Systems, methods, and servers for power management based on evolving routes

Abstract

The present disclosure relates to, but is not limited to, systems, methods, and servers for power management based on evolving routes. Power management is provided. A system for power management includes one or more processors. The processor is configured to detect sensor data indicative of an evolving route from a plurality of sensors for a vehicle. The processor is configured to provide sensor data to a remote server. The processor is configured to receive, from a remote server, a topographic map indicating a state of the evolving route. The processor is configured to predict a future power demand of the vehicle based on the topographic map. The processor is configured to share a power demand between the first power source and the second power source based on the prediction. The processor is configured to generate a control signal to adjust a power output of at least one of the first power source or the second power source according to the apportionment.

Description

BACKGROUND

[0001] The present disclosure relates generally to evolutional route based power management. A series of vehicles, such as off-road transport vehicles, can traverse an evolving route. Operation of the vehicles can depend on the evolution of the route. SUMMARY

[0002] Embodiments relate to a power management system. The power management system includes one or more processors to implement its operations. The power management system is configured to detect, from a plurality of sensors for a vehicle, sensor data indicative of an evolving route. The power management system is configured to provide the sensor data to a remote server. The power management system is configured to receive, from the remote server, a topographical map indicative of a state of the evolving route. The power management system is configured to predict, based on the topographical map, a future power demand of the vehicle. The power management system is configured to apportion, based on the prediction, the power demand between a first power source and a second power source. The power management system is configured to generate a control signal to adjust a power output of at least one of the first power source or the second power source in accordance with the apportioning.

[0003] In some embodiments, the one or more processors are coupled with one transducer of the plurality of transducers to receive first sensor data from the transducer. In some embodiments, the one or more processors are coupled with a virtual sensor to receive second sensor data of the sensor data, the virtual sensor being derived from a combination of two or more transducers of the plurality of transducers. In some embodiments, the evolving route comprises an off-road route comprising loose ground, the topographical map comprising an indication of a condition of the loose ground. The one or more processors can generate the control signal further based on the indication of the condition.

[0004] In some embodiments, the vehicle-based portion of the power management system is configured to cause the vehicle to traverse a first portion of the evolving route in accordance with the adjusted power output. The traversing can be after receiving the terrain map and before receiving the updated terrain map. The vehicle-based portion of the power management system can receive the updated terrain map indicating a second state of the evolving route. The vehicle-based portion of the power management system can predict a second power demand of the vehicle based on the updated terrain map. The vehicle-based portion of the power management system can apportion the second power demand between the first power source and the second power source. The vehicle-based portion of the power management system can adjust the control signal to adjust the power output of at least one of the first power source or the second power source in accordance with the apportioning of the second power demand. The vehicle-based portion of the power management system can cause the vehicle to traverse a second portion of the evolving route in accordance with the adjusted control signal. In some embodiments, the state of the evolving route and the second state of the evolving route each include a plurality of locations corresponding to the vehicle and one or more second vehicles, respectively. The one or more processors are to determine the prediction of the power demand and the second power demand based on the plurality of locations.

[0005] In some embodiments, the one or more processors are to determine the apportioning that satisfies the objective function. The objective function can include a first target value related to a fuel cost of fuel for the first power source. The objective function can include a second target value related to an emission output associated with the fuel. In some embodiments, the one or more processors are to evaluate the objective function using a third target value related to a power source condition related to at least one of a health of the battery or a health of the internal combustion engine.

[0006] In some embodiments, the one or more processors are to determine the prediction of the future power demand of the vehicle based on a vehicle type, a vehicle load, and a vehicle location along the evolving route. In some embodiments, the one or more processors are to determine the apportioning that includes apportioning a positive power output to one of the first power source or the second power source. In some embodiments, the one or more processors are to determine the apportioning that includes apportioning a negative power output to another of the first power source or the second power source.

[0007] Embodiments relate to a power management server comprising a controller coupled with a memory. The controller is configured to receive, from a plurality of telematics interfaces corresponding to a plurality of vehicles, sensor data for operation of a vehicle associated with an evolving route. The controller is configured to estimate, based on the sensor data, a state of the evolving route. The controller is configured to generate, based on the state of the evolving route, a terrain map configured to be extracted by a load prediction system of one of the plurality of vehicles. The controller is configured to transmit the terrain map to the telematics interface corresponding to the vehicle.

[0008] In some embodiments, the controller is configured to receive, from the vehicle, updated sensor data corresponding to operation of the vehicle along the evolving route according to the terrain map. The controller can extract the updated sensor data as a target value for a loss function to generate a loss score. The controller can update a model for predicting the state of the evolving route based on the loss score. The controller can generate an updated terrain map based on the updated model. In some embodiments, the controller can generate the terrain map comprising a location and a speed of the vehicle.

[0009] In some embodiments, the controller is configured to determine a location of at least one vehicle based on sensor data for operation of a power source of the vehicle. In some embodiments, the vehicle comprises a plurality of vehicle types including a transport truck and at least one additional vehicle type. The controller can distinguish between the vehicle types, generating the same terrain map. The controller can provide the same terrain map to each vehicle type.

[0010] In some embodiments, the controller is coupled with one of a plurality of transducers to receive first sensor data of the sensor data, and with a virtual sensor to receive second sensor data of the sensor data, the virtual sensor derived from a combination of two or more transducers. In some embodiments, the controller generates the second sensor data based on a plurality of data elements of the sensor data.

[0011] Embodiments relate to a method of power management. The method comprises receiving, locally at a plurality of telematics interfaces corresponding to a plurality of vehicles, sensor data indicative of operation of each vehicle. The method comprises transmitting, from each telematics interface, the sensor data to a remote server. The method comprises generating, by the remote server, a terrain map based on the sensor data from the telematics interfaces corresponding to the vehicles. The method comprises apportioning, by one of the plurality of vehicles, power between a first power source and a second power source based on the terrain map.

[0012] In some embodiments, the method includes embedding, by the remote server, information into the topographical map. The information can include first information indicative of a surface traveled in the evolving route and second information indicative of a location of the evolving route. The evolving route can include a time-varying component of at least one of the surface traveled or the location.

[0013] In some embodiments, apportioning the power includes predicting, based on the topographical map, a power demand at a future time. In some embodiments, apportioning the power includes generating, based on the power demand and prior to the future time, a control signal to adjust a power output of at least one of the first power source or the second power source. In some embodiments, the control signal includes a first control signal for the internal combustion engine based on a difference between a current exhaust temperature and a predicted exhaust temperature. In some embodiments, the control signal includes a second control signal for the energy storage device to receive or transmit power.

[0014] This summary is an overview of some of the teachings of the present application and is not intended to be an exclusive or exhaustive overview of the present disclosure. In this context the detailed description and examples should be consulted. BRIEF DESCRIPTION OF DRAWINGS

[0015] Figure 1 is a block diagram of a data processing system in accordance with some embodiments.

[0016] Figure 2 is a system diagram of a vehicle in communication with a server remote from the vehicle in accordance with some embodiments.

[0017] Figure 3 is a network diagram of a group of vehicles in communication with a server in accordance with some embodiments.

[0018] Figure 4 is a top view of an environment including an evolving route in accordance with some embodiments.

[0019] Figure 5 is a dataflow diagram of a method of power management in accordance with some embodiments.

[0020] Figure 6 is a flowchart of a method of power management in accordance with some embodiments.

[0021] Figure 7 is a block diagram illustrating an architecture of a computer system that can be used to implement the systems and methods described and illustrated herein. DETAILED DESCRIPTION

[0022] The following is a more detailed description of various concepts and implementations related to systems, servers, and methods related to power management. Before turning to the figures, which illustrate certain exemplary embodiments in detail, it should be understood that the disclosure is not limited to the details or methodology set forth in the description or illustrated in the figures. It should also be understood that the terminology used herein is for the purpose of describing only specific embodiments and is not intended to be limiting.

