Method and apparatus for estimating brake pad wear

Abstract

The present disclosure provides a method and apparatus for estimating brake pad wear. Methods and apparatus for estimating brake pad wear are disclosed. An example apparatus includes at least one processor circuit to: obtain temperature data and power data associated with a brake pad of a vehicle; executing a neural network based on the temperature data and the power data, the neural network being trained based on a loss value corresponding to a difference between: (a) a first rate of change in an output of the neural network when the neural network is executed based on training data, and (b) a second rate of change in an output of the neural network when the neural network is executed based on training data; the first rate of change is evaluated at a first temperature and a first power; and (b) a second rate of change proportional to the ratio between the first temperature and the first power; determining a brake wear metric corresponding to the brake pad based on the execution results of the neural network; and cause presentation of the brake pad metrics via a user interface.

Description

Technical Field

[0001] This disclosure relates generally to vehicles, and more specifically to methods and apparatus for estimating brake pad wear. Background Technology

[0002] Braking causes brake pads to wear over time, which can affect a vehicle's braking ability. Therefore, drivers schedule vehicle maintenance to replace brake pads, thereby reducing the likelihood of suboptimal braking conditions. Summary of the Invention

[0003] An example device includes at least one processor circuitry programmed by machine-readable instructions to: obtain temperature data and power data associated with brake pads of a vehicle; execute a neural network based on the temperature data and the power data, the neural network being trained based on a loss value corresponding to the difference between: (a) a first rate of change of the output of the neural network when the neural network is executed based on training data, the first rate of change being evaluated at a first temperature and a first power; and (b) a second rate of change proportional to the ratio between the first temperature and the first power; determine a brake wear metric corresponding to the brake pads based on the execution result of the neural network; and cause the brake pad metric to be presented via a user interface.

[0004] At least one example non-transitory machine-readable medium includes machine-readable instructions that cause at least one processor circuitry to at least: obtain temperature data and power data associated with brake pads of a vehicle; execute a neural network based on the temperature data and the power data, the neural network being trained based on a loss value corresponding to the difference between: (a) a first rate of change of the output of the neural network when the neural network is executed based on training data, the first rate of change being evaluated at a first temperature and a first power, and (b) a second rate of change proportional to the ratio between the first temperature and the first power; determine a brake wear metric corresponding to the brake pads based on the execution result of the neural network; and cause the brake pad metric to be presented via a user interface.

[0005] An example method includes: obtaining temperature data and power data associated with brake pads of a vehicle; executing a neural network based on the temperature data and the power data, the neural network being trained based on a loss value corresponding to the difference between: (a) a first rate of change of the output of the neural network when the neural network is executed based on training data, the first rate of change being evaluated at a first temperature and a first power, and (b) a second rate of change proportional to the ratio between the first temperature and the first power; determining a brake wear metric corresponding to the brake pads based on the execution result of the neural network; and causing the brake pad metric to be presented via a user interface. Attached Figure Description

[0006] Figure 1 It is a block diagram of an example environment in which an example vehicle analysis circuit system and an example model analysis circuit system can be implemented according to the teachings of this disclosure.

[0007] Figure 2 It shows Figure 1 Example brakes for vehicles.

[0008] Figure 3 yes Figure 1 A block diagram of an example implementation of a vehicle analysis circuit system.

[0009] Figure 4 It shows that it can be made by Figure 3 Example data matrix generated by the vehicle analysis circuit system.

[0010] Figure 5 yes Figure 1 A block diagram of an example implementation of a model analysis circuit system.

[0011] Figure 6 It means that it can be generated by Figure 5 The flowchart illustrates the process of implementing a model analysis circuit system to generate and / or train one or more brake wear prediction models, as shown in the example model training architecture.

[0012] Figure 7 It means that it can be generated by Figure 1 and / or Figure 3 Example vehicle analysis circuit system and Figure 1 and / or Figure 5 The example model analyzes the process flowchart of the example information exchange program implemented in the circuit system.

[0013] Figure 8 It means that it can be generated by Figure 1 and / or Figure 3 Vehicle analysis circuitry and / or Figure 1 and / or Figure 5The flowchart shows the process of an example brake wear prediction program implemented in the model analysis circuit system.

[0014] Figure 9 This is a flowchart illustrating example machine-readable instructions and / or example operations that can be executed, instantiated, and / or implemented by an example programmable circuit system. Figure 3 The vehicle analysis circuit system.

[0015] Figure 10 This is a flowchart illustrating example machine-readable instructions and / or example operations that can be executed, instantiated, and / or implemented by an example programmable circuit system. Figure 5 The model analysis circuit system is used to estimate brake pad wear.

[0016] Figure 11 This is a flowchart illustrating example machine-readable instructions and / or example operations that can be executed, instantiated, and / or implemented by an example programmable circuit system. Figure 5 The model analysis circuit system is used to generate and / or train one or more brake wear prediction models.

[0017] Figure 12 This is a block diagram of an example processing platform including a programmable circuit system structured to execute, instantiate, and / or implement example machine-readable instructions and / or perform... Figure 9 Example operations to implement Figure 3 The vehicle analysis circuit system.

[0018] Figure 13 This is a block diagram of an example processing platform including a programmable circuit system structured to execute, instantiate, and / or implement example machine-readable instructions and / or perform... Figure 10 and / or Figure 11 Example operations to implement Figure 5 The model analysis circuit system.

[0019] Generally, the same reference numerals will be used throughout the accompanying drawings and written description to refer to the same or similar parts. The drawings are not necessarily drawn to scale. Instead, the thickness of layers or areas may be magnified in the drawings. Although the drawings show layers and areas with simple lines and boundaries, some or all of these lines and / or boundaries may be idealized. In reality, boundaries and / or lines may be unobservable, mixed, and / or irregular. Detailed Implementation

[0020] Vehicle brakes typically include brake pads mounted on the brake calipers (e.g., brake caliper assemblies). In some examples, the brake pads consist of a rigid backing plate and friction pads attached thereto. In some examples, when the brakes are engaged, the brake calipers actuate toward the brake disc, causing the brake pads to contact the brake disc. The friction between the brake pads and the brake disc converts the kinetic energy of the rotating brake disc into heat energy, thereby slowing the vehicle's rotation.

[0021] During their service life, brake pads gradually wear down due to contact with the brake disc. When the brake pad thickness falls below a threshold (e.g., less than 2 mm), the brake pads may need to be replaced. Because brake pad wear varies based on vehicle usage (e.g., frequency of vehicle deceleration, intensity of brake application during braking events, etc.) and / or vehicle load (e.g., heavier vehicle load or greater brake pad wear, etc.), it can be difficult to estimate brake pad wear without visual inspection. Furthermore, while specific sensors (e.g., brake wear sensors) can be implemented on the vehicle to detect brake pad wear, adding such sensors may increase the complexity and / or weight associated with the vehicle.

[0022] Figure 1 This is a block diagram of an example environment 100 in which example vehicle analysis circuit system 102 and example model analysis circuit system 104 can be implemented according to the teachings of this disclosure. Figure 1 In the example shown, example vehicle 106 implements vehicle analysis circuitry system 102. Vehicle 106 is a motorized wheeled vehicle that includes a first wheel (e.g., left front wheel) 108A, a second wheel (e.g., right front wheel) 108B, a third wheel (e.g., left rear wheel) 108C, and a fourth wheel (e.g., right rear wheel) 108D (collectively referred to herein as wheels 108), wherein the first wheel 108A and the second wheel 108B correspond to the front wheels of vehicle 106, and the third wheel 108C and the fourth wheel 108D correspond to the rear wheels of the vehicle. Figure 1 In one example, vehicle 106 is a pickup truck. In other examples, vehicle 106 can be any type of vehicle with brakes (e.g., sedan, coupe, van, pickup truck, SUV, ATV, agricultural equipment, etc.). In some examples, vehicle 106 includes an internal combustion engine (e.g., non-electrified vehicle, partially electrified vehicle, etc.). In other examples, vehicle 106 can be implemented as a fully electric vehicle.

[0023] exist Figure 1In the illustrated example, a first example brake (e.g., a first vehicle brake) 110A is associated with a first wheel 108A, a second example brake (e.g., a second vehicle brake) 110B is associated with a second wheel 108B, a third example brake (e.g., a third vehicle brake) 110C is associated with a third wheel 108C, and a fourth example brake (e.g., a fourth vehicle brake) 110D is associated with a fourth wheel 108D. In this example, the first brake 110A, the second brake 110B, the third brake 110C, and the fourth brake 110D (collectively referred to herein as brake 110) are disc brakes. In some examples, one or more of the brakes 110 may be different types of brakes (e.g., electromechanical brakes (EMB), drum brakes, etc.).

[0024] Turning Figure 2 It shows what can be used for Figure 1 An example disc brake assembly 200 is one or more brakes in brake 110. Figure 2 In the example shown, the disc brake assembly 200 includes an example brake caliper assembly 202 and one or more example brake pads 204, which are relative to (e.g., toward and / or away from) the brake pads. Figure 1 The example brake disc 206 associated with one of the wheels 108 moves. In some examples, the brake caliper assembly 202 may be actuated (e.g., electrically and / or mechanically) to move the brake pads 204 toward the brake disc 206 to apply clamping pressure to the brake disc 206. In this example, the brake disc 206 is coupled (e.g., rigidly coupled) to... Figure 1 A disc that rotates with a corresponding wheel in wheel 108. The brake disc 206 may have any suitable shape (e.g., circular, etc.) and / or any suitable features (e.g., slots, holes, etc.).

[0025] During operation of the disc brake assembly 200, the brake caliper assembly 202 applies forces (e.g., frictional forces, clamping forces) to the brake disc 206 via brake pads 204 to slow the rotation of the brake disc 206 and the corresponding wheel 108, thereby reducing the travel speed of the vehicle 106. The brake pads 204 are designed to abrade (e.g., wear) upon contact with the brake disc 206, and thus reduce (e.g., limit, prevent) deformation and / or warping of the brake disc 206 and / or the brake caliper assembly 202. During the service life of the vehicle 106, the brake pads 204 may need to be replaced to maintain the braking efficiency of the disc brake assembly 200.

[0026] Return to Figure 1The vehicle analysis circuitry system 102 accesses and / or obtains example sensor data from one or more example sensors 112 of the vehicle 106. In some examples, the sensor 112 may include one or more example torque sensors, one or more example temperature sensors (e.g., ambient temperature sensors), and / or one or more example speed sensors implemented on and / or operatively coupled to the vehicle 106. In some examples, the torque sensor measures and / or detects the torque (e.g., braking torque) applied to the respective wheel of the wheel 108 of the vehicle 106. For example, torque corresponds to the force (e.g., in Newton-meters (Nm)) applied by the brakes in the brake 110 to the respective wheel 108 (e.g., on the brake disc of wheel 108). In some examples, the temperature sensor measures and / or detects the ambient temperature of the vehicle 106 (e.g., in degrees Celsius (°C)) (e.g., corresponding to the environment around the vehicle 106), and the speed sensor measures the speed of the vehicle 106 (e.g., travel speed) (e.g., in kilometers per hour (kph)). In some examples, the vehicle analysis circuitry system 102 acquires sensor data via the Controller Area Network (CAN) bus of the vehicle 106. In some examples, the vehicle analysis circuitry system 102 acquires sensor data periodically (e.g., every second, every 10 seconds, every 20 seconds, etc.).

[0027] In some examples, the vehicle analysis circuitry system 102 calculates example power data and / or example temperature data associated with the respective wheels in the wheel 108 based on sensor data. For example, when the brake 110 engages (e.g., contacts the brake disc of the respective wheel in the wheel 108), the brake 110 applies power to the respective wheel 108. Furthermore, heat may be generated due to friction between the brake pads of the brake 110 and the rotating brake disc, resulting in an increase in temperature at the respective wheel 108. In some examples, the power data represents the power applied to the respective wheel in the wheel 108 (e.g., front wheels 108A, 108B and / or rear wheels 108C, 108D), and / or represents the total (e.g., combined) power applied to the wheel 108 by the respective brake 110. In some examples, the temperature data represents the temperature at the interface between the brake pads of the brake 110 and the respective brake disc of the wheel 108.

[0028] exist Figure 1In the illustrated example, vehicle analysis circuitry 102 (e.g., via example network 114) is communicatively coupled to model analysis circuitry 104 to provide (e.g., send, transmit) calculated power and / or temperature data to model analysis circuitry 104 for estimating brake pad wear. In some examples, vehicle analysis circuitry 102 generates an example data matrix (e.g., a histogram matrix) based on the power and / or temperature data and provides the power and temperature data to model analysis circuitry 104 by transmitting the data matrix via network 114. In some of these examples, by utilizing the data matrix to store and / or transmit data, the disclosed examples can reduce the utilization of computational resources (e.g., memory, bandwidth) associated with the transmission and / or storage of temperature and / or power data.

[0029] exist Figure 1 In the illustrated example, model analysis circuitry 104 generates, trains, and / or executes one or more example brake wear prediction models to estimate example brake wear metrics associated with the corresponding brake 110 of vehicle 106. In this example, the brake wear prediction model is based on a Physical Information Neural Network (PINN). Model analysis circuitry 104 may generate and / or train the brake wear prediction model based on example training data 116 accessible to and / or preloaded in model analysis circuitry 104 (e.g., via network 114). In some examples, model analysis circuitry 104 accesses a data matrix from vehicle analysis circuitry 102, obtains and / or extracts temperature and dynamic data from the data matrix, and executes the brake wear prediction model based on the extracted data. Due to the execution of the brake wear prediction model, model analysis circuitry 104 determines and / or outputs the brake wear metric for the corresponding brake (e.g., and / or associated brake pad) of brake 110. Brake wear metrics may include at least one of the following: the thickness of the brake pad corresponding to brake 110, the mass of the brake pad, or a change in thickness or mass (e.g., relative to initial thickness and / or initial mass). In some examples, model analysis circuitry 104 may estimate an example remaining service life (RUL) of the corresponding brake 110 based on brake wear metrics. RUL may represent an estimated time until the corresponding brake pad of brake 110 needs to be replaced.

