain set; and generate a control signal based on the second vehicle control state.

[0023] Among the other features, the first model is a dynamic model, and the second model is a kinematic model.

[0024] Among the other features, the current driving condition is the first current driving condition, and the gain set of multiple optimized gain sets is the first gain set.

[0025] Among other features, the control module is configured to: determine a second current driving condition associated with the vehicle based on one or more vehicle characteristics; select a second gain set of multiple optimized gain sets based on the determined second driving condition associated with the vehicle; blend a first gain set into the second gain set to provide a smooth transition; and generate a control signal based on one or more values ​​of the second gain set for controlling at least one operation of the vehicle.

[0026] Among the other features, one or more vehicle characteristics include the vehicle's lateral acceleration value and the slip ratio value of each wheel of the vehicle, and driving conditions include one or more of the following: a decrease in the coefficient of surface friction, an increase in lateral acceleration, and normal driving conditions.

[0027] A control method for selecting a gain set for controlling a vehicle includes: optimizing multiple gain sets for multiple different driving conditions; determining current driving conditions associated with the vehicle based on one or more vehicle characteristics; selecting a gain set of multiple optimized gain sets based on the determined driving conditions associated with the vehicle; and generating a control signal based on one or more values ​​of the selected gain set for controlling at least one operation of the vehicle.

[0028] The following solutions are provided:

[0029] 1. A vehicle system for selecting a set of gain for controlling a vehicle, the vehicle system comprising:

[0030] Multiple sensors configured to detect one or more vehicle characteristics; and

[0031] The control module communicates with multiple sensors and is configured as follows:

[0032] Multiple gain sets are optimized for various driving conditions, each gain set including one or more values ​​representing the relationship between the input and output of the control module;

[0033] Determine the current driving conditions associated with the vehicle based on one or more vehicle characteristics;

[0034] A gain set is selected based on the current driving conditions associated with the vehicle, comprising multiple optimized gain sets; and

[0035] A control signal is generated based on one or more values ​​of a selected gain set to control at least one operation of the vehicle.

[0036] 2. The vehicle system according to Scheme 1, wherein the control module is configured to: receive a dataset associated with multiple driving conditions; and optimize the cost function using an optimization algorithm based on the received dataset and the performance of an estimator for multiple gain sets for multiple driving conditions.

[0037] 3. The vehicle system according to Scheme 2, wherein the dataset includes lateral vehicle acceleration values ​​and wheel slip ratio values ​​for each driving condition.

[0038] 4. The vehicle system according to Scheme 2, wherein the control module is configured to determine the optimal value of multiple gain sets based on a cost function equation having correlation coefficients and root mean square error (RMSE) components.

[0039] 5. The vehicle system according to Scheme 4, wherein the control module is configured to detect singularities in a portion of the dataset and prevent the optimization algorithm from utilizing that portion of the optimization space when determining the optimal value of multiple gain sets based on the cost function equation.

[0040] 6. The vehicle system according to Scheme 1, wherein:

[0041] The control module includes a first model and a second model; and

[0042] The control module is configured to optimize a first subset of multiple gain sets for the first model and a second subset of multiple gain sets for the second model, respectively.

[0043] 7. The vehicle system according to Scheme 6, wherein the first model is a dynamic model and the second model is a kinematic model.

[0044] 8. The vehicle system according to Scheme 6, wherein the control module is configured to freeze a first subset of multiple gain sets, and optimize a second subset of multiple gain sets for a second model while freezing the first subset.

[0045] 9. The vehicle system according to Scheme 6, wherein the control module is configured as follows:

[0046] The first vehicle control state is determined using a first model based on a first subset of multiple gain sets;

[0047] The second vehicle control state is determined using a second model based on the first vehicle control state and a second subset of multiple gain sets; and

[0048] Control signals are generated based on the second vehicle control state.

[0049] 10. The vehicle system according to Scheme 9, wherein the second vehicle control state is a lateral speed estimation.

[0050] 11. The vehicle system according to Scheme 1, wherein:

[0051] The current driving conditions are the primary current driving conditions;

[0052] The gain set of multiple optimized gain sets is the first gain set; and

[0053] The control module is configured as follows:

[0054] A second current driving condition associated with the vehicle is determined based on one or more vehicle characteristics;

[0055] A second gain set is selected based on the determined second driving conditions associated with the vehicle, which are multiple optimized gain sets.

[0056] Blend the first gain set into the second gain set to provide a smooth transition; and

[0057] A control signal is generated based on one or more values ​​of a second gain set to control at least one operation of the vehicle.

[0058] 12. The vehicle system according to Scheme 1, wherein one or more vehicle characteristics include the lateral acceleration value of the vehicle and the slip ratio value of each wheel of the vehicle.

[0059] 13. The vehicle system according to Scheme 1, wherein the driving conditions include one or more of the following: a decrease in surface friction coefficient, an increase in lateral acceleration, and normal driving conditions.

[0060] 14. A vehicle system for selecting a set of gain for controlling a vehicle, the vehicle system comprising:

[0061] Multiple sensors configured to detect one or more vehicle characteristics; and

[0062] The control module communicates with multiple sensors and is configured as follows:

[0063] Receive multiple optimized gain sets for multiple different driving conditions, each gain set including one or more values ​​representing the relationship between the input and output of the control module;

[0064] Determine the current driving conditions associated with the vehicle based on one or more vehicle characteristics;

[0065] A gain set is selected based on the current driving conditions associated with the vehicle, comprising multiple optimized gain sets; and

[0066] A control signal is generated based on one or more values ​​of a selected gain set to control at least one operation of the vehicle.

[0067] 15. The vehicle system according to claim 14, wherein:

[0068] The control module is the primary control module inside the vehicle.

[0069] The vehicle system also includes a second control module located outside the vehicle; and

[0070] The second control module is configured to optimize multiple gain sets for multiple different driving conditions and transmit the optimized gain sets to the first control module.

