Architecture and method for vehicle motion control health monitoring and mitigation

Abstract

A motor vehicle motion control health monitoring system includes sensors and actuators disposed on a motor vehicle. The sensors measure real-time static and dynamic telemetry data about the motor vehicle, and the actuators change the static and dynamic behavior of the motor vehicle. A control module has a processor, a memory, and input / output (I / O) ports. The processor executes portions of program code stored in the memory, the portions of program code including: an offline portion that collects telemetry data from the motor vehicle, performs fault analysis on the telemetry data, and assigns tasks based on the fault analysis; and an online portion that analyzes the telemetry data for faults within a particular sensor, actuator, or function utilizing the sensor and / or actuator system. The online portion mitigates deviations in the telemetry data by sending corrections to one or more sensors, actuators, and / or functions of the motor vehicle motion control system.

Description

Technical Field

[0001] This disclosure relates to systems and methods for vehicle motion control, and more particularly to systems and methods for managing the performance of vehicle motion control. Background Technology

[0002] Motor vehicle motion control systems manage vehicle performance in terms of longitudinal and lateral acceleration, pitch, roll, and yaw under various environmental and road conditions. These systems are complex, with numerous interacting functions. Therefore, potential performance degradation or even functional failure can propagate to other functions, leading to system performance degradation or even system failure.

[0003] Therefore, while current motor vehicle motion control systems achieve their intended purpose, there is a need for new and improved systems and methods for motor vehicle motion control management that reduce the burden on computational resources, increase system reliability, robustness, and redundancy, provide means to mitigate system component degradation and failures while maintaining or reducing costs and complexity, and improve motor vehicle motion control. Summary of the Invention

[0004] According to several aspects of this disclosure, a motor vehicle motion control health monitoring system includes one or more sensors mounted on a motor vehicle. The one or more sensors measure real-time static and dynamic telemetry data about the motor vehicle. The system further includes one or more actuators mounted on the motor vehicle that alter the static and dynamic behavior of the motor vehicle. A control module of the system has a processor, a memory, and input / output (I / O) ports for communicating with the one or more sensors and the one or more actuators. The control module executes a portion of program code stored in the memory, the portion including: an offline portion that collects telemetry data from the motor vehicle, performs fault analysis on the telemetry data, and assigns tasks based on the fault analysis; and an online portion that analyzes the telemetry data to locate faults within the functions of specific sensors and / or actuators and / or the sensor and / or actuator system, and mitigates deviations in the telemetry data by sending corrections to one or more sensors, actuators, and / or functions of the motor vehicle motion control system.

[0005] In another aspect of this disclosure, the offline program code portion further includes a first program code portion that collects real-time static and dynamic data from one or more sensors, one or more actuators, and one or more functions of the motor vehicle motion control system via I / O ports.

[0006] In another aspect of this disclosure, the offline program code portion further includes a second program code portion that performs fault mode determination for one or more sensors, one or more actuators, and one or more functions. Fault mode determination includes determining whether one or more sensors, one or more actuators, and / or one or more functions are experiencing a fault mode such as performance degradation, complete failure, and / or where telemetry data includes noise exceeding a predefined threshold noise value.

[0007] In another aspect of this disclosure, the offline program code portion further includes a third program code portion that performs data preparation, generation, and collection for one or more sensors, one or more actuators, and one or more functions. Data preparation further includes exclusion, filtering, and / or buffering processes.

[0008] In another aspect of this disclosure, the offline program code portion further includes a fourth program code portion that utilizes a machine learning architecture design to generate a task assignment scheme for telemetry data from one or more sensors, one or more actuators, and one or more functions. The machine learning architecture further includes: clustering methods based on machine learning and / or artificial intelligence for identifying and classifying failure modes. The fourth program code portion further applies temporary or short-term mitigations to one or more sensors, one or more actuators, and / or one or more functions and sends the raw telemetry data to a cloud computing system for further analysis.

[0009] In another aspect of this disclosure, the online program code portion further includes a fifth program code portion that performs data exclusion on telemetry data from one or more sensors, one or more actuators, and one or more functions.

[0010] In another aspect of this disclosure, the online program code portion further includes a sixth program code portion that detects and predicts potential faults within one or more sensors, one or more actuators, and one or more functions.

[0011] In another aspect of this disclosure, the online program code portion further includes a seventh program code portion that communicates telemetry data from one or more sensors, one or more actuators, and one or more functions via I / O ports between the motor vehicle and the I / O ports of a remote control module within a cloud computing system physically separated from the motor vehicle.

[0012] In another aspect of this disclosure, the online program code portion further includes an eighth program code portion, which mitigates biases in telemetry data from one or more sensors, one or more actuators, and one or more functions by applying a modified estimator algorithm and / or altered calibration corrections to one or more sensors, one or more actuators, and one or more functions.

[0013] In another aspect of this disclosure, a method for health monitoring of a motor vehicle motion control system includes: measuring real-time static and dynamic telemetry data about the motor vehicle using one or more sensors mounted on the motor vehicle; and altering the static and dynamic behavior of the motor vehicle using one or more actuators mounted on the motor vehicle. The method further includes utilizing a control module having a processor, memory, and input / output (I / O) ports for communication with one or more sensors and one or more actuators, the control module executing a portion of program code stored in the memory. The program code portion includes an offline program code portion and an online program code portion. The method further includes: collecting telemetry data from the motor vehicle via the offline code portion; performing fault analysis on the telemetry data via the offline code portion; and assigning tasks based on the fault analysis via the offline code portion. The method further includes: analyzing the telemetry data via the online code portion to locate faults in specific sensors, actuators, or functions utilizing the sensor and / or actuator system; and mitigating deviations in the telemetry data via the online code portion by sending corrections to one or more sensors, actuators, and / or functions of the motor vehicle motion control system.

[0014] In another aspect of this disclosure, the method includes collecting real-time static and dynamic data from one or more sensors, one or more actuators, and one or more functions of a motor vehicle motion control system via an offline code portion through an I / O port.

[0015] In another aspect of this disclosure, the method further includes performing fault mode determination for one or more sensors, one or more actuators, and one or more functions via an offline code portion, wherein fault mode determination includes determining whether one or more sensors, one or more actuators, and / or one or more functions are experiencing a fault mode of performance degradation, complete failure, and / or wherein the telemetry data includes noise exceeding a predefined threshold noise value.

[0016] In another aspect of this disclosure, the method further includes performing data preparation, generation, and collection for one or more sensors, one or more actuators, and one or more functions via an offline code portion, wherein the data preparation further includes: exclusion, filtering, and / or buffering processes.

