WO2022116203A1

WO2022116203A1 - Systems and methods for utilizing machine learning to monitor vehicle health

Info

Publication number: WO2022116203A1
Application number: PCT/CN2020/134075
Authority: WO
Inventors: Alex Cao; Wei Gao; Bruce Li; Jiarui Liu; Lixin Peng; Sun Shuai; Steven Tang; Xuewei Wang; Pingyu WU; Victor ZHONG
Original assignee: Cummins Inc.; Wei Gao
Priority date: 2020-12-04
Filing date: 2020-12-04
Publication date: 2022-06-09
Also published as: CN115461751A

Abstract

A system includes a controller coupled to a vehicle system. The controller is structured to determine a predicted parameter regarding operation of the vehicle system based on execution of a machine learning predictive model utilizing a vehicle parameter associated with the predicted parameter, compare the predicted parameter to an actual vehicle parameter of the same operation of the vehicle system to determine a relationship between the predicted parameter and the actual vehicle parameter, run an iterations of the machine learning predictive model to determine the relationship between the predicted parameter and the actual vehicle parameter, use the machine learning predictive model to detect an error value if a difference between the predicted parameter and the actual vehicle parameter is at or above a predetermined threshold, use the relationship to predict the operation of the vehicle system, and generate an alert based on the detected error value.

Description

SYSTEMS AND METHODS FOR UTILIZING MACHINE LEARNING TO MONITOR VEHICLE HEALTH

TECHNICAL FIELD

The present disclosure relates to systems and methods of monitoring vehicle health. More particularly, the present disclosure relates to utilizing models to predict and compare various vehicle parameters to monitor, diagnose, and facilitate servicing or repair of vehicle components.

BACKGROUND

Machine learning can be used to make predictions based on a set of algorithms. Machine learning allows “self-improvement” without complex programming, and thus produces continuously more accurate predictions, despite the traditional pitfalls of developing such a program. The cloud provides a platform to access and use machine learning. Often and with respect to vehicular applications, the detection and correction of various failure modes in vehicle engines and, for example, exhaust aftertreatment systems may be limited because on-road operating demands typically have priority over diagnostic and performance recovery procedures.

SUMMARY

One embodiment relates to a system including a controller coupled to a vehicle system. The controller is structured to determine a predicted parameter regarding operation of the vehicle system based on execution of a machine learning predictive model utilizing a vehicle parameter associated with the predicted parameter, compare the predicted parameter to an actual vehicle parameter of the same operation of the vehicle system to determine a relationship between the predicted parameter and the actual vehicle parameter, run iterations of the machine learning predictive model to determine the relationship between the predicted parameter and the actual vehicle parameter, use the machine learning predictive model to detect an error value if a difference between the predicted parameter and the actual vehicle parameter is at or above a predetermined threshold, use the relationship to predict the operation of the vehicle system, and generate an alert based on the detected error value.

Another embodiment relates to a system including a remote computing system coupled to a plurality of vehicles via a plurality of telematics units. The remote computing system is structured to gather a plurality of vehicle parameters regarding operation of the plurality of vehicles via the plurality of telematics units, use a machine learning predictive model to determine a predicted parameter regarding operation of the plurality of vehicles, wherein the machine learning predicted model utilizes the plurality of vehicle parameters associated with the predicted parameter, use the machine learning predictive model to compare the predicted parameter to an actual vehicle parameter of the same operation of the plurality of vehicles to determine a relationship between the predicted parameter and the actual vehicle parameter, run iterations of the machine learning predictive model to determine the relationship between the predicted parameter and the actual vehicle parameter, use the machine learning predictive model to detect an error value if a difference between the predicted parameter and the actual vehicle parameter is at or above a predetermined threshold, use the relationship to predict the operation of the vehicle, and generate an alert for at least one vehicle of the plurality of vehicles that is provided via the telematics unit associated with the at least one vehicle based on the detected error value.

Yet another embodiment relates to a method including determining a predicted parameter regarding operation of a vehicle system based on execution of a machine learning predictive model utilizing a vehicle parameter associated with the predicted parameter, comparing the predicted parameter to an actual vehicle parameter of the same operation of the vehicle system to determine a relationship between the predicted parameter and the actual vehicle parameter, running iterations of the machine learning predictive model to determine the relationship between the predicted parameter and the actual vehicle parameter, using the machine learning predictive model to detect an error value if a difference between the predicted parameter and the actual vehicle parameter is at or above a predetermined threshold, using the relationship to predict the operation of the vehicle system, and generating an alert based on the detected error value.

These and other features, together with the organization and manner of operation thereof, will become apparent from the following detailed description when taken in conjunction with the accompanying drawings.

BRIEF DESCRIPTION OF THE FIGURES

FIG. 1 is a schematic diagram of a vehicle system, according to an example embodiment.

FIG. 2 is a schematic diagram of the controller used with the vehicle of FIG. 1, according to an example embodiment.

FIG. 3 is a flow diagram of a logic that may be used with an oil diagnostic circuit of the controller of FIGS. 1-2, according to an example embodiment.

FIG. 4 is a chart of an error comparison between various vehicles based on the logic of FIG. 3, according to an example embodiment.

FIG. 5 is a flow diagram of a logic that may be used with a coolant diagnostic circuit of the controller of FIGS. 1-2, according to an example embodiment.

FIGS. 6A-6B are feature selection ranking charts for the logic of FIG. 5, according to an example embodiment.

FIG. 7 is a flow diagram of a logic that may be used with a battery diagnostic circuit of the controller of FIGS. 1-2, according to an example embodiment.

FIGS. 8A-8D are charts of battery health at various engine stages, according to an example embodiment.

FIG. 9 is a flow diagram of a method of generating an alert based on a predictive model, according to an example embodiment.

DETAILED DESCRIPTION

For the purposes of promoting an understanding of the principles of the disclosure, reference will now be made to the embodiments illustrated in the drawings and specific language will be used to describe the same. It will nevertheless be understood that no limitation of the scope of the disclosure is thereby intended, any alterations and further modifications in the illustrated embodiments, and any further applications of the principles of the disclosure as illustrated therein as would normally occur to one skilled in the art to which the disclosure relates are contemplated herein.

Referring to the Figures generally, the various embodiments disclosed herein relate to systems and methods of leveraging big data processing power in the cloud to develop precision models and provide customized alerts. According to the present disclosure, a controller is coupled to a vehicle system. The controller is structured to determine a predicted parameter regarding operation of the vehicle system based on execution of a machine learning predictive model which utilizes a vehicle parameter associated with the predicted parameter. The controller compares the predicted parameter to an actual vehicle parameter of the vehicle system to determine a relationship between the predicted parameter and the actual vehicle parameter. The controller runs or facilitates running of a predefined amount of iterations of the machine learning predictive model to determine the relationship between the predicted parameter and the actual vehicle parameter. The controller uses the machine learning predictive model to detect an error value if a difference between the predicted parameter and the actual vehicle parameter is at or above a predetermined threshold, and thus uses the relationship to predict the operation, or health, of the vehicle system. In operation and as described herein, the model may detect abnormal changes in oil pressure, detect the health of a vehicle coolant, and detect the health of a vehicle battery. Based on these detections, the controller may alert a user or operator of potential malfunctions within the system. This alert may effectively prevent the occurrence of more serious problems and reduce future costs. For example, the controller may predict abnormal changes in engine oil pressure. In another example, the controller may determine the health of the battery, factoring inputs such as the battery’s environment and the delta voltage experienced by the battery. Advantageously, with big data and machine learning, a user can identify issues faster and more accurately to reduce vehicle inefficiencies, downtime during repairs, and repair costs. Predictive machine learning and telematics technology provides customized service and problem solutions based on different user behavior. In this regard, the present disclosure utilizes a custom forecasting method for each vehicle to provide a more accurate alert if the vehicle system is not working properly and notifies the user remotely before or likely before a real failure happens.

Referring now to FIG. 1, a schematic diagram of a vehicle system is shown according to one embodiment. The vehicle system 50 is structured to provide an environment that facilitates and allows the exchange of information or data (e.g., communications) between a vehicle, such as vehicle 100, and one or more other components or sources. In this regard and for example, the vehicle system 50 may include telematics systems that facilitate the acquisition and transmission of data acquired regarding the operation of the vehicle 100. As shown and generally speaking, the vehicle system 50 includes a vehicle 100 communicably coupled via a network 51 to a remote computing system 35 and an external information source 170, where the term “external” refers to a component or system outside of the vehicle 100.