[0071] 16. The vehicle system according to claim 14, wherein:

[0072] The control module includes a first model and a second model;

[0073] The gain set includes a first subset for the first model and a second subset for the second model; and

[0074] The control module is configured to: determine a first vehicle control state using a first model based on a first subset of the gain set; determine a second vehicle control state using a second model based on the first vehicle control state and a second subset of the gain set; and generate a control signal based on the second vehicle control state.

[0075] 17. The vehicle system according to Scheme 16, wherein the first model is a dynamic model and the second model is a kinematic model.

[0076] 18. The vehicle system according to claim 14, wherein:

[0077] The current driving conditions are the primary current driving conditions;

[0078] The gain set of multiple optimized gain sets is the first gain set; and

[0079] The control module is configured as follows:

[0080] A second current driving condition associated with the vehicle is determined based on one or more vehicle characteristics;

[0081] A second gain set is selected based on the determined second driving conditions associated with the vehicle, which are multiple optimized gain sets.

[0082] Blend the first gain set into the second gain set to provide a smooth transition; and

[0083] A control signal is generated based on one or more values ​​of a second gain set to control at least one operation of the vehicle.

[0084] 19. The vehicle system according to claim 14, wherein:

[0085] One or more vehicle characteristics include the vehicle's lateral acceleration value and the slip ratio value of each wheel; and

[0086] Driving conditions include one or more of the following: conditions of decreased surface friction coefficient, conditions of increased lateral acceleration, and normal driving conditions.

[0087] 20. A control method for selecting a gain set for controlling a vehicle, the control method comprising:

[0088] Optimize multiple gain sets for various driving conditions;

[0089] Determine the current driving conditions associated with the vehicle based on one or more vehicle characteristics;

[0090] A gain set is selected from multiple optimized gain sets based on the determined driving conditions associated with the vehicle; and

[0091] A control signal is generated based on one or more values ​​of a selected gain set to control at least one operation of the vehicle.

[0092] Further applications of this disclosure will become apparent from the detailed description, claims, and accompanying drawings. The detailed description and specific examples are for illustrative purposes only and are not intended to limit the scope of this disclosure. Attached Figure Description

[0093] This disclosure will become more fully understood in light of the detailed description and accompanying drawings, in which:

[0094] Figure 1 The block diagram of the example vehicle system according to this disclosure is used to optimize the gain set of the control system across different driving conditions and select one of the gain set to control the vehicle for the current driving conditions.

[0095] Figure 2 This is a block diagram of an example control module comprising multiple models according to this disclosure;

[0096] Figure 3 This is a block diagram of an example control module according to the present disclosure, which includes a fusion module for smoothly switching between selected gains when driving conditions change;

[0097] Figure 4 This is a flowchart of an example control process for identifying driving conditions according to this disclosure; and

[0098] Figure 5-6 This is a flowchart of an example control process for selecting a gain set from an optimized gain set to control a vehicle, according to the present disclosure.

[0099] In the accompanying drawings, reference numerals may be used repeatedly to identify similar and / or identical elements. Detailed Implementation

[0100] Vehicles include control systems for controlling various vehicle operations. Traditionally, vehicles can utilize a set of system gains for this control across a wide range of driving conditions. For example, a vehicle control system may be provided with a single, optional set of system gains used to estimate the vehicle's control state (e.g., speed estimation) across different driving conditions, such as normal operating conditions, low surface friction conditions (e.g., wet, icy roads), high lateral acceleration conditions, etc. This one-size-fits-all approach with a single optional set of gains is unsatisfactory for multiple driving conditions. Therefore, relying on a single set of gains leads to suboptimal performance under some or all operating conditions for different driving conditions with varying characteristics.

[0101] The vehicle systems and methods disclosed herein provide solutions for optimizing multiple gain sets of a control system across different driving conditions, and then selecting one of the optimized gain sets based on the current driving conditions. For example, the vehicle systems and methods may apply particle swarm optimization (PSO) methods to optimize filter gains (e.g., Kalman filter gains, etc.) for vehicle speed estimation under different dynamic driving conditions, such as normal operating conditions, low surface friction conditions (e.g., wet, icy roads), high lateral acceleration conditions, etc. As further explained herein, this optimization can be performed on the vehicle itself or using an external module. Furthermore, in various embodiments, the vehicle systems and methods determine critical driving conditions for speed estimation or other vehicle control states before optimizing filter gains for speed estimation or other vehicle control states and / or smoothly transitioning between optimized gains for different driving conditions in real time. By optimizing multiple gain sets of the control system across different driving conditions and then selecting an appropriate gain set, robustness and optimal performance of the control system for speed estimation or other vehicle control states can be achieved.

[0102] Now for reference Figure 1 A block diagram of an example vehicle system 100 is presented, which optimizes the gain set of the control system across different driving conditions and selects a gain set for the current driving conditions to control vehicle 102. Figure 1 As shown, vehicle system 100 typically includes a control module 104 inside vehicle 102, multiple sensors 106, 108, 110 located on vehicle 102, and a vehicle control module 112 for controlling at least one operation of vehicle 102. Although Figure 1 Vehicle system 100 is shown as including a specific dedicated module; however, it should be understood that one or more other modules may be employed if needed. For example, any combination of modules (e.g., control module 104, vehicle control module 112, etc.) and / or their functions may be integrated into a single module or multiple different modules. Furthermore, although... Figure 1 Three sensors 106, 108, and 110 are shown, but it should be understood that any number of sensors can be arranged on vehicle 102 if needed.

[0103] exist Figure 1 In this example, sensors 106, 108, 110 and vehicle control module 112 communicate with control module 104. In such an example, modules and sensors of vehicle system 100 can share parameters via networks (e.g., Controller Area Network (CAN)) and signals. For example, in Figure 1 In the process, the control module 104 receives signals representing data from sensors 106, 108, and 110, and transmits the signals to the vehicle control module 112 for controlling vehicle operation.