[0017] In another aspect of this disclosure, the method further includes leveraging a machine learning architecture design via offline code to generate task assignment schemes for telemetry data from one or more sensors, one or more actuators, and one or more functions. The machine learning architecture further includes machine learning and / or artificial intelligence-based clustering methods for identifying and classifying failure modes.

[0018] In another aspect of this disclosure, the method further includes applying temporary or short-term mitigation to one or more sensors, one or more actuators, and / or one or more functions and sending the raw telemetry data to a cloud computing system for further analysis.

[0019] In another aspect of this disclosure, the method further includes implementing data exclusion on telemetry data from one or more sensors, one or more actuators, and one or more functions via an online code portion.

[0020] In another aspect of this disclosure, the method further includes detecting and predicting potential faults within one or more sensors, one or more actuators, and one or more functions via online code portions.

[0021] In another aspect of this disclosure, the method further includes communicating telemetry data from one or more sensors, one or more actuators, and one or more functions via an online code portion through I / O ports between the I / O ports of a remote control module within a motor vehicle and a cloud computing system physically separated from the motor vehicle.

[0022] In another aspect of this disclosure, the method further includes mitigating biases in telemetry data from one or more sensors, one or more actuators, and one or more functions by applying a modified estimator algorithm and / or altered calibration corrections to one or more sensors, one or more actuators, and one or more functions via an online code portion.

[0023] In another aspect of this disclosure, a method for health monitoring of a motor vehicle motion control system further includes: measuring real-time static and dynamic telemetry data about the motor vehicle using one or more sensors mounted on the motor vehicle; and altering the static and dynamic behavior of the motor vehicle using one or more actuators mounted on the motor vehicle. The method further includes utilizing a control module having a processor, memory, and input / output (I / O) ports for communication with one or more sensors and one or more actuators, the control module executing a portion of program code stored in the memory. The program code portion includes an offline program code portion and an online program code portion. The method further includes collecting telemetry data from the motor vehicle via the I / O ports through the offline code portion, including real-time static and dynamic data from one or more sensors, one or more actuators, and one or more functions of the motor vehicle motion control system. The method further includes performing data preparation, generation, and collection for one or more sensors, one or more actuators, and one or more functions through the offline code portion, wherein data preparation further includes exclusion, filtering, and / or buffering processes. The method further includes performing fault mode determination for one or more sensors, one or more actuators, and one or more functions through the offline code portion. Fault mode determination includes determining whether one or more sensors, one or more actuators, and / or one or more functions are experiencing performance degradation, complete failure, and / or where telemetry data includes noise exceeding a predefined threshold noise value. The method further includes assigning tasks based on fault analysis via a machine learning architecture within the offline code section. The machine learning architecture further includes: clustering methods based on machine learning and / or artificial intelligence for identifying and classifying fault modes. The online code section analyzes telemetry data to look for faults within specific sensors, actuators, or functions utilizing sensor and / or actuator systems. The online code section performs data exclusion on telemetry data from one or more sensors, one or more actuators, and one or more functions. The online code section detects and predicts potential faults within one or more sensors, one or more actuators, and one or more functions. The online code section communicates telemetry data from one or more sensors, one or more actuators, and one or more functions via I / O ports between the vehicle and the I / O ports of a remote control module within a cloud computing system physically separated from the vehicle. The online code section mitigates biases from telemetry data from one or more sensors, one or more actuators, and one or more functions by sending corrections, including applying modified estimator algorithms and / or altered calibration corrections to one or more sensors, one or more actuators, and one or more functions.

[0024] Other applicable areas will become apparent from the description provided herein. It should be understood that the descriptions and specific examples are intended for illustrative purposes only and are not intended to limit the scope of this disclosure.

[0025] The present invention also includes the following technical solutions.

[0026] Technical Solution 1. A motor vehicle motion control health monitoring system, comprising:

[0027] One or more sensors mounted on a motor vehicle, the one or more sensors measuring real-time static and dynamic telemetry data about the motor vehicle;

[0028] One or more actuators mounted on a motor vehicle, said one or more actuators altering the static and dynamic behavior of the motor vehicle;

[0029] A control module having a processor, memory, and input / output (I / O) ports for communicating with one or more sensors and one or more actuators, the control module executing a portion of program code stored in the memory, the program code including:

[0030] The offline program code section collects telemetry data from the vehicle, performs fault analysis on the telemetry data, and assigns tasks based on the fault analysis; and

[0031] The online program code section analyzes telemetry data to locate faults in specific sensors, actuators, or within the functions of sensor and / or actuator systems, and mitigates telemetry data bias by sending corrections to one or more sensors, actuators, and / or functions of the motor vehicle motion control system.

[0032] Technical Solution 2. The system according to Technical Solution 1, wherein the offline program code portion further includes:

[0033] The first program code portion collects real-time static and dynamic data from one or more sensors, one or more actuators, and one or more functions of the motor vehicle motion control system via I / O ports.

[0034] Technical Solution 3. The system according to Technical Solution 1, wherein the offline program code portion further includes:

[0035] The second program code section performs fault mode determination for one or more sensors, one or more actuators, and one or more functions, wherein the fault mode determination includes determining whether one or more sensors, one or more actuators, and / or one or more functions are experiencing a fault mode such as performance degradation, complete failure, and / or telemetry data including noise exceeding a predefined threshold noise value.

[0036] Technical Solution 4. The system according to Technical Solution 1, wherein the offline program code portion further includes:

[0037] The third program code portion performs data preparation, generation, and collection for one or more sensors, one or more actuators, and one or more functions, wherein the data preparation further includes: exclusion, filtering, and / or buffering processes.

[0038] Technical Solution 5. The system according to Technical Solution 1, wherein the offline program code portion further includes:

[0039] The fourth program code section utilizes a machine learning architecture design to generate a task allocation scheme for telemetry data from one or more sensors, one or more actuators, and one or more functions. The machine learning architecture further includes a clustering method based on machine learning and / or artificial intelligence for identifying and classifying fault modes. The fourth program code section further applies temporary or short-term mitigations to the one or more sensors, one or more actuators, and / or one or more functions and sends the raw telemetry data to a cloud computing system for further analysis.

[0040] Technical Solution 6. The system according to Technical Solution 1, wherein the online program code portion further includes:

[0041] The fifth program code section performs data exclusion on telemetry data from one or more sensors, one or more actuators, and one or more functions.

