Patent Application
20130046466
The present invention relates to a fuel optimization in vehicles. More specifically, the present invention relates to selecting an optimum route based on models and observations of vehicles, drivers, and routes.
Improving fuel efficiency in heavy-duty vehicles provides numerous benefits to the national and global communities. Heavy-duty vehicles consume a substantial amount of diesel fuel and gasoline, increasing dependence on fossil fuels. In the United States, medium and heavy-duty vehicles constitute the second largest contributor within the transportation sector to oil consumption. “EPA and NHTSA Adopt First-Ever Program to Reduce Greenhouse Gas Emissions and Improve Fuel Efficiency of Medium- and Heavy-Duty Vehicles”, Regulatory Announcement EPA-420-F-11-031, U.S. Environmental Protection Agency, August 2011 (hereinafter, “EPA Fact Sheet”). Currently, heavy-duty vehicles account for 17% of transportation oil use. “Annual Energy Outlook 2010”, U.S. Energy Information Admin., Report DOE/EIA-0382 (2010), April 2010. Demand for heavy-duty vehicles is expected to increase 37% between 2008 and 2035 (EPA Fact Sheet), making the need for more fuel-efficient vehicles even more apparent.
Heavy-duty vehicles also emit into the atmosphere carbon dioxide, particulates, and other by-products of burning fossil fuels. The EPA estimates that the transportation sector emitted 29% of all U.S. greenhouse gases in 2007 and has been the fastest growing source of U.S. greenhouse gas emissions since 1990. “Inventory of US Greenhouse Gas Emissions and Sinks: 1990-2009”, Report EPA 430-R-11-005, Apr. 15, 2011. By improving fuel efficiency in heavy-duty vehicles used in the U.S., the amount of greenhouse gases emitted could be drastically reduced. The benefits of improved fuel efficiency have prompted the Obama Administration to implement new regulations mandating stricter fuel efficiency standards for heavy-duty vehicles. In August 2011, the Environmental Protection Agency and the Department of Transportation's National Highway Traffic Safety Administration released the details of the Heavy Duty National Program, designed to reduce greenhouse gas emissions and improve fuel efficiency of heavy-duty trucks and buses. The Program will set forth requirements for fuel efficiency and emissions from heavy-duty vehicles between 2014 and 2018 in a first phase, and from 2018 and beyond in a second phase. The key initiatives targeted by this program are to reduce fuel consumption and thereby improve energy security, increase fuel savings, and reduce greenhouse gas emissions (EPA Fact Sheet). Creating sustainable processes for improving fuel efficiency of heavy-duty vehicles would allow vehicle owners to comply with the new emission standards, and would further the initiatives of the Heavy Duty National Program.
Poor fuel economy consumes resources that a vehicle operator might more profitably spend on opportunities that also benefit the economy as a whole. The EPA and Department of Transportation have estimated that the Heavy Duty National Program would result in savings of $35 billion in net benefits to truckers, or $41 billion total when societal benefits, such as reduced health care costs because of improved air quality, are taken into account. EPA Fact Sheet.
In the context of commercial vehicle fleets, a trip or mission often requires that a particular payload be moved from a point A to a point B at a particular time. The amount of fuel used for a mission will be affected by the particular choice of vehicle, by the geography (e.g., topography), by speed limits and other regulations, by traffic, and by the habits of the particular driver in operating the vehicle. Due to any or all of these factors, any mission can be expected to use more fuel than is optimal. Some of these factors, such as the choice of vehicle and how the driver operates the vehicle, can be manipulated, while others, such as regulations and traffic on a given route, cannot.
The inventor expects that the driver is often a major source of vehicle performance inefficiency. However, until now there has not been sufficient data to assess the magnitude of that inefficiency, an information gap that the data collection and analysis methodology of the invention will help to fill. Another goal of the invention is improving driver performance. By modeling vehicle dynamics and collecting and storing relevant data, factors subject to control of a driver or a fleet manager may be optimized.
Actual performance of a driver may be measured by one or more scoring functions. A scoring function may be based on indicia with regard to a “goodness” factor. For example, the fuel efficiency and the drivability of the vehicle are candidates for goodness factors that might each be rated by a respective scoring function. A given scoring function may be a composite of other scoring functions. Thus, an overall score might be a composite of a fuel efficiency score and a drivability score. A composite function may weight such scoring functions for individual goodness factors. The weighting may be constant, or might itself be a function of state of the vehicle. For example, acceleration (more specifically, positive acceleration) may be a factor in drivability, but the driver's need to accelerate is less at higher speeds. The overall scoring function might weight the vehicle's ability to accelerate more heavily, relative to fuel consumption, at slower speeds than at higher speeds.
