System and method for detecting road surface conditions

Abstract

The invention relates to a system and method for detecting road surface conditions. Various embodiments are disclosed for predicting road surface conditions based at least in part on tire resistance and vehicle data acquired from a controller area network (CAN) bus of a vehicle. In response to determining that the vehicle is in a steady state, traction may be estimated using vehicle data. Traction and vehicle speed may be input into the trained surface condition detection model. The surface condition detection model is trained to predict a probability that the surface is wet or dry based at least in part on the traction. The probability that the surface is wet or dry may be provided to other vehicle systems in order to adjust vehicle operation.

Description

Technical Field

[0001] The present disclosure relates to a system and method for detecting road surface conditions. Background Art

[0002] Road surface conditions may affect how a vehicle travels across the road surface. For example, road surface conditions such as rain, snow, gravel, or ice may affect the vehicle's traction, stability, braking distance, suspension, efficiency, and other operating characteristics of the vehicle. Knowing the road surface conditions may allow the vehicle control system to adjust accordingly to compensate for specific types of road surface conditions so that the vehicle operates safely, stably, and efficiently. Summary of the invention

[0003] The present invention includes the following technical solutions.

[0004] Solution 1. A method for determining a road surface condition, comprising:

[0005] acquiring, by at least one computing device, vehicle data collected from a vehicle controller area network (CAN) bus in communication with one or more vehicle systems of a vehicle traveling along a surface;

[0006] determining, by the at least one computing device, that the vehicle is in a stable state based at least in part on the vehicle data; and

[0007] In response to determining that the vehicle is operating in the stable condition:

[0008] estimating, by the at least one computing device, traction associated with at least one tire of the vehicle and the surface based at least in part on the vehicle data;

[0009] applying, by the at least one computing device, the traction and vehicle speed of the vehicle to a surface condition detection model; and

[0010] A probability of the surface being wet or dry is determined by the at least one computing device based at least in part on an output of the surface condition detection model.

[0011] Option 2. The method according to Option 1, wherein the vehicle data includes engine torque, engine revolutions per minute (RPM), vehicle speed, wheel speed, and vehicle acceleration.

[0012] Option 3. The method according to Option 2, wherein determining that the vehicle is operating in the stable state further comprises:

[0013] determining that the engine RPM exceeds an RPM threshold;

[0014] determining that the vehicle speed exceeds a speed threshold; and

[0015] It is determined that the vehicle acceleration exceeds an acceleration threshold.

[0016] Option 4. A method according to Option 1, wherein the surface condition detection model includes a trained classifier.

[0017] Solution 5. The method according to Solution 1, further comprising:

[0018] Determine total traction;

[0019] Determine road grade forces; and

[0020] Determine aerodynamic drag, and

[0021] The traction force or net drag is estimated by subtracting the road grade force and the aerodynamic drag from the total traction force.

[0022] Option 6. The method according to Option 1 further includes transmitting the probability to at least one vehicle system among the one or more vehicle systems, wherein the at least one vehicle system is configured to adjust the operation of the vehicle based at least in part on the probability.

[0023] Option 7. The method according to Option 1, wherein the at least one computing system is located within the vehicle or remotely located from the vehicle.

[0024] Solution 8. A road surface detection system, comprising:

[0025] a computing device comprising a processor and a memory; and

[0026] The machine-readable instructions stored in the memory, when executed by the processor, cause the computing device to at least:

[0027] acquiring vehicle data collected from a vehicle controller area network (CAN) bus in communication with one or more vehicle systems of the vehicle traveling along the surface;

[0028] determining that the vehicle is in a stable operating state based at least in part on the vehicle data; and

[0029] In response to determining that the vehicle is operating in a stable condition:

[0030] estimating at least one of the vehicle's

[0031] the traction associated with each tire and the surface;

[0032] The traction force and vehicle speed of the vehicle are applied to the surface condition detection model

[0033] type; and

[0034] A probability that the surface is wet or dry is determined based at least in part on an output of the surface condition detection model.

[0035] Option 9. The system of Option 8, wherein the vehicle data includes engine torque, engine revolutions per minute (RPM), vehicle speed, wheel speed, and vehicle acceleration.

[0036] Option 10. The system of Option 9, wherein determining that the vehicle is operating in a stable state further comprises:

[0037] determining that the engine RPM exceeds an RPM threshold;

[0038] determining that the vehicle speed exceeds a speed threshold; and

[0039] It is determined that the vehicle acceleration exceeds an acceleration threshold.

[0040] Option 11. A system according to Option 8, wherein the surface condition detection model includes a trained classifier.

[0041] Option 12. The system of Option 8, wherein the machine-readable instructions further cause the computing device to at least:

[0042] Determine total traction;

[0043] Determine road grade forces; and

[0044] Determine the aerodynamic drag,

[0045] The traction force is estimated by subtracting the road grade force and the aerodynamic drag from the total traction force.

[0046] Option 13. A system according to Option 8, wherein the machine-readable instructions also cause the computing device to transmit the probability to at least at least one vehicle system among the one or more vehicle systems, and the at least one vehicle system is configured to adjust the operation of the vehicle at least in part based on the probability.

[0047] Option 14. A system according to Option 8, wherein the at least one computing system is located within the vehicle or remotely located from the vehicle.

[0048] Embodiment 15. A non-transitory computer-readable medium comprising machine-readable instructions that, when executed by a processor of a computing device, cause the computing device to at least:

[0049] acquiring vehicle data collected from a vehicle controller area network (CAN) bus in communication with one or more vehicle systems of the vehicle traveling along the surface;

[0050] determining, based at least in part on the vehicle data, that the vehicle is operating in a steady state; and

[0051] In response to determining that the vehicle is operating in the stable condition:

[0052] estimating traction associated with at least one tire of the vehicle and the surface based at least in part on the vehicle data;

[0053] applying the traction force and vehicle speed of the vehicle to the surface condition detection model; and

[0054] A probability that the surface is wet or dry is determined based at least in part on an output of the surface condition detection model.