[0042] Technical Solution 7. The system according to Technical Solution 1, wherein the online program code portion further includes:

[0043] The sixth program code section detects and predicts potential faults within one or more sensors, one or more actuators, and one or more functions.

[0044] Technical Solution 8. The system according to Technical Solution 1, wherein the online program code portion further includes:

[0045] The seventh program code portion communicates telemetry data from one or more sensors, one or more actuators, and one or more functions via I / O ports between the motor vehicle and the I / O ports of a remote control module within a cloud computing system physically separated from the motor vehicle.

[0046] Technical Solution 9. The system according to Technical Solution 1, wherein the online program code portion further includes:

[0047] The eighth program code section mitigates biases in telemetry data from one or more sensors, one or more actuators, and one or more functions by applying a modified estimator algorithm and / or altered calibration corrections to one or more sensors, one or more actuators, and one or more functions.

[0048] Technical Solution 10. A method for health monitoring of a motor vehicle motion control system, the method comprising:

[0049] Real-time static and dynamic telemetry data about a motor vehicle are measured using one or more sensors mounted on the vehicle.

[0050] The static and dynamic behavior of a motor vehicle can be altered by means of one or more actuators mounted on the vehicle.

[0051] The control module has a processor, a memory, and input / output (I / O) ports for communicating with one or more sensors and one or more actuators. The control module executes a portion of program code stored in the memory, wherein the portion of program code includes an offline portion of program code and an online portion of program code.

[0052] Telemetry data is collected from motor vehicles via offline code segments;

[0053] Fault analysis is performed on telemetry data using offline code; and

[0054] Tasks are assigned based on fault analysis using offline code components;

[0055] Analyzing telemetry data via online code segments to locate faults within the functionality of specific sensors, actuators, or sensor and / or actuator systems; and

[0056] The bias in telemetry data is mitigated by sending corrections to one or more sensors, actuators, and / or functions of the vehicle's motion control system via an online code component.

[0057] Technical Solution 11. The method according to Technical Solution 10 further includes:

[0058] Real-time static and dynamic data are collected from one or more sensors, one or more actuators, and one or more functions of the vehicle motion control system via I / O ports through the offline code portion.

[0059] Technical Solution 12. The method according to Technical Solution 10 further includes:

[0060] Fault mode determination is performed for one or more sensors, one or more actuators, and one or more functions through an offline code section. Fault mode determination includes determining whether one or more sensors, one or more actuators, and / or one or more functions are experiencing a fault mode such as performance degradation, complete failure, and / or telemetry data including noise exceeding a predefined threshold noise value.

[0061] Technical Solution 13. The method according to Technical Solution 10 further includes:

[0062] The offline code section performs data preparation, generation, and collection for one or more sensors, one or more actuators, and one or more functions, wherein data preparation further includes: exclusion, filtering, and / or buffering processes.

[0063] Technical Solution 14. The method according to Technical Solution 10 further includes:

[0064] The machine learning architecture is designed to generate task assignment schemes for telemetry data from one or more sensors, one or more actuators, and one or more functions through offline code. The machine learning architecture further includes machine learning and / or artificial intelligence-based clustering methods for identifying and classifying failure modes.

[0065] Technical Solution 15. The method according to Technical Solution 10 further includes:

[0066] Apply temporary or short-term mitigation to one or more sensors, one or more actuators, and / or one or more functions and send the raw telemetry data to a cloud computing system for further analysis.

[0067] Technical Solution 16. The method according to Technical Solution 10 further includes:

[0068] Data exclusion is implemented through an online code section for telemetry data from one or more sensors, one or more actuators, and one or more functions.

[0069] Technical Solution 17. The method according to Technical Solution 10 further includes:

[0070] Detect and predict potential faults within one or more sensors, one or more actuators, and one or more functions through online code.

[0071] Technical Solution 18. The method according to Technical Solution 10 further includes:

[0072] Telemetry data from one or more sensors, one or more actuators, and one or more functions are communicated via I / O ports in a motor vehicle and a remote control module in a cloud computing system physically separated from the motor vehicle through an online code section.

[0073] Technical Solution 19. The method according to Technical Solution 10 further includes:

[0074] The online code section mitigates biases in telemetry data from one or more sensors, actuators, and functions by applying modified estimator algorithms and / or altered calibration corrections to one or more sensors, actuators, and functions.

[0075] Technical Solution 20. A method for health monitoring of a motor vehicle motion control system, the method comprising:

[0076] Real-time static and dynamic telemetry data about a motor vehicle are measured using one or more sensors mounted on the vehicle.

[0077] The static and dynamic behavior of a motor vehicle can be altered by means of one or more actuators mounted on the vehicle.

[0078] The control module has a processor, a memory, and input / output (I / O) ports for communicating with one or more sensors and one or more actuators. The control module executes a portion of program code stored in the memory, wherein the portion of program code includes an offline portion of program code and an online portion of program code.

[0079] Telemetry data is collected from the vehicle via I / O ports through the offline code section, including real-time static and dynamic data from one or more sensors, one or more actuators, and one or more functions of the vehicle's motion control system.

[0080] Data preparation, generation, and collection are performed for one or more sensors, one or more actuators, and one or more functions through an offline code component, wherein data preparation further includes exclusion, filtering, and / or buffering processes;

[0081] Fault mode determination is performed for one or more sensors, one or more actuators, and one or more functions through an offline code section. The fault mode determination includes determining whether one or more sensors, one or more actuators, and / or one or more functions are experiencing a fault mode such as performance degradation, complete failure, and / or telemetry data including noise exceeding a predefined threshold noise value.

[0082] Task assignment is determined based on failure modes through a machine learning architecture within the offline code section, wherein the machine learning architecture further includes: clustering methods based on machine learning and / or artificial intelligence for identifying and classifying failure modes;

[0083] Analyzing telemetry data via online code segments to locate faults within the functionality of specific sensors, actuators, or sensor and / or actuator systems; and

[0084] Data exclusion is performed on telemetry data from one or more sensors, one or more actuators, and one or more functions through an online code section;

[0085] Detects and predicts potential faults within one or more sensors, one or more actuators, and one or more functions through online code.

[0086] Telemetry data from one or more sensors, one or more actuators, and one or more functions are communicated via I / O ports in the online code portion between the vehicle and the I / O ports of a remote control module within a cloud computing system physically separated from the vehicle; and

[0087] The online code section mitigates biases in telemetry data from one or more sensors, one or more actuators, and one or more functions by sending corrections, including applying modified estimator algorithms and / or altered calibration corrections to one or more sensors, one or more actuators, and one or more functions. Attached Figure Description

[0088] The accompanying drawings described herein are for illustrative purposes only and are not intended to limit the scope of this disclosure in any way.