[0023] With general reference to the figures, various embodiments disclosed herein relate to systems and devices for power management and methods of use thereof. The ground traveled in a route (e.g., mud or gravel roads) can vary over time, such as including mud, dry ground, gravel, obstacles, etc. Further, the location of the ground traveled can itself change over time, such as to avoid deep ruts, or to correspond to changes in a destination or origin. For example, a route of a mining truck can vary according to changes in a destination as a mining area is pushed further down, or changes in environmental conditions. With respect to such dynamic conditions for a route traversal, the route traversed by a vehicle can include an evolving route, where the route can evolve in response to environmental conditions, vehicle operation, usage, or other conditions.

[0024] Energy usage management can improve vehicle operation when traversing a route based on a prediction of vehicle load. For example, a vehicle can adjust a mixture between a base fuel, such as petroleum-based diesel, and an alternative fuel, such as methane or natural gas, can recharge and deploy battery power in an electric hybrid system, and / or can manage power according to any combination of energy sources. However, energy management for an evolving route can be challenging. For example, for a first traversal of a route, it can be optimal to deploy battery power at a particular location, but it can be inappropriate to deploy battery power at the same location in a subsequent traversal (e.g., in the case of a climbing portion instead of continuing down a slope). Thus, static energy management for an evolving route can increase overall fuel usage, emissions, engine wear, or other avoidable parameters.

[0025] Any vehicle in a group traversing an evolving route can collect information related to the conditions of that evolving route. However, vehicles can include varying loads, traction systems, and sensors, making it difficult to compare energy usage between vehicles. For example, some vehicles in a group may include or omit position sensors and may include different energy sources and traction systems. Some vehicles may include single-fuel vehicles, hybrid vehicles, multiple tires, multiple treads, or other variations, allowing energy usage to vary widely among vehicles. According to this disclosure, various vehicles can collect data related to the evolving route, where a controller (e.g., at a remote server) can generate a topographic map based on data generated by the sensor suites of the various vehicles. The controller can distribute the topographic map to the vehicles, allowing them to deploy energy based on the map to manage energy usage. Such energy usage can, for example, extend refueling time, reduce carbon or pollutant emissions, maintain vehicle operation, and otherwise assist the operation of one or more vehicles in the group.

[0026] like Figure 1 As shown, the power management server includes a controller 102 coupled to memory. The controller 102 is configured to receive sensor data 122 from multiple telematics interfaces 104 corresponding to multiple vehicles for vehicle operation associated with an evolving route. The controller 102 is configured to estimate the state of the evolving route based on the sensor data. The controller 102 is configured to generate a topographic map 124 based on the state of the evolving route, the topographic map 124 being configured to be extracted by a load prediction system of one of the multiple vehicles. The controller 102 is configured to transmit the topographic map 124 to the telematics interface 104 corresponding to the vehicle.

[0027] The power management server may be hosted on or remotely to one or more vehicles. A system including components of the power management server may be referred to as data processing system 100. Data processing system 100 may include components of a vehicle-based server or other server (e.g., a remote server) or components that interface with a vehicle-based server or other server (e.g., a remote server). For example, data processing system 100 may include controller 102, which includes one or more processors of a remote server and one or more additional processors (e.g., processors coupled to various vehicles). The processors of the controller may communicate with each other via a network.

[0028] The controller 102, telematics interface 104, vehicle sensors 106, evolving route estimator 108, load predictor 110, power manager 112, and loss function adjuster 114 can each include at least one processing unit or other logic device (e.g., programmable logic array engine) or module configured to communicate with or interface with the data repository 120 or database, or at least one processing unit or other logic device (e.g., programmable logic array engine) or module configured to communicate with or interface with the data repository 120 or database. The controller 102, telematics interface 104, vehicle sensors 106, evolving route estimator 108, load predictor 110, power manager 112, and loss function adjuster 114 can be separate components, a single component, or part of one or more vehicle-based servers or other servers. The vehicle, servers, and various components thereof can include hardware elements, such as one or more processors, logic devices, or circuits. For example, the vehicle and servers can include Figure 7 one or more components or functionalities of the computing devices depicted in FIG. 1.

[0029] The data repository 120 can include one or more local databases or distributed databases, and can include a database management system. The data repository 120 can include computer data storage or memory, and can store one or more of sensor data 122 received from or otherwise determined from data from the sensors 106, a topographical map 124, or vehicle-specific data 126.

[0030] Sensor data 122 can refer to or include information derived from one or more transducers of a sensor (e.g., sensor-transducers); for example, controller 102 can be coupled with one or more transducers and / or virtual sensors. For example, sensor data 122 can include sensor data 122 received from one of a plurality of transducers (e.g., first sensor data), such as a wheel speed indication or position data. Sensor data 122 can include sensor data 122 from one or more virtual sensors (e.g., second sensor data) derived from a combination of two or more transducers. Sensor data 122 received from a transducer can be referred to as raw sensor data 122. Sensor data 122 generated via application of a transformation to raw sensor data 122 can be referred to as processed sensor data 122 (e.g., data processing system 100 can generate second sensor data 122 based on a plurality of data elements of sensor data 122). Processed sensor data 122 originating from a plurality of transducers can be referred to as originating from a virtual sensor. For example, an engine load virtual sensor can indicate engine load from a combination of throttle position sensor, turbo boost sensor, and manifold pressure sensor. A ground deformability virtual sensor can indicate the condition of evolving ground based on wheel speed sensors and position sensors such as an inertial management unit (IMU) or global navigation satellite system (GNSS) (e.g., global positioning system (GPS) or GLONASS). Any of the virtual or other sensor data 122 can be received from an ECU of the vehicle, or via additional sensors 106. For example, sensor data 122 can be received from sensors 106 used for other vehicle operations, or from dedicated sensors for energy management operations disclosed herein.

[0031] Sensor data 122 can include position data such as two- or three-dimensional position received from a position sensor such as a GPS or other GNSS sensor, or other indications of known position such as Wi-Fi signals, Bluetooth beacons, or inertial data of an IMU. Some non-position data can be used to determine a position of the vehicle (e.g., by evolving route estimator 108), as in the case of a vehicle traversing a known path, providing ECU signals indicative of position along the route. For example, a vehicle climbing a hill can provide indications of speed and transmission ratio reduction and engine load increase or wheel slip. Accordingly, ECU data can contain structures indicative of position of the vehicle or condition of the evolving route. Such structures can be extracted according to execution of a machine learning model of evolving route estimator 108 to determine aspects of terrain map 124.

[0032] The terrain map 124 can refer to or include a data map that includes information related to the evolving route. For example, the terrain map 124 can include an evolving path of the traveled evolving route. The traveled path can include physical coordinates of the traveled path along the evolving route, any changes or rate of changes to the traveled path. The traveled path can include a starting point, a destination, waypoints, or other locations along the route. The terrain map 124 can include an indication of the state of the ground traveled. For example, the evolving route can be an off-road route with different ground characteristics depending on usage, environmental conditions, or other time-varying characteristics. The terrain map 124 can include an indication of the level of friction, deformability, or other conditions of loose ground along the route. For example, the terrain map can include an indication of energy usage for one or more portions of the route (e.g., higher energy usage for a muddy portion or lower energy usage for a dry, tightly structured portion).

[0033] The vehicle-specific data 126 can refer to or include specific attributes of the vehicle. For example, the vehicle-specific data 126 can include an indication of the load carried by the vehicle, a state of health (SoH) or state of charge (SoC) of a battery, a fuel level, a number of driven wheels, or a gross tonnage of the vehicle.