[0055] Option 16. The non-transitory computer-readable medium of Option 15, wherein the vehicle data comprises engine torque, engine revolutions per minute (RPM), vehicle speed, wheel speed, and vehicle acceleration.

[0056] Embodiment 17. The non-transitory computer-readable medium of embodiment 16, wherein determining that the vehicle is operating in the stable state further comprises:

[0057] Determining that the engine RPM exceeds a RPM threshold;

[0058] determining that the vehicle speed exceeds a speed threshold; and

[0059] It is determined that the vehicle acceleration exceeds an acceleration threshold.

[0060] Embodiment 18. The non-transitory computer-readable medium of Embodiment 15, wherein the surface condition detection model comprises a trained classifier.

[0061] Option 19. A non-temporary computer-readable medium according to Option 15, wherein, when the processor executes the machine-readable instructions, it also causes the computing device to at least: transmit the probability to at least one vehicle system among the one or more vehicle systems, and the at least one vehicle system is configured to adjust the operation of the vehicle at least in part based on the probability.

[0062] Embodiment 20. The non-transitory computer-readable medium of embodiment 15, wherein when the machine-readable instructions are executed by the processor, the computing device is further caused to at least:

[0063] Determine total traction;

[0064] Determine road grade forces; and

[0065] Determine the aerodynamic drag,

[0066] The traction force is estimated by subtracting the road grade force and the aerodynamic drag from the total traction force. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Many aspects of the present disclosure may be better understood with reference to the following drawings. The components in the drawings are not necessarily drawn to scale, but emphasis is placed on clearly illustrating the principles of the present disclosure. Furthermore, in the drawings, the same reference numerals designate corresponding parts throughout the several views.

[0068] Figure 1A An example perspective view of a vehicle supported by tires is illustrated in accordance with various embodiments.

[0069] Figure 1B Pictured Figure 1A An example diagram of a top view of a vehicle illustrating vehicle components according to various embodiments of the present disclosure.

[0070] Figure 2A and Figure 2B is an example diagram of a network environment according to various embodiments of the present disclosure.

[0071] Figure 3 is a diagram illustrating known longitudinal vehicle dynamics associated with a vehicle.

[0072] Figure 4 is an example graphical representation illustrating the relationship between tire traction (under steady-state cruising conditions) versus vehicle speed on dry and wet surfaces in accordance with various embodiments of the present disclosure.

[0073] Figure 5A , 5B 5C are example graphical representations illustrating the relationship between tire traction (under steady state cruising conditions) versus vehicle speed on dry and wet surfaces for different tire type configurations in accordance with various embodiments of the present disclosure.

[0074] Figure 6 is an example graphical representation illustrating the relationship between tire traction (under steady state cruising conditions) versus vehicle speed on dry and wet surfaces for both new and fully worn tires in accordance with various embodiments of the present disclosure.

[0075] Figure 7 is an example graphical representation corresponding to the training of a surface condition detection model for classifying road conditions in accordance with various embodiments of the present disclosure.

[0076] Figure 8is an exemplary schematic diagram according to various embodiments of the present disclosure, which illustrates Figure 2A and 2B Some functions of applications executed in a computing environment in a network environment.

[0077] Fig.9A , 9B 9C are example graphical representations associated with real-time testing of a surface condition detection model trained to predict wet road conditions or dry road conditions. Fig.9A Illustrated are graphical representations associated with various data parameters that may be applied as input to a surface condition detection model. Fig. 9B A graphical representation associated with the predictions of a surface condition detection model relative to ground truth is illustrated. Fig. 9C A graphical representation of model confidence for a surface condition detection model is illustrated.

[0078] Fig.10 is a flowchart illustrating various embodiments according to the present disclosure, which illustrates the Figure 2A and 2B An example of a function implemented partially by an application executed in a computing environment in a network environment. DETAILED DESCRIPTION

[0079] Various methods for predicting road surface conditions based at least in part on tire resistance are disclosed. In various examples, the road surface condition may be predicted based at least in part on an estimate of tire resistance using vehicle data obtained from a controller area network (CAN) bus of the vehicle. For example, the vehicle data may include engine torque, engine revolutions per minute (RPM), vehicle speed, wheel speed, vehicle acceleration, and / or other types of vehicle-based data that may be obtained from the vehicle's CAN bus data. When it is determined that the vehicle is traveling in a steady state, traction may be estimated using the CAN bus data. The traction, along with the vehicle speed, may be applied as input to a trained road condition detection model. The output of the road condition detection model may include a probability that the road condition is wet or dry. For example, when the road is wet, the tire resistance is greater. Therefore, the probability that the road condition is wet or dry may be based at least in part on the real-time estimated tire resistance. Other vehicle control systems may use the probability and / or prediction that the road condition is wet or dry to make appropriate adjustments based on the type of road condition.

[0080] In the following discussion, a general description of the system and its components is provided, followed by a discussion of its operation. Although the following discussion provides illustrative examples of the operation of various components of the present disclosure, the use of the following illustrative examples does not exclude other embodiments consistent with the principles disclosed by the following illustrative examples.

[0081] Now turn to Figure 1A and 1B, showing an example view of a vehicle 103 according to various embodiments. Figure 1A An example perspective view of a vehicle 103 supported by a plurality of tires 106 is illustrated in accordance with various embodiments. Figure 1B An example diagram of a top view of a vehicle 103 is illustrated, which illustrates vehicle components, including a tire 106, a sensor unit 109 attached to the tire 106, a vehicle computing device 112, and a CAN bus 115. The vehicle computing system 112 and the CAN bus 115 are related to Figure 2A and 2B for a more detailed discussion.