[0089] Figure 1 This is a schematic diagram of a system and method for health monitoring and mitigation of vehicle motion control according to aspects of this disclosure;

[0090] Figure 2A It is a depiction of aspects according to this disclosure. Figure 1 A flowchart outlining the systems and methods for health monitoring and mitigation in vehicle motion control;

[0091] Figure 2B It is a depiction of aspects according to this disclosure. Figure 2A A flowchart of a set of logical steps for grouping telemetry data in a system and method for vehicle motion control health monitoring and mitigation; and

[0092] Figure 2C It is a diagram illustrating aspects of this disclosure for use in Figure 2A A flowchart of the training algorithm for a system and method for vehicle motion control health monitoring and mitigation. Detailed Implementation

[0093] The following description is exemplary in nature and is not intended to limit this disclosure, application, or use.

[0094] refer to Figure 1 This illustration shows a vehicle motion control (VMC) health monitoring system 10. System 10 operates on one or more motor vehicles 12. Motor vehicle 12 is shown as an automobile; however, it should be understood that motor vehicle 12 can be a minivan, bus, tractor-trailer, semi-trailer, SUV, truck, bicycle, electric bicycle, tricycle, motorcycle, aircraft, helicopter, amphibious vehicle, or any other such vehicle without departing from the scope or intent of this disclosure. Figure 1 In the example shown, the motor vehicle 12 is equipped with a powertrain 14 capable of transmitting prime mover power to the wheels 16 of the motor vehicle and the tires 18 fixed to the wheels 16. The powertrain 14 may include various components such as an internal combustion engine (ICE) 20 and / or an electric motor 22, and a transmission 24 capable of transmitting torque from the ICE 20 and / or the electric motor 22 to the wheels 16. In one example, the motor vehicle 12 may include an ICE 20 acting on the rear axle 26 of the motor vehicle 12 and one or more electric motors 22 acting on the front axle 28 of the motor vehicle 12. In another example, the motor vehicle 12 may use one or more ICEs 20 and / or one or more electric motors 22, arranged in a different configuration and providing torque to the front axle 28 or the rear axle 26, or even to individual wheels 16 of the motor vehicle 12, without departing from the scope or intent of this disclosure.

[0095] In several aspects, the powertrain 14 includes one or more in-plane actuators 30. The in-plane actuators 30 may include an all-wheel drive (AWD) system (including an electronically controlled or electric AWD (eAWD) 32 system) and a limited-slip differential (LSD) 34 (including an electronically controlled or electric LSD (eLSD) 36 system). The in-plane actuators 30 can generate or modify force generation at the tires 18 in the X and / or Y directions toward the road contact portion 38 within a predetermined capacity. The eAWD 32 system can transmit torque from the front to the rear of the vehicle 12 and / or from one side of the vehicle 12 to the other. Similarly, the eLSD 36 can transmit torque from one side of the vehicle 12 to the other. In some examples, the eAWD 32 and / or eLSD 36 can directly alter or manage torque delivery from the ICE 20 and / or electric motor 22, and / or the eAWD 32 and eLSD 36 can act on the braking system 40 to adjust the amount of torque delivered to each of the tires 18 of the vehicle 12. Additional in-plane actuators 30 may include an active steering or electronic power steering (EPS) system 42 at either or both of the front axle 28 and rear axle 26. The active steering system or EPS system 42 can actively adjust the angle of the wheels 16 relative to the longitudinal axis of the vehicle 12.

[0096] In a further example, the motor vehicle 12 may include means for altering the normal force on each of the tires 18 of the motor vehicle 12 via one or more out-of-plane actuators 44. The out-of-plane actuators 44 of the motor vehicle 12 may include any of a variety of actuators 44 capable of managing the vertical movement of the motor vehicle 12. In several aspects, the out-of-plane actuators 44 may include active aerodynamic actuators 46, active suspension actuators 48, etc. The active aerodynamic actuators 46 may actively or passively alter the aerodynamic profile of the motor vehicle via one or more active aerodynamic elements 49 (such as winglets, spoilers, fans or other suction devices, actively managed venturi wind tunnels, diffusers, etc.). The active suspension actuators 48 adjust suspension travel, spring stiffness, and damping characteristics. In some examples, the active suspension actuators 48 may include magnetorheological dampers, pneumatic dampers, or springs, or other such electrically, hydraulically, or pneumatically adjusted dampers or springs, without departing from the scope or intent of this disclosure.

[0097] The terms “forward,” “rear,” “inner,” “inward,” “outer,” “outer,” “above,” and “below” are terms used relative to the orientation of the motor vehicle 12, as shown in the accompanying drawings of this application. Thus, “forward” refers to the direction toward the front of the motor vehicle 12, and “rearward” refers to the direction toward the rear of the motor vehicle 12. “Left side” refers to the direction toward the left-hand side of the motor vehicle 12 relative to the front of the motor vehicle 12. Similarly, “right side” refers to the direction toward the right-hand side of the motor vehicle 12 relative to the front of the motor vehicle 12. “Inner” and “inward” refer to the direction toward the interior of the motor vehicle 12, and “outer” and “outer” refer to the direction toward the exterior of the motor vehicle 12. “Below” refers to the direction toward the bottom of the motor vehicle 12, and “above” refers to the direction toward the top of the motor vehicle 12. Further, the terms “top,” “above,” “bottom,” “side,” and “above” are terms used relative to the orientation of the actuator, and the motor vehicle 12 is shown more broadly in the accompanying drawings of this application. Therefore, although the orientation of actuator 52 or motor vehicle 12 may vary with respect to a given application, these terms are intended to remain applicable relative to the orientation of the components of system 10 and motor vehicle 12 shown in the accompanying drawings.

[0098] System 10 further includes one or more controllers 50. Controller 50 is a non-general-purpose electronic control device having a pre-programmed digital computer or processor 52, a non-transitory computer-readable medium or memory 54 for storing data (such as control logic, software applications, instructions, computer code, data, lookup tables, etc.), and input / output (I / O) ports 56. The computer-readable medium or memory 54 includes any type of media accessible by a computer, such as read-only memory (ROM), random access memory (RAM), hard disk drive, solid-state storage, compact optical disc (CD), digital video optical disc (DVD), or any other type of memory 54. The non-transitory computer-readable memory 54 excludes wired, wireless, optical, or other communication links that transmit transient electrical or other signals. The non-transitory computer-readable memory 54 includes media where data can be permanently stored and media where data can be stored and later rewritten, 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. Processor 52 is configured to execute code or instructions. The motor vehicle 12 may have a controller 50, including a dedicated Wi-Fi controller, an engine control module, a transmission control module, a body control module, a suspension control module, a braking control module, an infotainment control module, etc. I / O ports 56 may be configured to communicate via wired communication, or wirelessly via Wi-Fi protocols according to IEEE 802.11x, cellular links, satellite links, etc., without departing from the scope or intent of this disclosure.