[0104] Figure 1 The vehicle system 100 can be used in any suitable vehicle, such as an autonomous vehicle, a semi-autonomous vehicle, etc. Furthermore, the vehicle system 100 can be applied to electric vehicles (e.g., pure electric vehicles, plug-in hybrid electric vehicles, etc.) and internal combustion engine (ICE) vehicles. Figure 1 In the example, vehicle system 100 is used in vehicle 102 (e.g., autonomous vehicle).

[0105] exist Figure 1 In the examples, sensors 106, 108, and 110 may include any suitable sensor for detecting one or more vehicle characteristics. For example, one of sensors 106, 108, and 110 may be a lateral acceleration sensor for detecting the lateral acceleration of vehicle 102. In some examples, one or more of sensors 106, 108, and 110 may include a yaw rate sensor, a vehicle speed sensor, a steering angle sensor, a wheel speed sensor, etc.

[0106] In various embodiments, data from sensors 106, 108, and 110 may represent vehicle characteristics and / or be used by control module 104 to detect vehicle characteristics. For example, in some embodiments, control module 104 may receive lateral acceleration (α) from one of sensors 106, 108, and 110. y In addition, the control module 104 can receive wheel speed data and vehicle speed data from one or more of the sensors 106, 108, and 110, and then determine the wheel slip ratio value of each wheel 114, 116, 118, and 120 of the vehicle 102 based on the angular velocity and vehicle speed of each wheel.

[0107] exist Figure 1 In the example, vehicle system 100 optimizes multiple gain sets of a control system (e.g., control module 104 or a portion thereof). Typically, each gain set includes one or more values ​​representing the relationship between the inputs and outputs of control module 104. For example, each gain set may include the filter gains of filters in control module 104, such as Kalman filters, as further explained herein.

[0108] Vehicle system 100 optimizes the gain set for different driving conditions. For example, before gain optimization based on training and test data, various driving conditions can be identified (e.g., identified), as explained further below. In various embodiments, the number of different driving conditions can vary depending on system requirements. For example, vehicle system 100 can utilize three different driving conditions, ten driving conditions, etc. Furthermore, driving conditions can include any suitable type of driving condition, such as conditions with reduced or low surface friction coefficients (e.g., wet, icy roads), conditions with increased or high lateral acceleration (e.g., curves in the road), normal driving conditions, etc. In various embodiments, normal driving conditions can refer to low lateral acceleration on a high-friction road where the vehicle tires are in a linear region. The gain set can then be optimized for each specific driving condition.

[0109] In various embodiments, the gain set of the control system may be optimized on the vehicle 102 itself or external to the vehicle 102. For example, control module 104 may optimize the gain set of the control system to provide online / real-time optimization. In other examples, vehicle system 100 may include an optional control module 122 external to vehicle 102. In such an example, control module 122 may be implemented using cloud computing to optimize the gain set using online data collection. For example, recorded data may be sent to the cloud (e.g., represented by control module 122), and the optimization process may be run remotely. Control module 122 may then transmit the optimized gain set to control module 104 of vehicle 102 (or another specific vehicle) via one or more signals 124. Regardless of where the gain set is optimized, control module 104 may store the received gain set and / or the determined gain set along with their associated driving conditions for later use.

[0110] The gain set of the control system can be optimized by control module 104 and / or control module 122 in any suitable manner. For example, control module 104 and / or control module 122 can implement an optimization (or PSO) algorithm to optimize the gain set (with multiple gains) for each driving condition. In such an example, either control module can receive a dataset associated with the driving conditions and then implement an optimization algorithm based on the received dataset to optimize the gain set for the driving conditions. In various embodiments, the dataset comes from online data collection, such as crowdsourcing. By way of example only, the dataset may include similar or different types of data that will be used to determine the current driving conditions. For example, the dataset used to optimize the gains may include lateral vehicle acceleration values ​​and wheel slip rate values ​​for multiple driving conditions.

[0111] In various embodiments, control module 104 and / or control module 122 determine the optimal values ​​for different sets of gains based on one or more cost function equations. For example, when using an optimization algorithm to optimize the gains, points can be randomly distributed in an N-dimensional space (where N is an integer), and a cost function with equations for the control system can be formed to find the optimal value of the gain associated with the cost function. This cost function can be numerically solved for each particle in the swarm to optimize these values. For example, if the control system is used for speed estimation of vehicle 102, the cost function can be solved to determine the speed and position of each particle in the swarm relative to the global optimum in N-dimensional space. In other examples, the control system can be used for another vehicle control state, and the cost function can be solved to determine this parameter and associated position of each particle in the swarm.

[0112] In some examples, when implementing particle swarm optimization (e.g., non-Jacobi-based methods), certain combinations of gains can destabilize the system. For instance, during the random distribution of points in an N-dimensional space (e.g., a 15-dimensional space, etc.) and the learning-based adjustment of the number of points, certain combinations of gains may lead to instability. This can result in singularities when solving system equations and increase computational costs. In such examples, control module 104 and / or control module 122 can implement algorithms to monitor singularities in the equations and take mitigation actions in response to the detection of such singularities.

[0113] For example, one of the control modules 104 and 122 can detect singularities in a subset of the dataset and prevent the optimization algorithm from exploiting the optimization space when determining the optimal value of the gain set based on the cost function equation. In such an example, the control module can detect singularities based on the rank of one or more matrices that include representative data. If a singularity is detected, the control module can prevent the optimization algorithm from solving the cost function equation, keeping the algorithm away from regions that improve computational efficiency, and so on. For example, the gain values ​​of the cost functions associated with those particles can be adjusted so that they are far from regions where singularities are detected. In various embodiments, this can be achieved by moving towards globally optimal or locally optimal particles.