[0082] The tires 106 are of conventional construction and are each mounted on a corresponding wheel, as known to those skilled in the art. Each tire 106 includes a pair of sidewalls extending to a circumferential tread. An inner liner is disposed on the inner surface of the tire 106 and, when the tire is mounted on a wheel 118, forms an internal cavity that is filled with a pressurized fluid (such as air).

[0083] In various examples, the sensor unit 109 is mounted on or otherwise attached to each tire 106 to detect certain real-time parameters of the tire 106 (such as tire pressure and tire temperature). In various examples, the sensor unit 109 may include at least one of the following: a pressure sensor to sense the inflation pressure in the cavity; a temperature sensor to measure the temperature of the tire; an accelerometer to measure the acceleration of the wheel 118 on which the tire 106 is mounted; a tachometer to measure the rotation time of the wheel 118; and / or other types of sensors. The sensor unit 109 may be part of a commercially available tire pressure monitoring system (TPMS) module or sensing unit.

[0084] In various examples, the sensor unit 109 includes a processor and a memory to store vehicle tire information for each specific tire 106. For example, the tire vehicle tire information may include a tire identifier (ID), manufacturing information of the tire 106 (e.g., model, manufacturer name, etc.), size information (e.g., rim size, width, and outer diameter), manufacturing location, manufacturing date, tread cap code including or associated with a compound identification, mold code including or associated with a tread structure identification, and / or other information. The vehicle tire information may also include a maintenance history or other information to identify specific features and parameters of each tire 106. Alternatively or additionally, the vehicle tire information may be included in another sensor unit 109, or in a separate vehicle storage medium (such as a tire ID tag), which is preferably in electronic communication with the sensor unit 109.

[0085] The sensor unit 109 also includes an antenna (not shown) to wirelessly transmit the measured parameters and tire information data to a remote processor (such as a processor integrated into the vehicle computing device 112, a controller area network (CAN) bus 115 associated with the vehicle 103, and / or a cloud computing device) for analysis. According to various embodiments, each tire 106 on the vehicle 103 may include one or more sensor units 109.

[0086] Although Figure 1A and 1B The vehicle 103 in FIG. 1 is depicted as a passenger car, but the vehicle 103 may include any type of vehicle that uses tires, and a passenger car is shown as an example here. To this end, the vehicle 103 may include other vehicles belonging to various categories (such as commercial trucks and trailers, off-road vehicles, etc.), wherein such vehicles 103 include a greater number of tires 106 than the vehicle 103. Figure 1A and 1B A greater or lesser number of tires 106 may be shown.

[0087] Now turn to Figure 2A and 2B , illustrating an example networked environment 200 (eg, 200a, 200b) according to various embodiments. Figure 2A The networked environment 200a including the vehicle computing environment 203 is illustrated. The vehicle computing environment 203 may include the vehicle computing device 112, the CAN bus 115, one or more vehicle electronic systems 206, and the sensor unit 109, which may communicate data with each other via one or more networks. The vehicle computing device 112 includes a processor circuit that executes applications associated with the control and / or operation of the vehicle 103. For example, in Figure 2A In the embodiment of the present invention, the processor circuit of the vehicle computing device 112 may execute the surface condition detection service 209 and / or other applications.

[0088] The surface condition detection service 209 may be executed to analyze vehicle data 218 collected via the CAN bus 115 of the vehicle 103 to estimate traction and determine the surface condition associated with the surface on which the vehicle 103 is traveling. In various examples, the vehicle data 218 may include engine torque, engine revolutions per minute (RPM), vehicle speed, wheel speed, vehicle acceleration, and / or other types of vehicle-based data that may be obtained from the CAN bus 115 of the vehicle 103. In various examples, traction may be determined based at least in part on the engine torque, engine RPM, and vehicle speed, such as the surface condition associated with the surface on which the vehicle 103 is traveling. Figure 4 as discussed in further detail.

[0089] According to various examples, the surface condition detection service 209 may determine whether the vehicle 103 is in a steady state before estimating the traction required to determine the surface condition. In particular, the engine torque information contained in the vehicle data 218 is more reliable when the vehicle is considered to be in a steady state. In addition, estimating the force during steady state and constant speed may produce a high signal-to-noise ratio. In various examples, a steady state of the vehicle 103 may occur when the engine RPM is within a predetermined RPM threshold, the vehicle speed is within a predetermined speed threshold, and the vehicle acceleration is within a predetermined acceleration threshold. If all three conditions are met, the surface condition detection service 209 may trigger or otherwise activate the calculation of the traction estimate.

[0090] In various examples, the surface condition detection service 209 estimates traction based at least in part on the total traction, the road grade estimate, and aerodynamic drag. According to various examples, the traction used to predict the road surface condition is the total traction minus the road grade estimate and aerodynamic drag.

[0091] In various examples, the surface condition detection service 209 may execute a surface condition detection model 230 that is trained to predict or otherwise output a probability that a surface is wet or dry. In various examples, the surface condition detection model 230 is trained to predict or otherwise output a probability associated with a surface condition based at least in part on estimated traction and vehicle speed. In various examples, upon obtaining the output of the surface condition detection model 230, the surface condition detection service 209 may transmit the road condition probabilities to one or more vehicle electronic systems 206 or other applications within the vehicle computing device 112, which may be configured to adjust the operation of one or more components of the vehicle 103 based at least in part on the road condition probabilities.

[0092] In various examples, the vehicle computing device 112 may also include a receiver 212 to obtain sensor parameter data 215 transmitted from the sensor unit 109 on the tire 106 of the vehicle 103, vehicle data 218 obtained from the CAN bus 115 associated with the vehicle 103, and / or other data accessible via a network.