[0030] In some examples, the model analysis circuitry 104 may present brake wear measurements and / or estimated RUL to the operator of vehicle 106 via an example user interface 118 of vehicle 106. In some examples, the user interface 118 corresponds to a human-machine interface (HMI), display, etc., of vehicle 106. In some examples, the model analysis circuitry 104 may provide brake wear measurements and / or estimated RUL to one or more remote devices (e.g., mobile devices, devices separate from vehicle 106, etc.) via network 114 for presentation thereon. For example, brake wear measurements may be presented to the manufacturer and / or vehicle service provider via a remote device to notify and / or assist in maintenance activities of vehicle 106. In some examples, brake wear measurements may be used to notify the operator and / or vehicle service provider when to replace one or more brake pads of vehicle 106. Therefore, the disclosed examples may reduce deterioration and / or warping of one or more components of vehicle 106 (e.g., brake discs) due to delayed and / or postponed replacement of brake pads, and / or may reduce premature replacement of brake pads.

[0031] Figure 3 yes Figure 1 A block diagram of an example implementation of the vehicle analysis circuit system 102. Figure 3 The vehicle analysis circuit system 102 can be instantiated (e.g., instantiated, formed, materialized, implemented, etc.) by a programmable circuit system (such as a central processing unit (CPU) that executes first instructions). Alternatively or concurrently, Figure 3 The vehicle analysis circuit system 102 can be instantiated (e.g., instantiated, formed over any time period, materialized, implemented, etc.) by: (i) an application-specific integrated circuit (ASIC) and / or (ii) a field-programmable gate array (FPGA), said FPGA being structured and / or configured to perform operations corresponding to the first instruction in response to the execution of a second instruction. It should be understood that... Figure 3 Some or all of the circuit system can therefore be instantiated at the same or different times. Figure 3 Some or all of the circuitry can be instantiated, for example, in one or more threads that execute concurrently on hardware and / or serially on hardware. Furthermore, in some examples, Figure 3 Some or all of the circuitry in the system can be implemented by executing instructions through a microprocessor circuitry and / or operating through an FPGA circuitry to implement one or more virtual machines and / or containers.

[0032] exist Figure 3In the example shown, the vehicle analysis circuit system 102 includes an example data interface circuit system 302, an example power calculation circuit system 304, an example temperature calculation circuit system 306, an example matrix control circuit system 308, an example data transmission circuit system 310, and an example vehicle database 312.

[0033] Figure 3 The vehicle database 312 stores data used and / or obtained by the vehicle analysis circuit system 102. Figure 3 The example vehicle database 312 is implemented using any memory, storage device, and / or disk for storing data (such as flash memory, magnetic media, optical media, solid-state memory, hard disk drive, thumb drive, etc.). Furthermore, the data stored in the vehicle database 312 can be in any data format, such as binary data, comma-separated data, tab-separated data, Structured Query Language (SQL) structures, etc. Although the vehicle database 312 is shown as a single device in the example, the vehicle database 312 and / or any other data storage device described herein can be implemented using any number and / or type of memory.

[0034] Figure 3 The data interface circuitry 302 accesses, retrieves, and / or otherwise obtains sample data (e.g., sensor data) that will be utilized by the vehicle analysis circuitry 102. For example, the data interface circuitry 302 can access, retrieve, and / or otherwise obtain sample data (e.g., sensor data) that will be utilized by the vehicle analysis circuitry 102. Figure 1 Sensor 112 acquires example ambient temperature data 318, example torque data 320, and / or example speed data 322. In Figure 3 In the example, the ambient temperature data 318 represents Figure 1The data interface circuitry 302 acquires sensor data (e.g., ambient temperature data 318, torque data 320, and / or speed data 322) in the environment of vehicle 106. In some examples, torque data 320 represents the torque applied to the corresponding wheel of wheel 108 (e.g., Newton-meters). In some examples, speed data 322 represents the speed of vehicle 106 (e.g., travel speed) (e.g., km / h). In some examples, the data interface circuitry 302 acquires sensor data (e.g., ambient temperature data 318, torque data 320, and / or speed data 322) at a specific frequency (e.g., every 1 second, every 2 seconds, etc.), wherein the frequency may be preset in vehicle analysis circuitry 102 and / or may be selected and / or adjusted by the operator of vehicle 106. In some examples, the data interface circuitry 302 acquires sensor data during operation and / or travel of vehicle 106. In some examples, the data interface circuitry 302 provides sensor data to vehicle database 312, wherein the sensor data may be stored in association with the time of acquisition and / or acquisition of sensor data. In some examples, the data interface circuitry 302 is instantiated by and / or configured to operate by a programmable circuitry that executes data interface circuitry instructions, such as by... Figure 9 The flowchart represents those operations.

[0035] Figure 3 The power calculation circuit system 304 calculates example power data associated with the corresponding wheels 108 and / or brakes 110 of the vehicle 106 based on sensor data. For example, the power data may represent the power (e.g., in watts (W)) applied by the brakes 110 to one or more of the wheels 108. In some examples, the power data may include the total (e.g., combined) power (e.g., P) applied to the wheels 108, and / or may include multiple power values ​​corresponding to individual wheels of the wheels 108 (e.g., front wheels 108A, 108B and / or rear wheels 108C, 108D). In some examples, the power calculation circuit system 304 calculates the total power (e.g., P) based on Example Equation 1 below.

[0036]

[0037] In Example Equation 1 above, P represents the total power applied to wheel 108 at a specific time (e.g., in watts), τ represents the torque applied to wheel 108 (e.g., in Newton-meters (Nm), V represents the speed of vehicle 106 (e.g., in kilometers per hour (kph), c kph_to_mps The conversion factor is given from kilometers per hour (kph) to meters per second (mps), and r represents the tire radius of wheel 108 (e.g., in meters (m)).

[0038] In some examples, the power calculation circuit system 304 calculates individual power values ​​representing the power applied to the respective wheels in the wheel 108 based on the total power P. For example, the power calculation circuit system 304 calculates individual power values ​​based on: a first example power ratio (e.g., c 向左动力 The first example power ratio represents a first proportion of the total power applied to the left-hand side wheels (e.g., first wheel 108A and third wheel 108C) of vehicle 106; and the second example power ratio (e.g., c 向前动力 The second example power ratio represents a second proportion of the total power applied to the front wheels of vehicle 106 (e.g., first wheel 108A and second wheel 108B). In some examples, the power calculation circuit system 304 calculates the first power (e.g., left front power) applied to the first wheel 108A (e.g., P) based on the following example equations 2A, 2B, 2C, and 2D. 左前 ), and the second power applied to the second wheel 108B (e.g., right front power) (e.g., P) 右前 ), and the third power applied to the third wheel 108C (e.g., the left rear power) (e.g., P) 左后 ) and the fourth power applied to the fourth wheel 108D (e.g., the right rear power) (e.g., P 右后 ).

[0039] P 左前 =P*c 向左动力 *c 向前动力

[0040] (Equation 2A)

[0041] P 右前 =P*(1-c 向左动力 )*c 向前动力

[0042] (Equation 2B)

[0043] P 左后 =P*c 向左动力 *(1-c 向前动力 )

[0044] (Equation 2C)

[0045] P 右后 =P*(1-c 向左动力 )*(1-c 向前动力 )

[0046] (Equation 2D)

[0047] In the example equations 2A, 2B, 2C and / or 2D above, P 左前 Indicates left front drive, P 右前 Indicates right front power, P左后 Indicates left rear power, P 右后 This indicates the right rear power, P indicates the total power, and c indicates the right rear power. 向前动力 This indicates the proportion of the total power applied to the front wheels 108A and 108B, and c 向左动力 This indicates the proportion of total power applied to the left wheels 108A and 108C.

[0048] exist Figure 3 In the example, c 向前动力 =0.5, which means that the power applied to the front wheels 108A and 108B is equal to the power applied to the rear wheels 108C and 108D (e.g., the total power is evenly distributed between the front wheels 108A and 108B and the rear wheels 108C and 108D). Similarly, in this example, c 向左动力 =0.5, which means that the power applied to the left-hand wheels 108A and 108C is equal to the power applied to the right-hand wheels 108B and 108D (e.g., the total power is evenly distributed between the left-hand wheels 108A and 108C and the right-hand wheels 108B and 108D). In some examples, the first ratio (e.g., c) 向前动力 ) and / or a second proportion (e.g., c) 向左动力 The ratios can be different (e.g., at least 0, at most 1, and including 1). For example, the first ratio and / or the second ratio can be adjusted by the operator of vehicle 106 (e.g., via a...). Figure 1 The user interface 118 (user input) and / or may vary based on the model and / or type of vehicle 106. In some examples, after calculating power data (e.g., the power applied to the corresponding wheels in wheel 108), the power calculation circuit system 304 provides the power data to the matrix control circuit system 308 for generating and / or updating the data matrix. In some examples, the power calculation circuit system 304 is instantiated by a programmable circuit system that executes power calculation circuit system instructions and / or is configured to perform operations, such as by... Figure 9 The flowchart represents those operations.

[0049] Figure 3 The temperature calculation circuit system 306 calculates example temperature data associated with the corresponding wheel in wheel 108 based on sensor data and / or calculated power values ​​from the power calculation circuit system 304. For example, the temperature data may represent the temperature of the corresponding wheel 108 (e.g., at or near the interface between the brake 110 and the corresponding brake disc of the corresponding wheel 108). In some examples, the temperature calculation circuit system 306 calculates the front wheel temperature (e.g., T) of the front wheels 108A, 108B based on the example equation 3A below. 前 ).

[0050]

[0051] In the example equation 3A above, T 前 P represents the current temperature of the front wheels (e.g., in degrees Celsius). 前 The power applied to the front wheels, as determined by the power calculation circuit system 304 (e.g., in watts), is represented by m (e.g., in kilograms), and C represents the mass of the brake pads associated with the front wheels. 前 The specific heat capacity (e.g., in joules per kilogram per degree Celsius (J / kg℃)) of the material corresponding to the brake disc associated with the front wheels, cF 前 T represents the convective heat transfer coefficient (e.g., in watts per Kelvin (W / K)) corresponding to the material of the brake disc. 环境 Indicates ambient temperature (e.g., according to data 318), T 前，prev The previous front wheel temperature represents the temperature at a previous time (e.g., before the current time), and dt represents the difference (e.g., duration) between the previous and current times. In some examples, the ambient temperature is initially selected as the previous front wheel temperature (e.g., when the previous front wheel temperature is unavailable and / or unknown, when vehicle 106 is started, etc.). In this example, the front wheel brake discs are made of steel, such that the specific heat capacity (e.g., C) is... 前 The heat transfer coefficient is 420 J / kg℃, and the heat transfer coefficient (e.g., cF) is 420 J / kg℃. 前 The specific heat capacity and / or heat transfer coefficient are between 0.001 W / K and 0.005 W / K. In some examples, different values ​​may be used for the specific heat capacity and / or heat transfer coefficient (e.g., when different materials are used for the brake disc).

[0052] In some examples, the temperature calculation circuit system 306 can also be based on the following examples, etc.

[0053] Formula 3B is used to calculate the rear wheel temperature (e.g., T) of the rear wheels in rear wheels 108C and 108D. 后 ).

[0054]

[0055] Equation 3B above is similar to Equation 3A above, except that the subscripts of the variables in Equation 3B have been changed (e.g., relative to Equation 3A above) to correspond to the rear wheels 108C and 108D. Therefore, the description of the variables described in Equation 3A above regarding the rear wheels 108C and 108D can be equally applied to the variables shown in Equation 3B above. In some examples, after calculating the temperature data of the corresponding wheels in wheel 108 (e.g., front wheel temperature and / or rear wheel temperature), the temperature calculation circuit system 306 provides the temperature data to the matrix control circuit system 308 for generating and / or updating the data matrix. In some examples, the temperature calculation circuit system 306 is instantiated by and / or configured to operate by a programmable circuit system that executes temperature calculation circuit system instructions, such as by... Figure 9 The flowchart represents those operations.

[0056] The matrix control circuit system 308 can generate an example data matrix (e.g., a histogram matrix) 324 based on power data and temperature data for efficient storage and / or transmission of the power data and temperature data. For example, Figure 4 An example data matrix 324 that can be generated by the matrix control circuit system 308 is shown. To generate... Figure 4 The data matrix 324 and matrix control circuit system 308 select and / or define the example temperature range 402 and example power range 404. Temperature range 402 corresponds to a range of different temperature values ​​that can be represented in the temperature data. Figure 4 In this configuration, the matrix control circuit system 308 defines twelve temperature ranges 402, wherein eleven of the temperature ranges 402 represent a range from 0 degrees (e.g., degrees Celsius) up to (but not including) 300 degrees, and the twelfth temperature range 402 represents a range of 300 degrees and above. In other words, the temperature ranges 402 include a first temperature range corresponding to a first example temperature value range (e.g., from 0 degrees to but not including 25 degrees), a second temperature range corresponding to a second example temperature value range (e.g., from 25 degrees to but not including 50 degrees), and so on.

[0057] Furthermore, power range 404 corresponds to a range of different power values ​​that can be represented in the power data. Figure 4In this configuration, the matrix control circuit system 308 defines nine power ranges 404, eight of which represent a range from 0 watts (W) up to (but not including) 4500W, and the ninth power range 404 represents a range of 4500W and above. In other words, the power ranges 404 include a first power range corresponding to a first example power value range (e.g., from 0W up to but not including 500W), a second power range corresponding to a second example power value range (e.g., from 500W up to but not including 1000W), and so on. Although in Figure 4 The diagram shows twelve temperature ranges 402 and nine power ranges 404, but different numbers of temperature ranges 402 and / or power ranges 404 can be used alternatively. Additionally, Figure 4 One or more of the example ranges of the corresponding temperature range 402 and power range 404 shown may differ in some examples. In some examples, the number and / or associated ranges of ranges 402, 404 may be pre-loaded in the matrix control circuit system 308, and / or may be based on user input (e.g., via...). Figure 1 The user interface 118 provides options and / or adjustments for selection.