The reserve or available acceleration is the acceleration that the vehicle would have at the current speed if the vehicle were given full throttle; in other words, the accelerator pedal is 100 percent depressed. Because reserve acceleration may be more important to drivability than actual acceleration, reserve acceleration may be preferable as a goodness factor in scoring. Whether reserve acceleration or actual acceleration is intended will be distinguished in particular contexts in this document.
A scoring function, for a goodness factor such as fuel efficiency, might involve a comparison of a measured value with, or ratio to, one or more reference values. A reference value for fuel efficiency might be, for example, (1) the best fuel efficiency ever measured for this particular vehicle; (2) the average fuel efficiency recorded by drivers in a fleet for this model of vehicle; (3) a government or manufacturer estimate of average fuel efficiency for this model of vehicle; (4) the best fuel efficiency achieved by any vehicle available from any manufacturer within this class of vehicles; or (5) a target fuel efficiency, possibly set by an expected future regulation or by a company's goals.
When operating a vehicle, driver manipulates certain vehicle “controls”, such as a gear stick to control transmission gear, an accelerator pedal (or throttle pedal) to control fuel usage, and brake pedal to slow the vehicle. We may sometimes use “accelerator” or “throttle” as short for accelerator/throttle pedal; “gear” as short for “transmission gear stick”; and “brake” as short for brake pedal. If the vehicle has a manual transmission, the driver also controls the clutch position in order to shift gears. Because braking is dictated primarily by regulations and traffic, a driver's choices with respect to braking are unlikely to be much improved upon. Nor is it practical to change a driver's habits regarding the use of clutch and gear shift stick in moving from one gear to the next.
Drivability and fuel economy are dependent on accelerator position and transmission gear, and with regard to those particular vehicle controls, the driver usually has some choices. Consider exemplary individual scoring functions for drivability and fuel economy, and an overall scoring function that is a weighted average of them. At any given time while a vehicle is being driven, and for any given choice of transmission gear, there is expected to be an accelerator position that optimizes the overall scoring function (as well as accelerator positions that optimize the individual scores for the component factors). Thus, taken together, the optimal (with respect to the overall scoring function) gear-accelerator pair choices form a curve to which the driver may aspire. Each gear-accelerator optimal pair is associated with an efficiency score, a drivability score, and an overall score. One of the gear positions will have a highest overall score.
Depending on the formulation of the overall scoring function, the various scores, and hence the curve, may either be static for a particular mission, or change over time. For example, if weightings of component scores change with vehicle speed, then the shape of the curve may change frequently or even constantly. Environmental factors may also cause the curve to evolve, such as road rolling resistance, aerodynamic drag due to wind changes, road grade, temperature, elevation, rain or snow, and ice.
Indicia of driver performance include current values of variables relating to fuel-efficiency. By “current” we mean averaged over a short period, e.g., over an interval of 10 seconds or some shorter period. By “instantaneous” or “near real time” we mean a time no more than 1 second. variables may include some or all of the following: current gear and accelerator control positions; the actual drivability fuel-efficiency, and overall scores that the vehicle is presently achieving under control the driver; the optimal gear-accelerator pairs and their scores; and the evolving aspirational curve. The indicia may also include indicia spanning longer times than “current”, such as values averaged or integrated since the start of the mission. These may include, for example, average fuel consumption rate, total fuel used, total miles driven, and average values of various goodness scores.
Such indicia of driver performance may be shown through a user interface (UI) on a monitor or display. The vehicle may be equipped with such a UI to influence the driver's operation of the vehicle. A chart may display the current grid-accelerator pair and a curve of optimal grid-accelerator pairs, and include respective representations of scores for these various pairs. A driver, or a group of drivers, might be recognized for meeting or exceeding threshold values of one or more of the indicia during a single mission, or averaged over a set of missions in an awards program sponsored by a fleet manager.
Such indicia of driver performance may be collected in tangible electronic storage (e.g., memory, flash drive, solid state disk, rotational media drive). Such storage may be located on the vehicle itself, at some remote location, or some combination thereof. Data about the vehicle design, the state of the vehicle and its components (including, for example, driver controls, fuel consumption, powertrain state, payload, and environmental conditions) may also be saved to such storage. Data may be collected from various sources including, for example: a controller-area network (CAN) on the vehicle; other sensors on the vehicle, such as a global position system (GPS) sensor; environmental sensors on the vehicle; external sources such as weather stations; and manufacturers' specifications for the vehicle or its components. Physical dynamics models may calculate unknown parameters from such data, and use the results as feedback to guide a driver.