The network 51 may be a communication protocol that facilitates the exchange of information between and among the vehicle 100, the remote computing system 35, and the external information source 170. As shown, the network 51 is configured as a wireless network. In this regard, the vehicle 100 may wirelessly transmit and receive data from at least the external information source 170. The wireless network may be a wireless network, such as Wi-Fi, WiMax, Geographical Information System (GIS) , Internet, Radio, Bluetooth, Zigbee, satellite, radio, Cellular, Global System for Mobile Communications (GSM) , General Packet Radio Service (GPRS) , Long Term Evolution (LTE) , light signaling, etc. In an alternate embodiment, the network 51 may be configured as a wired network or a combination of wired and wireless protocol. For example, the controller 150 and/or telematics unit of the vehicle 100 may electrically and/or operatively couple via fiber optic cable (s) to the network 51 to selectively transmit and receive data wirelessly to and from at least the external information source 170 and/or remote computing system 35.

The external information source 170 may be any information (e.g., data, value, etc. ) provider capable of providing external information. Accordingly, the external information source 170 may include another vehicle to enable vehicle-to-vehicle communications. In this regard, the vehicle 100 may communicate with one or more other vehicles directly (e.g., via NFC, via Bluetooth, etc. ) to obtain data regarding one or more upcoming conditions for the vehicle 100. In another embodiment, the external information source 170 may include a vehicle-to-X configuration, where the “X” refers to any remote information providing source. For example, the remote information providing source may include one or more servers, computers, mobile devices, infrastructure components, etc. Further specific examples of the source may include, but are not limited to, one or more of a global positioning system satellite that provides latitude, longitude, and/or elevation data, satellites that provide dynamic external information, other vehicles that provide external information to the vehicle 100, external databases and processing systems, etc.

The external information source 170 may provide information or data external to the vehicle, which may include one or more map based databases, where the map based database includes information including, but not limited to, road grade data (e.g., the road grade at various spots along various routes) , speed limit data (e.g., posted speed limits in various road locations) , elevation or altitude data at various points along a route, curvature data at various points along a route, location of intersections along a route, etc. The external information may further include dynamic external data (i.e., information that may change as a function time) including, but not limited to, a traffic density at a particular location at a particular time, a weather condition at a particular location at a particular time, a fuel price at a particular location at a particular time, an electricity cost at a particular location at a particular time, etc. The external information may further include regulation information including, but not limited to, an emission regulation (e.g., a permissible NOx and CO amount, etc. ) , a braking regulation (e.g., no engine braking, etc. ) , a noise regulation, and so on.

In the embodiment depicted, the vehicle 100 includes an internal combustion engine 101 power source. In some embodiments, the vehicle 100 may be configured as any type of at least partially electric powered vehicle (e.g., a full electric vehicle, a plug-in hybrid vehicle, etc. ) . The vehicle 100 may be configured as an on-road or an off-road vehicle including, but not limited to, line-haul trucks, mid-range trucks (e.g., pick-up truck) , tanks, airplanes, and any other type of vehicle that utilizes a transmission. The vehicle 100 is shown to generally include a powertrain system 110, an exhaust aftertreatment system 120, a telematics unit 130, a diagnostic and prognostic system 135, an operator input/output (I/O) device 140, and a controller 150, where the controller 150 is communicably coupled to each of the aforementioned components.

The powertrain system 110 facilitates power transfer from the engine 101 and/or a motor generator to power and/or propel the vehicle 100. The powertrain system 110 includes an engine 101 operably coupled to a transmission 102 that is operatively coupled to a drive shaft 103, which is operatively coupled to a differential 104, where the differential 104 transfers power output from the engine 101 to the final drive (shown as wheels 105) to propel the vehicle 100. As a brief overview, the engine 101 receives a chemical energy input (e.g., a fuel such as gasoline or diesel) and combusts the fuel to generate mechanical energy, in the form of a rotating crankshaft. In comparison, if a motor generator is provided, a motor generator may be in a power receiving relationship with an energy source, such as battery 106 that provides an input energy (and stores generated electrical energy) to the motor generator for the motor generator to output in form of usable work or energy to in some instances propel the vehicle 100 alone or in combination with the engine 101. It should be understood, that other configurations of the vehicle 100 are intended to fall within the spirit and scope of the present disclosure (e.g., a series configuration and non-hybrid applications, such as a full electric vehicle, etc. ) . As a result of the power output from the engine 101, the transmission 102 may manipulate the speed of the rotating input shaft (e.g., the crankshaft) to effect a desired drive shaft 103 speed. The rotating drive shaft 103 is received by a differential 104, which provides the rotation energy of the drive shaft 103 to the final drive 105. The final drive 105 then propels or moves the vehicle 100.

The engine 101 may be structured as any internal combustion engine (e.g., compression-ignition or spark-ignition) , such that it can be powered by any fuel type (e.g., diesel, ethanol, gasoline, etc. ) . Furthermore, the transmission 102 may be structured as any type of transmission, such as a continuous variable transmission, a manual transmission, an automatic transmission, an automatic-manual transmission, a dual clutch transmission, etc. Accordingly, as transmissions vary from geared to continuous configurations (e.g., continuous variable transmission) , the transmission can include a variety of settings (gears, for a geared transmission) that affect different output speeds based on the engine speed. Like the engine 101 and the transmission 102, the drive shaft 103, differential 104, and final drive 105 may be structured in any configuration dependent on the application (e.g., the final drive 105 is structured as wheels in an automotive application and a propeller in an airplane application) . Further, the drive shaft 103 may be structured as a one-piece, two-piece, and a slip-in-tube driveshaft based on the application.

Moreover, the battery 106 may be configured as any type of rechargeable (i.e., primary) battery and of any size. That is to say, the battery 106 may be structured as any type of electrical energy storing and providing device, such as one or more capacitors (e.g., ultra capacitors, etc. ) and/or one or more batteries typically used or that may be used in hybrid vehicles (e.g., Lithium-ion batteries, Nickel-Metal Hydride batteries, Lead-acid batteries, etc. ) . The battery 106 may be operatively and communicably coupled to the controller 150 to provide data indicative of one or more operating conditions or traits of the battery 106. The data may include a temperature of the battery, a current into or out of the battery, a number of charge-discharge cycles, a battery voltage, etc. As such, the battery 106 may include one or more sensors coupled to the battery 106 that acquire such data. In this regard, the sensors may include, but are not limited to, voltage sensors, current sensors, temperature sensors, etc.

As also shown, the vehicle 100 includes an exhaust aftertreatment system 120 in fluid communication with the engine 101. The exhaust aftertreatment system 120 receives the exhaust from the combustion process in the engine 101 and reduces the emissions from the engine 101 to less environmentally harmful emissions (e.g., reduce the NOx amount, reduce the emitted particulate matter amount, etc. ) . The exhaust aftertreatment system 120 may include components that reduce diesel exhaust emissions, such as a selective catalytic reduction catalyst, a diesel oxidation catalyst, a diesel particulate filter, a diesel exhaust fluid doser with a supply of diesel exhaust fluid, and a plurality of sensors for monitoring the exhaust aftertreatment system 120 (e.g., a NOx sensor) . It should be understood that other embodiments may exclude an exhaust aftertreatment system and/or include different, less than, and/or additional components than that listed above. All such variations are intended to fall within the spirit and scope of the present disclosure.

The vehicle 100 is also shown to include a telematics unit 130. The telematics unit 130 may include, but is not limited to, one or more memory devices for storing tracked data, one or more electronic processing units for processing the tracked data, and a communications interface for facilitating the exchange of data between the telematics unit 130 and one or more remote devices (e.g., a provider/manufacturer of the telematics device, etc. ) . In this regard, the communications interface may be configured as any type of mobile communications interface or protocol including, but not limited to, Wi-Fi, WiMax, Internet, Radio, Bluetooth, Zigbee, satellite, radio, Cellular, GSM, GPRS, LTE, and the like. The telematics unit 130 may also include a communications interface for communicating with the controller 150 of the vehicle 100. The communication interface for communicating with the controller 150 may include any type and number of wired and wireless protocols (e.g., any standard under IEEE 802, etc. ) . For example, a wired connection may include a serial cable, a fiber optic cable, an SAE J1939 bus, a CAT5 cable, or any other form of wired connection. In comparison, a wireless connection may include the Internet, Wi-Fi, Bluetooth, Zigbee, cellular, radio, etc. In one embodiment, a controller area network (CAN) bus including any number of wired and wireless connections provides the exchange of signals, information, and/or data between the controller 150 and the telematics unit 130. In other embodiments, a local area network (LAN) , a wide area network (WAN) , or an external computer (for example, through the Internet using an Internet Service Provider) may provide, facilitate, and support communication between the telematics unit 130 and the controller 150. In still another embodiment, the communication between the telematics unit 130 and the controller 150 is via the unified diagnostic services (UDS) protocol. All such variations are intended to fall within the spirit and scope of the present disclosure.