[0034] The vehicle-specific data 126 can include aspects of one or more energy sources of the vehicle. The vehicle energy source can include an energy conversion device that consumes fuel, such as an internal combustion engine or a fuel cell. A reference to an energy source can refer to one of any number of fuels for the energy conversion device or the energy conversion device itself. For example, in a hybrid system that includes a dual-fuel engine and a hybrid power system, the energy source can refer to the first and second fuels of the engine or the engine and the battery of the hybrid power system. The energy source can also include the following illustrative embodiments: a flywheel or an electrified (e.g., battery or capacitor) energy storage system, a fuel cell, a compressed gas system, or an ammonia cracker (referring to a thermal storage device of energy or stored hydrogen or methane gas). The vehicle-specific data 126 can include information associated with the fuel, such as a price, a carbon intensity, or an energy output. Thus, some chemically similar fuels, such as petroleum-based diesel and biodiesel, can be referred to as different energy sources.

[0035] The data processing system 100 can include or interface with at least one controller 102 to perform operations to manage execution of the systems and methods described herein. The controller 102 can include one or more processors coupled with memory. The memory can include instructions or other components for performing various operations. The one or more processors of the controller 102 can include specialized circuitry for performing some operations. In some embodiments, portions of the instructions or circuitry can be coupled with different processors of the controller 102. The instructions and circuitry can include instructions for performing various operations disclosed herein as performed by any of the servers, vehicles, or other aspects of the energy management system provided herein. The controller 102 can cause one or more of the disclosed operations, for example, by employing another element of the system 100. Operations of other elements of the disclosed system 100 can be initiated, scheduled, or otherwise controlled by the controller 102. For example, as shown below, operations of the power management system can be performed according to instructions accessible to the controller 102.

[0036] The power management system can detect, from a plurality of sensors of the vehicle, sensor data 122 indicative of an evolving route. The power management system can provide the sensor data 122 to a remote server. The power management system can receive, from the remote server, a topographical map 124 indicative of a state of the evolving route. The power management system can predict a future power demand of the vehicle based on the topographical map 124. The power management system can apportion the power demand between a first power source and a second power source based on the prediction. The power management system can generate a control signal to adjust a power output of at least one of the first power source or the second power source according to the apportionment.

[0037] The controller 102 can include or be coupled with communication electronics. The communication electronics can conduct wired and / or wireless communications. For example, the communication electronics can include one or more wired (e.g., Ethernet, Modbus, PCIe, AXI, or CAN (e.g., J1939)) or wireless transceivers (e.g., Wi-Fi transceivers, Bluetooth transceivers, NFC transceivers, or cellular transceivers). The communication electronics can couple the controller 102 to one or more elements of the system 100. For example, the controller 102 can receive various sensor data 122 associated with the vehicle or transmit various control signals to adjust energy sources of the vehicle via the communication electronics. The controller 102 can exchange information (e.g., commands or status information) with the telematics interface 104 via the communication electronics.

[0038] The data processing system 100 can include or interface with at least one telematics interface 104. The telematics interface can include either: various software components (e.g., a network stack) or hardware components (e.g., a transceiver). For example, the telematics interface 104 can include software components coupled with transceivers of different subsystems or devices, or include dedicated communication hardware. The telematics interface 104 can couple with a vehicle or a remote server to facilitate communication between various components of the data processing system 100. For example, the telematics interface 104 can include instructions, executable code, or circuitry that includes or is configured to interface with communication electronics to enable communication between a vehicle and a server of a system (e.g., via a modem coupled with the telematics interface 104).

[0039] The telematics interface 104 of a vehicle can couple with one or more vehicle sensors 106 to receive sensor data 122 therefrom. For example, the telematics interface 104 can communicatively couple with an electronic control module (ECM) of a vehicle to retrieve sensor data 122 related to engine operation or other energy deployment. The telematics interface 104 can receive sensor data 122 related to vehicle location. For example, the telematics interface 104 can receive information from a location sensor, such as a GNSS, Wi-Fi, or other transceiver configured to receive signals indicative of a place. The telematics interface 104 can communicate any sensor data to another telematics interface 104 remote from the vehicle (e.g., at another vehicle or a remote server).

[0040] The telematics interface 104 of the server can distribute the terrain map 124 to one or more vehicles. For example, the telematics interface 104 of the remote server can generate the terrain map 124 based on sensor data 122 received from multiple vehicles in a set of vehicles, and then distribute the terrain map 124 to individual vehicles in the set of vehicles. The telematics interface 104 can distribute updates to the terrain map 124 to vehicles in response to receiving updated sensor data. In this way, the set of vehicles can receive updates from the collective sensor data 122 provided to the server. In some embodiments, the telematics interface 104 can detect a condition indicative of a loss of communication with the server or other vehicles. For example, the telematics interface 104 can detect a lack of a reply or heartbeat, a received signal strength indication (RSSI), or other non-communication metric. The telematics interface 104 can communicate the loss of communication to other components of the data processing system 100, which can base this to adjust operational modes. For example, the data processing system 100 can operate in an offline mode, or in a peer-to-peer mode with other vehicles in communication with the network.

[0041] The data processing system 100 can include or interface with at least one vehicle sensor 106. The vehicle sensor 106 can include a physical or virtual sensor 106 configured to generate sensor data 122. In some embodiments, the vehicle sensor 106 can be integrated or coupled with a vehicle. For example, the vehicle sensor 106 can be coupled with an ECM, or can be implemented separately from the ECM, as in the example of a GPS module integrated to the telematics interface 104 and separate from the ECM. In some embodiments, the vehicle sensor 106 is disposed separate from the vehicle. For example, the vehicle presence sensor 106 can be disposed at a fuel or recharging point or otherwise along a route to detect presence, speed, or other attributes of vehicles within a range of the sensor 106.

[0042] The data processing system 100 can include or interface with at least one evolving route estimator 108. The evolving route estimator 108 can receive sensor data 122 from various vehicles. The evolving route estimator 108 can be disposed on a server (e.g., a remote server) remote from one or more vehicles. In some embodiments, a vehicle can maintain a local instance of the evolving route estimator 108. For example, the local instance can operate in an offline mode based on a model received from the server or a local (e.g., ad hoc) network in communication with the vehicle.

[0043] The evolving route estimator 108 can receive sensor data 122 from individual vehicles in a set of vehicles in near real-time (e.g., with latency according to transmission of the telematics interface 104 or other in-flight data processing delays) or with timestamps, such that the evolving route estimator 108 can correlate various data 122 of sensors to each other or to locations along the route. For example, the evolving route estimator 108 can receive location information of a vehicle, such as GNSS location data or an indication of presence at a predefined location. The evolving route estimator 108 can receive operational information of a vehicle, such as an indication of power output. For example, the evolving route estimator 108 can receive an indication of engine load, battery deployment, fuel substitution, or other indicators of vehicle operation.

[0044] The evolving route estimator 108 can estimate aspects of the evolving route based on received sensor data. The evolving route estimator 108 can embed information into a terrain map 124 and distribute the terrain map 124 to individual vehicles. For example, the evolving route estimator 108 can provide an updated terrain map 124 at periodic intervals, upon determining an update to the terrain map 124, or based on a threshold change to the terrain map 124.