A trip dynamics “executor” (TDE) may collect data from a vehicle and external sources, analyze that data, and initiate appropriate actions, for example, to provide diagnostics to a driver. The TDE may include a logger to collect relevant data, a kernel for to analyze information and control execution, and a monitor to provide diagnostics to a user. These elements may include or utilize sensors, logic executed by processing hardware, and communications systems. The logic may include hardware logic, software logic based on instructions accessed from storage and executed by hardware, or any combination thereof. Data collection may use a device that connects to a CAN connector, such as a J1939 connector, on a vehicle. Sensors may be located, and the logic may be executed, by hardware on the vehicle and/or at one or more remote location. When some or all of the hardware for the logic, or the storage or sources for the data, is remote, then the one or more communication systems may be used to communicate relevant information as required. By the term “communication system”, we mean any system capable of transmitting and/or receiving information electronically; for example, alone or in combination, whether wired or wireless: a local area network (LAN), a wide area network (WAN), a personal area network (PAN), a hardware bus, or a cable.
Indicia of driver performance collected by one or more individual vehicles may be received over a communication system at some remote facility for display or analysis. Indicia might be averaged over a set of vehicles, and/or over some interval of time. A manufacturer might use such data to evaluate its vehicles or the vehicles of a competitor. A fleet operator might use such data for accountability of its drivers, or to make decisions about current environmental conditions.
Reserve acceleration (and hence drivability) depends on vehicle physical dynamics processes, and, in particular, on the net force applied to the vehicle. The net force on the vehicle depends on the vehicle load, environmental conditions, and fuel usage. Fuel usage, in turn, depends on the driver's operation of the gear and accelerator controls. Current fuel usage can be monitored, although accuracy may require some function fitting or estimation based on observation of the current state of the internal components of the vehicle. Fuel drives the engine, which produces torque. The torque is transmitted, albeit with some loss to heat and vibration, through the powertrain (e.g., clutch or torque converter; transmission; and rear axle), to the wheels and tires. Force on the vehicle due to fuel usage depends on torque, generated from fuel consumption, on the tires.
The logic combines a trip dynamics model of vehicle components and such physical dynamics processes, real-time observations about the vehicle and the environment, and data known about the vehicle from the manufacturer or previous data collection and analysis. The model uses mathematical and physical equations, which may be approximated (e.g., discretized or otherwise simplified), to calculate or estimate indicia of driver performance. Any or all of the data used in these calculations, as well as the results of the calculations, may be saved to and/or retrieved from tangible storage.
An exemplary model will be presented in the Detailed Description of this document. Each item contained in the display is a variable in the model, and those variables are organized herein into a set of variable tables, each table containing a group of variables that are related to a vehicle system or to a component of the TDE (e.g., the display). There are also a set of equation tables, each table containing a set of equations similarly grouped. Each variable table also gives one or more sources for how a variable may be obtained. A source is either a basic source—a generally known quantity (e.g., gravitational acceleration), a measurement or observation (e.g., engine speed, road grade), a specification provided by a manufacturer, a statistic based on historical observation of vehicles, or a user preference—or an equation in the equation tables. When the source is an equation, the variable will be related functionally to other variables in the variable tables. Each of these other variables can therefore be sourced analogously. All variables in the display, and indeed all variables in the particular model provided herein as exemplary, can be traced by the above process back to a set of basic sources. The tables, therefore, provide a complete (in an exemplary embodiment of the invention) set of processes for obtaining any variable in the exemplary model and in any of the figures.
In addition to coaching a real driver in a real vehicle, other applications of the trip dynamics model, and observations collected by TDEs in one or more vehicles, are possible within the scope of the invention. For example, (1) a real driver might be taught how to improve fuel efficiency with a simulated vehicle that displays indicia of driver performance; (2) a fleet manager might evaluate a particular vehicle by simulating a set of typical missions for that fleet with the vehicle to see how it compares with other vehicles; or (3) a manufacturer of a vehicle, or of a vehicle component, might evaluate various candidate configurations of design to predict performance and choose a best design.