The vehicle 100 is also shown to include a diagnostic and prognostic system 135. The diagnostic and prognostic system 135 may be configured as any type of diagnostic and prognostic system. Accordingly, the diagnostic and prognostic system 135 may be communicably coupled to one or more sensors, physical or virtual, positioned throughout the vehicle 100 such that the diagnostic and prognostic system 135 may receive data indicative of one or more fault conditions, potential symptoms, operating conditions to determine a status of a component (e.g., healthy, problematic, malfunctioning, etc. ) . If the diagnostic and prognostic system 135 detects a fault, the diagnostic and prognostic system 135 may trigger a fault code and provide an indication to the operator input/output device 140 of the vehicle (e.g., a check engine light, etc. ) .

The operator I/O device 140 may be communicably coupled to the controller 150, such that information may be exchanged between the controller 150 and the I/O device 140, wherein the information may relate to one or more components of FIG. 1 or determinations (described below) of the controller 150. The operator I/O device 140 enables an operator of the vehicle 100 to communicate with the controller 150 and one or more components of the vehicle 100 of FIG. 1. For example, the operator input/output device 140 may include, but is not limited to, an interactive display, a touchscreen device, one or more buttons and switches, voice command receivers, etc. The operator may include a driver, a remote operator, a fleet manager/operator, and/or other remote attendant for the vehicle 100 or a plurality of vehicles 100. For instance, fleet managers may receive automated service notifications for a fleet vehicles having the vehicle system 50 as described herein. In various alternate embodiments, the controller 150 and components described herein may be implemented with non-vehicular applications (e.g., a power generator) . Accordingly, the I/O device may be specific to those applications. For example, in those instances, the I/O device may include a laptop computer, a tablet computer, a desktop computer, a phone, a watch, a personal digital assistant, etc. Via the operator I/O device, the controller 150 may provide diagnostic information, a fault or service notification based on one or more determinations. For example, in some embodiments, the controller 150 may display, via the operator I/O device, a temperature of the engine 101 and the exhaust gas, and various other information.

The I/O device 140 also includes a graphical user interface (GUI) device (e.g., screen, touch screen, monitor, light emitting diode display, liquid crystal display, keypad, touchpad, buttons, etc. ) . The GUI is configured to display information to a user. For example, the GUI may display an indication (e.g., graphic, image, text, a Fault Code, a Fault Code description, etc. ) to a user as to whether the vehicle is functional or non-functional (e.g., a failure mode identifier, etc. ) . If the GUI displays an indication to the user that the vehicle is non-functional, the indication may convey a portion (e.g., the engine, battery, coolant, etc. ) of the vehicle that is non-functional. The GUI may be any indicator (e.g., dial, gauge, indicator lamp, etc. ) where a fault code is stored in a memory and is accessible via an onboard diagnostics tool. The GUI may be a single instrument or an instrument panel (e.g., a dashboard) . For example, the instrument (s) may include an electric oil pressure gauge and an electric coolant temperature gauge. The indication can be provided at the vehicle or remotely via computer integrated with a fleet manager. In some instances, the indicator can cause an engine de-rating (e.g., control the operation of the exhaust aftertreatment system 120 in order to prevent additional exhaust gas emissions) . For example, the engine speed may be governed to a limited speed until the user addresses the indication or alert.

As shown, the controller 150 is communicably coupled to the powertrain system 110, the exhaust aftertreatment system 120 (and various components of each system) , the telematics unit 130, the diagnostic and prognostic system 135, and the operator input/output device 140. Communication between and among the components may be via any number of wired or wireless connections. For example, a wired connection may include a serial cable, a fiber optic cable, a CAT5 cable, or any other form of wired connection. In comparison, a wireless connection may include the Internet, Wi-Fi, cellular, radio, etc. In one embodiment and as shown, a CAN bus provides the exchange of signals, information, and/or data. The CAN bus includes any number of wired and wireless connections. Because the controller 150 is communicably coupled to the systems and components in the vehicle 100 of FIG. 1, the controller 150 is structured to receive data (e.g., instructions, commands, signals, values, etc. ) from one or more of the components shown in FIG. 1. This may generally be referred to as internal vehicle information (e.g., data, values, etc. ) . The internal vehicle information represents determined, acquired, predicted, estimated, and/or gathered data regarding one or more components in vehicle 100.

Accordingly, the internal vehicle information may include data regarding the battery 106. The data regarding the battery 106 may include, but is not limited to, a temperature of the battery, a current into or out of the battery, a number of charge-discharge cycles, a battery voltage, a battery state of charge, etc. For instance, in a machine learning predictive model to estimate, predict, and detect anomalies in battery health, the variables include at least initial voltage, voltage delta, air temperature, air pressure, coolant temperature, oil temperature, etc. The internal vehicle information may also include information from the diagnostic and prognostic system 135, which may include, but is not limited to, one or more fault codes, data identifiers, diagnostic trouble codes, and so on. The internal vehicle information may also include other data regarding the powertrain system 110 (and other components in the vehicle 100) . For example, the data regarding the powertrain system 110 may include, but is not limited to, the vehicle speed, the current transmission gear/setting, the load on the vehicle/engine, the throttle position, a set cruise control speed, data relating to the exhaust aftertreatment system 120, output power, engine speed, fluid consumption rate (e.g., fuel consumption rate, diesel exhaust fluid consumption rate, etc. ) , any received engine/vehicle faults (e.g., a fault code indicating a low amount of diesel exhaust fluid) , engine operating characteristics (e.g., whether all the cylinders are activated or which cylinders are deactivated, etc. ) , etc. For instance, in a machine learning predictive model to estimate, predict, and detect anomalies in oil health, the variables include engine coolant temperature, oil temperature, wheel based vehicle speed, engine speed, engine intake manifold pressure, ambient air temperature, engine fuel rate, engine load percentage at current speed, actual engine torque percent, and barometric pressure. Further, in a machine learning predictive model to estimate, predict, and detect anomalies in coolant health, for instance, the variables include engine speed, engine oil temperature, engine fuel rate, engine load percentage at current speed, and ambient air temperature. Data relating to the exhaust aftertreatment system 120 includes, but is not limited to, NOx emissions, particulate matter emissions, and conversion efficiency of one or more catalysts in the exhaust aftertreatment system 120 (e.g., the selective catalytic reduction catalyst) .

The internal vehicle information may be stored by the controller 150 and selectively transmitted to one or more desired sources (e.g., another vehicle such as in a vehicle-to-vehicle communication session, a remote operator, etc. ) . In other embodiments, the controller 150 may provide the internal vehicle information to the telematics unit 130 whereby the telematics unit transmits the internal vehicle information to one or more desired sources (e.g., a remote device, an operator of the telematics unit, etc. ) . Thus, the controller 150 directly may communicate information to the remote computing system 35 and/or external information source, or via the telematics unit. All such variations are intended to fall within the spirit and scope of the present disclosure.

Because the components of FIG. 1 are shown to be embodied in a vehicle 100, the controller 150 may be structured as a one or more electronic control units or modules (ECM) . The ECM may include a transmission control unit and any other control unit included in a vehicle (e.g., exhaust aftertreatment control unit, engine control module, powertrain control module, etc. ) . The function and structure of the controller 150 are shown described in greater detail in FIG. 2. Alternatively, at least some or all of the operations described herein of the controller 150 may be performed by the remote computing system 35, in addition to the other operations performed by the remote computing system 35. The telematics unit 130 may transmit received or determined information about the vehicle 100 (e.g., engine operating parameter signals and/or aftertreatment system operating parameter signals) to the remote computing system 35 for performing at least some of the operations described herein remotely. The remote computing system 35 may include one or more servers, network interfaces, input/output devices, and so on.

As shown in FIG. 1, the remote computing system 35 is in communication with the vehicle 100 via a network 51, or a plurality of vehicles 100. The remote computing system 35 creates and curates a database of vehicle information that contains information related to vehicle performance (e.g., engine performance parameters) . The remote computing system 35 is configured to perform advanced analytics to determine and identify patterns in the information. These advanced analytics may be Artificial Intelligence (AI) , physics-based models, machine learning, etc. The determined and identified patterns may relate to repeated instances of similar parameter values (e.g., battery health, coolant health, etc. ) for a vehicle. In some embodiments, the remote computing system 35 may store and use some or more of the aforementioned external information.