[0045] The evolving route estimator 108 can correlate operational information with location information. For example, the evolving route estimator 108 can determine that a section of the route (e.g., a downhill section) is associated with braking, battery recharging, or high fuel substitution rates, and determine that another section (e.g., an uphill section) is associated with high engine load, high battery deployment, high exhaust temperatures, and low fuel substitution rates. The evolving route estimator 108 can generate the terrain map 124 based on sensor data received from or otherwise associated with multiple vehicles. For example, the evolving route estimator 108 can receive location data and operational data from a first vehicle, and receive operational data from a second vehicle. The evolving route estimator 108 can correlate the operational data of the second vehicle with the operational data of the first vehicle to determine a location of the second vehicle along the route. For example, upon receiving sensor data 122 from multiple vehicles indicating a right turn followed by a 12% grade climb for 100 meters, the evolving route estimator 108 can extract a turn angle, engine load, or other use of energy sources, and determine a location of the vehicle along the evolving route and a state of the evolving route.

[0046] In some embodiments, the evolving route estimator 108 can receive vehicle-specific data 126 for a vehicle and determine an evolving route based on the vehicle-specific data 126. For example, where a fleet of vehicles includes multiple vehicle types and the multiple vehicle types include transport trucks and other vehicle types, the evolving route estimator 108 can differentiate between the multiple vehicle types, generate the same terrain map, and provide the same terrain map to each of the vehicle types. The multiple vehicle types can include transport trucks and at least one additional vehicle type. The evolving route estimator 108 can determine deformation or resistance of loose ground based on vehicle weight and power. In some embodiments, the evolving route estimator 108 is not configured to receive vehicle-specific data 126. Models operating locally at the vehicle or on servers separate from the evolving route estimator 108 can de-conflate vehicle-specific attributes so that the evolving route estimator 108 can operate in a vehicle agnostic phase space.

[0047] The data processing system 100 can include or interface with at least one load predictor 110. For example, the load predictor 110 can be implemented at or by various vehicles. The load predictor 110 can predict a future load of a vehicle based on an evolving route. For example, the load predictor 110 can extract a portion of the evolving route and determine energy usage of a vehicle traversing the route. The prediction can be based on the evolving route as provided in the terrain map 124 and any vehicle-specific data 126. The prediction of future power demand of a vehicle can be based on vehicle type, vehicle load, and vehicle location along the evolving route.

[0048] The data processing system 100 can include or interface with at least one power manager 112. The power manager 112 can manage various power sources of the vehicle. For example, the power manager 112 can generate control signals to regulate power generated by various vehicle power sources. The power manager 112 can implement or determine a target function associated with operation of the vehicle to meet energy, speed, emissions output, or other demands. For example, the target function can include tight or soft constraints on speed, emissions targets, or fuel costs. The power manager 112 can determine one or more solutions (e.g., local minima) that satisfy the target function. References to optimal operation or solution / satisfaction of the target function refer to local minima. The local minima can be a global minimum or can not be a global minimum. The provided target function parameters are not intended to be limiting. For example, other terms can correspond to battery health or other device life or maintenance, total or continuous hours of operator presence with the vehicle (e.g., labor intensity), etc.

[0049] The power manager 112 can operate over one or more predefined time periods, which can be referred to as lookahead windows. For example, the power manager 112 can determine solutions to the target function based on a lookahead period of thirty seconds, five minutes, etc. In some embodiments, the power manager 112 can operate based on a weighted average of one or more windows based on a confidence associated with the window. For example, frequency changing cutback locations can be associated with a low confidence, while static climbing sections can be associated with a higher confidence.

[0050] The data processing system 100 can include or interface with at least a loss function adjuster 114. The loss function adjuster 114 can adjust various models based on a difference between predicted performance and implemented performance. For example, the loss function adjuster 114 can provide updates to models used by the evolving route estimator 108, the load predictor 110, and the power manager 112. In some embodiments, a vehicle can maintain a local instance of the loss function adjuster 114. For example, the local instance can operate in an offline mode based on models received from a server or a local (e.g., ad hoc) network in communication with the vehicle. A local instance of a model (e.g., the evolving route estimator 108 or the loss function adjuster 114) can operate with lower data resolution or less input data than a remote model coupled with another vehicle. The data processing system 100 can switch between operation based on inputs recovered from a remote model and a local model according to a communication connection with a source of the remote model.

[0051] The loss function adjuster 114 can receive updated sensor data from the vehicle corresponding to vehicle operation along the evolving route according to the terrain map 124. The loss function adjuster 114 can extract the updated sensor data as a target value for a loss function to generate a loss score. The loss function can operate based on a difference between expected data and achieved data, referred to as a loss. The loss function adjuster 114 can update the model used to predict the state of the evolving route based on the loss score. The loss function adjuster 114 can communicate the updated model to the evolving route estimator 108 to cause the evolving route estimator 108 to generate an updated terrain map based on the updated model.

[0052] By way of example, with reference in part to the load predictor 110, the loss function adjuster 114 can receive a load prediction and load data corresponding to the load prediction. The loss function adjuster 114 can determine a deviation between the prediction and a measured value corresponding to the prediction. The loss function adjuster 114 can retrain individual machine learning models using the measured value and provide the updated models to the models. For example, the loss function adjuster 114 can be disposed on a server remote from the vehicle and provide updated models for execution locally at one or more vehicles. In some embodiments, the models are independent of vehicle type. In some embodiments, the models are trained based on vehicle type.

[0053] Figure 2 An example data processing system is depicted including a vehicle 202 and a server 204 remote from the vehicle 202, the vehicle 202 having portions of a vehicle-based system. The vehicle 202 includes one or more first processors 206, and the server 204 includes one or more second processors 208 of the controller 102. Additional servers 204 or vehicles 202 can include additional processors, but for brevity of the present disclosure, only one processor is depicted in Figure 2 An illustrative example embodiment is depicted in which the vehicle 202 includes a first instance of a telematics interface 104A; the server 204 includes a second instance of the telematics interface 104B. The telematics interfaces 104 are communicatively coupled, for example, via direct communication, network communication, or other coupling. Information such as sensor data 122 and terrain map 210 that is shown as being exchanged between the vehicle 202 and the server 204 is communicated via the telematics interfaces 104. In Figure 2 Separate connections between components of the vehicle 202 and the server 204 are provided in to depict logical connections between the components, however, such data is communicated over one or more links of the telematics interfaces 104.

[0054] The vehicle includes vehicle sensors 106 to determine location data, operational data, or other sensor data 122 associated with the vehicle or the evolving route traversed by the vehicle 202. The sensor data 122 generated by the vehicle sensors 106 can be used locally at the vehicle to control the vehicle. The vehicle sensors 106 provide the sensor data 122 to the evolving route estimator 108 on the server 204. The evolving route estimator 108 generates a terrain map 210 based on the sensor data 122 (and any other sensor data 122 that can be received from additional vehicles 202). The evolving route estimator 108 provides the terrain map 210 to the load predictor 110, whereby the load predictor 110 uses the terrain map 210 to predict a future load used by the vehicle 202. The evolving route estimator 108 can provide the same terrain map 210 to various vehicles of one or more types, which can predict a future load based on the terrain map 210 and vehicle-specific data 126. The load predictor 110 provides the prediction to the power manager 112. For example, the load predictor 110 can provide a predicted power demand for a defined time period, such as thirty seconds, one minute, or five minutes. The defined time period can correspond to the operation of a power source. For example, a small battery or supercapacitor bank can correspond to a lookahead window of tens of seconds, while an ammonia cracker, fuel cell, or large battery can correspond to a lookahead window of minutes. In some embodiments, the lookahead window can depend on the confidence of the prediction or the impact on the objective function (e.g., where a longer lookahead window does not produce a threshold amount of positive impact on the solution to the objective function).