[0058] exist Figure 4 In the example shown, the matrix control circuit system 308 generates and / or initializes the data matrix 324 based on the temperature range 402 and the power range 404. For example, the matrix control circuit system 308 selects the size of the data matrix 324 (e.g., row count and column count) based on the range counts of the temperature range 402 and the power range 404. Figure 4 In the example, the matrix control circuit system 308 generates a data matrix 324 with twelve rows 406 (e.g., corresponding to temperature range 402) and nine columns 408 (e.g., corresponding to power range 404). Although in this example, row 406 corresponds to temperature range 402 and column 408 corresponds to power range 404, in some examples, row 406 and column 408 can be reversed (e.g., such that row 406 corresponds to power range 404 and column 408 corresponds to temperature range 404).

[0059] exist Figure 4 In the example, the example temperature index is used to represent the corresponding row 406 of data matrix 324, and the temperature index also corresponds to the corresponding temperature interval in temperature interval 402. Similarly, the example dynamic index is used to represent the corresponding column 408 of data matrix 324, and the dynamic index also corresponds to the corresponding dynamic interval in dynamic interval 404. In some examples, the matrix values ​​(e.g., elements) of data matrix 324 may be represented by corresponding temperature and dynamic indices. For example, the first matrix value (e.g., in...) Figure 4The matrix (represented as V21) corresponds to a temperature index of 2 (e.g., also corresponding to a temperature range of 50 to 75) and a dynamic index of 0 (e.g., also corresponding to a dynamic range of 0 to 500). In some examples, the matrix values ​​represent the corresponding number (e.g., count) of data samples corresponding to the respective temperature and dynamic ranges. In this example, the data samples correspond to sensor data captured and / or acquired at a given time (e.g., Figure 3 Ambient temperature data 318, torque data 320 and / or speed data 322).

[0060] exist Figure 4 In the example shown, the matrix control circuit system 308 is based on... Figure 1 and / or Figure 3 The vehicle analysis circuit system 102 updates the matrix values ​​of the data matrix 324 based on one or more data samples obtained. For example, during the operation and / or travel of the vehicle 106, the vehicle analysis circuit system 102 may collect and / or obtain data samples at a given frequency (e.g., every 1 second, every 2 seconds, etc.). In some examples, after calculating temperature and power values ​​for a given data sample (e.g., by temperature calculation circuit system 306 and power calculation circuit system 304, respectively), the matrix control circuit system 308 selects a temperature range corresponding to the calculated temperature value in temperature range 402 and a power range corresponding to the calculated power value in power range 404. For example, when the data sample corresponds to a temperature of 55 degrees and a power of 560W, the matrix control circuit system 308 determines that the data sample corresponds to a third temperature range in temperature range 402 (e.g., associated with a temperature range of 50 degrees to 75 degrees) and also corresponds to a second power range in power range 404 (e.g., associated with a power range of 500W to 1000W). In such an example, the matrix control circuit system 308 selects a temperature index and a power index (e.g., temperature index 2 and power index 1) corresponding to the selected temperature range 402 and power range 404, and then increments (e.g., increments by 1) a matrix value corresponding to the selected temperature index and power index.

[0061] In some examples, the matrix control circuitry 308 updates the data matrix 324 based on additional data samples collected and / or acquired during the operation of the vehicle 106 (e.g., by continuously and / or periodically incrementing the matrix values). For example, the matrix control circuitry 308 may monitor and / or update the data matrix 324 over the duration of the vehicle 106's journey and / or operation. In some examples, the matrix control circuitry 308 may monitor and / or update the data matrix 324 over a pre-selected duration (e.g., based on user input selection). In some examples, the matrix control circuitry 308 may generate and / or update multiple data matrices in the data matrix 324, where the multiple data matrices 324 may correspond to individual wheels among the wheels 108, such as front wheels 108A, 108B and rear wheels 108C, 108D, etc. In some examples, the matrix control circuitry 308 may provide the data matrix 324 (and / or multiple data matrices) to the vehicle database 312 for storage. In some examples, by storing data matrix 324 (e.g., instead of individual temperature and dynamic values ​​for corresponding data samples), the disclosed examples can reduce the utilization of computational resources (e.g., memory) for data storage. In some examples, matrix control circuitry 308 is instantiated by and / or configured to operate by a programmable circuitry that executes matrix control circuitry instructions, such as by... Figure 9 The flowchart represents those operations.

[0062] Return to Figure 3 The data transmission circuit system 310 sends, transmits, and / or otherwise provides data from the vehicle analysis circuit system 102 to... Figure 1 The model analysis circuit system 104. For example, the data transmission circuit system 310 can be transmitted via... Figure 1Network 114 transmits data matrix 324 (e.g., generated by matrix control circuitry 308) to model analysis circuitry 104. In some examples, data transmission circuitry 310 sends and / or transmits example metadata associated with data matrix 324, where the metadata may include the time when data matrix 324 was generated and / or the time when sensor data (e.g., ambient temperature data 318, torque data 320, and / or speed data 322) was acquired. In some examples, data transmission circuitry 310 transmits data matrix 324 (and / or associated metadata) periodically (e.g., once a day, etc.) and / or after vehicle 106 has completed a trip or operation. In some examples, by transmitting calculated temperature and power values ​​(e.g., temperature data and power data) on network 114 using data matrix 324, the disclosed examples reduce bandwidth utilization (e.g., compared to transmitting temperature and power data directly via network 114). In some examples, data transmission circuitry 310 is instantiated by and / or configured to perform operations such as by a programmable circuitry executing data transmission circuitry instructions. Figure 9 The flowchart represents those operations.

[0063] Figure 5 yes Figure 1 A block diagram of an example implementation of the model analysis circuit system 104. Figure 5 The model analysis circuit system 104 can be instantiated (e.g., instantiated, formed, materialized, implemented, etc.) by a programmable circuit system (such as a central processing unit (CPU) that executes first instructions). Alternatively or concurrently, Figure 5 The model analysis circuit system 104 can be instantiated (e.g., instantiated, formed over any time period, materialized, implemented, etc.) by: (i) an application-specific integrated circuit (ASIC) and / or (ii) a field-programmable gate array (FPGA), said FPGA being structured and / or configured to perform operations corresponding to the first instruction in response to the execution of a second instruction. It should be understood that... Figure 5 Some or all of the circuit system can therefore be instantiated at the same or different times. Figure 5 Some or all of the circuitry can be instantiated, for example, in one or more threads that execute concurrently on hardware and / or serially on hardware. Furthermore, in some examples, Figure 5 Some or all of the circuitry in the system can be implemented by executing instructions through a microprocessor circuitry and / or operating through an FPGA circuitry to implement one or more virtual machines and / or containers.

[0064] exist Figure 5In the example shown, the model analysis circuit system 104 includes an example input interface circuit system 502, an example data processing circuit system 504, an example model training circuit system 506, an example model execution circuit system 508, an example output circuit system 510, an example metric calculation circuit system 512, and an example cloud database 514.

[0065] Figure 5 The cloud database 514 stores data used and / or obtained by the model analysis circuit system 104. Figure 5 The example cloud database 514 is implemented using any memory, storage device, and / or storage disk (such as flash memory, magnetic media, optical media, solid-state storage, hard disk drive, thumb drive, etc.) for storing data. Furthermore, the data stored in the cloud database 514 can be in any data format, such as binary data, comma-separated data, tab-separated data, Structured Query Language (SQL) structures, etc. Although the cloud database 514 is shown as a single device in the example, the cloud database 514 and / or any other data storage device described herein can be implemented using any number and / or type of memory.

[0066] Figure 5 The input interface circuitry system 502 accesses, retrieves, and / or otherwise obtains data that will be utilized by the model analysis circuitry system 104. For example, the input interface circuitry system 502 via... Figure 1 Network 114 obtained by Figure 1 and / or Figure 3 The vehicle analysis circuit system 102 generates a data matrix 324. Additionally, the input interface circuit system 502 can obtain training data 116 via a network 114. In some examples, the training data 116 can be pre-loaded into the model analysis circuit system 104. In some examples, the input interface circuit system 502 provides the data matrix 324 and / or the training data 116 to a cloud database 514 for storage. In some examples, the input interface circuit system 502 is instantiated by a programmable circuit system that executes input interface circuit system instructions and / or configured to perform operations, such as those by... Figure 10 and / or Figure 11 The flowchart represents those operations.

[0067] Figure 5 The data processing circuit system 504 processes the data obtained from the input interface circuit system 502 (e.g., data matrix 324 and / or training data 116). For example, the data processing circuit system 504 can extract and / or determine data from the data matrix 324 related to... Figure 1The data processing circuitry 504 associates power and temperature values ​​(and / or ranges of power and temperature values) with the corresponding wheels 108 of vehicle 106. In some examples, the data processing circuitry 504 may preprocess training data 116 by, for example, removing duplicate data samples from training data 116. In some examples, the data processing circuitry 504 selects and / or generates juxtaposition points for training and / or evaluating one or more example brake wear prediction models (e.g., Physical Information Neural Network (PINN) models). For example, the data processing circuitry 504 selects juxtaposition points from an input space (e.g., a multidimensional space, a sample space) defined based on an input temperature range, an input power range, and an input time range. In such examples, the input temperature range represents a first range of expected (e.g., possible) temperature values ​​for the brake pads of brake 110, the input power range represents a second range of expected (e.g., possible) power values ​​that can be applied by the brake pads to the corresponding brake discs of wheel 108, and the input time range represents the expected (e.g., possible) time for which brake 110 is expected to operate (e.g., the duration for which brake 110 is applied or operated to decelerate vehicle 106). In some examples, the input temperature range, input power range, and input time range are selected based on user input (e.g., defining minimum and / or maximum values ​​for the respective ranges) and / or may be preloaded in the model analysis circuitry system 104 (e.g., in cloud database 514). In this example, the number of juxtaposition points corresponds to (e.g., equal to) the number of data samples represented in training data 116. In some examples, a different number of juxtaposition points may be used instead (e.g., greater than or less than the number of data samples represented in training data 116).

[0068] In some examples, for a given juxtaposition point, the data processing circuitry 504 selects a temperature value from an input temperature range, a power value from an input power range, and a time value from an input time range. In some examples, the data processing circuitry 504 selects juxtaposition points from the input space based on the Latin American hypercube sampling method. In some examples, different sampling methods (e.g., random sampling, full factorial sampling, Sobol sampling, etc.) may be used alternatively. In some examples, the data processing circuitry 504 provides the juxtaposition points to a cloud database 514 for storage. In some examples, the data processing circuitry 504 is instantiated by and / or configured to perform operations, such as by a programmable circuitry system executing data processing circuitry system instructions. Figure 10 The flowchart represents those operations.

[0069] Figure 5The model training circuit system 506 generates, trains, and / or retrains one or more example brake wear prediction models (e.g., Physical Information Neural Network (PINN) models) to predict brake pad wear (e.g., brake wear metrics) of the corresponding brake 110 of vehicle 106. For example, the model training circuit system 506 may generate and / or train brake wear prediction models based on training data 116.

[0070] Artificial intelligence (AI), including machine learning (ML), deep learning (DL), and / or other artificial machine-driven logic, enables machines (e.g., computers, logic circuits, etc.) to use models to process input data to generate outputs based on patterns and / or associations previously learned by the model through a training process. For example, a model can be trained with data to recognize patterns and / or associations, and when processing input data, it follows such patterns and / or associations so that other inputs produce outputs consistent with the recognized patterns and / or associations.

[0071] There are many different types of machine learning models and / or machine learning architectures. In the examples disclosed herein, a Physics Information Neural Network (PINN) model is used. The PINN model supports training based on available training data (e.g., training data 116) and further based on the known and / or expected physical behavior (e.g., dynamic behavior) of the underlying system. In some examples, because the PINN model is trained based on expected dynamic behavior (e.g., using one or more equations) in addition to the labeled training data, the PINN model can be used with inputs and / or outputs not represented in training data 116 (e.g., falling outside the range of inputs and / or outputs represented in training data 116). In other words, while training data 116 may represent a specific range of temperature and dynamic values, the PINN model can be trained to accurately predict brake wear metrics for temperature and dynamic values ​​falling outside (e.g., greater than or less than) that specific range. Therefore, compared to other machine learning models, the PINN model can be trained with less training data. The following is combined with… Figure 6 Further describe the training and / or generation of the PINN model.

[0072] Typically, a neural network is the machine learning model / architecture suitable for use in the example methods disclosed herein. However, other types of machine learning models may be used alternatively or in addition. Generally, implementing an ML / AI system involves two phases: a learning / training phase and an inference phase. In the learning / training phase, training algorithms are used to train the model based on, for example, training data (e.g., training data). Figure 5The training data (116) is used to operate based on patterns and / or associations. Typically, a model includes intrinsic parameters that guide how the input data is transformed into output data, such as transforming the input data into output data through a series of nodes and connections within the model. Additionally, hyperparameters are used as part of the training process to control how learning is performed (e.g., learning rate, number of layers to use in the machine learning model, etc.). Hyperparameters are defined as training parameters determined before the training process is initiated.

[0073] Different types of training can be implemented based on the type of ML / AI model and / or the expected output. For example, supervised training uses inputs and corresponding expected (e.g., labeled) outputs to select parameters for the ML / AI model that reduce model error (e.g., by iterating over selected combinations of parameters). As used herein, "label" refers to the expected output of the machine learning model (e.g., classification, expected output value, etc.). Alternatively, unsupervised training (e.g., for deep learning, subsets of machine learning, etc.) involves inferring patterns from inputs to select parameters for the ML / AI model (e.g., without the benefit of expected (e.g., labeled) outputs).

[0074] In the examples disclosed herein, stochastic gradient descent is used to train the ML / AI model. However, any other training algorithm may be used alternatively or in addition. In the examples disclosed herein, training is performed until an acceptable amount of error is achieved (e.g., one or more loss values ​​associated with the model's output meet a threshold). In the examples disclosed herein, training is performed at model analysis circuitry system 104, which may be implemented in a cloud-based environment. Training is performed using hyperparameters that control how learning is performed (e.g., learning rate, number of layers to be used in the machine learning model, etc.). In some examples, retraining may be performed. Such retraining may be performed in response to the availability of additional training data. For example, additional training data may increase the prediction error associated with the ML / AI model and may therefore trigger retraining of the ML / AI model.