In one embodiment, the remote computing system 35 is in continuous or near continuous communication with the controller 150 via network 51 and telematics unit 130 in order to provide real-time control over the vehicle 100 throughout use. This embodiment is referred to as an “on-line method, ” such that the controller 150 maintains an on-line relationship with the remote computing system 35 while the vehicle 100 is in use (i.e., operating along a route) . While in operation with the on-line method, the controller 150 is continuously sending operational information (e.g., exhaust temperature, engine speed, torque, etc. ) and location information from a positioning system to the remote computing system 35. The remote computing system 35 receives this information in real-time or near real-time. In some embodiments utilizing the on-line method, the remote computing system 35 assumes some of the duties of the controller 150, such that the remote computing system 35 performs the correlation between current data and predictive models described herein, and issues commands to the vehicle 100 components accordingly. Beneficially, this reduces the demand and load on the memory and processing circuit of the controller 150 to enable more efficient operation.

In another embodiment, the remote computing system 35 transmits to the vehicle 100 before the vehicle 100 departs on a route. This embodiment is referred to as an “off-line method, ” such that the controller 150 maintains an off-line relationship with the remote computing system 35 while the vehicle 100 is in use (i.e., operating along a route) . Because the controller 150 is not in active communication with the remote computing system 35 during operation of the vehicle, the remote computing system 35 transmits a predictive model to the controller 150 before the vehicle 100 departs (e.g., while the vehicle is in a service bay, in a parked non-mobile state, etc. ) , and the controller 150 downloads (e.g., stores in a memory, etc. ) the model. In some embodiments, the controller 150 receives and stores a predictive model based on a particular of the vehicle (e.g., in the memory 206) .

Accordingly, referring now to FIG. 2, a schematic diagram 200 of the remote computing system 35 and the controller 150 of the vehicle 100 of Figure 1 is shown according to an example embodiment. As mentioned above, the controller 150 may be structured as one or more electronic control units (ECU) . The controller 150 may be separate from or included with at least one of a transmission control unit, an exhaust aftertreatment control unit, a powertrain control module, an engine control module, etc. In one embodiment, the components of the controller 150 are combined into a single unit. In another embodiment, one or more of the components may be geographically dispersed throughout the system. All such variations are intended to fall within the scope of the disclosure. The controller 150 is shown to include a processing circuit 202 having a processor 204 and a memory device 206, a control system 208 having an oil diagnostic circuit 210, a coolant diagnostic circuit, a battery diagnostic circuit 214 and a communications interface 216.

The remote computing system 35 may be a computing system that is separate from the controller 150 and other computing system (s) contained within the vehicle. In an example embodiment, the remote computing system 35 is a cloud-based computing system hosted on at least one server. As shown in FIG. 2, the remote computing system 35 is in communication with the controller 150 via the telematics unit 130, which facilitates two-way transmission of data (i.e., from the remote computing system 35 to the controller 150, and from the controller 150 to the remote computing system 35. The remote computing system 35 is structured or configured to receive information from the controller 150 related to the vehicle, including sensed or determined operation information (e.g., route information, location information) and/or input from a user of the vehicle. The remote computing system 35 stores this received information for a plurality of vehicles in a database as historical data. The remote computing system 35 then determines a predictive model by applying advanced analytics to stored historical data associated with operation of vehicle. This historical data includes performance parameters indicative of vehicle performance. The advanced analytics includes artificial intelligence (AI) , physic-based models, machine learning that is employed to identify patterns and predict the performance parameters of the vehicle 100 to identify potential health issues.

In one configuration, the oil diagnostic circuit 210, the coolant diagnostic circuit, and the battery diagnostic circuit 214 are embodied as machine or computer-readable media storing instructions that are executable by a processor, such as processor 204. As described herein and amongst other uses, the machine-readable media facilitates performance of certain operations to enable reception and transmission of data. For example, the machine-readable media may provide an instruction (e.g., command, etc. ) to, e.g., acquire data. In this regard, the machine-readable media may include programmable logic that defines the frequency of acquisition of the data (or, transmission of the data) . The computer readable media may include code, which may be written in any programming language including, but not limited to, Java or the like and any conventional procedural programming languages, such as the "C" programming language or similar programming languages. The computer readable program code may be executed on one processor or multiple remote processors, such as housed in the remote computing system 35.

In another configuration, the oil diagnostic circuit 210, the coolant diagnostic circuit, and the battery diagnostic circuit 214 are embodied as hardware units, such as electronic control units. As such, the oil diagnostic circuit 210, the coolant diagnostic circuit, and the battery diagnostic circuit 214 may be embodied as one or more circuitry components including, but not limited to, processing circuitry, network interfaces, peripheral devices, input devices, output devices, sensors, etc. In some embodiments, the oil diagnostic circuit 210, the coolant diagnostic circuit, and the battery diagnostic circuit 214 may take the form of one or more analog circuits, electronic circuits (e.g., integrated circuits (IC) , discrete circuits, system on a chip (SOCs) circuits, microcontrollers, etc. ) , telecommunication circuits, hybrid circuits, etc. In this regard, the oil diagnostic circuit 210, the coolant diagnostic circuit, and the battery diagnostic circuit 214 may include any type of component for accomplishing or facilitating achievement of the operations described herein. For example, a circuit as described herein may include one or more transistors, logic gates (e.g., NAND, AND, NOR, OR, XOR, NOT, XNOR, etc. ) , resistors, multiplexers, registers, capacitors, inductors, diodes, wiring, and so on) . The oil diagnostic circuit 210, the coolant diagnostic circuit, and the battery diagnostic circuit 214 may also include programmable hardware devices such as field programmable gate arrays, programmable array logic, programmable logic devices or the like. The oil diagnostic circuit 210, the coolant diagnostic circuit, and the battery diagnostic circuit 214 may include one or more memory devices for storing instructions that are executable by the processor (s) of the oil diagnostic circuit 210, the coolant diagnostic circuit, and the battery diagnostic circuit 214. The one or more memory devices and processor (s) may have the same definition as provided below with respect to the memory device 206 and processor 204. In some hardware unit configurations and as described above, the oil diagnostic circuit 210, the coolant diagnostic circuit, and the battery diagnostic circuit 214 may be geographically dispersed throughout separate locations in the system. Alternatively and as shown, the oil diagnostic circuit 210, the coolant diagnostic circuit, and the battery diagnostic circuit 214 may be embodied in or within a single unit/housing, which is shown as the controller 150.

In the example shown, the controller 150 includes the processing circuit 202 having the processor 204 and the memory device 206. The processing circuit 202 may be structured or configured to execute or implement the instructions, commands, and/or control processes described herein with respect to the oil diagnostic circuit 210, the coolant diagnostic circuit, and the battery diagnostic circuit 214. The depicted configuration represents the oil diagnostic circuit 210, the coolant diagnostic circuit, and the battery diagnostic circuit 214 as machine or computer-readable media storing instructions executable by the processor 204. Alternatively, the logic/instructions may be stored by the memory device 206 and executable by the processor 204. However, as mentioned above, this illustration is not meant to be limiting as the present disclosure contemplates other embodiments where the oil diagnostic circuit 210, the coolant diagnostic circuit, and the battery diagnostic circuit 214, or at least one circuit of the circuits the oil diagnostic circuit 210, the coolant diagnostic circuit, and the battery diagnostic circuit 214, is configured as a hardware unit. All such combinations and variations are intended to fall within the scope of the present disclosure.

The processor 204 may be implemented as one or more general-purpose processor, an application specific integrated circuit (ASIC) , one or more field programmable gate arrays (FPGAs) , a digital signal processor (DSP) , a group of processing components, or other suitable electronic processing components. In some embodiments, the one or more processors may be shared by multiple circuits (e.g., oil diagnostic circuit 210, the coolant diagnostic circuit, and the battery diagnostic circuit 214 may comprise or otherwise share the same processor which, in some example embodiments, may execute instructions stored, or otherwise accessed, via different areas of memory) . Alternatively or additionally, the one or more processors may be structured to perform or otherwise execute certain operations independent of one or more co-processors. In other example embodiments, two or more processors may be coupled via a bus to enable independent, parallel, pipelined, or multi-threaded instruction execution. All such variations are intended to fall within the scope of the present disclosure.

The memory device 206 (e.g., memory, memory unit, storage device) may include one or more devices (e.g., RAM, ROM, Flash memory, hard disk storage) for storing data and/or computer code for completing or facilitating the various processes, layers and modules described in the present disclosure. The memory device 206 may be communicably coupled to the processor 204 to provide computer code or instructions to the processor 204 for executing at least some of the processes described herein. Moreover, the memory device 206 may be or include tangible, non-transient volatile memory or non-volatile memory. Accordingly, the memory device 206 may include database components, object code components, script components, or any other type of information structure for supporting the various activities and information structures described herein.