[0093] Furthermore, various data is stored in a vehicle data store 233, which is accessible by the vehicle computing device 112. As may be appreciated, the data store 233 may represent a plurality of data stores 233. For example, the data stored in the data store 233 is associated with the operation of various applications and / or functional entities associated with the vehicle 103. For example, the data store 233 may include vehicle tire data 236, vehicle data 218, surface condition detection model 230, surface condition detection rules 239, and / or other information.

[0094] The vehicle tire data 236 may include information for each specific tire 106. For example, the vehicle tire data 236 may include a tire identifier, manufacturing information for the tire 106 (e.g., manufacturer name, tire model, etc.), tire size information (e.g., rim size, width, and outer diameter, etc.), manufacturing location, manufacturing date, a tread cap code including or associated with a compound identification, a mold code including or associated with a tread structure identification, and / or other information. The vehicle tire data 236 may also include a maintenance history or other information to identify specific features and parameters of each tire 106. The vehicle tire data 236 may also include a sensor unit ID 242, sensor parameter data 215, and / or other data.

[0095] The sensor unit ID 242 may include an alphanumeric identifier that may be used to identify the sensor unit 109 on a given tire 106. In some examples, the sensor unit ID 242 may include additional information about the sensor unit 109, including the location of the tire, on which axle the tire 106 is mounted, on which side of the vehicle 103 the tire 106 is located (e.g., left, right), and / or other information. The sensor parameter data 215 may include data collected from the sensor unit 109 on a given tire 106. As previously discussed, the sensor unit 109 may include at least one of a pressure sensor, a temperature sensor, an accelerometer, a tachometer, and / or other types of sensors. Accordingly, the sensor parameter data 215 may include temperature data, pressure data, accelerometer data, tachometer data, and / or other types of data.

[0096] The vehicle data 218 may include data associated with the vehicle 103 supported by the tires 106. The vehicle data 218 may be obtained from a vehicle CAN bus 115 that communicates with one or more vehicle systems 206 of the vehicle 103 supported by the tires 106. The vehicle data 218 may include engine torque, engine revolutions per minute (RPM), vehicle speed, wheel speed, vehicle acceleration, vehicle load, odometer values, brake cylinder pressure values, and / or other types of vehicle data.

[0097] The surface condition detection model 230 is a machine learning model that is trained to predict the probability of a surface being wet or dry based at least in part on traction and vehicle speed. The surface condition detection model 230 may include a decision tree classifier, a gradient boosting classifier, a Gaussian naive Bayes classifier, a reinforcement learning algorithm, a logistic regression classifier, a random forest classifier, a multi-layer perceptron classifier, a recurrent neural network, a neural network, a label-specific attention network, an ensemble model, and / or any other type of training model that can be understood. Although the surface condition detection model 230 may include a neural network, it is noted that the computational overhead associated with the neural network can be minimized by implementing a non-neural network classifier model.

[0098] The surface condition detection rules 239 include rules, models, and / or configuration data for various algorithms or methods employed by the surface condition detection service 209 and / or other applications or devices. For example, the surface condition detection rules 239 may include predetermined thresholds (e.g., RPM thresholds, speed thresholds, acceleration thresholds) associated with determining whether the vehicle is in a stable state. In addition, the surface condition detection rules 239 may include various models, formulas, equations, and / or algorithms for determining traction. In various examples, the surface condition detection rules 239 may include rules associated with which vehicle electronic systems 206 and / or other applications will be provided with predicted surface conditions of the road on which the vehicle 103 is traveling.

[0099] In various examples, the vehicle computing device 112 may include a communication system to facilitate communication with vehicle components of the vehicle 103, via the network 224 ( Figure 2B )'s remote computing environment 221( Figure 2B )、client device 227( Figure 2B ) or other types of computing devices (e.g., obtaining sensor parameter data 215 transmitted from the sensor unit 109 on the tire 106, vehicle data 218 obtained from the controller area network (CAN) bus 115 via one or more vehicle electronic systems 206 of the vehicle 203, etc.). In this regard, the vehicle computing device 112 may include appropriate communication capabilities to connect to a cellular network, a Wi-Fi network, network, microwave transmission network, radio broadcast network or other communications network.

[0100] although Figure 2A 103 as being separate from the vehicle electronic system 206, but in some examples, the vehicle computing device 112 can be integrated with other vehicle electronic systems 206 in the vehicle 103. In various examples, the vehicle computing device 112 can be coupled to the CAN bus 115 and can communicate with other vehicle electronic systems 206 included on the CAN bus 115.

[0101] In various examples, the vehicle electronic systems 206 may include electronic components of the vehicle 103 that are responsible for managing and regulating various operations of the vehicle 103. For example, the vehicle electronic systems 206 may include an engine control system, a transmission control system, a braking system, a suspension system, a steering system, and / or other types of vehicle systems. The vehicle electronic systems 206 may be coupled to the CAN bus 115 to allow communication between different vehicle electronic systems 206, as can be appreciated. In various examples, the vehicle electronic systems 206 may include processor circuitry to execute applications to adjust operational aspects associated with a particular vehicle electronic system 206.

[0102] refer to Figure 2B , illustrates an alternative network environment 200b according to various embodiments. Figure 2B and Figure 2A The difference is that the functionality associated with predicting the road surface condition occurs in the remote computing environment 221, rather than in the vehicle computing environment 203. In particular, the vehicle computing environment 203 can transmit the vehicle data 218 to the remote computing environment 221 in real time. In addition, the surface condition detection service 209 of the remote computing environment 221 can transmit the predicted road surface condition to the vehicle computing environment 203 and / or one or more vehicle electronic systems 206 of the vehicle. Figure 2B The network environment 200 b may include a remote computing environment 221 , a vehicle computing environment 203 , and a client device 227 , which may perform data communication with each other via a network 224 .