[0075] Training is performed using training data 116. In the examples disclosed herein, training data 116 is derived from simulation data and / or test data. For example, training data 116 may be obtained based on computer simulation data based on finite element analysis of brake pads subjected to varying conditions (e.g., varying applied force, varying temperature, etc.). Alternatively, training data 116 may be obtained based on dynamic testing of one or more brakes, during which the brakes are subjected to varying braking conditions (e.g., applied force, temperature, etc.), and the resulting brake wear metrics (e.g., brake pad mass and / or width) are measured at selected intervals. In some examples, training data 116 may be obtained based on vehicle testing of a vehicle and measuring vehicle-related brake wear metrics at selected intervals. Because supervised training is used, training data 116 is labeled. For example, data samples represented in training data 116 (e.g., including temperature values, force values, and / or time) may be labeled with indications of corresponding brake wear metrics (e.g., corresponding brake pad mass and / or width). Labels are applied to the training data manually (e.g., by one or more operators) and / or automatically (e.g., by the model training circuit system 506). In some examples, the training data is subdivided into a training dataset and a validation dataset.

[0076] Once training is complete, the model is deployed as an executable construct that processes inputs and provides outputs based on the network of nodes and connections defined in the model. In some examples, the model is stored in a cloud-based storage environment (e.g., in...). Figure 5 (The cloud database is located at 514). Then, the model can be executed by the model execution circuit system 508 to predict brake pad wear.

[0077] Once trained, the deployed model can be manipulated to process data during the inference phase. In the inference phase, the data to be analyzed (e.g., real-time data) is fed into the model, and the model is executed to create output. This inference phase can be thought of as the AI ​​"thinking" to generate output based on what it has learned from training (e.g., by executing the model to apply learned patterns and / or correlations to real-time data). In some examples, the input data undergoes preprocessing before being used as input to the machine learning model. Furthermore, in some examples, the output data may undergo post-processing after being generated by the AI ​​model to transform the output into a useful result (e.g., a display of data, instructions to be executed by the machine, etc.).

[0078] In some examples, the output of the deployed model can be captured and provided as feedback. By analyzing the feedback, the accuracy of the deployed model can be determined. If the feedback indicates that the accuracy of the deployed model is less than a threshold or other criterion, the feedback, along with an updated training dataset, hyperparameters, etc., can be used to trigger the training of an updated model to generate an updated deployment model.

[0079] Figure 6 This is a flowchart illustrating the process of generating and / or training an example model training architecture (e.g., the PINN model architecture) 600, which can be implemented by the model analysis circuit system 506 to generate and / or train a brake wear prediction model. Figure 6 In the example shown, the model training circuit system 506 initializes the example neural network 602 to be used for the brake wear prediction model (e.g., selects initial weights for it). In this example, neural network 602 is a feedforward neural network (FNN) with example nodes 604 arranged in multiple layers. Figure 6 In the diagram, a portion of the neural network 602 is shown as having two layers and four nodes 604 in each layer. In some examples, different numbers of layers and / or nodes 604 may be used instead (e.g., four layers with fifty nodes 604 in each layer, etc.).

[0080] exist Figure 6 In the example shown, the model training circuit system 506 provides a portion of the training data 116 as example input 606 to the neural network 602. For example, input 606 may include one or more example temperature values ​​(e.g., T) 608, corresponding dynamic values ​​(e.g., P) 610, and corresponding times (e.g., t) 612 represented in the training data 116 (e.g., the time at which the temperature value 608 and / or dynamic value 610 are measured and / or obtained (e.g., relative to the start time)). In some examples, the temperature value 608, dynamic value 610, and / or time 612 may be represented in vector and / or matrix format.

[0081] In some examples, the model training circuit system 506 executes the neural network 602 based on the input 606. As a result of this execution, the model training circuit system 506 determines one or more example prediction quality values ​​corresponding to the input 606 (e.g., )616. For example, the predicted mass value 616 represents the predicted mass of the brake pad at corresponding time 612. In Figure 6In this model training circuit system 506, partial differentiation (e.g., automatic differentiation) can be performed to evaluate the partial derivative (e.g., gradient) of the predicted quality value 616 with respect to time. For example, the model training circuit system 506 performs automatic differentiation to determine and / or estimate the partial derivative 618 (e.g., with respect to time) corresponding to the predicted quality 616. Furthermore, the model training circuit system 506 determines the product of the predicted quality 616 and the identity function 620.

[0082] exist Figure 6 In the illustrated example, model training circuitry 506 determines a first example loss value (e.g., a data-based loss value) 622 and a second example loss value (e.g., a physics-based loss value, a partial differential equation (PDE) loss value) 624 associated with neural network 602. In some examples, the first loss value 622 is based on the difference between the predicted quality value 616 and the corresponding measured and / or simulated quality value (e.g., ground-based quality value) represented in training data 116. In some examples, model training circuitry 506 determines the first loss value 622 based on example equation 4 below.

[0083]

[0084] In example equation 4 above, L represents the first loss value of 622, m i This represents the measurement and / or simulation quality corresponding to the i-th data sample. 616 represents the prediction quality corresponding to the i-th data sample (e.g., prediction based on the input value 606 from the i-th data sample), and N represents the total number of data samples represented in the training data 116 (e.g., the total number).

[0085] exist Figure 6 In the illustrated example, the model training circuitry 506 determines a second loss value 624 based on the difference between: a first result based on an evaluation of the partial derivative 618 at a juxtaposition point (e.g., a juxtaposition point selected and / or generated by the data processing circuitry 504); and a second result based on an evaluation of an example physics-based equation 626 at the juxtaposition point. In some examples, the physics-based equation 626 represents the known and / or expected dynamics of material wear. In this example, the physics-based equation 626 is based on the Archard equation for wear dynamics, where the physics-based equation is a function of temperature and dynamics. In some examples, different physics-based equations may be used alternatively. Figure 6 In the example, the model training circuit system 506 evaluates at the juxtaposition point a physics-based equation 626 corresponding to the example equation 5 below.

[0086]

[0087] In example equation 5 above, The calculated rate of change of mass is represented by P, where P represents the dynamic value (e.g., corresponding to a given juxtaposition point), T represents the temperature value (e.g., corresponding to a given juxtaposition point), and v represents the vehicle speed (e.g., ...). Figure 1 The speed of vehicle 106 is given by K, and K and α are constants corresponding to the material of the brake pads of brake 110. In other words, for a given juxtaposition point, the calculated rate of change of mass is proportional to the ratio between the dynamic value and the temperature value. In some examples, K corresponds to a value of at least 1 and at most 4, and α corresponds to a value of at least 10,000 and at most 30,000.

[0088] exist Figure 6 In the example shown, the model training circuit system 506 evaluates the example equation 5 above at the corresponding juxtaposition point to determine the rate of change in the computational quality corresponding to the respective juxtaposition point (e.g., Furthermore, the model training circuit system 506 evaluates the partial derivatives 618 at the corresponding juxtaposition points to determine the rate of change in the quality of the predictions corresponding to the respective juxtaposition points (e.g., In some examples, the model training circuit system 506 determines a second loss value 624 based on the difference (e.g., average difference) between the calculated quality rate and the corresponding predicted quality rate. Furthermore, the model training circuit system 506 determines an example combined loss value 628 based on the first loss value 622 and the second loss value 624. For example, the model training circuit system 506 may determine the combined loss value 628 based on a combination (e.g., total, sum) of the first loss value 622 and the second loss value 624. Figure 6 In the example, the model training circuit system 506 adjusts the weights of the neural network 602 based on the combined loss value 628.

[0089] In some examples, the model training circuit system 506 continues to train the neural network 602 (e.g., by adjusting the weights of the neural network 602) until the combined loss value 628 determined based on the execution of the neural network 602 meets an example threshold (e.g., an error threshold). For example, when the combined loss value 628 does not meet (e.g., is greater than) the error threshold, the model training circuit system 506 adjusts the weights of the neural network 602 and re-executes the updated neural network 602 with the adjusted weights. Conversely, when the combined loss value 628 meets (e.g., is less than or equal to) the error threshold, the model training circuit system 506 generates a brake wear prediction model based on the trained neural network 602, and then provides the brake wear prediction model to the cloud database 514 for storage and / or execution by the model execution circuit system 508. In some examples, the model training circuit system 506 is instantiated by a programmable circuit system that executes model training circuit system instructions and / or configured to perform operations, such as by... Figure 11 The flowchart represents those operations.

[0090] Return to Figure 4 The model execution circuit system 508 executes the brake wear prediction model generated and / or trained by the model training circuit system 506. For example, the model execution circuit system 508 may execute the brake wear prediction model based on temperature data and / or power data extracted from the data matrix 324. Due to this execution, the model execution circuit system 508 determines and / or estimates the brake wear prediction model corresponding to… Figure 1 One or more example brake wear metrics (e.g., brake pad wear metrics) for a corresponding brake in brake 110. For example, brake wear metrics may include at least one of the mass, thickness (e.g., width), and / or variation in mass and / or thickness (e.g., relative to initial mass and / or thickness) of the brake pad associated with a corresponding brake in brake 110. In some examples, the mass corresponds to the output of an executed brake wear prediction model, and the model execution circuitry 508 also determines the thickness based on the mass, the density of the material used for the brake pad, and / or the geometry (e.g., surface area) of the brake pad. In some examples, the model execution circuitry 508 provides brake wear metrics to a cloud database 514 for storage, wherein the brake wear metrics may be stored in association with an identifier corresponding to the corresponding brake 110 and / or in association with the time at which the brake wear metrics were determined. In some examples, the model execution circuitry 508 is instantiated by and / or configured to perform operations, such as by a programmable circuitry system executing model execution circuitry system instructions. Figure 10 The flowchart represents those operations.

[0091] Figure 5The measurement calculation circuitry system 512 can predict and / or determine the remaining service life (RUL) (e.g., RUL metric) associated with the brake pads of the corresponding brake 110 based on brake wear metrics. In some examples, RUL represents the remaining width of the brake pad. In some examples, RUL represents the duration of time the brake pads can be utilized before replacement is required (e.g., to avoid deterioration of braking efficiency of brake 110). In some examples, the measurement calculation circuitry system 512 calculates RUL based on the example equation 6 below.

[0092] RUL=w 电流 -Δw

[0093] (Equation 6)

[0094] In example equation 6 above, RUL means corresponding to Figure 1 RUL, w of one or more brake pads corresponding to one brake in brake 110 电流 This represents the current (e.g., measured) width of the brake pad (e.g., the last measured width of the brake pad from the brake pad), and Δw represents the estimated wear (e.g., width loss) of the brake pad predicted based on the execution results of a brake wear prediction model. In some examples, the measurement calculation circuit system 512 calculates the RUL corresponding to the respective brake in brake 110 and provides the calculated RUL to the cloud database 514 for storage. In some examples, the measurement calculation circuit system 512 is instantiated by a programmable circuit system that executes measurement calculation circuit system instructions and / or configured to perform operations, such as by... Figure 10 The flowchart represents those operations.

[0095] The output circuitry system 510 outputs information generated, determined, and / or obtained by the model analysis circuitry system 104 (e.g., example brake wear information 516). For example, brake wear information 516 may include a brake wear metric determined for a specific brake in brake 110. In some examples, brake wear information 516 may include one or more timestamps associated with the brake wear metric (e.g., indicating the time when the brake wear metric was determined, the time when sensor data used to determine the brake wear metric was collected, etc.). In some examples, the output circuitry system 510 may (e.g., via...) Figure 1 Network 114) sends and / or transmits brake wear information 516 to Figure 1The brake wear information 516 is presented (e.g., displayed) via the user interface 118 of the vehicle 106. Alternatively, the output circuitry system 510 may send and / or transmit the brake wear information 516 to the manufacturer of the vehicle 106, the service provider of the vehicle 106, etc. In such examples, the brake wear information 516 may be used (e.g., by the owner of the vehicle 106, by the manufacturer and / or service provider of the vehicle 106, etc.) to plan and / or adjust maintenance activities of the vehicle 106 (e.g., replacing the brake pads of the brake 110).

[0096] In some examples, the output circuit system 510 periodically sends and / or transmits brake wear information 516 (e.g., once a day, once every two days, etc.). In some examples, the output circuit system 510 sends and / or transmits brake wear information 516 when the brake wear metric and / or predicted RUL does not meet example criteria (e.g., brake performance criteria, brake wear threshold). For example, the output circuit system 510 may send brake wear information 516 when the predicted quality of the brake pad is less than the threshold quality, when the predicted width of the brake pad is less than the threshold width, when the predicted RUL is less than the threshold RUL, etc. In some examples, the output circuit system 510 is instantiated by and / or configured to perform operations such as by a programmable circuit system executing output circuit system instructions. Figure 10 The flowchart represents those operations.

[0097] Figure 7 It means that it can be generated by Figure 1 and / or Figure 3 Example vehicle analysis circuit system 102 and Figure 1 and / or Figure 5 The example model analysis circuit system 104 implements an example information exchange program process flowchart 700. In Figure 7 In the illustrated example, at example vehicle frame 702, vehicle analysis circuitry system 102 receives example input 704 including ambient temperature data 318, torque data 320, and / or speed data 322. Furthermore, vehicle analysis circuitry system 102 calculates and / or determines example power data and example temperature data based on input 704, and then generates and / or updates one or more data matrices (e.g., data matrix 324) based on said power data and said temperature data. In some examples, vehicle analysis circuitry system 102 provides data matrix 324 to model analysis circuitry system 104 (e.g., implemented in a cloud-based environment).

[0098] At example cloud 706, model analysis circuitry 104 executes one or more brake wear prediction models (e.g., PINN models) based on dynamic and temperature data obtained from one or more data matrices from vehicle analysis circuitry 102. In some examples, based on the results of the execution, model analysis circuitry 104 determines and / or estimates... Figure 1 An example mass (e.g., brake pad mass) 708 of one or more brake pads of vehicle 106. In some examples, model analysis circuitry system 104 provides the determined mass 708 to vehicle analysis circuitry system 102 for calculating updated dynamic and temperature data based on new (e.g., incoming) inputs 704. In some examples, this is repeated and / or periodically performed during the lifespan of the brake pads of vehicle 106. Figure 7 Information exchange procedures are used to monitor brake pad wear and thus notify brake pad maintenance and / or replacement activities.