[0114] In various embodiments, the cost function equation may include one or more correlation coefficients and root mean square error (RMSE) components. Combining these components in the cost function equation leads to better estimator performance when dealing with larger-dimensional state dynamics and larger measurement datasets. For example, equation (1) below represents an example cost function equation (J) for speed estimation of vehicle 102 that may be generated by control module 104 and / or control module 122. In equation (1), This represents the lateral velocity estimate (e.g., the output of the first model, as explained further below). This represents the longitudinal velocity estimate (e.g., the output of the first model). This represents another lateral velocity estimate (e.g., the output of a second model, as explained further below). and This represents the gain used for the first model. and v represents the gain used in the second model. y,RYK and v x,RTK The base signal is the real signal (e.g., a reference signal), and U is the number of available data points in the signal. For example, if a 1-second velocity signal with a sampling rate of 0.01 seconds is used, then U would be 100. In this example, the correlation coefficient is the Pearson product-moment correlation coefficient (PPMC), which is a measure of the correlation between two signals (e.g., how similar the signatures of two signals are). Additionally, an RMSE component is provided, represented by a square root function. The value of each term can be adjusted depending on the optimization problem.

[0115] Equation (1)

[0116] Then, after the gain value for each identified driving condition is optimized, the control module 104 can implement scheduling techniques to select an appropriate gain value for vehicle control. For example, after the optimized gain value for the identified driving condition is known, the control module 104 determines the current driving condition associated with the vehicle 102 (e.g., when the vehicle 102 is moving). In various embodiments, the current driving condition can be determined based on one or more vehicle characteristics received or otherwise detected by one or more of the sensors 106, 108, 110. For example, the control module 104 can be based on the lateral acceleration (α) provided by one of the sensors 106, 108, 110. y The values ​​of slip ratio for each wheel (114, 116, 118, 120) determine whether the current driving conditions are low surface friction coefficient driving conditions or another suitable driving condition.

[0117] Once the current driving conditions are determined, the control module 104 selects an optimized set of gains. For example, the control module 104 may store the driving conditions for each identifier and its associated optimized gain. Then, after determining the driving conditions, the control module 104 may select a specific set of optimized gains (e.g., gain values) for the determined driving conditions. In this way, the selected set of gains is specific to the current driving conditions.

[0118] Vehicle system 100 can then control one or more vehicle operations of vehicle 102 based on optimized gain values ​​from a selected gain set. For example, the selected optimized gain set can then be used in a specific control system to determine the characteristics of vehicle 102. For instance, the optimized gain set can be used as filter gains (e.g., Kalman filter gains) in a control system for speed estimation of vehicle 102 or another control parameter of vehicle 102. In such an example, control module 104 can generate control signals for vehicle control module 112 based on control parameters of vehicle 102 (e.g., speed estimation of vehicle 102, etc.). Vehicle control module 112 can then control the operation of vehicle 102, such as the motion and / or trajectory of vehicle 102, based on the control signals.

[0119] In various embodiments, as explained above, the complexity of determining the optimal values ​​for different gain sets can increase exponentially by adding more dimensions to the N-dimensional space. In other words, more gain values ​​are optimized during the optimization process when more dimensions are added. Therefore, in some examples, the system can be decomposed into smaller, independent subsystems, where the gain value of each subsystem can be optimized individually. As an example, the independent subsystems can be represented by two or more models, such as kinematic and dynamic models. By relying on smaller subsystems rather than a single system to optimize the gain values, the optimization process can experience reduced computational cost and converge to the global optimum more quickly when using, for example, an optimization procedure.

[0120] For example, Figure 2 An example of a control module 204 is depicted, which can be used as... Figure 1 The control module 204 and / or control module 122 or a portion thereof. As shown, control module 204 includes two models 226, 228 (e.g., two subsystems). Although Figure 2 Two models are shown, but it should be understood that in some examples, additional models or subsystems may be used if needed.

[0121] Models 226 and 228 can be any suitable model type. For example, model 226 can be a Kalman filter with a dynamic model, and model 228 can be a kinematic model. In other examples, any other suitable model can be used.

[0122] exist Figure 2In the example, control module 204 optimizes a first subset of the gain set for model 226 and another second subset of the gain set for model 228, respectively. For example, the first subset of the gain set can be optimized as explained above. During this time, control module 204 can freeze (e.g., maintain, not change, etc.) the gain values ​​of the second subset (if known). Similarly, when the second subset is optimized, control module 204 can freeze the gain values ​​of the first subset (if known). Using this method, equation (1) above can be decomposed into two cost function equations (J1, J2) represented by equations (2) and (3) below.

[0123] Equation (2)

[0124] Equation (3)

[0125] exist Figure 2 In the example, control module 204 can rely on the output of one of models 226 and 228 (e.g., subsystems) to determine the vehicle control state. For example, and as explained above, the current driving conditions can be determined, and an optimized gain value for those conditions can be selected. In this example, the gain value is decomposed into two subsets, which are optimized for two different models 226 and 228 respectively.

[0126] Then and as Figure 2 As shown, model 226 receives input 230, for example, from... Figure 1 One or more of sensors 106, 108, and 110 receive or otherwise detect one or more vehicle characteristics. Control module 204 can then use model 226 to determine a vehicle control state (e.g., lateral speed estimation and longitudinal speed estimation) based on a subset of selected gain values ​​for model 226 and the received inputs. In this example, model 226 provides an output 232 representing the vehicle control state to model 228. Control module 204 then uses model 228 to determine a vehicle control state (e.g., lateral speed estimation) based on the vehicle control state from model 226 and a subset of selected gain values ​​for model 228. Model 228 can then provide an output 234 representing the vehicle control state (e.g., lateral speed estimation). As explained above, control module 204 can generate control signals based on the vehicle control state and / or parameters for controlling one or more vehicle operations.