[0103] The network 224 may include a wide area network (WAN), a local area network (LAN), a personal area network (PAN), or a combination thereof. These networks may include wired or wireless components or a combination thereof. Wired networks may include Ethernet, cable networks, fiber optic networks, and telephone networks such as dial-up, digital subscriber line (DSL), and integrated services digital network (ISDN) networks. Wireless networks may include cellular networks, satellite networks, Institute of Electrical and Electronics Engineers (IEEE) 802.11 wireless networks (i.e., wireless networks). ), Networks 224 may include microwave transmission networks, microwave transmission networks, and other networks that rely on radio broadcasting. Network 224 may also include a combination of two or more networks 224. Examples of network 224 may include the Internet, an intranet, an extranet, a virtual private network (VPN), and the like.

[0104] Remote computing environment 221 may include one or more computing devices, including processors, memory, and / or network interfaces. For example, a computing device may be configured to perform computations on behalf of other computing devices or applications. As another example, such a computing device may host and / or provide content to other computing devices in response to requests for content.

[0105] Furthermore, the remote computing environment 221 may employ a plurality of computing devices, which may be arranged in one or more server groups or computer groups or other arrangements. Such computing devices may be located in a single facility or may be distributed in many different geographical locations. For example, the remote computing environment 221 may include a plurality of computing devices, which together may comprise a managed computing resource, a grid computing resource, or any other distributed computing arrangement. In some cases, the remote computing environment 221 may correspond to an elastic computing resource, in which the capacity of the allocated processing, network, storage, or other computing-related resources may vary over time.

[0106] Various applications or other functions may be executed in the remote computing environment 221. Components executing on the remote computing environment 221 include the vehicle management system 245, the surface condition detection service 209, and other applications, services, processes, systems, engines, or functions not discussed in detail herein.

[0107] A vehicle management system 245 may be implemented to track the location and status of vehicles 103. Such a vehicle management system 245 may track hundreds or even thousands of vehicles 103. The vehicle management system 245 indicates to an operator when a vehicle 103 may require maintenance, replacement, and / or other information.

[0108] Moreover, various data are stored in a remote data store 248, which can be accessed by a remote computing environment 221. The remote data store 248 can represent a plurality of remote data stores 248, which can include relational databases or non-relational databases, such as object-oriented databases, hierarchical databases, hash tables or similar key-value data stores, as well as other data storage applications or data structures. Furthermore, combinations of these databases, data storage applications and / or data structures can be used together to provide a single logical data store. The data stored in the remote data store 221 is associated with the operation of various applications or functional entities described below. The data may include vehicle tire data 236, vehicle data 218, surface condition detection model 230, surface condition detection rules 238, and possibly other data.

[0109] Client device 227 represents a plurality of client devices that can be connected to network 224. Client device 227 may include a processor-based system, such as a computer system. Such a computer system can be implemented in the form of a personal computer (e.g., a desktop computer, a laptop computer, or a similar device), a mobile computing device (e.g., a personal digital assistant, a cellular phone, a smart phone, a web tablet, a tablet computer system, a music player, a portable game console, an e-book reader, and a similar device), a media playback device (e.g., a media streaming device, a Blu-ray player, a The client device 227 may include one or more displays 251, such as a liquid crystal display (LCD), a gas plasma-based flat panel display, an organic light emitting diode (OLED) display, an electrophoretic electronic ink (“E-ink”) display, a projector, or other type of display device. In some cases, the display 251 may be a component of the client device 227, or may be connected to the client device 227 via a wired or wireless connection.

[0110] The client device 227 may be configured to execute various applications, such as the client application 254 or other applications. The client application 254 may be executed in the client device 227 to access network content provided by the remote computing environment 221, the vehicle computing environment 203, or other servers, thereby presenting a user interface 257 on the display 251. To this end, the client application 254 may include a browser, a dedicated application, or other executable file, and the user interface 257 may include a network page, an application screen, or other user mechanism for obtaining user input. The client device 227 may be configured to execute applications other than the client application 254, such as an email application, a social networking application, a word processor, a spreadsheet, or other applications.

[0111] Next, about Figure 3 -5, provides a general description of the operation of the various components of the network environment 200. First, Figure 3 The known longitudinal vehicle dynamics associated with vehicle 103 are illustrated. In particular, Figure 3 Different forces associated with the vehicle 103 traveling along a surface are illustrated, along with corresponding equations for determining traction based at least in part on the values ​​of the different forces. Figure 3 The forces shown in the figure include aerodynamic forces (F aero ), acceleration force (F acc ), driving force (F drive ), slope force (F grade ), rolling force (F roll ) and load (mg). Figure 3 As illustrated, traction corresponds to the sum of the force used to overcome the acceleration resistance of the car, the force used to overcome the gradient, the force used to overcome the rolling resistance, and the force associated with aerodynamic drag. In various examples, standard dynamics for calculating traction may be calculated.

[0112] With respect to prediction of road surface conditions, the surface condition detection model 230 is trained to account for the effect of water on tire rolling resistance relative to estimated traction and vehicle speed. For example, when a tire rolls on a wet road, the tire must move water out before directly contacting the road surface, resulting in additional rolling resistance. Similarly, water can affect the contact temperature of the tire 106, thereby affecting rolling resistance (e.g., lower temperatures produce higher rolling resistance).

[0113] Continue to Figure 4 , shows an example graphical representation illustrating the relationship between tire traction (under steady-state cruising conditions) versus vehicle speed on dry and wet surfaces. In particular, Figure 4 The figure shows that wet road surfaces produce higher drag. In additional tests, such as Figures 5A-5CAs shown, the same behavior (e.g., higher drag on wet road surfaces) was observed in all tire configurations tested. Figure 5A A graphical representation showing the relationship between tire traction on dry and wet surfaces versus vehicle speed for winter tires. Figure 5B A graphical representation of tire traction versus vehicle speed on dry and wet surfaces for an all-season tire. Figure 5C A graphical representation showing the relationship between tire traction versus vehicle speed on dry and wet surfaces for a summer tire.