[0099] Figure 8 It means that it can be generated by Figure 1 and / or Figure 3 Vehicle analysis circuit system 102 and / or Figure 1 and / or Figure 5 The flowchart 800 shows the process of an example brake wear prediction program implemented in the model analysis circuit system 104. Figure 8 In the illustrated example, the vehicle analysis circuitry system 102 accesses and / or obtains sensor data including ambient temperature data 318, torque data 320, and speed data 322, and generates a data matrix 324 based on the sensor data. In this example, the vehicle analysis circuitry system 102 via... Figure 1 The example telematics control unit (TCU) 802 of vehicle 106 is communicatively connected to an example cloud-based environment 804, in which the model analysis circuitry system 104 can be implemented. In this example, the vehicle analysis circuitry system 102 provides a data matrix 324 and associated metadata to the model analysis circuitry system 104 in the cloud-based environment 804 via the TCU 802.

[0100] exist Figure 8 In the example shown, the model analysis circuit system 104 performs a data-based analysis based on data from the data matrix 324. Figure 6One or more example brake wear prediction models 806 are generated by the neural network 602. As a result of execution, the model analysis circuitry system 104 determines and / or predicts example brake wear metrics (e.g., mass, width, etc.) associated with the brake pads of vehicle 106. In this example, the model analysis circuitry system 104 may provide the predicted brake wear metrics to an example dealer (e.g., manufacturer) 808 of vehicle 106 to determine whether the brake pads need to be inspected and / or replaced. Additionally, the model analysis circuitry system 104 may provide the predicted brake wear metrics to an example customer 810 of vehicle 106 via, for example, a mobile application, webpage, portal, etc. In return, the customer 810 may provide example measurement data to the model analysis circuitry system 104 for training and / or execution of the brake wear prediction model 806. For example, the measurement data may include manually measured values ​​(e.g., measured width and / or mass) corresponding to the respective brake pads in the brake pads. In some examples, the model analysis circuitry 104 may provide the vehicle analysis circuitry 102 with predicted and / or measured brake pad measurements as example configuration data 812 via the TCU 802 for updating the data matrix 324 and / or generating one or more additional data matrices.

[0101] In some examples, the vehicle analysis circuit system 102 includes means for acquiring data, means for calculating power, means for calculating temperature, means for generating a data matrix, and means for transmitting data. For example, the means for acquiring data may be implemented by a data interface circuit system 302, the means for calculating power may be implemented by a power calculation circuit system 304, the means for calculating temperature may be implemented by a temperature calculation circuit system 306, the means for generating a matrix may be implemented by a matrix control circuit system 308, and the means for transmitting data may be implemented by a data transmission circuit system 310. In some examples, the data interface circuit system 302, the power calculation circuit system 304, the temperature calculation circuit system 306, the matrix control circuit system 308, and the data transmission circuit system 310 may be implemented by a programmable circuit system (such as...). Figure 12Example programmable circuit system 1212) is instantiated. Alternatively or concurrently, data interface circuit system 302, power calculation circuit system 304, temperature calculation circuit system 306, matrix control circuit system 308, and data transmission circuit system 310 can be instantiated by any other combination of hardware, software, and / or firmware. For example, data interface circuit system 302, power calculation circuit system 304, temperature calculation circuit system 306, matrix control circuit system 308, and data transmission circuit system 310 can be implemented by at least one or more hardware circuits (e.g., processor circuit systems, discrete and / or integrated analog and / or digital circuit systems, FPGAs, ASICs, XPUs, comparators, operational amplifiers (op-amps), logic circuits, etc.) configured and / or structured to execute some or all of machine-readable instructions, and / or perform some or all of the operations corresponding to machine-readable instructions, without executing software or firmware, but other structures are equally applicable.

[0102] In some examples, the model analysis circuit system 104 includes means for acquiring input, means for processing, means for training, means for execution, means for output, and means for calculating metrics. For example, the means for acquiring input may be implemented by an input interface circuit system 502, the means for processing may be implemented by a data processing circuit system 504, the means for training may be implemented by a model training circuit system 506, the means for execution may be implemented by a model execution circuit system 508, the means for output may be implemented by an output circuit system 510, and the means for calculating metrics may be implemented by a metric calculation circuit system 512. In some examples, the input interface circuit system 502, data processing circuit system 504, model training circuit system 506, model execution circuit system 508, output circuit system 510, and metric calculation circuit system 512 may be implemented by a programmable circuit system (such as...). Figure 12Example programmable circuit system 1212) instantiated. Alternatively or concurrently, input interface circuit system 502, data processing circuit system 504, model training circuit system 506, model execution circuit system 508, output circuit system 510, and metric calculation circuit system 512 can be instantiated by any other combination of hardware, software, and / or firmware. For example, input interface circuit system 502, data processing circuit system 504, model training circuit system 506, model execution circuit system 508, output circuit system 510, and metric calculation circuit system 512 can be implemented by at least one or more hardware circuits (e.g., processor circuit systems, discrete and / or integrated analog and / or digital circuit systems, FPGAs, ASICs, XPUs, comparators, operational amplifiers (op-amps), logic circuits, etc.) configured and / or structured to execute some or all of machine-readable instructions, and / or perform some or all of the operations corresponding to machine-readable instructions, without executing software or firmware, but other structures are equally applicable.

[0103] Although Figure 3 The implementation is shown in the figure. Figure 1 The vehicle analysis circuit system 102 is an example of a method, but Figure 3 One or more of the elements, processes, and / or devices shown may be combined, divided, rearranged, omitted, eliminated, and / or implemented in any other way. Furthermore, Figure 3 The example data interface circuit system 302, example power calculation circuit system 304, example temperature calculation circuit system 306, example matrix control circuit system 308, example data transmission circuit system 310, example vehicle database 312, and / or more generally, example vehicle analysis circuit system 102 can be implemented by hardware only, by a combination of hardware and software and / or firmware. Therefore, for example, any of the example data interface circuit system 302, example power calculation circuit system 304, example temperature calculation circuit system 306, example torque control circuit system 308, example data transmission circuit system 310, example vehicle database 312, and / or more generally, example vehicle analysis circuit system 102 can be implemented by a programmable circuit system combined with machine-readable instructions (e.g., firmware or software), processor circuit system, analog circuitry, digital circuitry, logic circuitry, programmable processor, programmable microcontroller, graphics processing unit (GPU), digital signal processor (DSP), ASIC, programmable logic device (PLD), and / or field-programmable logic device (FPLD) (such as FPGA). Furthermore, Figure 3 Example vehicle analysis circuit system 102 may include one or more components, processes and / or devices, as a means of... Figure 3The elements, processes and / or devices shown may be supplemented or substituted, and / or may include more than one of any or all of the elements, processes and devices shown.

[0104] Although Figure 5 The implementation is shown in the figure. Figure 1 The example method for analyzing the circuit system 104 using a model, but Figure 5 One or more of the elements, processes, and / or devices shown may be combined, divided, rearranged, omitted, eliminated, and / or implemented in any other way. Furthermore, Figure 5 The example input interface circuit system 502, example data processing circuit system 504, example model training circuit system 506, example model execution circuit system 508, example output circuit system 510, example metric calculation circuit system 512, example cloud database 514, and / or more generally, example model analysis circuit system 104 can be implemented by hardware only, by a combination of hardware and software and / or firmware. Therefore, for example, any of the example input interface circuit system 502, example data processing circuit system 504, example model training circuit system 506, example model execution circuit system 508, example output circuit system 510, example metric calculation circuit system 512, example cloud database 514, and / or more generally, example model analysis circuit system 104 can be implemented by a programmable circuit system in combination with machine-readable instructions (e.g., firmware or software), processor circuit system, analog circuitry, digital circuitry, logic circuitry, programmable processor, programmable microcontroller, graphics processing unit (GPU), digital signal processor (DSP), ASIC, programmable logic device (PLD), and / or field-programmable logic device (FPLD) (such as FPGA). also, Figure 5 Example model analysis circuit system 104 may include one or more elements, processes and / or devices, as a reference for... Figure 5 The elements, processes and / or devices shown may be supplemented or substituted, and / or may include more than one of any or all of the elements, processes and devices shown.

[0105] exist Figure 9 , Figure 10 and / or Figure 11 The flowchart shown represents example machine-readable instructions that can be executed by a programmable circuit system to implement and / or instantiate. Figure 3 Vehicle analysis circuit system 102 and / or Figure 5 The model analysis circuit system 104, and / or represents example operations, which can be implemented and / or instantiated by a programmable circuit system. Figure 3 Vehicle analysis circuit system 102 and / or Figure 5The model analysis circuit system 104. Machine-readable instructions can be generated by a programmable circuit system (such as those combined below). Figure 12 The programmable circuit system 1212 shown in the example processor platform 1200 discussed below and / or in conjunction with the following Figure 13 The programmable circuit system 1312 shown in the example processor platform 1300 discussed here executes one or more executable programs or portions of one or more executable programs. In some examples, machine-readable instructions cause operations, tasks, etc., to be performed and / or carried out in an automated manner in the real world. As used herein, “automation” means without human intervention.

[0106] The program may be embodied in instructions (e.g., software and / or firmware) stored in one or more non-transitory computer-readable and / or machine-readable storage media, such as cache memory, magnetic storage devices or disks (e.g., floppy disks, hard disk drives (HDDs), etc.), optical storage devices or optical discs (e.g., Blu-ray discs, compact discs (CDs), digital versatile discs (DVDs), etc.), redundant arrays of independent disks (RAID), registers, ROM, solid-state drives (SSDs), SSD memory, non-volatile memory (e.g., electrically erasable programmable read-only memory (EEPROM), flash memory, etc.), volatile memory (e.g., random access memory (RAM) of any type), and / or any other storage device or disk. The instructions of the non-transitory computer-readable and / or machine-readable media may be programmed and / or executed by a programmable circuit system located in one or more hardware devices, but the entire program and / or portions thereof may alternatively be executed and / or instantiated by one or more hardware devices rather than a programmable circuit system, and / or embodied in dedicated hardware. Machine-readable instructions may be distributed across multiple hardware devices and / or executed by two or more hardware devices (e.g., server and client hardware devices). For example, client hardware devices may be implemented by endpoint client hardware devices (e.g., hardware devices associated with human and / or machine users) or by an intermediate client hardware device gateway (e.g., a radio access network (RAN)) that facilitates communication between the server and endpoint client hardware devices. Similarly, non-transitory computer-readable storage media may include one or more media. Furthermore, although references... Figure 9 , Figure 10 and / or Figure 11The flowchart shown describes an example program, but many other methods of implementing the example vehicle analysis circuit system 102 and / or model analysis circuit system 104 can also be used alternatively. For example, the execution order of the flowchart blocks can be changed, and / or some of the described blocks can be changed, eliminated, or combined. Additionally or alternatively, any or all of the flowchart blocks can be implemented by one or more hardware circuits (e.g., processor circuit systems, discrete and / or integrated analog and / or digital circuit systems, FPGAs, ASICs, comparators, operational amplifiers (op-amps), logic circuits, etc.) that are structured to perform corresponding operations without executing software or firmware. The programmable circuit system can be distributed across different network locations and / or local to one or more hardware devices (e.g., single-core processors (e.g., single-core CPUs), multi-core processors (e.g., multi-core CPUs, XPUs, etc.)). For example, a programmable circuit system may be a CPU and / or FPGA located in the same package (e.g., the same integrated circuit (IC) package or in two or more separate housings), one or more processors in a single machine, multiple processors distributed across multiple servers in a server rack, multiple processors distributed across one or more server racks, and / or any combination thereof.

[0107] The machine-readable instructions described herein may be stored in one or more of the following formats: compressed format, encrypted format, segmented format, compiled format, executable format, and packaged format. As described herein, machine-readable instructions may be stored as data (e.g., computer-readable data, machine-readable data, one or more bits (e.g., one or more computer-readable bits, one or more machine-readable bits, etc.), bit streams (e.g., computer-readable bit streams, machine-readable bit streams, etc.)) or data structures (e.g., stored as parts of instructions, code, representations of code, etc.). For example, machine-readable instructions may be segmented and stored on one or more storage devices, disks, and / or computing devices (e.g., servers) located in the same or different locations (e.g., in the cloud, at an edge device, etc.) within a network or network set. Machine-readable instructions may require one or more of the following processes: installation, modification, rewriting, updating, combination, supplementation, configuration, decryption, decompression, unpacking, distribution, reallocation, compilation, etc., to enable them to be directly read, interpreted, and / or executed by computing devices and / or other machines. For example, machine-readable instructions may be stored in multiple parts, which are individually compressed, encrypted, and / or stored on separate computing devices, wherein the parts, when decrypted, decompressed, and / or combined, form a set of computer-executable and / or machine-executable instructions that implement one or more functions and / or operations of a program that together form a program such as the program described herein.

[0108] In another example, machine-readable instructions may be stored in a state that can be read by a programmable circuit system, but require the addition of libraries (e.g., dynamic link libraries (DLLs)), software development kits (SDKs), application programming interfaces (APIs), etc., to execute the machine-readable instructions on a specific computing device or other device. In another example, it may be necessary to configure the machine-readable instructions (e.g., store settings, input data, record network addresses, etc.) before they can be executed in whole or in part. Therefore, as used herein, machine-readable, computer-readable media, and / or machine-readable media may include instructions and / or programs, regardless of their specific format or state.

[0109] The machine-readable instructions described in this article can be represented by any past, present, or future instruction language, scripting language, programming language, etc. For example, any of the following languages ​​can be used to represent machine-readable instructions: C, C++, Java, C#, Perl, Python, JavaScript, Hypertext Markup Language (HTML), Structured Query Language (SQL), Swift, etc.