The communications interface 216 may include any combination of wired and/or wireless interfaces (e.g., jacks, antennas, transmitters, receivers, transceivers, wire terminals) for conducting data communications with various systems, devices, or networks structured to enable in-vehicle communications (e.g., between and among the components of the vehicle) and out-of- vehicle communications (e.g., with a remote server) . For example and regarding out-of-vehicle/system communications, the communications interface 216 may include an Ethernet card and port for sending and receiving data via an Ethernet-based communications network and/or a Wi-Fi transceiver for communicating via a wireless communications network. The communications interface 216 may be structured to communicate via local area networks or wide area networks (e.g., the Internet) and may use a variety of communications protocols (e.g., IP, LON, Bluetooth, ZigBee, radio, cellular, near field communication) . Thus, this description indicates that the controller 150 may directly communicate with the remote computing system 35 via the network 51. In other embodiments, communication may be via the telematics unit 130.

The communications interface 216 may facilitate communication between and among the controller 150 and one or more components of the vehicle 100 (e.g., the engine 101, the transmission 102, the aftertreatment system 120, the sensors etc. ) . Communication between and among the controller 150 and the components of the vehicle 100 may be via any number of wired or wireless connections (e.g., any standard under IEEE) . For example, a wired connection may include a serial cable, a fiber optic cable, a CAT5 cable, or any other form of wired connection. In comparison, a wireless connection may include the Internet, Wi-Fi, cellular, Bluetooth, ZigBee, radio, etc. In one embodiment, a controller area network (CAN) bus provides the exchange of signals, information, and/or data. The CAN bus can include any number of wired and wireless connections that provide the exchange of signals, information, and/or data. The CAN bus may include a local area network (LAN) , or a wide area network (WAN) , or the connection may be made to an external computer (for example, through the Internet using an Internet Service Provider) .

An internal information circuit 203 is structured to receive, gather, and/or acquire internal vehicle information. In one embodiment, the internal information circuit 203 includes one or more data acquisition devices within the vehicle 100, such as the diagnostic and prognostic system 135 that facilitate acquisition of the internal vehicle information. In another embodiment, the internal information circuit 203 includes communication circuitry for facilitating reception of the internal information. In still another embodiment, the internal information circuit 203 includes machine-readable content for receiving and storing the internal information. In yet another embodiment, the internal information circuit 203 includes any combination of data acquisition devices, communication circuitry, and machine readable content. As mentioned above, the internal information may include any type of internal information regarding the vehicle 100 and from the vehicle 100 itself (e.g., a vehicle speed, a battery state of charge (SOC) , oil temperature, oil pressure, coolant temperature, a load on the vehicle, a torque output, a transmission setting, an engine temperature, one or more fault codes or a history of fault codes, etc. ) . The internal information circuit 203 is structured to provide the acquired and/or gathered internal information to the oil diagnostic circuit 210, the coolant diagnostic circuit, and the battery diagnostic circuit 214.

The external information circuit 205 is structured to receive, gather, and/or acquire external information from one or more external information sources 170 (e.g., a remote device, an infrastructure component, etc. ) and provide or transmit the external information to the oil diagnostic circuit 210, the coolant diagnostic circuit, and the battery diagnostic circuit 214. The external information circuit 205 may also store or facilitate storing the received external information (e.g., in the memory device) , where the storage configuration may be variable from application-to-application (e.g., store external information for the past thirty days, etc. ) . Advantageously, this information may be recalled by the external information circuit 205 to provide to the oil diagnostic circuit 210, the coolant diagnostic circuit, and the battery diagnostic circuit 214 on-demand. The external information may include data external to the vehicle such as road grade data, altitude data, traffic density, weather conditions, emissions regulation, etc. The external information circuit 205 may update the stored information upon a manual update from the operator (e.g., a refresh input received via the I/O device 140) and/or upon a configuration that dictates or defines how often the external data is provided to the controller 150. This may change as the vehicle is operated.

The oil diagnostic circuit 210 is structured to monitor the oil pressure based on internal vehicle information (e.g., engine speed, oil temperature, oil pressure, and fuel rate) and/or external information (e.g., ambient air temperature) received from each of the internal information circuit 203 and the external information circuit 205, respectively. In particular, the oil diagnostic circuit 210 is structured to predict an oil pressure, compare the predicted oil pressure to an actual oil pressure, and generate an alert based if the difference is at or above a predefined threshold value.

The oil diagnostic circuit 210 is structured to collect raw data from the telematics unit 130. The telematics data may be calculated in cloud (e.g., Azure server to support and build) . For instance, the remote computing system 35 may be the Azure cloud computing service, such that the Azure cloud hosts, support, and/or operates as the remote computing system 35. The calculated data from the telematics unit 130 includes internal information about the vehicle 100. For instance, the internal information may be oil pressure, oil temperature, and engine speed. The oil diagnostic circuit 210 is structured to collect raw data from an external information source 170 (i.e., external static information and external dynamic information) . The external information may include predefined vehicle information, warranty information, performance history, stored values from the internal information, etc. For instance, the external information may be performance history stored external to the vehicle, the performance history based on the oil health (e.g., oil pressure, oil temperature, etc. ) . Further, the external information may be a predefined oil health parameter relative to the particular vehicle (e.g., expected oil pressure, expected oil life, etc. ) .

The oil diagnostic circuit 210 is structured to determine a predictive model to predict a relationship between a first parameter and a second parameter. The predictive model may predict a relationship between a first, second, and third parameter (and so on, in some embodiments) . For instance, the predictive model may be represented by:

Such a multiple linear regression attempts to model the relationship between two or more explanatory variables and a response variable by fitting a linear equation to observed data. Every value of the independent variable x is associated with a value of the dependent variable y, with the population regression line for p explanatory variables x ₁, x ₂, ..., x _p and the regression line y _i for n observations. The parameters x and y may be based on the internal information and external information received by the telematics unit 130 and the external information source 170, respectively. For instance, x is the engine speed and the oil temperature, and y is the oil pressure. As such, the oil diagnostic circuit 210 is structured to predict the oil pressure based on oil temperature and engine speed. Applicant has determined that there are several possible features that are more contributive in the predictive modeling by using XGBoost model and plot_importance function. Those features include engine speed, engine oil temperature, engine fuel rate, engine load percentage at current speed, and ambient air temperature. Based on experimentation, the most important features to engine oil pressure are the engine speed and the oil temperature.

The oil diagnostic circuit 210 is further structured to compare the relationship between the predicted oil pressure and an actual oil pressure. If the oil diagnostic circuit 210 detects an error, a corresponding alert is sent to a terminal (e.g., a GUI, an electric oil pressure gauge on the vehicle dashboard) to notify the user to take action. For instance, an error alert may occur when the oil pressure is below a predefined threshold, thus indicating the vehicle may have leaked or burned oil. Similarly, an error alert may occur when the oil pressure is above a predefined threshold, indicating the user may have added too much oil.

As shown in FIG. 3, customized logic 300 for constructing a model for the oil diagnostic circuit 210 that is unique to each vehicle to detect abnormal performance. For each engine serial number (ESN) (thus, each unit) , the oil diagnostic circuit 210 monitors the engine oil pressure under certain constraints and determines abnormal changes based on the ESN’s historical data. If there is an obvious exception, or a particular fault code, such as the oil pressure being above a predefined threshold, the model will output the corresponding alert on the electric oil pressure gauge (e.g., I/O device) . A predictive model for an ESN may already be implemented because of a set up model, or a normal model. This model may be used to predict the oil pressure, or the model may be customized.

At the start of the process, the predictive model determines if the ESN already has a model, at 302. If the ESN does not, the process moves to 304 and determines if the sum (records) is greater than or equal to a preset threshold (e.g., 5000) . At 306, the process determines a model and saves the model at 308 to implement the model at a subsequent instance or time. A training model is used if the ESN is running the process for the first time. Once the model is trained, it may be saved to the cloud and directly loaded for future iterations. At 310, the process loads the model for specific ESN that has been saved at 308. At 312, the process is implemented as described herein to predict a parameter. For instance, the parameter is the oil pressure. The parameter is then used to determine an error value at 314 between the predicted parameter and an actual parameter (e.g., a predicted oil pressure and an actual oil pressure) . At 316, the process runs iterations of steps 312-314 in order to determine mean values for each of a preset number of records (e.g., 10 records) . At 318, the process considers the abnormal data and reports an alert at 320 based on the abnormal data detected.

Referring now to FIG. 4, a chart 350 depicting an error comparison between various vehicles is shown, thus illustrating the advantages of this logic described in FIG. 3 with regard to the predictive model of the oil diagnostic circuit 210. The x axis represents an index of time and the y axis represents absolute error between a predicted value and a measured value. The upper half of the coordinate axis represents a situation where the oil pressure is lower than the normal value. As such, the vehicle may have leaked or burned oil. The lower half of the coordinate axis represents a situation where the oil pressure is higher than the normal value. Thus, too much oil may have been added to the vehicle. Thus, by utilizing the predictive model, abnormalities may be detected prior to any obvious functionality defects or damage to the car occurs.