[0114] Next reference Figure 6 , an example graphical representation is shown that illustrates the relationship between tire traction (under steady state cruising conditions) on dry and wet surfaces versus vehicle speed for both new and fully worn tires. Figure 6 As shown, the classification boundary between dry and wet conditions remains unchanged for new and worn tires.

[0115] Continue to Figure 7 , shows an example graphical representation corresponding to the training of the surface condition detection model 230. In particular, the surface condition detection model 230 includes Figure 7 A graphical representation of a machine learning model trained to correlate the relationship between tire traction on dry and wet surfaces (under steady-state cruising conditions) versus vehicle speed. Figure 7 In a non-limiting example, the surface condition detection model 230 may be trained as a classification model using the following non-limiting example parameters:

[0116] Coefficient.Im_icept = 4.550804

[0117] Coefficient.Im_slope1 = -0.266266

[0118] Coefficient.Im_slope2 = 0.03163149

[0119] X = coefficient.Im_icept*1+coefficient.Im_slope1*x1+coefficient.Im_slope2*x2

[0120] The output of the model (eg, sigmoid(x)) may correspond to the probability that the road surface is wet.

[0121] Next reference Figure 8, shows an example schematic diagram illustrating some functions of the surface condition detection service 209. According to various embodiments of the present disclosure, the total traction force can be calculated using the vehicle data 218 obtained from the vehicle 103. For example, the total traction force can be estimated using the engine torque, engine RPM, and vehicle speed information contained in the vehicle data 218. In various examples, the total traction force can be calculated as follows:

[0122]

[0123] Total drive train ratio = gear ratio * final drive ratio, where the gear ratio corresponds to the gear ratio of the engaged gears and the final drive ratio corresponds to the final drive ratio at the differential. Therefore, the following formula applies:

[0124]

[0125] Therefore, tire traction may be calculated using vehicle data 218 as follows:

[0126]

[0127] In various examples, the surface condition detection model 230 is trained using data assuming zero road grade force and zero aerodynamic drag. Thus, the road grade force may be estimated and the aerodynamic drag may be estimated so that the road grade force and the aerodynamic drag may be deducted from the calculated total traction. The road grade drag may be estimated based at least in part on an estimate of the road grade and an estimate of the vehicle mass. In various examples, the road grade may be estimated using vehicle acceleration and wheel speed, which may be included in the vehicle data 218. In some examples, the wheel speed and / or the vehicle acceleration may be included in the sensor parameter data 215 acquired from the sensor unit 109 of the tire 106. Similarly, the aerodynamic drag may be estimated based at least in part on the vehicle speed included in the vehicle data 218. The net drag (e.g., traction) corresponds to the total traction minus the estimated road grade force and aerodynamic drag. The net drag corresponding to the traction may be applied as an input to the surface condition detection model 230 together with the vehicle speed from the vehicle data 218.

[0128] However, as previously discussed, before estimating the traction required to determine the surface condition (e.g., net drag), the surface condition detection service 209 determines whether the vehicle 103 is in a stable state. In particular, the engine torque information contained in the vehicle data 218 is more reliable when the vehicle is considered to be in a stable state. In various examples, a stable state of the vehicle 103 may occur when the engine RPM is within a predetermined RPM threshold, the vehicle speed is within a predetermined speed threshold, and the vehicle acceleration is within a predetermined acceleration threshold. If all three conditions are met, the surface condition detection service 209 can trigger or otherwise activate the calculation of the traction estimation discussed above.

[0129] When estimating traction using vehicle data 218, surface condition detection service 209 can apply the estimated traction and vehicle speed as inputs to surface condition detection model 230. Surface condition detection model 230 can be trained and optimized to output road surface probability 403, which can be used to indicate whether the road surface is dry or wet. In various examples, probability 403 can be determined as follows:

[0130]

[0131] where x 1 =1 (e.g. intercept term), x 1 = vehicle speed (kph), and x 1 = traction force (N).

[0132] Go to Fig.9A , 9B 9C, shows an example graphical representation associated with real-time testing of the surface condition detection model 230. In particular, Fig.9A Illustrated are graphical representations associated with various data parameters that may be applied as input to the surface condition detection model 230. For example, the data parameters may include optical sensor surface type, vehicle speed, traction, road grade, and / or other characteristics. Fig. 9B 2 illustrates a graphical representation associated with the predictions of the surface condition detection model 230 relative to the ground truth. Fig. 9B As shown, the dots represent the moments when the estimator is activated and makes a prediction. The graphical representation illustrates a near-perfect classification accuracy of about 99%. Fig. 9C A graphical representation of the model confidence of the surface condition detection model 230 is illustrated.

[0133] Next reference Fig.10 , illustrates a flow chart that provides one example of the operation of a portion of the surface condition detection model 230 . Fig.10 The flowchart of provides only an example of many different types of functional arrangements that may be employed to implement the operation of the depicted portion of the surface condition detection model 230. As an alternative, Fig.10 The flowcharts of FIG. 200a and FIG. 200b may be viewed as examples depicting elements of a method implemented within the network environments 200a, 200b.

[0134] Starting from block 1003, the surface condition detection model 230 acquires the vehicle data 218. In various examples, the surface condition detection model 230 may communicate with the CAN bus 115 and receive the vehicle data 218 from the CAN bus 115. In other examples, the vehicle data 218 from the CAN bus 115 may be stored in the vehicle data memory 233 and / or transmitted and stored in the remote data memory 248, and the surface condition detection model 230 may acquire the vehicle data 218 from the vehicle data memory 233 and / or the remote data memory 248. In various examples, the vehicle data 218 is acquired in real time.