[0110] As mentioned above, it can be implemented using executable instructions (e.g., computer-readable instructions and / or machine-readable instructions) stored on one or more non-transitory computer-readable and / or machine-readable media. Figure 9 , Figure 10 and / or Figure 11Example operation. As used herein, the terms non-transitory computer-readable medium, non-transitory computer-readable storage medium, non-transitory machine-readable medium, and / or non-transitory machine-readable storage medium are expressly defined to include any type of computer-readable storage device and / or storage disk, excluding propagated signals and transmission media. Examples of such non-transitory computer-readable medium, non-transitory computer-readable storage medium, non-transitory machine-readable medium, and / or non-transitory machine-readable storage medium include optical storage devices, magnetic storage devices, HDDs, flash memory, read-only memory (ROM), CDs, DVDs, caches, any type of RAM, registers, and / or any other storage device or storage disk in which information is stored for any duration (e.g., extended time period, permanently, for short-term cases, for temporary buffering, and / or for caching information). As used herein, the terms "non-transitory computer-readable storage device" and "non-transitory machine-readable storage device" are defined to include any physical (mechanical, magnetic, and / or electrical) hardware for retaining information for a period of time, excluding propagated signals and transmission media. Examples of non-transitory computer-readable storage devices and / or non-transitory machine-readable storage devices include any type of random access memory, any type of read-only memory, solid-state memory, flash memory, optical disk, magnetic disk, disk drive, and / or redundant array of independent disks (RAID) system. As used herein, the term "device" refers to a physical structure, such as mechanical and / or electrical equipment, hardware, and / or circuitry, that can be configured and / or may be manufactured or may not be manufactured to execute computer-readable instructions, machine-readable instructions, etc., whether or not by computer-readable instructions, machine-readable instructions, etc.

[0111] Figure 9 This is a flowchart illustrating example machine-readable instructions and / or example operations 900, which can be provided by... Figure 3 The vehicle analysis circuitry system 102 executes, instantiates, and / or implements to generate one or more sample data matrices (e.g., Figure 3 and / or Figure 4 Data matrix 324). Figure 9 Example machine-readable instructions and / or example operations 900 begin at box 902, where the example vehicle analysis circuitry system 102 initializes data matrix 324 and corresponding intervals (e.g., temperature interval 402 and / or power interval 404). For example, Figure 3Example matrix control circuit system 308 selects a first quantity (e.g., a first interval count) of temperature interval 402, a second quantity (e.g., a second interval count) of power interval 404, and / or an example range (e.g., a range of values) corresponding to the respective intervals in temperature interval 402 and power interval 404. In some examples, matrix control circuit system 308 selects based on user input (e.g., via...). Figure 1 The user interface 118 provides the selection of the first quantity, the second quantity, and / or the range based on selections preloaded in the vehicle analysis circuitry system 102. Furthermore, based on the first and second quantities, the matrix control circuitry system 308 generates and / or initializes a data matrix 324 having an example row count corresponding to the first quantity and an example column count corresponding to the second quantity. In some examples, the matrix control circuitry system 308 selects initial values ​​(e.g., 0) for the corresponding matrix values ​​of the data matrix 324.

[0112] At box 904, the example vehicle analysis circuitry 102 accesses example sensor data including torque data 320, speed data 322, and ambient temperature data 318. For example, Figure 3 Data interface circuit system 302 access from Figure 1 Sensor data from one or more of the sensors 112 of vehicle 106.

[0113] At box 906, the example vehicle analysis circuitry 102 calculates the power applied to the corresponding wheels 108 of the vehicle 106 based on torque data 320 and speed data 322. For example, Figure 3 The example dynamic calculation circuit system 304 calculates example dynamic data based on the example equations 1, 2A, 2B, 2C and / or 2D above, wherein the dynamic data includes data derived from... Figure 1 The brake 110 applies power to the brake disc of the corresponding wheel 108.

[0114] At box 908, the example vehicle analysis circuitry 102 calculates the temperature at the corresponding wheel 108 based on ambient temperature data 318 and the power applied to the corresponding wheel 108 (e.g., power data). For example, Figure 3 The example temperature calculation circuit system 306 calculates example temperature data based on the example equations 3A and / or 3B above, wherein the temperature data includes the temperature at the interface between the brake pads of the brake 110 and the corresponding brake discs of the corresponding wheels 108.

[0115] At box 910, the example vehicle analysis circuit system 102 selects the intervals corresponding to the calculated temperature and power values. For example, the matrix control circuit system 308 selects the temperature interval 402 corresponding to the calculated temperature value, and further selects the power interval 404 corresponding to the calculated power value.

[0116] At box 912, the example vehicle analysis circuit system 102 updates one or more example matrix values ​​in data matrix 324 corresponding to selected intervals. For example, matrix control circuit system 308 selects matrix values ​​corresponding to selected temperature intervals and power intervals in temperature interval 402 and power interval 404, and updates (e.g., increases, increments (e.g., increases by 1)) the selected matrix values.

[0117] At box 914, the example vehicle analysis circuitry 102 determines whether to continue monitoring. For example, when the vehicle 106 is moving and / or operating and / or when additional sensor data is available from… Figure 1 When the sensor in sensor 112 acquires an acquisition, data interface circuitry 302 determines to continue monitoring. In response to data interface circuitry 302 determining to continue monitoring (e.g., block 914 returns a result of "Yes"), control returns to block 904. Alternatively, in response to data interface circuitry 302 determining not to continue monitoring (e.g., block 914 returns a result of "No"), control proceeds to block 916.

[0118] At box 916, the example vehicle analysis circuitry 102 transmits and / or causes the stored data matrix 324 to be executed. For example, Figure 3 Example data transmission circuit system 310 via Figure 1 Example network 114 transmits data matrix 324 to Figure 1 and / or Figure 5 Example model analysis circuit system 104 is used to determine one or more example brake wear metrics. Alternatively, data transmission circuit system 310 can transmit data to... Figure 3 The vehicle database 312 provides a data matrix 324 for storage.

[0119] Figure 10 It means that it can be generated by Figure 1 and / or Figure 5 The model analysis circuit system 104 executes, instantiates, and / or implements flowcharts of example machine-readable instructions and / or example operations 1000 to determine one or more example brake wear metrics. Figure 10 Example machine-readable instructions and / or example operations 1000 begin at box 1002, where example model analysis circuit system 104 obtains Figure 3 and / or Figure 4 Example data matrix 324. For example, Figure 5Example input interface circuit system 502 accesses, obtains, and / or retrieves data from vehicle analysis circuit system 102. Figure 1 The data matrix 324 generated from the corresponding wheels 108 of vehicle 106.

[0120] At box 1004, the example model analysis circuit system 104 extracts example temperature data and example dynamic data from data matrix 324. For example, Figure 5 Example data processing circuitry 504 extracts and / or obtains temperature data and power data from data matrix 324, wherein the temperature data includes the temperature corresponding to the respective wheel 108 (e.g., at the interface between the brake pads of brake 110 and the corresponding brake disc of the respective wheel 108), and the power data includes the power applied by the brake pads to the corresponding brake disc of the respective wheel 108.

[0121] At box 1006, the example model analysis circuit system 104 executes one or more example brake wear prediction models (e.g., physical information neural network models) based on temperature and dynamic data. For example, Figure 5 The example model execution circuit system 508 provides temperature data and dynamic data as inputs to the brake wear prediction model, and executes the brake wear prediction model based on said inputs. The following is in conjunction with... Figure 11 Further description of the generation and / or training of the brake wear prediction model.

[0122] At box 1008, example model analysis circuitry 104 estimates brake wear metrics based on the results of its execution. For example, model execution circuitry 508 determines and / or estimates at least one of the mass or width (e.g., thickness) of the brake pad corresponding to the respective wheel 108 based on the results of its execution. Alternatively or additionally, model execution circuitry 508 may determine variations in the mass and / or width of the brake pad (e.g., relative to the initial mass and / or initial width of the brake pad).

[0123] At box 1010, example model analysis circuitry 104 calculates the remaining useful life (RUL) of the corresponding brake pads based on brake wear metrics. For example, example metric calculation circuitry 512 calculates and / or estimates the RUL of the brake pads based on example equation 6 above.

[0124] At box 1012, the example model analysis circuit system 104 determines whether one or more metrics (e.g., brake wear metric and / or RUL) meet associated example criteria. For example, when the brake pad mass is less than a threshold mass, the brake pad width is less than a threshold width, and / or RUL is less than a threshold RUL, Figure 5Example output circuitry 510 determines that the metric is not met. In response to output circuitry 510 determining that the metric is not met (e.g., block 1012 returns a result "No"), control proceeds to block 1014. Alternatively, in response to output circuitry 510 determining that the metric is met (e.g., block 1012 returns a result "Yes"), control proceeds to block 1016.

[0125] At box 1014, example model analysis circuitry 104 generates and / or outputs example brake wear information 516. For example, output circuitry 510 generates and / or outputs brake wear information 516, which includes brake wear measurements of the corresponding brake pads of vehicle 106 and / or the determined RUL (Range Wear Limit). In some examples, output circuitry 510 provides brake wear information 516 to vehicle 106 such that the brake wear information 516 is transmitted via... Figure 1 The user interface 118 is presented. In some examples, the output circuit system 510 provides brake wear information 516 to the manufacturer and / or dealer of vehicle 106, the service provider of vehicle 106, the owner of vehicle 106, etc., to help plan the maintenance activities of vehicle 106.

[0126] At box 1016, the example model analysis circuitry 104 determines whether at least one additional data matrix exists to be analyzed. In response to the input interface circuitry 502 determining that at least one additional data matrix exists to be analyzed (e.g., box 1016 returns a "Yes" result), control returns to box 1002. Alternatively, in response to the input interface circuitry 502 determining that no additional data matrix exists to be analyzed (e.g., box 1016 returns a "No" result), control ends.

[0127] Figure 11 This is a flowchart illustrating example machine-readable instructions and / or example operations 1100, which can be provided by... Figure 1 and / or Figure 5 The model analysis circuit system 104 executes, instantiates, and / or implements to generate and / or train one or more example brake wear prediction models. Figure 11 Example machine-readable instructions and / or example operations 1100 begin at box 1102, where example model analysis circuit system 104 obtains Figure 1 Example training data 116. For example, example input interface circuit system 502 obtains training data 116, which includes measured and / or simulated mass values ​​(e.g., brake mass values) and corresponding input values ​​(e.g., temperature, power and / or time) determined and / or obtained based on simulated data, vehicle and / or brake test data, etc.

[0128] At box 1104, the example model analysis circuit system 104 initializes the example neural network (e.g., Figure 6 (Neural network 602). For example, Figure 5 The example model training circuit system 506 generates a neural network 602 with a selected number of layers and / or nodes per layer, and selects initial weights for the neural network 602.

[0129] At box 1106, example model analysis circuitry 104 selects and / or generates one or more example juxtaposition points. For example, example data processing circuitry 504 selects juxtaposition points from an input space defined by an input temperature range, an input dynamic range, and an input time range using a selected sampling method (e.g., Latin hypercube sampling).

[0130] At box 1108, the example model analysis circuit system 104 executes a neural network 602 based on training data 116 to determine one or more example prediction quality values ​​(e.g., 616. For example, the model training circuit system 506 selects an input value 606 represented in the training data 116 (e.g., temperature value 608, corresponding dynamic value 610, and / or corresponding time value 612), and executes the neural network 602 based on the input value 606. In some examples, due to the execution, the model training circuit system 506 determines and / or outputs a predicted quality value 616 corresponding to the input value 606.

[0131] At box 1110, the example model analyzes the circuit system 104 based on the partial derivative with respect to the predicted quality value 616 at the juxtaposition point (e.g., The model training circuit system 506 determines one or more predicted quality rates (e.g., predicted quality change rates) by evaluating the predicted quality value 616. For example, based on the automatic differentiation of the predicted quality value 616, the model training circuit system 506 determines the partial derivative 618 of the predicted quality value 616 (e.g., gradient, rate of change), and evaluates the partial derivative 618 at the juxtaposition point to determine the predicted quality rate.

[0132] At box 1112, example model analysis circuit system 104 determines the quality rate (e.g., the rate of change of computational quality) of one or more computations based on an evaluation of a physics-based equation at the juxtaposition point (e.g., physics-based equation 626 corresponding to example equation 5 above). For example, model training circuit system 506 evaluates example equation 5 above at the juxtaposition point to determine the rate of computational quality.

[0133] At box 1114, example model analysis circuitry 104 determines a first example loss value (e.g., first loss value 622) based on the difference between the predicted quality value and the corresponding measured and / or simulated quality value. For example, model training circuitry 506 determines the first loss value 622 based on example equation 4 above.

[0134] At box 1116, example model analysis circuitry 104 determines a second example loss value (e.g., second loss value 624) based on the difference between the predicted quality rate and the corresponding calculated quality rate. For example, model training circuitry 506 calculates the difference between the predicted quality rate and the calculated quality rate (e.g., average difference) for corresponding juxtaposition points in the juxtaposition.

[0135] At box 1118, example model analysis circuitry 104 determines a combined loss value 628 based on a first loss value 622 and a second loss value 624. For example, model training circuitry 506 determines the combined loss value 628 by summing (e.g., adding together) the first loss value 622 and the second loss value 624.

[0136] At box 1120, example model analysis circuitry 104 determines whether the combined loss value 628 meets a threshold (e.g., an error threshold). For example, model training circuitry 506 determines whether the combined loss value 628 meets (e.g., is less than or equal to) the threshold. In response to model training circuitry 506 determining that the combined loss value 628 meets the threshold (e.g., box 1120 returns a result "Yes"), control proceeds to box 1124. Alternatively, in response to model training circuitry 506 determining that the combined loss value 628 does not meet the threshold (e.g., box 1120 returns a result "No"), control proceeds to box 1122.

[0137] At box 1122, example model analysis circuitry 104 adjusts one or more weights of neural network 602. For example, model training circuitry 506 adjusts said weights based on combined loss value 628. After model training circuitry 506 adjusts the weights, control returns to box 1108 to further train neural network 602.

[0138] At box 1124, example model analysis circuitry 104 causes the storage of neural network 602. For example, model training circuitry 506 generates one or more brake wear prediction models based on the trained network 602, and causes the brake wear prediction models to be stored... Figure 5 In the cloud database 514.