[0127] In various embodiments, fusion logic can be implemented to smoothly transition between different driving conditions and provide superior performance in real time. For example, fusion logic can be used to smooth the transition between control states for different driving conditions and avoid abrupt changes in gain that could affect estimated performance. In some examples, fusion logic can be implemented using a low-pass filter.

[0128] Figure 3 An example control module 304 implementing this gain scheduling logic is shown, and control module 304 can be used as... Figure 1 The control module 104 or a part thereof. For example... Figure 3 As shown, the control module 304 includes a driving state module 306 for determining the current driving conditions and corresponding control states, a fusion module 308 for smoothly transitioning between selected gains when driving conditions change, and models 326 and 328. Figure 3 In the examples, models 326 and 328 can be similar to Figure 2 Models 226 and 228. For example, model 326 can be a kinematic model, and model 328 can be a dynamic model.

[0129] exist Figure 3 In the example, the driving state module 306 receives various inputs 330, such as those from... Figure 1 One or more of the sensors 106, 108, 110 receive or otherwise detect one or more vehicle characteristics. In various embodiments, input 330 may include, for example, lateral acceleration (α). y Values ​​such as slip ratio of each wheel and slip ratio of each wheel.

[0130] Then, based on the received input 330, the driving state module 306 determines the current driving conditions and the corresponding control state. For example, the driving state module 306 is shown as including three driving control states 332, 334, and 336 corresponding to three driving conditions. In various embodiments, driving conditions may include normal operating conditions, low surface friction conditions, high lateral acceleration conditions, or other suitable driving conditions. Although Figure 3 The example shows three driving control states 332, 334, and 336 corresponding to three driving conditions. However, it should be understood that more or fewer driving control states and corresponding driving conditions may be used if necessary.

[0131] In such an example, driving state module 306 may initially begin with one of driving states 332, 334, or 336, for example, a driving state corresponding to normal operating conditions. In such an example, control module 304 may select a suitable set of optimized gain values ​​for this driving condition for use by models 326 and 328, as explained above. For example, control module 304 may provide model 326 with a subset of the selected optimized gain values ​​and model 328 with another subset of the selected optimized gain values, as indicated by the arrows between fusion module 308 and models 326 and 338. In various embodiments, the gain values ​​selected for the driving condition may pass through fusion module 308 without interference.

[0132] However, if driving conditions change, control module 304 transitions to another of driving states 332, 334, and 336. For example, when new driving conditions are determined based on received input 330, driving state module 304 can then transition to a different driving control state 332, 334, or 336. For instance, if received input 330 satisfies a specific set of conditions (e.g., thresholds), driving state module 304 can determine new current driving conditions and transition from one state (e.g., driving control state 332) to another state (e.g., driving control state 334 or driving control state 336). Once the new current driving conditions are determined, control module 304 selects a new set of optimized gain values ​​for those conditions. In such an example, driving state module 304 can pass the new driving control state and / or the newly selected optimized gain values ​​to fusion module 308.

[0133] The fusion module 308 can then blend a first optimized set of gain values ​​for the previous driving conditions into a new optimized set of gain values ​​for the new driving conditions to provide a smooth transition. In various embodiments, the fusion module 308 may include one or more components, such as a low-pass filter, for incrementally changing the gain from one set of values ​​to another.

[0134] Once the fusion module 308 has blended the optimized gain values ​​into a new set of optimized gain values, it provides a subset of the new optimized gain values ​​to models 326 and 338. This is indicated by the arrows between the fusion module 308 and models 326 and 338. For example, a subset of the selected optimized gain values ​​is provided to model 326, and another subset of the selected optimized gain values ​​is provided to model 328.

[0135] Models 326 and 328, then with Figure 2Models 226 and 228 determine the vehicle control state in a similar manner. For example, model 326 determines the vehicle control state (e.g., lateral speed estimate and longitudinal speed estimate) based on a subset of its optimized gain values ​​and outputs the vehicle control state to model 328. Model 328 then determines the vehicle control state (e.g., lateral speed estimate) based on the vehicle control state from model 326 and a subset of its optimized gain values. Model 328 then provides an output representing the vehicle control state (e.g., lateral speed estimate), which can be used to generate control signals for controlling one or more vehicle operations, as explained above.

[0136] Figure 4 It shows Figure 1 The vehicle system 100 can be used to identify or determine driving conditions through an example control process 400, and Figure 5-6 It shows Figure 1 The vehicle system 100 can be used for example control processes 500 and 600 to select a gain set from an optimized gain set to control the vehicle. Although example control processes 400, 500, and 600 are relative to... Figure 1 The vehicle system 100, control module 104, and vehicle 102 are described, but any one of the control processes 400, 500, and 600 may be another suitable vehicle system, control module, and / or vehicle that may be adopted.

[0137] like Figure 4 As shown, control process 400 begins at 402, setting multiple driving conditions into one. Control process 400 then proceeds to 404. At 404, the control determines the optimal gain for the driving conditions (e.g., normal operating conditions). In various embodiments, as explained above, Figure 1 Control module 104 and / or control module 122 can be used to determine the optimal gain value for the driving condition. Control process 400 then proceeds to 406.

[0138] At 406, control quantization is based on an estimated performance determined by an optimal gain value for the driving conditions. In various embodiments, the estimated performance can be quantized according to conventional techniques. Control process 400 then proceeds to 408, where control determines whether an accuracy threshold for the estimated performance is met. For example, control module 104 may compare the quantized estimated performance obtained in 406 with one or more thresholds. If the accuracy threshold is not met (e.g., the quantized estimated performance is less than the threshold), control proceeds to 410.