The modeling approach has much wider applicability than the trip dynamics display. Suppose, by way of illustration, that an equation specifies A as a function of B and C, and suppose that function is not known publicly. For example, a vehicle or component manufacturer might know the function, but might not be willing to reveal it for competitive or legal reasons. Using the vast amounts of data that can be collected by the TDEs from operation of real vehicles and from sources of environmental data, mathematical fitting of the equations of the model can be used to infer such relationships quite accurately.
The equations in the model may be used in different sequences for different purposes. If the source of variable B in the source tables is an equation that shows B to be a function of A, then A is also mathematically related to B, but A might not be a function of B. For a given value of B, there may be more than one value of A. In such a case, data collection can be used to eliminate the ambiguities, allowing such a relationship to find the correct value of A in particular situations.
As already mentioned, the models described herein can be used to evaluate and rank routes as well as drivers. A vast amount of data may be collected and stored by a trip dynamics logger from a single vehicle. Some data may be static, such as the type of vehicle itself, and the characteristics of its components. Other data may change dynamically as the vehicle moves, such as the gear selected by the driver (see, for example,
Data from multiple routes, drivers, vehicles, vehicle components, and weather and road conditions may be aggregated and analyzed. Typically, the data would be transmitted by the vehicles across one or more communication systems to a processing facility. The data might also include data from sources other than a monitored vehicle, such as environmental information from the National Weather Service or from nearby snow removal vehicles.
The processing facility might use the data to improve the physical dynamics model by adjusting various parameters. Information collected and analyzed by the facility might be of interest to vehicle and component manufacturers; to fleet managers; to drivers; and to research and development teams. Results of such data collection and analysis might be available for communication by wireless or wireless means to any electronic device with a user interface. Such a device (e.g., a computer system or a handheld electronic device) may have a processor, tangible storage, and a display.
The data for analysis might include, for example, particular vehicles and/or sets of vehicles; routes and/or sets of routes; drivers and/or sets of drivers; and environmental conditions. For a given driver-trip-vehicle-load combination, the data might include detailed time series of: key elements of driver behavior, the physical state of the vehicle and its important internal components (e.g., engine, clutch, transmission, rear axle, tires); and the route/environment (e.g., grade, rolling resistance coefficient, wind speed, traffic, regulatory restrictions).
Techniques familiar to practitioners of the statistical arts can use such data to make various kinds of forecasts and predictions. Such techniques include regression, discriminant analysis, time series analysis, spectral analysis, and atmospheric modeling. There is substantial literature on these topics, such as Hastie et al., “The Elements of Statistical Learning Data Mining, Inference, and Prediction, (Springer, 2nd ed. 2009), which is hereby incorporated by reference.
An exemplary application of such techniques is to predict when a driver will shift gears based on the state of the vehicle. For example, a characteristic shift schedule might be calculated for a given class of drivers, such as good drivers, average drivers, or drivers at some particular percentile rank in a distribution of all drivers. A shift schedule predicts whether a driver will shift gears, either to a lower or to a higher gear, under certain circumstances, such as a given percentage of full throttle and vehicle speed. A shift schedule is one example of a driver model or aspect of a driver model. A driver model predicts the state of the various controls available to the driver for ongoing vehicle operation, such as accelerator position, brake pedal position, clutch position, and transmission gear. A driver model can be used to simulate a real driver.
A route model can be constructed as a sequence of states, changing either at discrete locations or continuously over a particular route. The state of the route might include variables such as rolling resistance coefficient, friction coefficient, grade, minimum speed, maximum speed, elevation, temperature, and head wind speed. These variables may be changed at discrete locations, or in some cases may be interpolated to obtain values for intermediate locations.
A vehicle model can be constructed based on its components (e.g., engine, transmission, tires) and information known about the vehicle, either known from a source such as the manufacturer, arbitrarily specified, or measured from one or more actual vehicles in operation. The vehicle model might include variables such as in the tables of
Using techniques known to practitioners of the statistical arts, various goodness factors can be predicted to evaluate performance of a proposed configuration for a route. For a given trip or mission, a goodness score may be a function that depends on various factors, such as fuel economy, drivability, and time to complete a mission over a given route under certain environmental conditions. The goodness score may be calculated from simulations using virtual drivers, vehicles, and routes, based on statistics—and possibly formulas derived from such statistics—measured from real world missions. Some or all of the data used may have been collected by a trip dynamics logger 1361, where the driver 1350 was guided by a trip dynamics display 1100.