[0139] Figure 12 It is structured for execution and / or instantiation. Figure 9 Example machine-readable instructions and / or example operations for implementation Figure 3The block diagram illustrates an example programmable circuit system platform 1200 for a vehicle analysis circuit system 102. The programmable circuit platform 1200 can be, for example, a server, personal computer, workstation, self-learning machine (e.g., neural network), or mobile device (e.g., mobile phone, smartphone, such as iPad). TM Tablet computers, personal digital assistants (PDAs), internet devices, DVD players, CD players, digital video recorders, Blu-ray players, game consoles, personal video recorders, set-top boxes, headsets (e.g., augmented reality (AR) headsets, virtual reality (VR) headsets, etc.) or other wearable devices or any other type of computing and / or electronic device.

[0140] The illustrated programmable circuit system platform 1200 includes a programmable circuit system 1212. The illustrated programmable circuit system 1212 is hardware. For example, the programmable circuit system 1212 may be implemented by one or more integrated circuits, logic circuits, FPGAs, microprocessors, CPUs, GPUs, DSPs, and / or microcontrollers from any desired series or manufacturer. The programmable circuit system 1212 may be implemented by one or more semiconductor-based (e.g., silicon-based) devices. In this example, the programmable circuit system 1212 implements a data interface circuit system 302, a power calculation circuit system 304, a temperature calculation circuit system 306, a matrix control circuit system 308, a data transmission circuit system 310, and a vehicle database 312.

[0141] The programmable circuit system 1212 of the example shown includes local memory 1213 (e.g., cache, registers, etc.). The programmable circuit system 1212 of the example shown communicates via bus 1218 with main memory 1214, 1216, which includes volatile memory 1214 and non-volatile memory 1216. Volatile memory 1214 may be synchronous dynamic random access memory (SDRAM), dynamic random access memory (DRAM), etc. Dynamic Random Access Memory The non-volatile memory 1216 may be implemented using flash memory and / or any other desired type of memory device. Access to the main memory 1214, 1216 of the illustrated example is controlled by the memory controller 1217. In some examples, the memory controller 1217 may be implemented by one or more integrated circuits, logic circuits, microcontrollers, or any other type of circuit system from any desired series or manufacturer to manage the data flow to and from the main memory 1214, 1216.

[0142] The programmable circuit system platform 1200 illustrated also includes an interface circuit system 1220. The interface circuit system 1220 can be implemented in hardware according to any type of interface standard, such as an Ethernet interface, a Universal Serial Bus (USB) interface, etc. Interfaces, Near Field Communication (NFC) interfaces, Peripheral Component Interconnect (PCI) interfaces, and / or Peripheral Component Interconnect High Speed ​​(PCIe) interfaces.

[0143] In the illustrated example, one or more input devices 1222 are connected to the interface circuitry 1220. The input devices 1222 allow users (e.g., human users, machine users, etc.) to input data and / or commands into the programmable circuitry 1212. The input devices 1222 can be implemented, for example, audio sensors, microphones, cameras (still or video), keyboards, buttons, mice, touchscreens, trackpads, trackballs, isotope devices, and / or voice recognition systems.

[0144] One or more output devices 1224 are also connected to the interface circuitry 1220 of the illustrated example. The output devices 1224 may be implemented, for example, via display devices (e.g., light-emitting diode (LED), organic light-emitting diode (OLED), liquid crystal display (LCD), cathode ray tube (CRT) display, in-situ switch (IPS) display, touchscreen, etc.), haptic output devices, printers, and / or speakers. Therefore, the interface circuitry 1220 of the illustrated example typically includes a graphics driver card, a graphics driver chip, and / or a graphics processor circuitry, such as a GPU.

[0145] The interface circuit system 1220 of the example shown also includes communication devices, such as transmitters, receivers, transceivers, modems, residential gateways, wireless access points, and / or network interfaces, to facilitate data exchange with external machines (e.g., any kind of computing device) via network 1226. Communication can be carried out via, for example, Ethernet connections, digital subscriber line (DSL) connections, telephone line connections, coaxial cable systems, satellite systems, beyond-line-of-sight wireless systems, line-of-sight wireless systems, mobile phone systems, optical connections, etc.

[0146] The programmable circuit system platform 1200 illustrated also includes one or more mass storage disks or devices 1228 for storing firmware, software, and / or data. Examples of such mass storage disks or devices 1228 include magnetic storage devices (e.g., floppy disks, drives, HDDs, etc.), optical storage devices (e.g., Blu-ray discs, CDs, DVDs, etc.), RAID systems, and / or solid-state storage disks or devices, such as flash memory devices and / or SSDs.

[0147] It can be by Figure 9The machine-readable instructions 1232 implemented by the machine-readable instructions may be stored in mass storage device 1228, in volatile memory 1214, in non-volatile memory 1216 and / or on at least one non-transitory computer-readable storage medium (such as a CD or DVD) that may be removable.

[0148] Figure 13 It is structured for execution and / or instantiation. Figure 10 and / or Figure 11 Example machine-readable instructions and / or example operations for implementation Figure 5 The block diagram of an example programmable circuit system platform 1300 for model analysis circuit system 104 is provided. Programmable circuit system platform 1300 can be, for example, a server, personal computer, workstation, self-learning machine (e.g., neural network), mobile device (e.g., mobile phone, smartphone, such as iPad). TM Tablet computers, personal digital assistants (PDAs), internet devices, DVD players, CD players, digital video recorders, Blu-ray players, game consoles, personal video recorders, set-top boxes, headsets (e.g., augmented reality (AR) headsets, virtual reality (VR) headsets, etc.) or other wearable devices or any other type of computing and / or electronic device.

[0149] The illustrated programmable circuit system platform 1300 includes a programmable circuit system 1312. The illustrated programmable circuit system 1312 is hardware. For example, the programmable circuit system 1312 may be implemented by one or more integrated circuits, logic circuits, FPGAs, microprocessors, CPUs, GPUs, DSPs, and / or microcontrollers from any desired family or manufacturer. The programmable circuit system 1312 may be implemented by one or more semiconductor-based (e.g., silicon-based) devices. In this example, the programmable circuit system 1312 implements an input interface circuit system 502, a data processing circuit system 504, a model training circuit system 506, a model execution circuit system 508, an output circuit system 510, a metric calculation circuit system 512, and a cloud database 514.

[0150] The programmable circuit system 1312 shown in the example includes local memory 1313 (e.g., cache, registers, etc.). The programmable circuit system 1312 shown in the example communicates via bus 1318 with main memory 1314, 1316, which includes volatile memory 1314 and non-volatile memory 1316. Volatile memory 1314 may be synchronous dynamic random access memory (SDRAM), dynamic random access memory (DRAM), etc. Dynamic Random Access Memory And / or any other type of RAM device. The non-volatile memory 1316 may be implemented by flash memory and / or any other desired type of memory device. Access to the main memories 1314, 1316 of the illustrated examples is controlled by the memory controller 1317. In some examples, the memory controller 1317 may be implemented by one or more integrated circuits, logic circuits, microcontrollers, or any other type of circuit system from any desired family or manufacturer to manage the data flow to and from the main memories 1314, 1316.

[0151] The programmable circuit system platform 1300 shown in the example also includes an interface circuit system 1320. The interface circuit system 1320 can be implemented in hardware according to any type of interface standard, such as an Ethernet interface, a Universal Serial Bus (USB) interface, etc. Interfaces, Near Field Communication (NFC) interfaces, Peripheral Component Interconnect (PCI) interfaces, and / or Peripheral Component Interconnect High Speed ​​(PCIe) interfaces.

[0152] In the illustrated example, one or more input devices 1322 are connected to the interface circuitry 1320. The input devices 1322 allow users (e.g., human users, machine users, etc.) to input data and / or commands into the programmable circuitry 1312. The input devices 1322 can be implemented, for example, audio sensors, microphones, cameras (still or video), keyboards, buttons, mice, touchscreens, trackpads, trackballs, dot devices, and / or voice recognition systems.

[0153] One or more output devices 1324 are also connected to the interface circuitry 1320 of the illustrated example. The output devices 1324 may be implemented, for example, via display devices (e.g., light-emitting diode (LED), organic light-emitting diode (OLED), liquid crystal display (LCD), cathode ray tube (CRT) display, in-situ switch (IPS) display, touchscreen, etc.), haptic output devices, printers, and / or speakers. Therefore, the interface circuitry 1320 of the illustrated example typically includes a graphics driver card, a graphics driver chip, and / or a graphics processor circuitry, such as a GPU.

[0154] The interface circuit system 1320 of the example shown also includes communication devices, such as transmitters, receivers, transceivers, modems, residential gateways, wireless access points, and / or network interfaces, to facilitate data exchange with external machines (e.g., any kind of computing device) via network 1326. Communication can be carried out via, for example, Ethernet connections, digital subscriber line (DSL) connections, telephone line connections, coaxial cable systems, satellite systems, beyond-line-of-sight wireless systems, line-of-sight wireless systems, mobile phone systems, optical connections, etc.

[0155] The programmable circuit system platform 1300 illustrated also includes one or more mass storage disks or devices 1328 for storing firmware, software, and / or data. Examples of such mass storage disks or devices 1328 include magnetic storage devices (e.g., floppy disks, drives, HDDs, etc.), optical storage devices (e.g., Blu-ray discs, CDs, DVDs, etc.), RAID systems, and / or solid-state storage disks or devices, such as flash memory devices and / or SSDs.

[0156] can be Figure 10 and / or Figure 11 The machine-readable instructions 1332 implemented by the machine-readable instructions may be stored in mass storage device 1328, in volatile memory 1314, in non-volatile memory 1316, and / or on at least one non-transitory computer-readable storage medium (such as a CD or DVD) that may be removable.

[0157] "Comprising" and "including" (and all their forms and tenses) are used herein as open-ended terms. Therefore, whenever a claim uses any form of "comprising" or "including" (e.g., including, containing, encompassing, covering, having, etc.) as a preamble or within any kind of claim statement, it should be understood that additional elements, items, etc., may be present without falling outside the scope of the corresponding claim or statement. As used herein, when the phrase "at least" is used as a transitional term in the preamble of a claim, it becomes an open-ended term in the same way that the terms "comprising" and "including" become open-ended terms. The term "and / or," when used, for example, in forms such as A, B, and / or C, refers to any combination or subset of A, B, and C, such as (1) only A, (2) only B, (3) only C, (4) A and B, (5) A and C, (6) B and C, and (7) A and B and C. As used herein in the context of describing structures, components, articles, objects, and / or things, the phrase "at least one of A and B" is intended to refer to an implementation comprising (1) at least one A, (2) at least one B, or (3) at least one A and at least one B. Similarly, as used herein in the context of describing structures, components, articles, objects, and / or things, the phrase "at least one of A and B" is intended to refer to an implementation comprising (1) at least one A, (2) at least one B, or (3) at least one A and at least one B. As used herein in the context of describing the implementation or execution of processes, instructions, actions, activities, etc., the phrase "at least one of A and B" is intended to refer to an implementation comprising (1) at least one A, (2) at least one B, or (3) at least one A and at least one B. Similarly, as used herein in the context of describing the implementation or execution of processes, instructions, actions, activities, etc., the phrase “at least one of A or B” is intended to refer to an implementation that includes (1) at least one A, (2) at least one B, or (3) at least one A and at least one B.

[0158] As used herein, singular references (e.g., "a," "an," "first," "second," etc.) do not exclude plurals. As used herein, the term "a / an" refers to one or more of the same object. The terms "a / an," "one or more," and "at least one" are used interchangeably herein. Furthermore, although listed separately, multiple means, elements, or actions may be implemented by, for example, the same entity or object. Additionally, although individual features may be included in different examples or claims, they may be combined, and inclusion in different examples or claims does not imply that a combination of features is impractical and / or advantageous.

[0159] As used herein, unless otherwise stated, the term "above" describes the relationship of two parts relative to the Earth. The first part is above the second part if the second part has at least one portion between the Earth and the first part. Similarly, as used herein, the first part is "below" the second part when the first part is closer to the Earth than the second part. As stated above, the first part may be above or below the second part in one or more of the following ways: when there are other parts between them, when there are no other parts between them, when the first and second parts are in contact, or when the first and second parts are not in direct contact with each other.

[0160] As used in this patent, a statement that any part (e.g., layer, film, region, area, or plate) is located on another part in any way (e.g., positioned on it, situated on it, disposed on it, or formed on it, etc.) indicates that the referenced part is in contact with the other part, or that the referenced part is above the other part, wherein one or more intermediate parts are located between them.

[0161] As used herein, unless otherwise indicated, a connection reference (e.g., attachment, coupling, linking, and linking) may include intermediate components between the elements referenced by the connection reference and / or relative movement between these elements. Therefore, a connection reference does not necessarily imply that two elements are directly connected and / or fixed to each other. As used herein, a statement that any part is “in contact” with another part is limited to meaning that there is no intermediate part between the two parts.

[0162] Unless otherwise specifically stated, descriptors such as “first,” “second,” and “third” as used herein do not in any way impose or otherwise indicate any meaning of priority, physical order, arrangement in a list, and / or sorting, but are merely used as labels and / or arbitrary names to distinguish elements for the purpose of understanding the disclosed examples. In some examples, the descriptor “first” may be used to refer to an element in a detailed description, while in the claims, different descriptors such as “second” or “third” may refer to the same element. In such cases, it should be understood that such descriptors are only used to clearly identify these elements in the context of the discussion (e.g., within the claims), in which the elements may otherwise share the same name.

[0163] As used herein, “approximately” and “about” modify their subject / value to identify the potential presence of variations that occur in real-world applications. For example, “approximately” and “about” may modify dimensions that may be imprecise due to manufacturing tolerances and / or other real-world defects, as will be understood by one of ordinary skill in the art. For example, unless otherwise stated herein, “approximately” and “about” may indicate that such dimensions are within tolerances of + / - 10%.

[0164] As used in this article, “substantially real-time” means occurring in a near-instantaneous manner, recognizing that there may be real-world delays for calculations, transmissions, etc. Therefore, unless otherwise stated, “substantially real-time” means real-time + / - 1 second.