[0099] The vehicle controller 50 further includes one or more applications 68. An application 68 is a software program configured to perform a specific function or set of functions. An application 68 may include one or more computer programs, software components, instruction sets, processes, functions, objects, classes, examples, related data, or portions thereof suitable for implementation in suitable computer-readable program code. Applications 68 may be stored within memory 54 or in additional or separate memory 54. In several aspects, applications 68 may manage powertrain system functions, suspension system functions, braking system functions, aerodynamic system functions, and / or body control system functions in the exemplary motor vehicle 12.

[0100] Application 68, which manages the functions of the powertrain system, suspension system, braking system, aerodynamic system, and body control system in the motor vehicle 12, receives static and / or dynamic vehicle state information or sensing data from a set of sensors 70 mounted on the motor vehicle 12. Sensors 70 may include any of a variety of sensors, including an inertial measurement unit (IMU) 72, a suspension control unit such as a semi-active damped suspension (SADS) 74, a global positioning system (GPS) sensor 76, wheel speed sensors 78, throttle position sensors 80, accelerator pedal position sensors 82, brake pedal position sensors 84, steering position sensors 86, tire pressure monitoring sensors 88, aerodynamic element position sensors 90, and so on. The IMU 72 measures movement, acceleration, etc., with several degrees of freedom. In a specific example, the IMU 72 may measure position, movement, acceleration, etc., with three or more degrees of freedom. Similarly, the SADS 74 sensor may be an IMU 72 capable of measuring with three or more degrees of freedom. In some more specific examples, SADS 74 could be a suspension hub accelerometer, etc. Therefore, the sensor data can include, but is not limited to: wheel speed data, SADS 74 and IMU 72 data, including attitude, acceleration, etc.

[0101] The sensor data is transmitted to the controller 50 via I / O port 56. At the controller 50, the application 68 uses the sensor data about the vehicle 12 to determine the positions of the in-plane actuator 30 and the out-of-plane actuator 44 to achieve a specific state of the vehicle 12.

[0102] System 10 further includes a non-vehicle or remote computing system 92. The remote computing system 92 may include one or more non-vehicle or remote controllers 50' whose architectural design is generally similar to those described herein. Specifically, the remote controller 50' includes a digital processor 52', a computer-readable storage 54', and an I / O port 56'. The memory 54' of the remote computing system 92 stores one or more applications 68' executable by the digital processor 52', and the I / O port 56' of the remote computing system 92 communicates with the I / O port 56 of the controller 50 of the motor vehicle 12.

[0103] Turn now Figure 2A and Figure 2B And continue to refer to Figure 1System 10 for motor VMC health monitoring collects data about motor vehicle 12 at block 100. More specifically, at block 100, system 10 utilizes sensor data from one or more onboard applications 68 and one or more offboard or remote applications 68' to detect or predict potential faults in various sensors 70 and / or in-plane and out-of-plane actuators 30, 44. To address large-scale VMC, a structured machine learning (ML)-based architecture is utilized within the onboard and remote applications 68, 68'. The ML architecture generally consists of an offline phase 102 or offline program code portion and an online phase 104 or online code portion. Offline phase 102 implements a first step 106 of data preparation, collection, and labeling of sensor data from various sensors 70 and in-plane and out-of-plane actuators 30, 44. In the first step 106 of offline phase 102, system 10 sets fault modes for each function of the motor vehicle. For example, the IMU 72, wheel speed sensor 78, or actuators (such as electric motor 22, EPS 42, transmission 24, ICE 20, etc.) have personalized and system failure modes based on predefined parameters or domain knowledge. In several aspects, the first step 106 of the offline phase 102 also utilizes clustering methods based on machine learning (ML) and / or artificial intelligence (AI) to identify and classify failure modes, even when a single physical system may not be in a failure state, but the broader VMC system may be in a failure state when combined with the state of other physical systems or software functions. In the example of software failure, the failure mode might be performance degradation, which can be further subdivided into multiple degradation levels, or the failure mode might be a complete failure, or the failure mode might simply be noise in the data exceeding a predefined threshold. The failure modes can then be classified based on causality. For example, software failures can be attributed to input errors, algorithmic problems such as unmodeled dynamics, calibration problems, etc. System 10 then analyzes the data, including system input signals, system 10 output signals, and checkpoint signals from internal system 10 check steps.

[0104] Subsequently, system 10 implements the second step 108 of offline phase 102, in which the system collects a set of data for an idealized scenario in which no faults occur. Based on the failure mode of each specific function, system 10 applies human error to each input and / or model. Then, system 10 collects the inputs and outputs of each function or feature and utilizes these inputs and outputs in the third step 110 of offline phase 102.

[0105] In the third step 110 of the offline phase 102, a structured multi-group ML / AI architecture is used to monitor the status of VMC system functions, features, and hardware. The third step 110 further categorizes and groups the functions, features, and hardware based on their impact on the operation of the motor vehicle 12, and more specifically, their criticality to the VMC. Among several aspects, features may include torque vectorization between the front axle 28 and rear axle 26 and / or between individual wheels 16 via AWD or eAWD 32, LSD 34, and / or eLSD 36, etc.

[0106] The input to each group consists of direct measurements or estimates of those measurements from various sensors 70 and in-plane and out-of-plane actuators 30, 44. Estimates of the measurements can be considered as virtual sensors. That is, additional VMC information can be indirectly collected by interpreting data from one or more sensors 70 and in-plane and out-of-plane actuators 30, 44. The indirectly collected information may be related to specific physical or functional conditions of the vehicle 12 and the VMC system. Additionally, a confidence level for the accuracy of each group is determined. The confidence level provides a measure of the accuracy for each member of the group and is used to provide weights in weighted calculations involving each member of the group.

[0107] Priorities and criticality determine which functions, features, and hardware are placed in each group. More specifically, prioritizing functions, features, and hardware allows system 10 to determine whether calculations about potential failures, etc., should be performed locally within one or more vehicle 12 controllers 50 or one or more remote controllers 50' (such as controllers 50' located within a cloud computing system).