[0055] The power manager 112 is configured to apportion the predicted power demand between the power sources 218, 220 of the vehicle 202. The apportioning can include apportioning of positive power output (e.g., drawing power from a fuel cell, internal combustion engine, or battery). The apportioning can include apportioning of negative power output (e.g., engine braking of an internal combustion engine or storing power in a battery, ammonia cracker, or other energy source). The apportioning can apportion power between power sources. For example, the apportioning can apportion power from an internal combustion engine to a battery storage device, or between fuel sources of an internal combustion engine. The power manager 112 can adjust control signals for various energy sources to meet the objective function based on the apportioning by the load predictor 110. For example, the power manager 112 can adjust the operation of an engine, fuel cell, battery, flywheel, or other energy source. The adjusting can include drawing or absorbing power from a power source. For example, the adjusting can charge a battery or deploy energy, or cause an internal combustion engine to deploy energy or perform engine braking.

[0056] The vehicle 202 can host various models for local execution. The models can be implemented according to any of a variety of architectures or a weighted set thereof. For example, the data repository 120 local to the vehicle 202 can include a sensor transformation model 212 to determine vehicle operating parameters (e.g., total power output, traction, or position information) based on sensor data 122. The data repository 120 local to the vehicle 202 can include a load prediction model 214 to predict future power demands based on a terrain map 210 and vehicle-specific data 126. The data repository 120 local to the vehicle 202 can include an objective function 216 to be satisfied according to parameters such as fuel cost, speed, emissions, engine wear, or other aspects of vehicle operation. The emissions output (e.g., mass volume or flow of gaseous or particulate matter) can be referred to as an objective function value or objective value.

[0057] The vehicle 202 can receive various models from the server 204. For example, the server can iterate the models based on a difference between a predicted result and an achieved result. Specifically, the loss function adjuster 114 can receive prediction data and achieved data 222 for each of the sensor transformation model 212, the load prediction model 214, and the objective function 216, and provide an updated instance of the sensor transformation model 212, the load prediction model 214, and the objective function 216 to one or more vehicles in the set of vehicles (e.g., each vehicle 202 in the set of vehicles).

[0058] In some embodiments, the vehicle 202 includes a local instance of the evolving route estimator 108A or a local instance of the loss function adjuster 114A. The telematics interface 104A for the vehicle can retrieve models from the server 204 to operate the local models and operate based on the local models upon detecting a loss or degradation of the communicative coupling to the server 204. Upon reestablishing the communicative coupling, the telematics interface 104A for the vehicle can transfer stored data to the server 204 to assist the server in generating an updated terrain map 210 or model. That is, in response to detection of a condition of the communicative coupling, sensor data 122 derived from the vehicle sensors 106 can be transferred to the local evolving route estimator 108A. The local evolving route estimator 108A can generate a local instance of the terrain map 210 for extraction by the load predictor 110 and the power manager 112, which can communicate achieved data 222A to the local instance of the loss function adjuster 114A.

[0059] As Figure 3As shown, the system 300 can include various vehicles, such as a first vehicle 202A, a second vehicle 202B, and a third vehicle 202C (collectively, vehicles 202). The system 300 can be referred to as a fleet of vehicles. The vehicles 202 are in network communication with a remote server 204. For example, the data processing system 100 can include or interface with network devices in a network 301 to exchange information between the vehicles 202 and any number of servers 204. For example, the telematics interface 104 of each of the vehicles 202 can interface with the telematics interface 104 of at least one server 204. The telematics interface 104 can include hardware or software components configured to communicate with other telematics interfaces 104 via any network architecture, such as a star or mesh topology. The network 301 can include a computer network, such as an Ethernet, controller area network 301 (CAN), local interconnect network 301 (LIN), peripheral component interconnect express (PCIe), the Internet, a local, wide, metro, or other area network 301, an intranet, a cellular network 301, a satellite network 301, and other communication networks 301 such as Bluetooth or data mobile telephone networks 301. The network 301 can be public or private. The various elements of the data processing system 100 can communicate over the network 301.

[0060] The vehicles 202 can be one or more types. For example, the vehicles 202 can vary according to gross weight, payload, drivetrain, or energy source. At least one of the vehicles 202 can include multiple power sources, such as an internal combustion engine 302 and a battery 304 of the first vehicle. Any of the vehicles 202 can include any combination of one or more fuel-consuming sources or other energy sources, such as internal combustion engines, fuel cells, accumulators, ammonia crackers, etc. The fuel-consuming energy sources can be configured to operate with one or more fuels. For example, an internal combustion engine or fuel cell can operate with two or more fuels having different energy densities, carbon intensity of emissions, or other attributes. For example, a hydrogen fuel cell can operate with “blue” or “green” hydrogen associated with different emission profiles, or an internal combustion engine can be configured to receive one or more chemically distinct fuels, such as diesel fuel and an alternative fuel (e.g., hydrogen, methane, or natural gas), whereby the power management system 112 of the vehicle 202 can apportion energy between the energy sources, for example, replacing a relatively large amount of low-cost or low-emission fuel during low-load periods of time and a higher-cost or higher-emission fuel during higher-load periods of time.

[0061] Each of the vehicles 202 can traverse an evolving route extending between an origin and a destination (including an envisioned evolving route for round trips that share a spatial location for the origin and the destination). Each of the vehicles 202 can include a suite of sensors including, for example, location sensors, fuel level sensors, exhaust temperature sensors, wheel speed or ground speed sensors, cylinder compression sensors, etc. In some embodiments, at least a portion of the sensors are coupled with the ECM. The vehicles 202 can be configured to determine properties of the ground traveled upon locally (e.g., in conjunction with operation of a traction control system or in processes performed to provide data to a remote server to aid the systems and methods described herein).

[0062] In some embodiments, a server 204 remote from the vehicles 202 can receive raw or processed sensor data 122 (e.g., traction control information) from the vehicles 202. The server 204 can generate a terrain map 124 based on the received data. For example, the server 204 can generate a terrain map 124 including conditions of the ground traveled upon in the evolving route and locations of the ground traveled upon, where either of the conditions of the ground traveled upon and the locations of the ground traveled upon can vary over time. In some embodiments, the terrain map 124 is vehicle-agnostic, such as including variability or slope of the route, whereby the load predictor 110 executed locally at each vehicle 202 is configured to predict power demand based on vehicle-specific data 126 (e.g., vehicle-specific properties stored or sensed locally at the vehicle). In some embodiments, the terrain map 124 includes information related to a specific type of vehicle. For example, the terrain map 124 can include an indication of energy spent or received by a loaded or unloaded vehicle 202 traversing the route or based on the vehicle type, whereby the load predictor 110 at each vehicle 202 is configured to predict load based on the received information.

[0063] Figure 4 A top view of an environment 400 including an evolving route 410 is provided in accordance with some embodiments. The depicted environment includes a mining site, although embodiments of the present disclosure are not limited to such a site. The environment can include infrastructure sites such as fueling or recharging points, as well as any number of vehicles 202 or other objects of interest. One or more of the vehicles 202 can include different types of vehicles. For example, the vehicles 202 can include loaded or unloaded vehicles 202, haul trucks, water trucks, tractor trailer trucks, dump trucks, fuel trucks, etc. For example, the environment 400 can include a group of vehicles 202 in communication with a server 204 (e.g., a fleet of vehicles 202), where the server 204 is configured to generate a terrain map 124 based on sensor data 122 received from the vehicles 202. In some embodiments, the server 204 is configured to generate a terrain map 124 based on sensor data 122 received from a plurality of vehicles 202. In some embodiments, the server 204 is configured to generate a terrain map 124 based on sensor data 122 received from a plurality of vehicles 202 of a same type. In some embodiments, the server 204 is configured to generate a terrain map 124 based on sensor data 122 received from a plurality of vehicles 202 of different types. Figure 3The server 204 can update the terrain map 124 based on the various vehicles 202 of the fleet of vehicles). The server 204 can update the terrain map 124 based on information received from the fleet of vehicles, which can generate more frequent or more accurate instances of the terrain map 124 relative to implementations of individual vehicles 202.