Turning now to the coolant diagnostic circuit 212, the coolant diagnostic circuit 212 is structured to monitor the health of the coolant based on internal vehicle information (e.g., engine speed, oil temperature, oil pressure, fan speed, engine coolant level, and fuel rate) , external information received from each of the internal information circuit 203 and external information circuit 205, respectively. In particular, the coolant diagnostic circuit 212 is structured predict a coolant health issue based on its coolant performance history data and calculate the probability of issues, and generate an alert if the difference is at or above a predefined threshold value.

The coolant diagnostic circuit 212 is structured to collect raw data from telematics unit and send it to the remote computing system 34 (e.g., cloud computing) . The data from the telematics unit 130 includes internal information about the vehicle. For instance, the internal information may be coolant temperature, engine speed, torque, air temperature, and vehicle speed at various times of operation of the vehicle. The coolant diagnostic circuit 212 is structured to collect raw data from an external information source. The external information may include predefined vehicle information, warranty information, performance history, stored values from the internal information, etc. The coolant diagnostic circuit 212 is structured to determine a predictive model to predict a probability of coolant performance error based on a first parameter and a second parameter. The predictive model may predict a relationship between a first, second, and third parameter. For instance, the predictive model may be represented by:

The parameters may be based on the internal information and external information received by the telematics unit 130 and the external information source, respectively. Based on machine learning (e.g., XGBoost, linear regression, etc. ) , the predictive model determines engine coolant system performance and runs in the cloud to monitor the system, thus identifying any issues if the system performance changes. The model calculates a risk level based on the predictive result and a warning threshold. If the coolant diagnostic circuit 212 detects an error, a corresponding alert is sent to a terminal (e.g., a GUI, an electric coolant temperature gauge) to notify the user to take action. The terminal may be a user interface such as an electric coolant temperature gauge dial on a phone or web app.

As shown in FIG. 5, a logic 400 for constructing a model for the coolant diagnostic circuit 212 includes a data pipeline for data transmission, data cleaning and data matching. The data pipeline may also utilize Azure server, for example, to support and build. Typically, the data is not ready for the model immediately after collection. Thus, at 402, the data (e.g., the data associated with at least one of the vehicle parameter associated with the predicted parameter and the actual vehicle parameter and external information from an external information source) is cleaned, filtered, or otherwise reformatted, from what was provided and before utilization in the model. The process may then slice each trip into stages to look at the specific periods of data that has been acquired at 404. At 406, a daily monitoring occurs through the three following aspects based on specific periods of data. The process can detect abnormalities every day and choosing a suitable frequency to do so, according to the specific requirement (e.g., desired information) at that time. For instance, one aspect includes the process 408, a second aspect includes the process at 410 to 414, and a third aspect is the process at 411 to 412 to 416. At 408, the state of a coolant sensor is monitored (e.g., healthy or operational, on/off, faulty, etc. ) . At 410, for the stable stage of coolant temperature data, a threshold is set to determine if the coolant temperature is abnormal, then at 414 the stable data state is acquired based on the result from 410. At 411, the process determines a predictive model for the ESN, either training a model as in FIG. 3 or implementing the model already determined for the ESN. For each trip, the trend of coolant temperature usually can be divided into two parts: a warm-up stage 412 (e.g., the coolant temperature rises as the engine runs, and a stable stage 410 (e.g., the coolant temperature stabilized within a range) . Abnormal detection may be based on the characteristics of these two parts. For instance, if the daily outlier ratio in two consecutive records exceeds the sensitive range (e.g., a predefined threshold) there will be an alert at 414, 416. The alert provided may be a fault code or indicator lap that gets transmitted over the telematics unit and back to the vehicle. At 418, the process determines whether the predictive model needs to repeat or if the proper alert has been communicated to the user.

The predictive model determined by the coolant diagnostic circuit 212 is unique to each vehicle and is a customized method to detect the abnormal performance. For each engine serial number (ESN) , the coolant diagnostic circuit 212 monitors the engine coolant health under certain constraints and finds abnormal changes based on the ESN’s historical data. If there is an obvious exception, or a particular fault code, the model outputs the corresponding alert on the electric coolant temperature gauge. A predictive model for an ESN may already be implemented because of a set up model, or a normal model. This model may be used to predict the coolant health risk, or the model may be customized.

Referring now to FIGS. 6A-6B, feature

selection ranking charts

450, 452 were used to determine the features used in the predictive model of the coolant diagnostic circuit 212. As shown in FIG. 6A, the top ten features in the predictive modelling (after adjusting the parameters) by using sklearn feature_selection include the engine coolant temperature, oil temperature, wheel based vehicle speed, engine speed (e.g., revolutions per minute of a crankshaft of the engine) , engine intake manifold pressure, ambient air temperature, engine fuel rate, engine load percentage at current speed, actual engine torque percent, and barometric pressure. However, after removing highly relevant variables, such as engine oil temperature, wheel based vehicle speed, and actual engine percent torque, the most important features include the coolant temperature, engine speed, ambient air temperature, actual engine torque percentage, barometric pressure, engine fuel used, and engine coolant level.

Turning now to the battery diagnostic circuit 214, the battery diagnostic circuit 214 is structured to monitor the battery health based on internal vehicle information (e.g., battery voltage, coolant temperature, oil temperature, clutch status and engine speed) and external information (e.g., ambient air temperature, ambient air pressure, etc. ) received from each of the internal information circuit 203 and the external information circuit 205, respectively. In particular, the battery diagnostic circuit 214 is structured to predict battery health, compare the predicted battery health to an actual battery health, and generate an alert based if the difference is at or above a predefined threshold value. In one embodiment, the battery diagnostic circuit 214 may include communication circuitry to provide one or more commands to the controller 150. In another embodiment, the battery diagnostic circuit 214 may include machine-readable content. In yet another embodiment, the battery diagnostic circuit 214 may include any combination of communication circuitry and machine-readable content.

The battery diagnostic circuit 214 is structured to collect raw data from telematics unit. The telematics data are calculated in the cloud (e.g., Azure server to support and build) . The data from the telematics unit 130 includes internal information about the vehicle. For instance, the internal information may be battery voltage, coolant temperature, oil temperature, clutch status and engine speed. The battery diagnostic circuit 214 is structured to collect raw data from an external information source. The external information may include predefined vehicle information, warranty information, performance history, stored values from the internal information, etc. The battery diagnostic circuit 214 is structured to determine a predictive model to predict a relationship between a first parameter and a second parameter. The predictive model may predict a relationship between a first, second, and third parameter. For instance, the predictive model may be represented by:

The parameters may be based on the internal information and external information received by the telematics unit 130 and the external information source, respectively. As such, the battery diagnostic circuit 214 is structured to predict the battery health based on the engine, battery, and environment. Additionally, the battery diagnostic circuit 214 is configured to receive data from existing engine system sensors. Thus, these parameters are used in the predictive model to estimate battery health. Any number of parameters, or variables, in the model may be used (e.g., x ₁-x ₆, x ₁-x ₄, etc. ) If the battery diagnostic circuit 214 detects an error, a corresponding alert is sent to a terminal (e.g., user interface) to notify the user to take action. If the difference of the actual voltage compared to the predicted voltage (e.g., based on a maxim, minimum, and/or average voltage from key-on to a successful start) is above a threshold, an alert may be generated.

Further, the battery health may be predicted at various stages. For instance, FIG. 7 shows a logic 500 for constructing a model at the cranking phase. The cranking phase is the initial point when the ignition is triggered (e.g., by a key) and the crankshaft begins to rotate. For instance, the cranking phase is shown in chart 800 of FIG. 8A, where line 802 is the engine speed threshold for a successful start. At the cranking phase, voltage can be predicted where x ₀ is the initial voltage delta at key-on, x ₁ is air temperature, x ₂ is air pressure, x ₃ is coolant temperature or oil temperature, x ₄ is S _clutch (e.g., the clutch status, connect and disconnect) , x ₅ is revolutions per minute (RPM) of engine speed, and X ₆ is the maximum. An anomaly can be detected based on the predicted value compared to the V _delta observed. At 502, the cranking speed is determined. If the cranking speed is in range, the battery anomaly detection algorithm (i.e., the machine learning predicted model of the battery diagnostic circuit 214) is initiated at 504. If a battery anomaly is detected, a battery alert is generated at 506. If the cranking speed is abnormally high, the cranking delta voltage is determined at 508. If the cranking delta voltage is abnormally high, a warning regarding the battery and the starter is generated at 510. If the cranking delta voltage is in range or abnormally low, the battery is operating normally and no alert is generated at 512. Back at 502, if the cranking speed is abnormally low, the cranking delta voltage is determined at 514. If the cranking delta voltage is abnormally high or abnormally low, a warning regarding the battery and the starter is generated at 510. If the cranking voltage is in range, the battery is operating normally and no alert is generated at 512.