[0135] At box 1006, the surface condition detection model 230 determines whether the vehicle 103 is in a stable state. In various examples, the stable state of the vehicle 103 may occur when the engine RPM is within a predetermined RPM threshold, the vehicle speed is within a predetermined speed threshold, and the vehicle acceleration is within a predetermined acceleration threshold. In various examples, the thresholds may be included in the surface condition detection rules 239. The surface condition detection model 230 may compare the values ​​of the engine RPM, vehicle speed, and vehicle acceleration with corresponding thresholds to determine whether all three conditions are met. In some examples, the values ​​of the engine RPM, vehicle speed, and vehicle acceleration correspond to window standard deviation values ​​over a period of time. If all three conditions are met, the surface condition detection service 209 may trigger or otherwise activate the calculation of the traction estimation and continue to box 1009. Otherwise, the surface condition detection service 209 returns to box 1003.

[0136] At box 1009, the surface condition detection service 209 determines traction based at least in part on the vehicle data 218. In various examples, the traction is estimated by first estimating the total traction using the engine torque, engine RPM, and vehicle speed information contained in the vehicle data 218. In various examples, the surface condition detection model 230 is trained using data assuming zero road grade force and zero acceleration. Therefore, the road grade force can be estimated and the aerodynamic drag can be estimated so that the road grade force and the aerodynamic drag can be deducted from the calculated total traction. The road grade force can be estimated based at least in part on an estimate of the road grade and an estimate of the vehicle mass. In various examples, the road grade can be estimated using the vehicle acceleration and wheel speed that can be included in the vehicle data 218. In some examples, the wheel speed and / or the vehicle acceleration can be included in the sensor parameter data 215 obtained from the sensor unit 109 of the tire 106. Similarly, the aerodynamic drag can be estimated based at least in part on the vehicle speed contained in the vehicle data 218. In various examples, the surface condition detection service 209 can estimate a net drag (eg, traction) based at least in part on subtracting an estimated road grade force value and an aerodynamic drag value from a total traction value.

[0137] At block 1012, the surface condition detection service 209 applies the traction and vehicle speed as inputs to a surface condition detection model 230. The surface condition detection model 230 is a machine learning model that is trained to predict the probability of a surface being wet or dry based at least in part on the traction and vehicle speed. The surface condition detection model 230 is trained to account for the effect of water on tire rolling resistance relative to the estimated traction and vehicle speed.

[0138] At block 1015 , the surface condition detection service 209 determines the road condition probability 403 . The road condition probability 403 represents the probability that the surface is wet or dry. The road condition probability 403 corresponds to the output of the surface condition detection model 230 .

[0139] At block 1018, the surface condition detection service 209 transmits the road condition probability 403 to the vehicle system 206. In various examples, the vehicle electronic system 206 may adjust operational aspects associated with the particular vehicle electronic system 206 based at least in part on the road condition probability 403. Thereafter, this portion of the process proceeds to completion.

[0140] Several software components discussed previously are stored in the memory of the corresponding computing device and can be executed by the processor of the corresponding computing device. In this regard, the term "executable" refers to a program file in a form that can ultimately be run by the processor. An example of an executable program can be a compiler that can be converted into machine code in a format that can be loaded into a random access portion of the memory and run by the processor; a source code that can be represented in an appropriate format, such as an object code that can be loaded into a random access portion of the memory and executed by the processor; or a source code that can be interpreted by another executable program to generate instructions in a random access portion of the memory for execution by the processor. The executable program can be stored in any part or component of the memory, including a random access memory (RAM), a read-only memory (ROM), a hard drive, a solid-state drive, a universal serial bus (USB) flash drive, a memory card, an optical disk (such as a compact disc (CD) or a digital versatile disc (DVD), a floppy disk, a tape, or other storage component.

[0141] The memory includes both volatile and non-volatile memory and data storage components. Volatile components are those components that do not retain data values ​​when the power is off. Non-volatile components are those components that retain data when the power is off. Therefore, the memory may include random access memory (RAM), read-only memory (ROM), hard disk drive, solid state drive, USB flash drive, memory card accessed via a memory card reader, floppy disk accessed via an associated floppy disk drive, optical disk accessed via an optical drive, magnetic tape accessed via an appropriate tape drive, or other storage components, or any combination of two or more of these storage components. In addition, RAM may include static random access memory (SRAM), dynamic random access memory (DRAM) or magnetic random access memory (MRAM) and other such devices. ROM may include programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM) or other similar storage devices.

[0142] Although the applications and systems described herein may be implemented in code or software executed by general-purpose hardware as discussed above, as an alternative, they may also be implemented in dedicated hardware or a combination of software / general-purpose hardware and dedicated hardware. If implemented in dedicated hardware, each may be implemented as a circuit or state machine that uses a combination of any one or more of a variety of technologies. These technologies may include, but are not limited to, discrete logic circuits with logic gates (which are used to implement various logic functions when one or more data signals are applied), application-specific integrated circuits (ASICs) with appropriate logic gates, field programmable gate arrays (FPGAs), or other components, etc. Such technologies are generally well known to those skilled in the art and are therefore not described in detail herein.

[0143] The flow chart shows the functions and operations of the implementation scheme of the part of various embodiments of the present disclosure. If implemented in software form, each box can represent a module, section or code including program instructions to implement a specific logic function. The program instruction can be implemented in the form of source code, and the source code includes a human-readable statement or machine code written in a programming language, and the machine code includes digital instructions that can be recognized by a suitable execution system (such as a processor in a computer system). Machine code can be converted from source code by various processes. For example, a compiler can be used to generate machine code from source code before executing the corresponding application. As another example, machine code can be generated from source code while using an interpreter to execute. Other methods can also be used. If implemented in hardware form, each box can represent a circuit or multiple interconnected circuits to implement a specific logic function.