[0064] The evolving route 410 can extend from a starting location 402 at an upper portion of the mine site and a destination location 404 at a lower portion of the mine site, including the starting location 402 and the destination location 404. The traversed ground in the evolving route 410 is depicted as a substantially helical route connecting the upper and lower portions. The traversed ground can extend further downward as material is removed from the lower portion, such that each subsequent traversal of the evolving route 410 extends slightly further. Further, the traveled ground of the evolving route 410 can adjust over time in response to vehicle traffic, environmental conditions, ground re-paving, grading operations, or other route reconstruction activities. Accordingly, the composition or location of the traveled ground can vary between traversals of the evolving route 410 by various vehicles 202. The server 204 can generate updated instances of the terrain map 124 in response to updated sensor data 122 received from the vehicles indicating changes in the state.

[0065] The vehicle 202 can traverse a first portion 410A of the evolving route 410 according to a first adjusted output (e.g., based on the first terrain map 124). For example, the traversal of the first portion 410A can be after receiving the first terrain map 124 and before receiving the updated terrain map 124. The vehicle-based portion of the power management system 100 can cause the traversal of the vehicle 202. The vehicle 202 can receive an updated terrain map 124 indicating a second state of the evolving route 410. The vehicle 202 can predict a second power demand of the vehicle based on the updated terrain map 124 and apportion the second power demand between the first power source and the second power source. The vehicle 202 can adjust the control signal to adjust a power output of at least one of the first power source or the second power source according to the apportioning of the second power demand and cause the vehicle 202 to traverse a second portion 410B of the evolving route 410 according to the adjusted control signal.

[0066] The state of the evolving route 410 can include a location corresponding to a plurality of vehicles traversing the evolving route 410, including the at least one vehicle 202 receiving the updated terrain map 124. In some embodiments, the prediction of the power demand is based on the location of the various vehicles.

[0067] The power manager 112 can determine an apportionment that satisfies a target function that includes target values regarding fuel cost for a power source, emissions associated with the fuel, or power source conditions related to SoH of a battery or internal combustion engine.

[0068] The route traversal can depend on the position or speed of other vehicles 202 along the evolving route 410, such as in cases where a side line 406 can not be available for one vehicle 202 to overtake another, or in cases where a vehicle 202 is waiting to be loaded. The presence or speed of various vehicles 202 and the condition or position of the ground being traveled in the evolving route 410 can be included in the evolving route 410 (e.g., in the terrain map 124). Thus, the load predictor 110 can make predictions based on congestion of the evolving route 410. For example, in cases where vehicles are queued to be loaded, such as for a loading operation at a lower level, the predicted load of vehicles 202 approaching the queue 408 can be strongly associated with low fuel usage, as the speed of arrival at the queue can not affect the completion of traversing the evolving route 410.

[0069] Figure 5 A dataflow diagram of a method 500 of energy management according to some embodiments is provided. The method 500 can be performed by a data processing system 100 including components hosted by at least one of a vehicle 202 and a server 204 remote from the vehicle 202. For example, the method 500 can be performed for multiple vehicles 202 of the same type or different types.

[0070] At operation 502, the method 500 includes collecting data from various sensors-transducers of the vehicle 202. For example, the data collection can include raw sensor data 122. Operation 502 can be performed asynchronously by various vehicles 202 in a fleet of vehicles. The performance of operation 502 can be referred to as a data collection layer.

[0071] At operation 504, the method 500 includes data interpolation from various sensors-transducers of the vehicles 202. For example, a model can interpolate engine load or other data in the raw sensor data 122 to generate processed sensor data 122 (e.g., virtual sensor data 122). The model can be locally disposed on any of the individual vehicles 202 in the set of vehicles or disposed remotely therefrom (e.g., on a server 204 remote from the vehicles 202). The model can include a lookup table (LUT), a discrete function, or a machine learning model trained to determine aspects of operation of the vehicle 202 based on the sensor data 122, such as engine load, location, or a state of a surface traveled upon. Execution of operation 504 can be referred to as a data pre-processing layer, and can include operations performed locally at each vehicle 202 in the set of vehicles or performed at a server remote from one or more (e.g., all) vehicles 202.

[0072] At operation 506, the method 500 includes generating a terrain map 124 based on data collected at operation 502 or processed at operation 504. The terrain map 124 can be generated via one or more machine learning models. For example, separate models can be executed to generate different aspects of the terrain map 124, such as a location model, a state of a surface traveled upon model, a traffic condition model, etc. Execution of operation 506 can be performed at a server 204 remote from one or more vehicles 202.

[0073] At operation 508, the method 500 includes predicting power demand based at least on the terrain map 124. In some embodiments, operation 508 is performed locally at the individual vehicles 202 in the set of vehicles, and can also be based on vehicle-specific data 126 generated at or locally stored at the vehicles 202. The prediction can be made using one or more machine learning models. The machine learning models can be shared across the individual vehicles 202. One or more models can be trained or updated based on a loss function between predicted data and implementation data. For example, the predicted data and implementation data can be communicated from each of the plurality of vehicles 202 to a loss function adjuster 114 executing at a server 204 to update a common model and distribute the updated model to the individual vehicles 202.

[0074] At operation 510, the method 500 includes outputting the predicted power demand to a power manager 112. The power manager 112 can refer to a local instance of a power management system (PMS) that is executed locally at each vehicle in the set of vehicles. The power manager 112, in turn, can provide control signals to various energy sources of the vehicle 202 based on the predicted demand. The performance of operation 510 can be performed locally at the various vehicles 202. The control signals can be generated according to an objective function or other model that can be updated by the loss function adjuster 114 on the server.

[0075] Figure 6 A flowchart of a method 600 of power management according to some embodiments is provided. The method 600 can be performed by the data processing system 100 including components hosted by at least one of the vehicles 202 and the server 204 remote from the vehicles 202. The method 600 can be performed by the remote server 204 according to the method 500 of power management performed by a system including the remote server 204 and one or more vehicles 202. For example, the method 600 can be performed for a plurality of vehicles 202 of the same type or different types. Briefly, the method 600 includes receiving sensor data 122 indicative of operation of each vehicle 202 locally at telematics interfaces 104 corresponding to the plurality of vehicles 202 at operation 602. At operation 604, the method 600 includes transmitting the sensor data 122 from each of the telematics interfaces 104 to the remote server 204. At operation 606, the method 600 includes generating a terrain map 124 by the remote server 204 based on the sensor data 122 from the telematics interfaces 104 corresponding to the vehicles 202. At operation 608, the method 600 includes apportioning power between a first power source 218 and a second power source 220 by at least one of the vehicles 202 based on the terrain map 124. Figure 5

[0076] Referring again to operation 602, the method includes receiving sensor data 122 indicative of operation of each vehicle 202 at a plurality of telematics interfaces 104 corresponding to the corresponding vehicles 202. For example, the sensor data 122 can include raw sensor data 122 received by the sensor-transducers or processed sensor data 122 derived or otherwise processed (e.g., smoothed or averaged) from the plurality of sensor-transducers.