The battery health can also be detected at the pre-heat phase. The preheat phase is the duration of the initial cranking before the engine starts or reaches a certain temperature (e.g., a predefined temperature) after the battery sends power to the engine. For instance, the pre-heat phase is shown in chart 810 of FIG. 8B. At the pre-heat phase, voltage can be predicted where, x ₀ is the delta voltage at the pre-heating phase, x ₁ is air temperature, x ₂ is coolant temperature or oil temperature, x ₃ is pre-heat temperature, and x ₄ is the V _max, for instance. The delta voltage is the difference in voltage between the start and end of the pre-heating phase (e.g., during a pre-heating process to increase the charge temperature in order to successfully start the engine in lower temperature areas) . An anomaly can be detected based on the predicted value compared to the pre-heat V _delta observed. At the post-heating phase (after reaching the certain predefined temperature) , after the engine starts and the battery is charging (see chart 820 of FIG. 8C) , an anomaly may be detected based on the predicted voltage compared to the post-heat V _delta observed, where X ₀ is the delta voltage of the post-heating phase, x ₁ is air temperature, x ₂ is coolant temperature or oil temperature, x ₃ is S _clutch, x ₄ is RPM, x ₅ is post-heat temperature, and X ₆ is V _max. The delta voltage is the difference in voltage between the start and end of the post-heating phase (e.g., during a post-heating is a process after engine start to prevent the engine from stalling due to the temperature being too low) . Lastly, when the engine is shut off, the voltage has little discharge, as shown in chart 830 of FIG. 8D. An anomaly here may be detected based on the predicted voltage compared to the observed key-on voltage, where X ₀ is the voltage at the key-on, or cranking, x ₁ is air temperature, x ₂ is coolant temperature or oil temperature, x ₃ is last trip key-off voltage, and x ₄ is delta time between last trip key-off to this trip key-on. In the above description, x ₀-x ₆, for instance may be any of the parameters explained herein.

These four phases as inputs can create the voltage model to estimate battery health. At the same conditions of other variables, if the difference between the observed delta voltage and the predicted delta voltage, a battery anomaly can be determined and an alert generated. For instance, the alert could be sent to the driver, the fleet manager, or a service technician.

The predictive model determined by the battery diagnostic circuit 214 is unique to each vehicle and is a customized method to detect the abnormal performance. For each ESN, the battery diagnostic circuit 214 monitors the battery health under certain constraints and finds abnormal changes based on the ESN’s historical data. If there is an obvious exception, or a particular fault code, the model will output the corresponding alert on the electric oil pressure gauge. A predictive model for an ESN may already be implemented because of a set up model, or a normal model. This model may be used to predict the oil pressure, or the model may be customized. For instance, the telematics unit and edge computing may be used to remotely estimate the health of engine battery. The edge computing technology may clean the data and/or or train the model.

Referring now to FIG. 9, a flow diagram of a method 600 of generating an alert based on a predictive model is shown according to one embodiment. Because the method 600 may be implemented with the controller 150 and in the vehicle 100, reference may be made to one or more features of the controller 150 and the vehicle 100 to explain method 600. Further, method 600 may be implemented with a plurality of vehicles 100 coupled to a plurality of telematics units 130, as such there may be a second customized predictive model for a second engine, for example.

At step 601, internal information and external information for a vehicle with an engine and an aftertreatment system is received. The internal information and the external information may have the same definition as described herein above. At step 602, a predictive model is determined to predict a relationship between a predicted parameter and an actual parameter. The actual parameter may be the internal information from a telematics unit and/or the external information from an external information source. At step 603, a predicted parameter is determined based on the predictive model. For example, the controller may determine the predicted parameter regarding operation of the vehicle system 50 based on execution of the predicted model (e.g., a machine learning predictive model) utilizing a vehicle parameter associated with the predicted parameter. The predicted parameter may be predicted oil pressure associated with an actual parameter, such as at least one of oil pressure, oil temperature, and engine speed. In another example, the predicted parameter may be a probability of coolant performance error, such that the actual parameter includes at least one of coolant temperature, engine speed, torque, air temperature, vehicle speed, and/or performance history of the vehicle in regard to coolant temperature, engine speed, torque, air temperature, vehicle speed over a predetermined duration of time. In another example, the predicted parameter may be a predicated battery health, such that the first and second parameters include at least one of engine battery state of charge, battery voltage, ambient air temperature, etc.

At step 604, the predicted parameter is compared to the actual parameter of the engine to determine a relationship between the predicted parameter and the actual parameter. For instance, if the predicted parameter is the predicted oil pressure, a predicted coolant performance error probability or the predicted battery health, the actual parameter is an actual observed oil pressure, an actual observed error rate, and an actual observed battery health (e.g., voltage) , respectively. A predefined number of iterations may occur to determine the relationship. Based on this relationship using the predictive model, an error value can be detected if a difference between the predicted parameter and the actual vehicle parameter is at or above a predetermined threshold. The error detection and relationship determined can be used to predict the operation of the vehicle system. Thus, at step 605, an alert is generated at a terminal (e.g., the GUI, dial, gauge, etc. ) to notify the user of potential action needed to remedy the issues.

As utilized herein, the terms “approximately, ” “about, ” “substantially” , and similar terms are intended to have a broad meaning in harmony with the common and accepted usage by those of ordinary skill in the art to which the subject matter of this disclosure pertains. It should be understood by those of skill in the art who review this disclosure that these terms are intended to allow a description of certain features described and claimed without restricting the scope of these features to the precise numerical ranges provided. Accordingly, these terms should be interpreted as indicating that insubstantial or inconsequential modifications or alterations of the subject matter described and claimed are considered to be within the scope of the disclosure as recited in the appended claims.

It should be noted that the term “exemplary” and variations thereof, as used herein to describe various embodiments, are intended to indicate that such embodiments are possible examples, representations, or illustrations of possible embodiments (and such terms are not intended to connote that such embodiments are necessarily extraordinary or superlative examples) .

The term “coupled” and variations thereof, as used herein, means the joining of two members directly or indirectly to one another. Such joining may be stationary (e.g., permanent or fixed) or moveable (e.g., removable or releasable) . Such joining may be achieved with the two members coupled directly to each other, with the two members coupled to each other using one or more separate intervening members, or with the two members coupled to each other using an intervening member that is integrally formed as a single unitary body with one of the two members. If “coupled” or variations thereof are modified by an additional term (e.g., directly coupled) , the generic definition of “coupled” provided above is modified by the plain language meaning of the additional term (e.g., “directly coupled” means the joining of two members without any separate intervening member) , resulting in a narrower definition than the generic definition of “coupled” provided above. Such coupling may be mechanical, electrical, or fluidic. For example, circuit A “coupled” to circuit B may signify that the circuit A communicates directly with circuit B (i.e., no intermediary) or communicates indirectly with circuit B (e.g., through one or more intermediaries) .

While various circuits with particular functionality are shown, it should be understood that the controller 150 may include any number of circuits for completing the functions described herein. For example, the activities and functionalities of the oil diagnostic circuit 210, the coolant diagnostic circuit, and the battery diagnostic circuit 214 may be combined in multiple circuits or as a single circuit. Additional circuits with additional functionality may also be included. Further, the controller 150 may further control other activity beyond the scope of the present disclosure.

As mentioned above and in one configuration, the “circuits” may be implemented in machine-readable medium storing instructions for execution by various types of processors, such as the processor 204 of Figure 2. An identified circuit of executable code may, for instance, comprise one or more physical or logical blocks of computer instructions, which may, for instance, be organized as an object, procedure, or function. Nevertheless, the executables of an identified circuit need not be physically located together, but may comprise disparate instructions stored in different locations which, when joined logically together, comprise the circuit and achieve the stated purpose for the circuit. Indeed, a circuit of computer readable program code may be a single instruction, or many instructions, and may even be distributed over several different code segments, among different programs, and across several memory devices. Similarly, operational data may be identified and illustrated herein within circuits, and may be embodied in any suitable form and organized within any suitable type of data structure. The operational data may be collected as a single data set, or may be distributed over different locations including over different storage devices, and may exist, at least partially, merely as electronic signals on a system or network.