[0139] At 410, the control adds another driving condition to the setting. For example, if the initial driving condition is a normal operating condition, a low surface friction condition and / or a high surface friction condition can be added to differentiate it from the normal operating condition. The control process 400 then returns to 404, where the control determines the optimal gain for each driving condition (e.g., the normal operating condition and the low surface friction condition). The control process 400 then proceeds again to 406, where the control quantifies the estimated performance determined based on the optimal gain value determined for the driving condition. The control process 400 then proceeds again to 408, where the control determines whether a precision threshold for the estimated performance is met, as explained above. This loop of adding new driving conditions repeats until the precision threshold is met. Once the precision threshold is met (e.g., the quantified estimated performance is less than the threshold), the control proceeds to 412, where the driving condition can be stored for later use. The control process 400 then ends. Figure 4 As shown.

[0140] exist Figure 5 In the process 500, control procedure 502 begins by receiving an optimized gain set for different driving conditions. In various embodiments, driving conditions can be... Figure 4 The control process 400 is determined and can be predefined, etc. Furthermore, as explained above, Figure 1 Control module 122 can optimize a set of gain values ​​for each driving condition and then transmit the optimized gain values ​​for different driving conditions to control module 104 of vehicle 102. In such an example, the gain values ​​can be optimized based on datasets collected from multiple different vehicles, including or excluding vehicle 102. With this configuration, the gain values ​​can be optimized offline and remotely in vehicle 102. In other examples, control module 104 can receive an optimized set of gain values ​​(or gain values) from an internal source (e.g., another control module in vehicle 102 similar to control module 104). In this case, the gain values ​​can be optimized based on datasets collected from vehicle 102 and in real time. In either example, the gain values ​​can be optimized by implementing an optimization algorithm, dividing the control system into independent subsystems, etc., as explained above. Control process 500 then proceeds to 504.

[0141] At 504, control module 104 receives vehicle characteristics. For example, and as explained above, Figure 1Sensors 106, 108, 110 and / or other suitable sensors associated with vehicle 102 can detect vehicle characteristics such as acceleration (e.g., lateral acceleration, longitudinal acceleration, etc.), angular velocities of each wheel 114, 116, 118, 120, etc. Control module 104 can then use the received vehicle characteristics and / or portions of other known characteristics to determine other vehicle characteristics, such as wheel slip ratios of each wheel 114, 116, 118, 120. Control process 500 then proceeds to 506.

[0142] At 506, control module 104 determines the current driving conditions associated with vehicle 102 based on the received and / or determined vehicle characteristics. For example, based on the lateral acceleration of vehicle 102, the wheel slip ratio of wheels 114, 116, 118, 120, and / or other possible vehicle characteristics, control module 104 can determine the current driving conditions, such as normal operating conditions, low surface friction conditions, high lateral acceleration conditions, or another suitable driving condition. In various embodiments, the current driving conditions are determined using... Figure 4 The control process 400 determines one of the driving conditions, one of the predefined driving conditions, etc. Control process 500 then proceeds to 508.

[0143] At 508, control module 104 selects a gain set from the optimized gain set based on the driving conditions associated with vehicle 102. For example, if the current driving condition is a first driving condition (e.g., low surface friction condition), control module 104 can select a first optimized gain value set (e.g., filter gain). However, if the current driving condition is a second driving condition (e.g., high lateral acceleration condition), control module 104 can select a second optimized gain value set different from the first optimized gain value set. Control process 500 then proceeds to 510.

[0144] At 510, control module 104 determines one or more vehicle control states (e.g., lateral speed estimation) of vehicle 102. For example, control module 104 and / or one or more filters in another control system may use optimized gain values ​​and vehicle characteristics (e.g., received and / or determined vehicle characteristics) to determine the vehicle control state (e.g., lateral speed estimation) of vehicle 102.

[0145] In various embodiments, as explained above, the control system may be divided into two or more independent subsystems. In such an example, at 508, control module 104 may select a specific set of optimized gain values, with each subsystem (e.g., each filter) employing a subset of the optimized gain values. For example, if two subsystems are employed (e.g., during optimization and vehicle control), at 510, the first subsystem or model may determine one or more vehicle control states (e.g., lateral speed estimates and longitudinal speed estimates) based on a selected subset of gain values ​​used for the first model and vehicle characteristics. Then, at 510, the second subsystem or model may determine a vehicle control state (e.g., lateral speed estimates) based on a selected subset of gain values ​​used for the second model and the output of the first model (e.g., one or more vehicle control states).

[0146] Control process 500 then proceeds to 512, where control module 104 generates a control signal based on the vehicle control state (e.g., lateral speed estimation). Control module 104 can then transmit the control signal to vehicle control module 112. Control process 500 then proceeds to 514, where vehicle control module 112 controls at least one operation of vehicle 102 based on the control signal. Control process 500 then proceeds to 516.

[0147] At 516, control module 104 determines whether the current driving conditions have changed. This determination may be based on received and / or determined vehicle characteristics. For example, control module 104 may determine the current driving conditions associated with vehicle 102 periodically, randomly, substantially continuously, etc., as explained above. If it is determined that the current or new driving conditions are different from the previous driving conditions (e.g., a change in driving conditions is detected), control process 500 proceeds to 518. However, if no change is detected, control process 500 proceeds to 522.

[0148] At 518, control module 104 selects a new set of optimized gain values ​​based on new driving conditions associated with vehicle 102. In various embodiments and as explained above, the optimized gain values ​​may include subsets for individual subsystems. Control process 500 proceeds to 520.

[0149] At 520, control module 104 blends the optimized gain value for the previous driving conditions into a new optimized gain value for the new driving conditions to provide a smooth transition. In various embodiments, control module 104 may include one or more components, such as a low-pass filter, for incrementally changing the gain from one set of values ​​to another. Control process 500 then returns to 510.

[0150] At 522, control module 104 determines whether the vehicle is operable. For example, when vehicle 102 is in energized mode, moving, etc., control module 104 can receive a signal. If control module 104 determines that the vehicle is operable (e.g., not in energized mode), control process 500 returns to 516. Otherwise, if control module 104 determines that the vehicle is not operable, control process 500 can end, such as... Figure 5 As shown.