[0077] ​Referring again to operation 604, the method 600 includes transmitting the sensor data 122 from each of the telematics interfaces 104 to the server 204. The vehicles 202 can traverse an ever-evolving route (also referred to as an evolving route 410, but without limiting effect). The server 204 is disposed remote from at least one of the vehicles 202. For example, the server 204 can be hosted locally at one of the vehicles 202 remote from the other vehicles 202, or hosted at a remote data site (e.g., a back office or data center location). The (e.g., remote) server 204 can embed information into the topographical map 124. The information can include information indicative of the ground traveled in the ever-evolving route (e.g., information related to load on the engine as compared to distance traversed by the vehicle 202). The information can include information indicative of a location of the ever-evolving route, such as a GPS or other location sensor of the vehicle 202, or operational data such as a pattern of engine load data matching an operational pattern of the vehicle 202 traversing the ever-evolving route (e.g., the evolving route estimator 108 can determine a location of at least one vehicle 202 based on sensor data of operation of a power source for the vehicle 202, such as engine load). The ever-evolving route indicated by the topographical map can include a time-varying component with respect to the ground traveled or the location along the ever-evolving route. For example, the ever-evolving route can adjust to change a distance or location of one or more segments of the ever-evolving route, or change a coefficient of friction, resistance, traction, roughness, plasticity, load bearing capacity, permeability, or other aspect of the ground traveled.

[0078] Referring again to operation 606, the method includes generating, by the server 204, a topographical map 124 based on the sensor data 122. For example, the topographical map 124 can include an indication of a location of the ever-evolving route, a location and speed of the vehicle 202 along the ever-evolving route, or a ground state of the ground traveled along the ever-evolving route. The ever-evolving route indicated in the instance of the topographical map 210 can be determined based on a previous instance of the topographical map 210. The evolving route estimator 108 can determine an updated topographical map 210 based on a change from a previous instance of the topographical map 124, via a maximum or minimum threshold for change between maps.

[0079] Referring again to operation 608, the method 600 includes apportioning power between the first power source 218 and the second power source 220 based on the terrain map 210 by one of the vehicles 202. The apportioning can be based on predicted power demands at future times (e.g., in one or more look-ahead windows). The predicted power demands can be based on the terrain map 124. For example, the predicted power demands can be based on the slope, distance, composition, or state of the ground traveled upon. The predicted power demands can be based on vehicle-specific data 126, such as the load carried by the vehicle 202, other aspects of the vehicle weight or vehicle type, or energy sources (e.g., energy intensity or emissions associated with the energy sources). The power manager 112 can generate control signals to adjust the power output of at least one of the first power source 218 or the second power source 220 based on the power demands and prior to the future times. For example, the power manager 112 can adjust the use of one or more fuels, recharge of energy storage devices (e.g., batteries, compressed gas systems, flywheels, or chemical processes such as cracking ammonia to generate hydrogen gas), or the like.

[0080] In some embodiments, the generation of the control signals includes generating control signals for the internal combustion engine based on a difference between a current exhaust temperature and a predicted exhaust temperature. For example, the control signals can maintain an exhaust temperature associated with an efficiency band of the engine, or according to an emission level associated with an exhaust temperature range (e.g., to affect operation of an aftertreatment system). The illustrative examples provided are not intended to limit the present disclosure; the control signals can be generated based on manifold intake pressure, NOx emission measurements, or any other sensor data 122 available to the controller 102. For example, the current exhaust temperature can be substituted for other measured or otherwise derived sensor data 122, and the predicted exhaust temperature can be substituted for another prediction of future sensor data (e.g., according to the load predictor 108). In some embodiments, the generation of the control signals includes generating control signals for the energy storage device to receive or transmit power. For example, the energy storage device can include an energy source of a battery, whereby the control signals can cause the battery to charge (e.g., based on regenerative braking or via operation of an energy source that consumes fuel).

[0081] Figure 7is a block diagram illustrating an architecture of a computer system that can be used to implement elements of the systems and methods described and illustrated herein. The computer system or computing device 700 can include or be used to implement the controller 102 or components thereof, as well as components of the vehicle 202. For example, an instance of the computing device 700 can be implemented at any of the various vehicles 202 or the server 204 remote therefrom. The computing system 700 includes at least one bus 705 or other communication device for communicating information and at least one processor 710 or processing circuit coupled to the bus 705 for processing information. The computing system 700 also can include one or more processors 710 or processing circuits coupled to the bus for processing information. The computing system 700 also includes at least one main memory 715, such as a random access memory (RAM) or other dynamic storage device, coupled to the bus 705 for storing information and instructions to be executed by the processor 710. The main memory 715 can be used for storing information during processing by the processor 710. The computing system 700 further can include at least one read only memory (ROM) 720 or other static storage device coupled to the bus 705 for storing static information and instructions for the processor 710. A storage device 725, such as a solid state device, magnetic disk or optical disk, can be coupled to the bus 705 for persistently storing information and instructions (e.g., for the data repository 120).

[0082] The computing system 700 can be coupled via the bus 705 to a display 735, such as a liquid crystal display or active matrix display. An input device 730, such as a keyboard or mouse, can be coupled to the bus 705 for communicating information and commands to the processor 710. The input device 730 can include a touchscreen display 735.

[0083] The processes, systems and methods described herein can be implemented by the computing system 700 responsive to the processor 710 executing an ordered listing of instructions contained in a main memory 715. Such instructions can be read into the main memory 715 from another computer-readable medium, such as the storage device 725. Execution of the ordered listing of instructions contained in the main memory 715 causes the computing system 700 to implement the illustrative processes described herein. One or more processors in multiple processing arrangement can also be used to execute the instructions contained in the main memory 715. Hard-wired circuitry can be used in place of or in combination with software instructions for implementation of the systems and methods described herein. The systems and methods described herein are not limited to any specific combination of hardware circuitry and software.

[0084] Although the systems and methods described and illustrated herein have been described in Figure 7Example computing systems are described, but the subject matter including the operations described in this specification can be implemented in other types of digital electronic circuitry, or in computer software, firmware, or hardware, including the structures disclosed in this specification and their structural equivalents, or in combinations of one or more of the same.

[0085] References to “or” can be construed as inclusive so that any terms described using “or” can indicate any of a single, more than one, and all of the described terms. A reference to at least one of a conjunctive list of items can be construed as an inclusive “or” to indicate any of the single, more than one, and all of the items in the conjunctive list. For example, a reference to “at least one of “A” and “B” can include only “A”, only “B”, as well as both “A” and “B.” Such references used in conjunction with “comprising” or other open terminology can include additional items.

[0086] As used herein, the terms “approximately,” “about,” “substantially,” and like terms are intended to have a broad meaning in harmony with the common and accepted usage of these terms by intending to be aligned with the broad meaning commonly used and understood by those having ordinary skill in the areas of the subject matter of the present disclosure. It is noted that some articulations used herein are used for the purpose of depiction rather than limitation, and that skilled artisans who are familiar with the subject matter that is the focus of the present disclosure should understand that the terms are intended to have meanings that are consistent with common usage by those in the relevant art and that the scope of the description is not intended to be limited to the precise ranges provided. Accordingly, these terms should be interpreted as indicating a possible non-substantial or immaterial modification or change can be made to the subject matter described and claimed without departing from the scope of the disclosure set forth in the appended claims.

[0087] It should be noted that the terms “exemplary” and variations thereof, as used herein to describe various embodiments, are intended to indicate that these embodiments are possible examples, representations, or illustrations of possible embodiments (and such terms are not intended to imply that any of the described embodiments necessarily constitute a special or preferred example of the described subject matter). It should be noted that the terms “exemplary” and variations thereof, as used herein to describe various embodiments, are intended to indicate that these embodiments are possible examples, representations, or illustrations of possible embodiments (and such terms are not intended to imply that any of the described embodiments necessarily constitute a special or preferred example of the described subject matter).