[0108] In the prioritization example, system 10 receives sensing data from one or more of the following: wheel speed sensor 78, steering position sensor 86, IMU 72, and functions such as vehicle speed function, wheel 16 control function, etc. Since each of the aforementioned sensors, features, and functions is directly related to the control and safety of the motor vehicle 12, these sensors, features, and functions can be designated to belong to priority group #1. Similarly, sensors such as the EPS 42 torque sensor, functions such as tire 18 force estimation, and features such as AWD 32, LSD 34, eLSD 36, etc., can be designated to belong to priority group #2. Data from additional sensors, features, and functions can be similarly grouped into lower priority groups, depending on various factors indicated by priority group #N, such as those including rain sensors (not shown), spring stiffness degradation functions, and features such as agility enhancement.

[0109] Turn now Figure 2C Special and continue to be referenced Figures 1-2B The diagram illustrates a flowchart of a training algorithm 200 that divides the computational burden between an onboard controller 50 and a remote or cloud-based controller 50'. The training algorithm 200 is based on applications 68, 68' described above. More specifically, at block 202, the training algorithm 200 is initialized or begins training. At block 204, the training algorithm 200 reads raw data from sensors 70 on the vehicle 12 within a snapshot covering a predetermined amount of time. The predetermined amount of time can vary, but should generally be understood to include sufficient time for the sensors 70 and the in-plane and out-of-plane actuators 30, 44 to generate data. That is, the response time of each of the sensors 70 and the in-plane and out-of-plane actuators 30, 44 is used to determine the length of the predetermined snapshot. The snapshot is temporarily stored in memory 54 within the onboard controller 50, such as in a circular buffer in RAM, before being transmitted or processed. At block 206, the training algorithm 200 prepares the data for further analysis. Data preparation may include various processes, such as exclusion and / or filtering and / or buffering processes, etc. At box 208, prepared or processed data is obtained from data preparation by training algorithm 200. At box 210, training algorithm 200 determines whether the data of one or more of the following—sensor 70, in-plane and out-of-plane actuators 30, 44, one or more features, and one or more functions—have a deviation below a predefined or ML-set or adjusted deviation threshold. If, at box 210, the data is below the deviation threshold, training algorithm 200 returns to box 204, where the original sensor data is promptly reread in a new snapshot.

[0110] However, if the data exceeds the deviation threshold, training algorithm 200 proceeds to box 212, where it determines whether the data deviation is equal to or higher than the deviation threshold, and additionally determines whether the deviation is lower than a predefined safety threshold. The safety threshold may include any of a variety of parameters from emission control, air quality within the vehicle 12, lateral and / or longitudinal and / or vertical acceleration and speed, steering angle, braking characteristics, torque vectoring settings, etc., without departing from the scope or intent of this disclosure. If the data deviation is equal to or greater than the deviation threshold but lower than the safety threshold, at box 214, the raw sensor data of the snapshot is sent to a remote or cloud-based controller 50' for further processing. Conversely, if the data deviation is higher than the safety threshold and / or lower than the deviation threshold, training algorithm 200 proceeds to box 216. At box 216, the training algorithm selects and applies mitigation methods that attempt to address the deviation and bring the raw sensor data back within the expected range. At box 218, training algorithm 200 runs multiple test scenarios to determine the effectiveness of the selected mitigation method, and then proceeds to box 214, where the raw sensor data, along with test scenario information and mitigation method information, is sent to the cloud-based controller 50'. From box 214, the training algorithm returns to box 204, where training algorithm 200 continues to continuously read data from the sensors 70 on the vehicle 12, as well as the in-plane and out-of-plane actuators 30, 44. It should be understood that... Figure 2C The training algorithm 200 can be run in online or offline mode and / or in real time without departing from the scope or intent of this disclosure.

[0111] The decision to proceed from box 212 to box 214 or to box 216 is based on deviation and safety thresholds. It should be understood that, generally, VMC systems generate large amounts of data, and the on-board controller 50 has limited storage and processing capabilities, while the controller 50' in the cloud has significantly greater storage and processing capabilities. However, the movement of data from the vehicle 12 to the cloud is limited by the communication means and bandwidth between the vehicle 12 and the controller 50' in the cloud. Therefore, some processing tasks are assigned to the cloud-based controller 50' in the online phase 104 of data processing, while other processing tasks are assigned to and executed by the on-board controller 50 in the offline phase 102.

[0112] In some specific examples, the communication scheme 112 in the online phase 104 is determined, partly based on the type of deviation (i.e., whether it is associated with safety considerations) and partly based on the amount of computational resources required to resolve the deviation in the data, to process the deviation via the onboard controller 50 or the cloud-based controller 50'. Specifically, the computer processing power within the vehicle 12 is limited by the hardware of the onboard controller 50, while significantly more processing resources are available in the cloud-based controller 50'. However, transmitting deviation data to the cloud-based controller 50' is limited by the bandwidth of the data connection between the onboard controller 50 and the cloud-based controller 50'. Therefore, in some cases, transmitting deviation data to the cloud-based controller 50' may take too long to make this solution feasible and / or to achieve the fast VMC response required to maintain the safe operation of the vehicle 12. In cases where the deviation is safety-critical, the system 10 applies temporary or short-term mitigation strategies to the data and continues to send the raw data to the cloud for further analysis. Temporary or short-term mitigation strategies are essentially predefined outputs that address the data deviation reaching the sensor 70, in-plane and out-of-plane actuators 30, 44, functions, characteristics, etc. The predefined output is obtained from a lookup table or other such database with predefined values. In contrast, when a non-safety-critical deviation is detected, system 10 simply sends the raw data to the cloud-based controller 50' for further analysis without applying short-term mitigation strategies. That is, for non-safety-critical data deviations, the computational resources required to accurately and precisely correct the data deviations can reside in the cloud.

[0113] In some further examples, data bias can be addressed in a hybrid manner. That is, some processing, such as regression, curve fitting, or parity algorithms, can be performed by the onboard controller 50, while additional processes can be performed on the regression or curve fitting data in the cloud-based controller 50'. Additionally, when snapshot data and / or bias data are uploaded to the cloud-based controller 50' for non-onboard evaluation, data from each snapshot can be enhanced with real-time data, such as telemetry data from the vehicle 12 from the onboard sensors 70.