[0151] exist Figure 6 In the middle, the control process 600 is similar to Figure 5 The control process is 500, but includes additional steps. For example, and such as Figure 6 As shown, process 600 begins at 602, where the gain set for different driving conditions is optimized. More specifically, at 602, control module 104 optimizes a first subset of gain values ​​for a first model (e.g., a first subsystem) in control module 104 for each driving condition. Control process 600 then proceeds to 604 and 606. At 604, control module 104 freezes the first subset of optimized gain values, and at 606, control module 104 optimizes a second subset of gain values ​​for a second model (e.g., a second subsystem) in control module 104 for each driving condition. In such an example, driving conditions can be used... Figure 4 The control process 400 is determined, can be predefined, etc., and the gain value can be optimized (respectively) by implementing optimization algorithms, etc., as explained above.

[0152] Figure 6 The control process 600 then proceeds to the above relative to Figure 5 The explanations for 504, 506, and 508. Next, control process 600 proceeds to 608 and 610.

[0153] At 608, control module 104 determines one or more first vehicle control states based on a first subset of optimized gain values. For example, and as explained above, the first model of control module 104 (e.g., Figure 2 Model 226 Figure 3 Model 326, etc., can determine one or more vehicle control states (e.g., lateral speed estimation and longitudinal speed estimation) based on a first subset of optimized gain values ​​and vehicle characteristics.

[0154] At 610, control module 104 determines one or more second vehicle control states. For example, and as explained above, the second model of control module 104 (e.g., Figure 2 Model 228 Figure 3Model 328, etc., can determine one or more vehicle control states (e.g., lateral speed estimation) based on a second subset of optimized gain values ​​and one or more first vehicle control states.

[0155] Then, Figure 6 The control process 600 is carried out as described above relative to... Figure 5 The numbers 512, 514, 516, 518, 520, and 522 are explained.

[0156] The vehicle system and method presented in this paper offer numerous benefits. For example, instead of a single gain set, multiple gain sets are optimized and applied to different driving conditions (e.g., in real-time). This allows individual gain sets to cover the edge cases of most (if not all) vehicle dynamics. Furthermore, because PSO optimization techniques can be employed to optimize the extended Kalman filter gain using experimental data from different conditions, each gain set can be optimized with high accuracy for each vehicle dynamic. Moreover, because the control system can be decomposed into smaller, independent subsystems, where the gain value of each subsystem can be optimized individually, reduced computational cost and faster convergence to the global optimum are achieved compared to a single control system. Furthermore, the vehicle system and method presented in this paper provide smooth transitions between optimized gain values ​​for different driving conditions, enabling smooth transitions between operating conditions and providing superior and robust performance in real-time. Additionally, the computational efficiency of the optimization algorithm is enhanced by incorporating singularity checks into the data and avoiding potentially problematic regions. This results in a reduction in the computational burden associated with the optimization algorithm and faster convergence to the global minimum due to the consideration of more "good" particles in the computation. Furthermore, the vehicle systems and methods described herein offer flexibility because some or all of the aspects described herein can be performed on the vehicle (e.g., in real time) and / or outside the vehicle, thereby enabling the acquisition of customized parameters to suit specific vehicle requirements. In other words, this flexibility provides customizability of gains for a particular vehicle.

[0157] The foregoing description is merely illustrative in nature and is by no means intended to limit this disclosure, its application, or its use. The broad teachings of this disclosure can be implemented in various forms. Therefore, while this disclosure includes specific examples, its true scope should not be so limited, as other modifications will become apparent upon examination of the drawings, specification, and appended claims. It should be understood that one or more steps in the method may be performed in a different order (or concurrently) without altering the principles of this disclosure. Furthermore, while each embodiment is described above as having certain features, any one or more of those features described with respect to any embodiment of this disclosure may be implemented in and / or combined with features of any other embodiment, even if such combination is not explicitly described. In other words, the described embodiments are not mutually exclusive, and substitutions of one or more embodiments for each other remain within the scope of this disclosure.

[0158] Spatial and functional relationships between components (e.g., between modules, circuit elements, semiconductor layers, etc.) are described using various terms, including “connected,” “joined,” “coupled,” “adjacent,” “next to,” “on top,” “above,” “below,” and “set.” Unless explicitly described as “direct,” when describing the relationship between first and second components in the above disclosure, the relationship can be a direct relationship where no other intervening components exist between the first and second components, but it can also be an indirect relationship (spatially or functionally) between the first and second components. As used herein, the phrases A, B, and C at least one should be interpreted as meaning the use of a non-exclusive logical OR (A or B or C) and should not be interpreted as meaning “at least one of A, at least one of B, and at least one of C.”

[0159] In a diagram, the direction of the arrow usually indicates the flow of information (such as data or instructions) that is of interest to the diagram. For example, when components A and B exchange various kinds of information, but the information transmitted from component A to component B is relevant to the diagram, the arrow can point from component A to component B. This unidirectional arrow does not mean that no other information is transmitted from component B to component A. Furthermore, for information sent from component A to component B, component B can send a request for or confirmation of receipt of that information to component A.

[0160] In this application, including the definitions below, the term "module" or "controller" may be replaced by the term "circuit". The term "module" may refer to or include a portion of the following: application-specific integrated circuit (ASIC); digital, analog, or mixed-signal analog / digital discrete circuit; digital, analog, or mixed-signal analog / digital integrated circuit; combinational logic circuit; field-programmable gate array (FPGA); processor circuitry (shared, dedicated, or grouped) that executes code; memory circuitry (shared, dedicated, or grouped) that stores code executed by the processor circuitry; other suitable hardware components that provide the described functionality; or some or all of the foregoing, such as in a system-on-a-chip.