The virtual driver of a mission might be the “best” driver, for example, a driver that optimizes a score based on fuel economy and drivability for that mission. Or the virtual driver might be a “typical” or average driver. Scoring might take into account financial factors, such as the total cost of fuel, the value of completing the mission, the dependence of costs and benefits upon completion time, and depreciation on the vehicle.
Given a particular vehicle and a choice of driver, a fleet operator or driver wants to choose an optimum route for a given mission. To do this, we may run a set of simulations for a variety of route choices, using virtualizations of the vehicle and the driver. These virtualizations may be based upon models of the vehicle and of driver behavior. Candidate routes may be selected using techniques, such as those used by tools like Google Maps, Map Quest, or TeleNav GPS Navigator, or manually. Once a set of candidate routes have been selected, simulations can be run for a fixed vehicle/driver pair to select a best route. The best route optimizes a scoring function, which might be based on fuel economy, drivability, or route completion time, alone or in combination. The same route might be repeated for a variety of weather conditions, and an average, possibly weighted by expected weather condition frequencies, over the repetitions might be used to select a single best route. On the other hand, the route might be varied by a driver or fleet manager depending on weather conditions predicted over the actual duration of the trip.
This description provides embodiments intended as exemplary applications of the invention. The reader of ordinary skill in the art will realize that the invention has broader scope than the particular examples described here. Although many of the concepts and innovations apply to any motor vehicle, the primary area of applicability of teachings herein is heavy-duty vehicles, especially commercial trucks.
All of the variables tables have the same column headings, so only the column headings in the first variables table have been given reference numerals. The first column in each variables table is reference numeral (REF. 130). The second column is the symbol (SYM. 131) for the variable. The third column is a definition of the variable. The next four columns (columns 4-7) give a source or sources for the variable in the model. A variable may have one or more source, and not all possible sources are listed in the tables. A variable may be measured (MEAS. 133), obtained from an equation (EQN. 134), specified (SPEC. 135), or simply a quantity or function that may vary (VBL. 136), such as time or throttle pedal position. The MEAS. 133 column contains the following entries: CAN (a network on a vehicle); History (statistics from previously collected data); ECU (a controller in a vehicle); GPS (a locating device); internet sources (WWW); or Scale (to measure weight). The EQN. 134 column refers to an equation, by equation number in the equations table, from which the variable may be calculated. Sources in the SPEC. 135 column are means of specification. These include “User” for user-specified; “Mfr.” for a value specified by a vehicle or component manufacturer; “Mfr map” for a mapping, table, or function from a manufacturer; “Tire mfr. map” for such a map, specifically from a tire manufacturer; or “Const.” for a known constant. The VBL. 136 is checked with an “x” for variable quantities. The USED 137 column lists numbers for equations in which the particular variable appears.
These variables are related to each other in the exemplary system of model equations shown in the equations tables: driver performance scoring (
For every transmission gear number 601, there may be a best throttle position 120, which is “best” objectively because it maximizes (or minimizes) some user-selected score function 115. The resulting score is the best score 116 for that transmission gear. The pair of a transmission gear number 601 and the best throttle position 120 for that gear describe a point 1106 on the grid 1140. The set of all such best points 1106 lie on a curve 1103, and may be indicated by circles in the display. As illustrated, the diameter 1105 of each such circle is proportional to the score 113 for that point 1106. Similarly, the size of the symbol (in this case, a square) for the current gear-throttle pair 1102 is correspondingly proportional to its score 113. The pair of best gear number 121 and best throttle position 120 correspond to the point best grid-throttle pair 1104 on the curve 1103 having the highest overall best score for any gear 117 is emphasized, in this example by shading. Other means of emphasis might be used, such as color, crosshatching, or animation. For esthetic reasons, a dashed line is shown passing through the circled points on the curve 1103, although obviously transmission gear numbers have only integer values.
Note that there are many other ways that regions of relatively good or bad scores 113 on the grid might be displayed. One such method would be a color contour plot of the scoring function, which can be regarded as describing a surface above the grid 1140. The invention encompasses all approaches of representing scoring information to the driver 1350 for guidance.
The driver 1350 might improve the performance score 113 by adjusting the throttle position 119 and/or shifting to a different transmission gear number 601 to move to a point on the grid 1140 where the goodness 113 is higher. For example, by simply shifting from 3rd to 6th or 7th gear, performance will be improved. Ideally, the driver 1350 in the illustrated situation would be in 9th gear and have the throttle 83% depressed.