[0114] In a specific example, a user of vehicle 12 can operate vehicle 12 by manipulating one or more of the accelerator pedal, steering wheel, brake pedal, etc. Inputs arriving at the accelerator pedal, steering wheel, or brake pedal are read by the accelerator pedal position sensor 82, steering position sensor 86, and brake pedal position sensor 84 and sent to the onboard controller 50. The onboard controller 50 makes a determination within itself regarding whether the signal is within a known threshold. Additional information from other sensors 70 on vehicle 12, as well as, for example, GPS location, is also reported to the onboard controller 50. This data defines the event information processed on-vehicle by an ML algorithm or model-based algorithm residing in the memory 54 of the onboard controller 50. Data inputs arriving at the ML algorithm and / or model-based algorithm may also include trip statistics, such as minimum and maximum values, average values, and standard deviations reported by each sensor 70. The onboard controller 50 then determines whether to issue signals and / or data based on default settings. In some cases, the onboard controller 50 may communicate with a remote controller 50' to increase the signal rate or data rate used for communication to address particularly critical situations. That is, the on-board controller 50 can negotiate with the additional controller 50 in the motor vehicle 12 and / or the remote controller 50' in the cloud to meet communication speed and bandwidth requirements, increase signal or data rates, and take on-board mitigation actions at box 114.

[0115] For each function or a given set of functions, multiple solutions are designed such that each solution can reliably perform the given task. This redundancy provides robustness and reduces the likelihood of mitigating failures. Each of the multiple solutions is based on a different set of signals, resulting in different signal estimation methods. Certain failure modes are associated with each function, and when system 10 confirms the presence of a specific failure mode, system 10 provides mitigation outputs at block 114 to the associated in-plane and / or out-of-plane actuators 30, 44, based on a mixture of confidence levels for each function. More specifically, all degradations, deviations, or failures communicate with remote controller 50', and remote controller 50' sends mitigation outputs to controller 50 within vehicle 12, where the mitigation outputs are applied to the associated VMC system. The mitigation outputs can take various forms, including but not limited to each associated sensor 70, in-plane and / or out-of-plane actuators 30, 44, modified estimator algorithm parameters for functions and / or characteristics, altered calibrations, etc.

[0116] The vehicle motion control health monitoring system disclosed herein offers several advantages. These include the ability to proactively, continuously, accurately, and precisely manage vehicle performance in terms of longitudinal and lateral acceleration, pitch, roll, and yaw under various environmental and road conditions. Furthermore, the VMC health monitoring system 10 of this disclosure allows for cost-effective resource optimization of proactively and continuously managing the health of complex VMC systems with numerous interactive functions. More specifically, the VMC health monitoring system 10 of this disclosure provides means to safely mitigate performance degradation or deterioration of system component performance, or even VMC system failure, while maintaining or reducing cost and complexity and improving the motion control of the motor vehicle 12.

[0117] The description in this disclosure is exemplary in nature only, and variations thereof that do not depart from the spirit and scope of this disclosure are intended to fall within its scope. Such variations should not be considered as departing from the spirit and scope of this disclosure.

Claims

1. A motor vehicle motion control health monitoring system, comprising: One or more sensors mounted on a motor vehicle, the one or more sensors measuring real-time static and dynamic telemetry data about the motor vehicle; One or more actuators mounted on a motor vehicle, said one or more actuators altering the static and dynamic behavior of the motor vehicle; A control module having a processor, memory, and input / output (I / O) ports for communicating with one or more sensors and one or more actuators, the control module executing a portion of program code stored in the memory, the program code including: The offline program code section collects telemetry data from the vehicle, performs fault analysis on the telemetry data, and assigns tasks based on the fault analysis; and The online program code section analyzes telemetry data to find faults in specific sensors, actuators, or within the functions of sensor and / or actuator systems and mitigates deviations in the telemetry data by sending corrections to one or more sensors, actuators, and / or functions of the motor vehicle motion control system. The correction includes applying a modified estimator algorithm and / or altered calibration correction to one or more sensors, one or more actuators, and one or more functions to mitigate biases in telemetry data from these sensors, actuators, and functions. The system determines priorities and criticality to decide which functions, features, and hardware to place in each group, thereby determining whether calculations regarding potential faults should be performed locally within one or more vehicle controllers or one or more remote controllers, wherein each of the sensors, features, and functions directly related to the control and safety of the vehicle is designated to belong to a high-priority group. For each function or a given set of functions, multiple solutions are designed such that each solution can reliably perform the given task, wherein each of the multiple solutions is based on a different set of signals, resulting in different signal estimation methods, some specific failure modes are associated with each function, and when the system confirms the presence of a specific failure mode, a mitigation output is provided to the relevant in-plane and / or out-of-plane actuators based on a mixture of confidence levels for each function.

2. The system according to claim 1, wherein, The offline program code section further includes: The first program code portion collects real-time static and dynamic data from one or more sensors, one or more actuators, and one or more functions of the motor vehicle motion control system via I / O ports.

3. The system according to claim 1, wherein, The offline program code section further includes: The second program code section performs fault mode determination for one or more sensors, one or more actuators, and one or more functions, wherein the fault mode determination includes determining whether one or more sensors, one or more actuators, and / or one or more functions are experiencing a fault mode such as performance degradation, complete failure, and / or telemetry data including noise exceeding a predefined threshold noise value.

4. The system according to claim 1, wherein, The offline program code section further includes: The third program code portion performs data preparation, generation, and collection for one or more sensors, one or more actuators, and one or more functions, wherein the data preparation further includes: exclusion, filtering, and / or buffering processes.

5. The system according to claim 1, wherein, The offline program code section further includes: The fourth program code section utilizes a machine learning architecture design to generate a task allocation scheme for telemetry data from one or more sensors, one or more actuators, and one or more functions. The machine learning architecture further includes a clustering method based on machine learning and / or artificial intelligence for identifying and classifying fault modes. The fourth program code section further applies temporary or short-term mitigations to the one or more sensors, one or more actuators, and / or one or more functions and sends the raw telemetry data to a cloud computing system for further analysis.

6. The system according to claim 1, wherein, The online program code section further includes: The fifth program code section performs data exclusion on telemetry data from one or more sensors, one or more actuators, and one or more functions.

7. The system according to claim 1, wherein, The online program code section further includes: The sixth program code section detects and predicts potential faults within one or more sensors, one or more actuators, and one or more functions.

8. The system according to claim 1, wherein, The online program code section further includes: The seventh program code portion communicates telemetry data from one or more sensors, one or more actuators, and one or more functions via I / O ports between the motor vehicle and the I / O ports of a remote control module within a cloud computing system physically separated from the motor vehicle.