[0161] A module may include one or more interface circuits. In some examples, the interface circuits may include wired or wireless interfaces that connect to a local area network (LAN), the Internet, a wide area network (WAN), or a combination thereof. The functionality of any given module disclosed herein can be distributed among multiple modules connected via the interface circuits. For example, multiple modules can allow for load balancing. In another example, a server (also referred to as a remote or cloud) module may perform some functions on behalf of a client module.

[0162] The term "code" as used above can include software, firmware, and / or microcode, and can refer to programs, routines, functions, classes, data structures, and / or objects. The term "shared processor circuit" covers a single processor circuit that executes some or all of the code from multiple modules. The term "grouped processor circuit" covers a processor circuit that, in combination with additional processor circuits, executes some or all of the code from one or more modules. References to multiple processor circuits cover multiple processor circuits on a discrete die, multiple processor circuits on a single die, multiple cores of a single processor circuit, multiple threads of a single processor circuit, or a combination of the above. The term "shared memory circuit" covers a single memory circuit that stores some or all of the code from multiple modules. The term "grouped memory circuit" covers a memory circuit that, in combination with additional memory, stores some or all of the code from one or more modules.

[0163] The term memory circuit is a subset of the term computer-readable medium. As used herein, the term computer-readable medium does not cover transient electrical or electromagnetic signals propagating through a medium (e.g., on a carrier wave); therefore, the term computer-readable medium can be considered tangible and non-transitory. Non-limiting examples of non-transitory tangible computer-readable media are non-volatile memory circuits (e.g., flash memory circuits, erasable programmable read-only memory circuits, or mask read-only memory circuits), volatile memory circuits (e.g., static random access memory circuits or dynamic random access memory circuits), magnetic storage media (e.g., analog or digital magnetic tape or hard disk drives), and optical storage media (e.g., CDs, DVDs, or Blu-ray discs).

[0164] The apparatus and methods described in this application can be implemented, in part or in whole, by a special-purpose computer created by configuring a general-purpose computer to perform one or more specific functions embodied in a computer program. The function blocks, flowchart components, and other elements described above serve as software specifications that can be translated into computer programs through the routine work of a skilled technician or programmer.

[0165] A computer program includes processor-executable instructions stored on at least one non-transitory tangible computer-readable medium. A computer program may also include or depend on stored data. A computer program may encompass a basic input / output system (BIOS) for interacting with the hardware of a special-purpose computer, device drivers for interacting with specific devices of the special-purpose computer, one or more operating systems, user applications, background services, background applications, etc.

[0166] Computer programs may include: (i) descriptive text to be parsed, such as HTML (Hypertext Markup Language), XML (Extensible Markup Language), or JSON (JavaScript Object Notation); (ii) assembly code; (iii) object code generated from source code by a compiler; (iv) source code executed by an interpreter; (v) source code compiled and executed by a real-time compiler, etc. As an example only, source code may come from languages ​​including C, C++, C#, Objective-C, Swift, Haskell, Go, SQL, R, Lisp, etc. Fortran, Perl, Pascal, Curl, OCaml, HTML5 (Hypertext Markup Language, Fifth Revision), Ada, ASP (Active Server Pages), PHP (PHP: Hypertext Preprocessor), Scala, Eiffel, Smalltalk, Erlang, Ruby, Visual Lua, MATLAB, SIMULINK and It is written using the syntax of the language.

Claims

1. A vehicle system for selectively controlling a gain set of a vehicle, the vehicle system comprising: a plurality of sensors configured to detect one or more vehicle characteristics; and a control module in communication with the plurality of sensors, the control module configured to: optimize a plurality of gain sets for a plurality of different driving conditions, each gain set comprising one or more values representing a relationship between an input to the control module and an output of the control module; determine a current driving condition associated with the vehicle based on the one or more vehicle characteristics; select a gain set of the plurality of optimized gain sets based on the current driving condition associated with the vehicle; and generate a control signal based on the one or more values of the selected gain set for controlling at least one operation of the vehicle.

2. The vehicle system of claim 1, wherein the control module is configured to: receive a data set associated with the plurality of driving conditions; and based on the received data set, optimize a cost function with an optimization algorithm based on a performance of an estimator of the plurality of gain sets for the plurality of driving conditions.

3. The vehicle system of claim 2, wherein the data set comprises lateral vehicle acceleration values and wheel slip ratio values for each driving condition.

4. The vehicle system of claim 2, wherein the control module is configured to determine optimal values of the plurality of gain sets based on a cost function equation having a correlation coefficient and a root mean square error (RMSE) component.

5. The vehicle system of claim 4, wherein the control module is configured to detect a singularity in a portion of the data set and prevent the optimization algorithm from utilizing that portion of the optimization space in determining the optimal values of the plurality of gain sets based on the cost function equation.

6. The vehicle system of claim 1, wherein: the control module comprises a first model and a second model; and the control module is configured to optimize a first subset of the plurality of gain sets for the first model and a second subset of the plurality of gain sets for the second model, respectively.

7. The vehicle system of claim 6, wherein the first model is a dynamic model and the second model is a kinematic model.

8. The vehicle system of claim 6, wherein the control module is configured to freeze the first subset of the plurality of gain sets and optimize the second subset of the plurality of gain sets for the second model while the first subset is frozen.

9. The vehicle system of claim 6, wherein the control module is configured to: determine a first vehicle control state using the first model based on the first subset of the plurality of gain sets; determine a second vehicle control state using the second model based on the first vehicle control state and the second subset of the plurality of gain sets; and generate the control signal based on the second vehicle control state.

10. The vehicle system of claim 1, wherein: the current driving condition is a first current driving condition; the gain set of the plurality of optimized gain sets is a first gain set; and the control module is configured to: determine a second current driving condition associated with the vehicle based on the one or more vehicle characteristics; select a second gain set of the plurality of optimized gain sets based on the determined second driving condition associated with the vehicle; blend the first gain set into the second gain set to provide a smooth transition; and ​ ​ generate a control signal based on the one or more values of the second set of gains for controlling at least one operation of the vehicle.

Images