[0088] As used herein, the term "coupled" and variations thereof, means the joining of two members directly or indirectly to one another. Such joining can be stationary (e.g., permanent or fixed) or moveable (e.g., removable or releasable). Such joining can be achieved either with or without using an intervening intermediary member. If "coupled" or variations thereof are modified with an additional term such as "directly", then the "coupled" or variations thereof is modified such that the plain language meaning of the additional term is retained (e.g., "directly coupled" means that two members are joined to one another without any intermediary member). Such joining can be mechanical, electrical, or fluidic. For example, circuit A is communicatively "coupled" to circuit B can mean that circuit A is in direct communication with circuit B (i.e., no intermediaries) or in indirect communication with circuit B (e.g., through one or more intermediaries).

[0089] References herein to the position of an element (e.g., "top," "bottom," "above," "below") are merely used for descriptive purposes. It is noted that the orientation of the various elements can differ according to other example embodiments, and such variations are intended to be encompassed by the present disclosure. The term "or" is used in its inclusive sense (and not in its exclusive sense) as used in the "either-or" sense to mean one, some, or all of the elements in the list.

[0090] Although the drawings and description can show a specific order of method steps, the order of the steps can differ from what is depicted and described unless specified differently above. Also, two or more steps can be performed concurrently or with partial concurrence. Such variation can depend, for example, on the software and hardware systems chosen and on designer choice. All such variations are within the scope of the disclosure. Likewise, software implementations could be accomplished with standard programming techniques with rule-based logic and other logic to accomplish various connection steps, processing steps, comparison steps, and decision steps.

[0091] It is important to note that the construction and arrangement of the systems and methods shown in the various example embodiments is illustrative only. Although only one or two examples from among a wide range of variants have been described, these are also merely illustrative and in no way limit the scope of the application. One of ordinary skill in the art will appreciate the many possible variations and alternatives within the scope of the many embodiments disclosed herein. Further, many of the components described herein have a hardware and software counterpart. Accordingly, but not by way of limitation, these components can be implemented using standard programming techniques with rule-based logic and other logic to accomplish various connection steps, processing steps, comparison steps, and decision steps.

Claims

1. A system for power management, the system comprising: - one or more processors configured to: -- detect, from a plurality of sensors for a vehicle, sensor data indicative of an evolving route; -- provide the sensor data to a remote server; -- receive, from the remote server, a topographic map indicative of a state of the evolving route; -- predict, based on the topographic map, a future power demand of the vehicle; -- apportion, based on the prediction, the power demand between a first power source and a second power source; and -- generate a control signal to adjust a power output of at least one of the first power source or the second power source in accordance with the apportionment.

2. The system of claim 1, wherein: the one or more processors are coupled with one transducer of a plurality of transducers to receive first sensor data from the one transducer; and the one or more processors are coupled with a virtual sensor to receive second sensor data of the sensor data, the virtual sensor being derived from a combination of two or more transducers of the plurality of transducers. the evolving route comprises an off-road route, the off-road route comprising loose ground, the topographic map comprises an indication of a condition of the loose ground, and the one or more processors are configured to generate the control signal further based on the indication of the condition.

3. The system of claim 1, wherein, a vehicle-based portion of the power management system is configured to:

4. The system of claim 1, wherein, after receiving the topographic map and before receiving an updated topographic map, cause the vehicle to traverse a first portion of the evolving route in accordance with an adjusted power output; receive the updated topographic map indicative of a second state of the evolving route; based on the updated topographic map, predict a second power demand of the vehicle; apportion the second power demand between the first power source and the second power source; adjust the control signal to adjust a power output of at least one of the first power source or the second power source in accordance with the apportionment of the second power demand; and cause the vehicle to traverse a second portion of the evolving route in accordance with the adjusted control signal.

5. The system of claim 4, wherein: the state of the evolving route and the second state of the evolving route each comprise a plurality of locations corresponding to: the vehicle; and one or more second vehicles; and the one or more processors are to determine the prediction of the power demand and the second power demand based on the plurality of locations. the one or more processors are to determine the apportionment that satisfies an objective function, the objective function comprising: a first objective value related to a fuel cost of fuel for the first power source; and a second objective value related to an emission output associated with the fuel. the one or more processors are to evaluate the objective function using:

6. The system of claim 1, wherein, a third objective value related to a power source condition, the power source condition related to at least one of a health condition of a battery or a health condition of an internal combustion engine. the one or more processors are to evaluate the objective function using: a third objective value related to a power source condition, the power source condition related to at least one of a health condition of a battery or a health condition of an internal combustion engine.

7. The system of claim 6, wherein, ​ ​ 8. The system of claim 1, wherein, The one or more processors are to determine a prediction of the future power demand of the vehicle based on: a vehicle type; a vehicle load; and a vehicle position along the evolving route.

9. The system of claim 1, wherein, The one or more processors are to determine the apportionment, the apportionment comprising: apportioning positive power output to one of the first power source or the second power source; and apportioning negative power output to the other of the first power source or the second power source.

10. A vehicle power management server comprising: a controller coupled with a memory, the controller to: receive sensor data for vehicle operation associated with an evolving route from a plurality of telematics interfaces corresponding to a plurality of vehicles; estimate a state of the evolving route based on the sensor data; generate a terrain map configured for extraction by a load prediction system of one of the plurality of vehicles based on the state of the evolving route; and transmit the terrain map to a telematics interface corresponding to the vehicle. The controller is to:

11. The vehicle power management server of claim 10, wherein, receive updated sensor data from the vehicle corresponding to vehicle operation along the evolving route according to the terrain map; extract the updated sensor data as a target value of a loss function to generate a loss score; update a model for predicting the state of the evolving route based on the loss score; and generate an updated terrain map based on the updated model. The controller is to generate the terrain map comprising positions and velocities of the plurality of vehicles. The controller is to:

12. The vehicle power management server of claim 10, wherein, determine a position of at least one of the plurality of vehicles based on sensor data for operation of a power source of the vehicle.

13. The vehicle power management server of claim 10, wherein, The plurality of vehicles comprises a plurality of vehicle types, the types comprising a transport truck and at least one additional vehicle type, and wherein the controller is to: distinguish the plurality of vehicle types, generate a same terrain map; and 14. The vehicle power management server of claim 10, wherein, provide the same terrain map for each vehicle type. The controller is coupled with one of a plurality of transducers to receive first of the sensor data, and the controller is coupled with a virtual sensor to receive second of the sensor data, the virtual sensor derived from a combination of two or more of the plurality of transducers. The controller is to:

15. The vehicle power management server of claim 10, wherein, generate the second sensor data based on a plurality of data elements of the sensor data.

16. The vehicle power management server of claim 10, wherein, 17. A method of power management, the method comprising: receiving sensor data indicative of operation of each of a plurality of vehicles locally at a plurality of telematics interfaces corresponding to the plurality of vehicles; transmitting the sensor data from each of the plurality of telematics interfaces to a remote server; generating, by the remote server, a terrain map based on the sensor data from the plurality of telematics interfaces corresponding to the plurality of vehicles; and ​ ​ ​ splitting, by one vehicle of the plurality of vehicles, power between a first power source and a second power source based on the terrain map.

18. The method of claim 17, the method comprising: - embedding, by the remote server, into the terrain map: -- first information indicative of a surface of a ground traveled on in an evolving route; and -- second information indicative of a location of the evolving route, wherein the evolving route comprises a time-varying component for at least one of: the surface of the ground traveled on or the location. splitting the power comprises:

19. The method of claim 17, wherein, predicting, based on the terrain map, a power demand at a future time; and generating, based on the power demand and prior to the future time, a control signal to adjust a power output of at least one of the first power source or the second power source. generating the control signal comprises:

20. The method of claim 19, wherein, generating a first control signal for an internal combustion engine based on a difference between the sensor data and a predicted change in the sensor data; and generating a second control signal for an energy storage device to receive or transmit power. ​

Images