9. A method for health monitoring of a motor vehicle motion control system, the method comprising: Real-time static and dynamic telemetry data about a motor vehicle are measured using one or more sensors mounted on the vehicle. The static and dynamic behavior of a motor vehicle can be altered by means of one or more actuators mounted on the vehicle. The control module has a processor, a memory, and input / output (I / O) ports for communicating with one or more sensors and one or more actuators. The control module executes a portion of program code stored in the memory, wherein the portion of program code includes an offline portion of program code and an online portion of program code. Telemetry data is collected from motor vehicles via offline code segments; Fault analysis is performed on telemetry data using offline code; and Tasks are assigned based on fault analysis using offline code components; Analyzing telemetry data via online code segments to locate faults within the functionality of specific sensors, actuators, or sensor and / or actuator systems; and The bias in telemetry data is mitigated by sending corrections to one or more sensors, actuators, and / or functions of the vehicle motion control system through the online code section. The correction includes applying a modified estimator algorithm and / or altered calibration correction to one or more sensors, one or more actuators, and one or more functions to mitigate biases in telemetry data from these sensors, actuators, and functions. The method also includes determining priorities and criticality to decide which functions, features, and hardware to place in each group, thereby determining whether calculations regarding potential failures should be performed locally within one or more vehicle controllers or one or more remote controllers, wherein each of the sensors, features, and functions directly associated with the control and safety of the vehicle is designated to belong to a high-priority group. For each function or a given set of functions, multiple solutions are designed such that each solution can reliably perform the given task, wherein each of the multiple solutions is based on a different set of signals, resulting in different signal estimation methods, some specific failure modes are associated with each function, and when the presence of a specific failure mode is confirmed, a mitigation output is provided to the relevant in-plane and / or out-of-plane actuators based on the confidence level of each function.

10. The method of claim 9, further comprising: Real-time static and dynamic data are collected from one or more sensors, one or more actuators, and one or more functions of the vehicle motion control system via I / O ports through the offline code portion.

11. The method of claim 9, further comprising: Fault mode determination is performed for one or more sensors, one or more actuators, and one or more functions through an offline code section. Fault mode determination includes determining whether one or more sensors, one or more actuators, and / or one or more functions are experiencing a fault mode such as performance degradation, complete failure, and / or telemetry data including noise exceeding a predefined threshold noise value.

12. The method of claim 9, further comprising: The offline code section performs data preparation, generation, and collection for one or more sensors, one or more actuators, and one or more functions, wherein data preparation further includes: exclusion, filtering, and / or buffering processes.

13. The method of claim 9, further comprising: The machine learning architecture is designed to generate task assignment schemes for telemetry data from one or more sensors, one or more actuators, and one or more functions through offline code. The machine learning architecture further includes machine learning and / or artificial intelligence-based clustering methods for identifying and classifying failure modes.

14. The method of claim 9, further comprising: Apply temporary or short-term mitigation to one or more sensors, one or more actuators, and / or one or more functions and send the raw telemetry data to a cloud computing system for further analysis.

15. The method of claim 9, further comprising: Data exclusion is implemented through an online code section for telemetry data from one or more sensors, one or more actuators, and one or more functions.

16. The method of claim 9, further comprising: Detect and predict potential faults within one or more sensors, one or more actuators, and one or more functions through online code.

17. The method of claim 9, further comprising: Telemetry data from one or more sensors, one or more actuators, and one or more functions are communicated via I / O ports in a motor vehicle and a remote control module in a cloud computing system physically separated from the motor vehicle through an online code section.

18. A method for health monitoring of a motor vehicle motion control system, the method comprising: Real-time static and dynamic telemetry data about a motor vehicle are measured using one or more sensors mounted on the vehicle. The static and dynamic behavior of a motor vehicle can be altered by means of one or more actuators mounted on the vehicle. The control module has a processor, a memory, and input / output (I / O) ports for communicating with one or more sensors and one or more actuators. The control module executes a portion of program code stored in the memory, wherein the portion of program code includes an offline portion of program code and an online portion of program code. Telemetry data is collected from the vehicle via I / O ports through the offline code section, including real-time static and dynamic data from one or more sensors, one or more actuators, and one or more functions of the vehicle's motion control system. Data preparation, generation, and collection are performed for one or more sensors, one or more actuators, and one or more functions through an offline code component, wherein data preparation further includes exclusion, filtering, and / or buffering processes; Fault mode determination is performed for one or more sensors, one or more actuators, and one or more functions through an offline code section. The fault mode determination includes determining whether one or more sensors, one or more actuators, and / or one or more functions are experiencing a fault mode such as performance degradation, complete failure, and / or telemetry data including noise exceeding a predefined threshold noise value. Task assignment is determined based on failure modes through a machine learning architecture within the offline code section, wherein the machine learning architecture further includes: clustering methods based on machine learning and / or artificial intelligence for identifying and classifying failure modes; Analyzing telemetry data via online code segments to locate faults within the functionality of specific sensors, actuators, or sensor and / or actuator systems; and Data exclusion is performed on telemetry data from one or more sensors, one or more actuators, and one or more functions through an online code section; Detects and predicts potential faults within one or more sensors, one or more actuators, and one or more functions through online code. Telemetry data from one or more sensors, one or more actuators, and one or more functions are communicated via I / O ports in the online code portion between the vehicle and the I / O ports of a remote control module within a cloud computing system physically separated from the vehicle; and The online code section mitigates biases in telemetry data from one or more sensors, one or more actuators, and one or more functions by sending corrections, including applying modified estimator algorithms and / or altered calibration corrections to one or more sensors, one or more actuators, and one or more functions. The correction includes applying a modified estimator algorithm and / or altered calibration correction to one or more sensors, one or more actuators, and one or more functions to mitigate biases in telemetry data from these sensors, actuators, and functions. The method further includes determining priorities and criticality to decide which functions, features, and hardware to place in each group, thereby determining whether calculations regarding potential failures should be performed locally within one or more vehicle controllers or one or more remote controllers, wherein each of the sensors, features, and functions directly associated with the control and safety of the vehicle is designated to belong to a high-priority group. For each function or a given set of functions, multiple solutions are designed such that each solution can reliably perform the given task, wherein each of the multiple solutions is based on a different set of signals, resulting in different signal estimation methods, some specific failure modes are associated with each function, and when the presence of a specific failure mode is confirmed, a mitigation output is provided to the relevant in-plane and / or out-of-plane actuators based on the confidence level of each function.

Images