[0165] As used herein, the phrase “to communicate” (including its variations) encompasses direct communication and / or indirect communication via one or more intermediate components, and does not require direct physical (e.g., wired) communication and / or continuous communication, but additionally includes selective communication at periodic intervals, scheduled intervals, aperiodic intervals and / or one-off events.

[0166] As used herein, “programmable circuit system” is defined to include: (i) one or more special-purpose circuits (e.g., application-specific integrated circuits (ASICs)) that are structured to perform specific operations and include one or more semiconductor-based logic devices (e.g., electrical hardware implemented by one or more transistors); and / or (ii) one or more general-purpose semiconductor-based circuits that can be programmed with instructions to perform specific functions and / or operations and include one or more semiconductor-based logic devices (e.g., electrical hardware implemented by one or more transistors). Examples of programmable circuit systems include programmable microprocessors, such as a central processing unit (CPU) that executes first instructions to perform one or more operations and / or functions, a field-programmable gate array (FPGA) (which can be programmed with second instructions to instantiate one or more operations and / or functions corresponding to the first instructions by configuring and / or structuring the FPGA), a graphics processing unit (GPU) that executes first instructions to perform one or more operations and / or functions, a digital signal processor (DSP) that executes first instructions to perform one or more operations and / or functions, an XPU, a network processing unit (NPU), one or more microcontrollers that execute first instructions to perform one or more operations and / or functions, and / or integrated circuits such as application-specific integrated circuits (ASICs). For example, an XPU may be implemented by a heterogeneous computing system that includes multiple types of programmable circuit systems (e.g., one or more FPGAs, one or more CPUs, one or more GPUs, one or more NPUs, one or more DSPs, etc., and / or any combination thereof) and orchestration techniques (e.g., application programming interfaces (APIs)) that can assign computational tasks to one or more types of programmable circuit systems suitable for and available for performing the computational tasks.

[0167] As used herein, an integrated circuit / circuit system is defined as one or more semiconductor packages containing one or more circuit elements, such as transistors, capacitors, inductors, resistors, current paths, diodes, etc. For example, an integrated circuit can be implemented as one or more of an ASIC, FPGA, chip, microchip, programmable circuit system, semiconductor substrate connecting multiple circuit elements, system-on-a-chip (SoC), etc.

[0168] As can be seen from the foregoing, example systems, apparatuses, articles, and methods for estimating brake pad wear in vehicles have been disclosed. The examples disclosed herein obtain sensor data (e.g., ambient temperature data, torque data, and / or speed data) from one or more vehicle sensors and calculate example temperature data and example dynamic data corresponding to the respective brakes and / or brake pads of the vehicle based on said sensor data. The disclosed examples generate example data matrices (e.g., histogram matrices) based on the temperature and dynamic data, and then store and / or transmit said data matrices to an example cloud-based environment. In some examples, by using data matrices to store and / or transmit data, the disclosed examples reduce the utilization of computational resources (e.g., memory, bandwidth) used for storage and / or transmission. Furthermore, the disclosed examples generate and / or train one or more example brake wear prediction models for estimating brake pad wear based on the data matrices in a cloud-based environment. For example, the brake wear prediction model is a Physical Information Neural Network (PINN) trained based on labeled training data and on known and / or expected material wear dynamics (e.g., represented using one or more physics-based equations). In some examples, by training the model based on known and / or expected dynamics, the disclosed examples can improve the accuracy of the model's output predictions when the model's inputs are outside the range represented in the training data. Therefore, the disclosed examples can reduce the amount of training data required to generate and / or train the model, thus reducing the utilization of computational resources (e.g., memory, bandwidth) used to store and / or transmit the training data. Consequently, the disclosed systems, apparatus, articles of art, and methods improve the efficiency of using computing devices.

[0169] Furthermore, the examples disclosed herein can perform models to predict and / or estimate brake wear metrics (e.g., mass and / or width) of brake pads. These brake wear metrics can be used to inform and / or plan brake pad maintenance activities (e.g., replacement). Therefore, the disclosed examples can reduce premature brake pad replacement and / or reduce the degradation of one or more brake discs of a vehicle due to delayed and / or postponed brake pad replacement. Thus, the disclosed systems, apparatus, articles, and methods relate to one or more improvements in the operation of machines or other electronic and / or mechanical devices.

[0170] This document discloses example methods, apparatus, systems, and articles for estimating brake pad wear. Further examples, and combinations thereof, include the following:

[0171] Example 1 includes an apparatus comprising an interface circuitry system, machine-readable instructions, and at least one processor circuitry programmed by the machine-readable instructions to: obtain temperature data and power data associated with brake pads of a vehicle; execute a neural network based on the temperature data and the power data, the neural network being trained based on a loss value corresponding to the difference between: (a) a first rate of change of the output of the neural network when the neural network is executed based on training data, the first rate of change being evaluated at a first temperature and a first power; and (b) a second rate of change proportional to the ratio between the first temperature and the first power; determine a brake wear metric corresponding to the brake pads based on the execution result of the neural network; and cause the brake pad metric to be presented via a user interface.

[0172] Example 2 includes the device of Example 1, wherein the brake wear metric includes at least one of the following: the width of the brake pad, the mass of the brake pad, or a change in at least one of the width or mass.

[0173] Example 3 includes the device of Example 1, wherein one or more of the at least one processor circuitry is used to determine the remaining service life of the brake pads based on the brake wear metric.

[0174] Example 4 includes the device of Example 1, wherein the loss value is a first loss value, the difference is a first difference, and wherein one or more of the at least one processor circuitry is configured to: determine a second loss value based on a second difference between the output of the neural network and a measured value included in the training data; and adjust the weights of the neural network based on a combination of the first loss value and the second loss value.

[0175] Example 5 includes the device of Example 1, wherein the second rate of change is based on the material of the brake pads and the vehicle speed.

[0176] Example 6 includes the device of Example 1, wherein one or more of the at least one processor circuitry is configured to obtain the temperature data and the power data based on a matrix generated at the vehicle, the matrix values ​​of the matrix corresponding to a first interval value and a second interval value, the first interval value corresponding to the temperature data and the second interval value corresponding to the power data.

[0177] Example 7 includes the device of Example 1, wherein one or more of the at least one processor circuitry is used to cause the presentation of the brake wear metric when the brake wear metric does not meet a threshold.

[0178] Example 8 includes at least one non-transitory machine-readable medium comprising machine-readable instructions such that at least one processor circuitry at least: obtains temperature data and power data associated with brake pads of a vehicle; executes a neural network based on the temperature data and the power data, the neural network being trained based on a loss value corresponding to the difference between: (a) a first rate of change of the output of the neural network when the neural network is executed based on training data, the first rate of change being evaluated at a first temperature and a first power, and (b) a second rate of change proportional to the ratio between the first temperature and the first power; determines a brake wear metric corresponding to the brake pads based on the execution result of the neural network; and causes the brake pad metric to be presented via a user interface.

[0179] Example 9 includes at least one non-transitory machine-readable medium of Example 8, wherein the brake wear metric includes at least one of the following: the width of the brake pad, the mass of the brake pad, or a change in at least one of the width or mass.

[0180] Example 10 includes at least one non-transitory machine-readable medium of Example 8, wherein the machine-readable instructions cause one or more of the at least one processor circuitry to determine the remaining service life of the brake pad based on the brake wear metric.

[0181] Example 11 includes at least one non-transitory machine-readable medium of Example 8, wherein the loss value is a first loss value, the difference is a first difference, and wherein the machine-readable instructions cause one or more of the at least one processor circuitry to: determine a second loss value based on a second difference between the output of the neural network and a measurement value included in the training data; and adjust the weights of the neural network based on a combination of the first loss value and the second loss value.

[0182] Example 12 includes at least one non-transitory machine-readable medium of Example 8, wherein the second rate of change is based on the material of the brake pad and the vehicle speed.

[0183] Example 13 includes at least one non-transitory machine-readable medium of Example 8, wherein the machine-readable instructions cause one or more of the at least one processor circuitry to obtain the temperature data and the power data based on a matrix generated at the vehicle, the matrix values ​​of which correspond to a first interval value and a second interval value, the first interval value corresponding to the temperature data and the second interval value corresponding to the power data.

[0184] Example 14 includes at least one non-transitory machine-readable medium of Example 8, wherein the machine-readable instructions cause one or more of the at least one processor circuitry to present the brake wear metric when the brake wear metric does not meet a threshold.

[0185] Example 15 includes a method comprising: obtaining temperature data and power data associated with brake pads of a vehicle; executing a neural network based on the temperature data and the power data, the neural network being trained based on a loss value corresponding to the difference between: (a) a first rate of change of the output of the neural network when the neural network is executed based on training data, the first rate of change being evaluated at a first temperature and a first power, and (b) a second rate of change proportional to the ratio between the first temperature and the first power; determining a brake wear metric corresponding to the brake pads based on the execution result of the neural network; and causing the brake pad metric to be presented via a user interface.

[0186] Example 16 includes the method of Example 15, wherein determining the brake wear metric includes determining at least one of the following: the width of the brake pad, the mass of the brake pad, or a change in at least one of the width or mass.

[0187] Example 17 includes the method of Example 15, which further includes determining the remaining service life of the brake pad based on the brake wear metric.

[0188] Example 18 includes the method of Example 15, wherein the loss value is a first loss value, the difference is a first difference, and the method further includes: determining a second loss value based on a second difference between the output of the neural network and a measurement value included in the training data; and adjusting the weights of the neural network based on a combination of the first loss value and the second loss value.

[0189] Example 19 includes the method of Example 15, wherein the second rate of change is based on the material of the brake pad and the vehicle speed.

[0190] Example 20 includes the method of Example 15, the method further comprising: obtaining the temperature data and the power data based on a matrix generated at the vehicle, the matrix values ​​of the matrix corresponding to a first interval value and a second interval value, the first interval value corresponding to the temperature data and the second interval value corresponding to the power data.

[0191] The appended claims are hereby incorporated by reference into this specific embodiment. While certain example systems, apparatuses, articles of manufacture, and methods have been disclosed herein, the scope of this patent is limited to…

[0192] The scope is not limited to this. Instead, this patent covers all items that fall entirely within the scope of the claims of this patent.

[0193] It includes systems, equipment, products, and methods.

Claims

1. An apparatus comprising: Interface circuit system; Machine-readable instructions; as well as At least one processor circuit, said at least one processor circuit being programmed by said machine-readable instructions to: Obtain temperature and dynamic data associated with the vehicle's brake pads; A neural network is executed based on the temperature data and the power data, the neural network being trained based on a loss value corresponding to the difference between the two: (a) a first rate of change of the output of the neural network when the neural network is executed based on the training data, the first rate of change being evaluated at a first temperature and a first power. And (b) a second rate of change that is proportional to the ratio between the first temperature and the first power; Based on the execution results of the neural network, a brake wear metric corresponding to the brake pad is determined; and This causes the brake pad measurements to be displayed via the user interface.

2. The device of claim 1, wherein the brake wear measurement includes at least one of the following: the width of the brake pad, the mass of the brake pad, or a change in at least one of the width or the mass.

3. The device as claimed in any one of claims 1 or 2, wherein one or more of the at least one processor circuitry is used to determine the remaining service life of the brake pads based on the brake wear metric.

4. The device as claimed in any one of claims 1 to 3, wherein the loss value is a first loss value, the difference is a first difference, and wherein one or more of the at least one processor circuitry is used for: A second loss value is determined based on a second difference between the output of the neural network and a measured value included in the training data; and The weights of the neural network are adjusted based on a combination of the first loss value and the second loss value.

5. The device as claimed in any one of claims 1 to 4, wherein the second rate of change is based on the material of the brake pads and the vehicle speed.

6. The device of any one of claims 1 to 5, wherein one or more of the at least one processor circuitry is configured to obtain the temperature data and the power data based on a matrix generated at the vehicle, wherein matrix values ​​of the matrix correspond to a first interval value and a second interval value, the first interval value corresponding to the temperature data and the second interval value corresponding to the power data.

7. The device of any one of claims 1 to 6, wherein one or more of the at least one processor circuitry is configured to cause the presentation of the brake wear measurement when the brake wear measurement does not meet a threshold.

8. The device as claimed in any one of claims 1 to 7, wherein the temperature data is first temperature data, the power data is first power data, and the brake pad is a first brake pad, and one or more of the at least one processor circuit is configured to obtain second temperature data and second power data associated with the second brake pad of the vehicle.

9. A method comprising: Obtain temperature and dynamic data associated with the vehicle's brake pads; A neural network is executed based on the temperature data and the power data, the neural network being trained based on a loss value corresponding to the difference between the two: (a) a first rate of change of the output of the neural network when the neural network is executed based on the training data, the first rate of change being evaluated at a first temperature and a first power. And (b) a second rate of change that is proportional to the ratio between the first temperature and the first power; The brake wear metric corresponding to the brake pad is determined based on the execution result of the neural network. as well as This causes the brake pad measurements to be presented via the user interface.

10. The method of claim 9, wherein determining the brake wear metric comprises determining at least one of the following: the width of the brake pad, the mass of the brake pad, or a change in at least one of the width or the mass.

11. The method of any one of claims 9 or 10, further comprising determining the remaining service life of the brake pads based on the brake wear metric.

12. The method of any one of claims 9 to 11, wherein the loss value is a first loss value, the difference is a first difference, and the method further comprises: A second loss value is determined based on a second difference between the output of the neural network and a measured value, the measured value being included in the training data; as well as The weights of the neural network are adjusted based on a combination of the first loss value and the second loss value.

13. The method of any one of claims 9 to 12, wherein the second rate of change is based on the material of the brake pad and the vehicle speed.

14. The method of any one of claims 9 to 13, further comprising obtaining the temperature data and the power data based on a matrix generated at the vehicle, wherein matrix values ​​of the matrix correspond to a first interval value and a second interval value, the first interval value corresponding to the temperature data and the second interval value corresponding to the power data.

15. The method of any one of claims 9 to 14, further comprising causing the presentation of the brake wear metric when the brake wear metric does not meet the threshold.

Images