While the term “processor” is briefly defined above, the term “processor” and “processing circuit” are meant to be broadly interpreted. In this regard and as mentioned above, the “processor” may be implemented as one or more general-purpose processors, application specific integrated circuits (ASICs) , field programmable gate arrays (FPGAs) , digital signal processors (DSPs) , or other suitable electronic data processing components structured to execute instructions provided by memory. The one or more processors may take the form of a single core processor, multi-core processor (e.g., a dual core processor, triple core processor, quad core processor, etc. ) , microprocessor, etc. In some embodiments, the one or more processors may be external to the apparatus, for example the one or more processors may be a remote processor (e.g., a cloud based processor) . Alternatively or additionally, the one or more processors may be internal and/or local to the apparatus. In this regard, a given circuit or components thereof may be disposed locally (e.g., as part of a local server, a local computing system, etc. ) or remotely (e.g., as part of a remote server such as a cloud based server) . To that end, a “circuit” as described herein may include components that are distributed across one or more locations.

Although the figures and description may illustrate a specific order of method steps, the order of such steps may differ from what is depicted and described, unless specified differently above. Also, two or more steps may be performed concurrently or with partial concurrence, unless specified differently above. Such variation may depend, for example, on the software and hardware systems chosen and on designer choice. All such variations are within the scope of the disclosure.

The foregoing description of embodiments has been presented for purposes of illustration and description. It is not intended to be exhaustive or to limit the disclosure to the precise form disclosed, and modifications and variations are possible in light of the above teachings or may be acquired from this disclosure. The embodiments were chosen and described in order to explain the principals of the disclosure and its practical application to enable one skilled in the art to utilize the various embodiments and with various modifications as are suited to the particular use contemplated. Other substitutions, modifications, changes and omissions may be made in the design, operating conditions and arrangement of the embodiments without departing from the scope of the present disclosure as expressed in the appended claims.

Accordingly, the present disclosure may be embodied in other specific forms without departing from its spirit or essential characteristics. The described embodiments are to be considered in all respects only as illustrative and not restrictive. The scope of the disclosure is, therefore, indicated by the appended claims rather than by the foregoing description. All changes which come within the meaning and range of equivalency of the claims are to be embraced within their scope.

Claims

A system, comprising:

a controller coupled to a vehicle system, the controller structured to:

determine a predicted parameter regarding operation of the vehicle system based on execution of a machine learning predictive model utilizing a vehicle parameter associated with the predicted parameter;

compare the predicted parameter to an actual vehicle parameter of the same operation of the vehicle system to begin to determine a relationship between the predicted parameter and the actual vehicle parameter;

run iterations of the machine learning predictive model to determine the relationship between the predicted parameter and the actual vehicle parameter;

use the machine learning predictive model to detect an error value if a difference between the predicted parameter and the actual vehicle parameter is at or above a predetermined threshold;

use the relationship to predict the operation of the vehicle system; and

generate an alert based on the detected error value.
The system of claim 1, wherein the predicted parameter includes a predicted oil pressure, wherein the vehicle parameter associated with the predicted oil pressure includes at least one of an oil pressure, an oil temperature, an engine speed, an engine fuel rate, an engine load percentage at a current speed, and an ambient air temperature.
The system of claim 2, wherein the engine oil pressure, the engine speed, and the oil temperature are weighted more valuable than other parameters.
The system of claim 1, wherein the predicted parameter includes a probability of coolant performance error, wherein the vehicle parameter associated with the probability of coolant performance error includes at least one of a coolant temperature, a coolant level, an engine speed, an air temperature, a vehicle speed, an oil temperature, an engine intake manifold pressure, an engine fuel rate, an engine load percentage at a current speed, an engine torque, and a barometric pressure.
The system of claim 4, wherein the coolant temperature, the engine speed, the ambient air temperature, the engine torque, the barometric pressure, the engine fuel rate, and the coolant level are weighted more valuable than other parameters.
The system of claim 1, wherein the predicted parameter includes a predicated battery health, wherein the vehicle parameter associated with the predicted battery health includes at least one of an engine battery state of charge, a battery voltage, an ambient air temperature, an ambient air pressure, a coolant temperature, an oil temperature, a clutch status, or an engine speed.
The system of claim 6, wherein the predicted battery health is detected at an engine cranking phase, wherein the parameters used during the cranking phase include the ambient air temperature, the air pressure, the coolant temperature, the oil temperature, the clutch status, and the engine speed.
The system of claim 6, wherein the predicted battery health is detected at an engine pre-heat phase, wherein the parameters used during the pre-heat phase include a delta voltage from a start of the pre-heat phase to an end of the pre-heat phase, the ambient air temperature, the coolant temperature, the oil temperature, a pre-heat temperature, and a maximum voltage.
The system of claim 6, wherein the predicted battery health is detected at an engine post-heat phase, wherein the parameters include a delta voltage from a start of the post-heat phase to an end of the pre-heat phase, the ambient air temperature, the coolant temperature, the oil temperature, the clutch status, the engine speed, a post-heat temperature, and a maximum voltage.
The system of claim 6, wherein the predicted battery health is detected at an engine-off phase, wherein the parameters include initial voltage at key-on, the ambient air temperature, the coolant temperature, the oil temperature, a last trip key-off voltage, and a delta time between last trip key-off to present trip key-on.
The system of claim 1, wherein the alert is provided via a graphical user interface, the graphical user interface being at least one of an electric oil gauge and an electric coolant temperature gauge of the vehicle.
The system of claim 1, wherein the machine learning predicted model is structured to filter data associated with at least one of the vehicle parameter associated with the predicted parameter and the actual vehicle parameter and external information from an external information source.
The system of claim 1, wherein the machine learning predictive model is customized for the vehicle system such that the controller is structured to determine a second predictive model for a second engine.
A system, comprising:

a remote computing system coupled to a plurality of vehicles via a plurality of telematics units, the remote computing system structured to:

gather a plurality of vehicle parameters regarding operation of the plurality of vehicles via the plurality of telematics units;

use a machine learning predictive model to determine a predicted parameter regarding operation of the plurality of vehicles, wherein the machine learning predicted model utilizes the plurality of vehicle parameters associated with the predicted parameter;

use the machine learning predictive model to compare the predicted parameter to an actual vehicle parameter of the same operation of the plurality of vehicles to determine a relationship between the predicted parameter and the actual vehicle parameter;

run iterations of the machine learning predictive model to determine the relationship between the predicted parameter and the actual vehicle parameter;

use the machine learning predictive model to detect an error value if a difference between the predicted parameter and the actual vehicle parameter is at or above a predetermined threshold;

use the relationship to predict the operation of the vehicle; and

generate an alert for at least one vehicle of the plurality of vehicles that is provided via the telematics unit associated with the at least one vehicle based on the detected error value.
The vehicle system of claim 15, wherein the predicted parameter includes at least one of a predicted oil pressure, a probability of coolant performance error, and a predicated battery health.
A method, comprising:

determining a predicted parameter regarding operation of a vehicle system based on execution of a machine learning predictive model utilizing a vehicle parameter associated with the predicted parameter;

comparing the predicted parameter to an actual vehicle parameter of the same operation of the vehicle system to determine a relationship between the predicted parameter and the actual vehicle parameter;

running iterations of the machine learning predictive model to determine the relationship between the predicted parameter and the actual vehicle parameter;

using the machine learning predictive model to detect an error value if a difference between the predicted parameter and the actual vehicle parameter is at or above a predetermined threshold;

using the relationship to predict the operation of the vehicle system; and

generating an alert based on the detected error value.
The method of claim 16, wherein the predicted parameter includes a predicted oil pressure, wherein the vehicle parameter associated with the predicted oil pressure includes at least one of an oil pressure, an oil temperature, an engine speed, an engine fuel rate, an engine load percentage at current speed, and an ambient air temperature.
The method of claim 16, wherein the predicted parameter includes a probability of coolant performance error, wherein the vehicle parameter associated with the probability of coolant performance error includes at least one of a coolant temperature, an engine speed, a torque, an air temperature, a wheel based vehicle speed, an oil temperature, an engine intake manifold pressure, an engine fuel rate, an engine load percentage at current speed, an actual engine torque percent, a barometric pressure, and a performance history of the vehicle system in regard to any of the parameters over a predetermined duration of time.
The method of claim 16, wherein the predicted parameter includes a predicated battery health, wherein the vehicle parameter associated with the predicted battery health includes at least one of an engine battery state of charge, a battery voltage, an ambient air temperature, an ambient air pressure, a coolant temperature, an oil temperature, a clutch status, an RPM, and an engine speed.
The method of claim 16, wherein the predicted battery health is detected at at least one of an engine cranking phase, an engine pre-heat phase, an engine post-heat phase, and an engine-off phase.