[0144] Although the flowchart shows a specific execution order, it should be understood that the execution order may be different from the depicted order. For example, the execution order of two or more boxes may be disrupted relative to the shown order. Moreover, two or more boxes shown in succession may be executed simultaneously or partially simultaneously. In addition, in some embodiments, one or more boxes shown in the flowchart may be skipped or omitted. In addition, for the purpose of enhancing practicality, accounting, performance measurement, or providing troubleshooting assistance, any number of counters, state variables, warning semaphores, or messages may be added to the logic flow described herein. It is understood that all such changes are within the scope of the present disclosure.

[0145] Moreover, any logic or application described herein, including software or code, may be implemented in any non-transient computer-readable medium for use by or in connection with an instruction execution system (such as a processor in a computer system or other system). In this sense, logic may include statements including instructions and declarations that may be obtained from a computer-readable medium and executed by an instruction execution system. In the context of the present disclosure, a "computer-readable medium" may be any medium that may contain, store, or maintain the logic or application described herein for use by or in connection with an instruction execution system. Furthermore, a collection of distributed computer-readable media located on multiple computing devices (such as a storage area network or a distributed or clustered file system or database) may also be considered collectively as a single non-transient computer-readable medium.

[0146] Computer readable media may include any of many physical media, such as magnetic, optical or semiconductor media. More specific examples of suitable computer readable media will include, but are not limited to, magnetic tape, magnetic floppy disk, magnetic hard drive, memory card, solid state drive, USB flash drive or optical disk. Moreover, the computer readable medium may be a random access memory (RAM), including a static random access memory (SRAM) and a dynamic random access memory (DRAM), or a magnetic random access memory (MRAM). In addition, the computer readable medium may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM) or other types of storage devices.

[0147] In addition, any logic or application described herein can be implemented and constructed in a variety of ways. For example, one or more applications described herein can be implemented as modules or components of a single application. In addition, one or more applications described herein can be executed in a shared or separate computing device or a combination thereof. For example, multiple applications described herein can be executed in the same computing device, or in multiple computing devices in the same computing environment 203, 221.

[0148] Unless expressly stated otherwise, disjunctive language, such as the phrase "at least one of X, Y, or Z," should be understood as generally used to indicate that an item, term, etc., may be any one of X, Y, or Z, or any combination thereof (e.g., X; Y; Z; X or Y; X or Z; Y or Z; X, Y, or Z; etc.), depending on the context. Thus, such disjunctive language is generally not intended to, and should not, imply that certain embodiments require that each of at least one X, at least one Y, or at least one Z be present.

[0149] It should be emphasized that the above-mentioned embodiments of the present disclosure are merely possible examples of implementation methods proposed for a clear understanding of the principles of the present disclosure. Many modifications and variations may be made to the above-mentioned embodiments without substantially departing from the spirit and principles of the present disclosure. All such modifications and variations are intended to be included herein within the scope of the present disclosure and protected by the following claims.

Claims

1. A method for determining a road surface condition, comprising: acquiring, by at least one computing device, vehicle data collected from a vehicle controller area network (CAN) bus in communication with one or more vehicle systems of a vehicle traveling along a surface; determining, by the at least one computing device, that the vehicle is in a stable state based at least in part on the vehicle data; as well as In response to determining that the vehicle is operating in the stable condition: estimating, by the at least one computing device, traction associated with at least one tire of the vehicle and the surface based at least in part on the vehicle data; applying, by the at least one computing device, the traction force and the vehicle speed of the vehicle to a surface condition detection model; as well as A probability of the surface being wet or dry is determined by the at least one computing device based at least in part on an output of the surface condition detection model.

2. The method according to claim 1, wherein: The vehicle data includes engine torque, engine revolutions per minute (RPM), vehicle speed, wheel speed, and vehicle acceleration.

3. The method according to claim 2, wherein: Determining that the vehicle is operating in the stable state further comprises: determining that the engine RPM exceeds an RPM threshold; determining that the vehicle speed exceeds a speed threshold; and It is determined that the vehicle acceleration exceeds an acceleration threshold.

4. The method according to claim 1, wherein: The surface condition detection model includes a trained classifier.

5. The method according to claim 1, further comprising: Determine total traction; Determine road grade forces; as well as Determine aerodynamic drag, and The traction force or net drag is estimated by subtracting the road grade force and the aerodynamic drag from the total traction force.

6. The method of claim 1 further comprising transmitting the probability to at least one vehicle system of the one or more vehicle systems, the at least one vehicle system being configured to adjust operation of the vehicle based at least in part on the probability.

7. The method according to claim 1, wherein: The at least one computing system is located within the vehicle or remotely from the vehicle.

8. A road surface detection system, comprising: A computing device comprising a processor and a memory; as well as The machine-readable instructions stored in the memory, when executed by the processor, cause the computing device to at least: acquiring vehicle data collected from a vehicle controller area network (CAN) bus in communication with one or more vehicle systems of the vehicle traveling along the surface; determining that the vehicle is in a stable operating state based at least in part on the vehicle data; and In response to determining that the vehicle is operating in a stable condition: estimating traction associated with at least one tire of the vehicle and the surface based at least in part on the vehicle data; applying the traction force and vehicle speed of the vehicle to a surface condition detection model; and A probability that the surface is wet or dry is determined based at least in part on an output of the surface condition detection model.

9. The system according to claim 8, wherein: The vehicle data includes engine torque, engine revolutions per minute (RPM), vehicle speed, wheel speed, and vehicle acceleration.

10. The system according to claim 9, wherein: Determining that the vehicle is operating in a stable state further comprises: determining that the engine RPM exceeds an RPM threshold; determining that the vehicle speed exceeds a speed threshold; and It is determined that the vehicle acceleration exceeds an acceleration threshold.

Images