One might ask why the grid 1140 shows any points on the curve 1103 other than the best grid-throttle pair 1104. We note in response that ambient traffic and regulatory conditions might preclude the driver 1350 from operating the vehicle 1300 at the best point. Consequently, the driver 1350 needs more information than the best grid-throttle pair 1104 to optimize performance under such constraints. A more sophisticated scoring system in an embodiment of the invention might take such constraints imposed upon the driver 1350 into account in more fairly rating performance. A constraint might be known (e.g., a speed limit or a construction zone) or inferred (e.g., the vehicle 1300 is determined based upon observations by the trip dynamics logger 1361) to be moving slower than posted speeds on a highway segment known for stop-and-go rush hour traffic). Real time traffic data from external sources might also be taken into account. The scope of the invention includes any scoring system that utilizes a model of vehicle dynamics to estimate driver performance scoring parameters and, hence, includes such more sophisticated systems.
The performance statistics 1120 fall into two categories, trip diagnostics 1121 and current diagnostics 1122. The current diagnostics 1122 include current values of fuel economy score 104; drivability score 110; and overall score 113; and instantaneous fuel economy at steady state 303. The trip diagnostics 1121 include time-averaged (typically, over a trip or mission) values: time-averaged fuel economy score 105; time-averaged drivability score 111; and overall time-averaged score 123; and average fuel economy 304, as well as total distance traveled 203 and trip fuel 301. A fleet manager might provide a driver with an incentive or reward for achieving a score (whether fuel, drivability, or overall) in some specified range.
A purpose of the chart 1101 and diagnostics 1120 in some embodiments of the invention is to improve performance by the driver 1350 of a vehicle 1300. As shown in
As shown by
As mentioned previously, a driver 1350 might be a simulated or virtual driver rather than a human. Collection of data by a TDE over time will allow drivers 1350 of various types (e.g., having a specified number of years of experience; employed by a particular fleet manager; or assigned certain metropolitan areas) to be simulated with statistical accuracy. A typical statistical distribution of such driver 1350 types might be used to evaluate how a vehicle 1300 or a fleet might perform over a suite of varying conditions (e.g., load, distance, environment). When optimizing a score function or other reference function, we are in effect operating the vehicle 1300 with a virtual driver 1350, using our models to determine which combination of choices or actions by such a virtual driver 1350 are the optimum set of choices. A virtual vehicle 1300 might be used to compare various choices of vehicles to determine which vehicle, or suite of vehicles, is optimal for a particular task or suite of tasks.
Physiological 1391 inputs from the driver 1350 is transferred to the engine control unit (ECU, also known as the power-train control module) 1321 over the CAN 1380, as indicated by arrow 1383, to set the fuel mass flow rate 302 to the engine 1322. Information about the state of systems in the vehicle 1300, such as engine angular speed 306 and engine brake torque 313, are transferred to the ECU 1321, and may be accessed by the TDE 1360 over the CAN 1380, as indicated by arrow 1381.
Resulting engine brake torque 313 is transferred to the engine-to-transmission coupling 1323 (a clutch for a manual transmission 1331 or a torque converter for an automatic). The output torque from the coupling 1323 is transferred to the driveline 1330 (including the transmission 1331, the drive shafts 1332, and the rear axle 1333) as transmission input torque 609. Output torque from the driveline 1330 is transferred to the rear wheels and the rear tires 1340 as rear axle output torque 707.
Information about the environment 1351 in which the vehicle 1300 is operating is transferred over the CAN 1380 to the vehicle 1300, as indicated by arrow 1382. Such environmental data may be available to the TDE 1360 over the CAN 1380 as well.
Environmental conditions 1371 and the payload 1341 exert a load torque 1342 on the rear tires 1340. The combined torque on the rear tires 1340 results in a tractive force 802 on the vehicle 1300, causing it to accelerate. The reserve acceleration is calculated by assuming the application of full throttle starting from a vehicle 1300 moving at steady state in the current transmission gear number 103.
Like the driver 1350, a vehicle 1300 may be real or simulated. Simulated vehicles are useful at least for vehicle, system, and component design; driver training; fleet cost estimation; and mission route selection. Likewise, the evolution of an environment 1351 can be simulated, based on statistics or a dynamic model of the atmosphere, and geographic information systems when convenient for some purpose at hand.
The trip dynamics kernel 1362 may analyze data, communicate information, and cause actions to be taken. The trip dynamics kernel 1362 may compute the variables such as those in the tables of
Note in
Hardware components of a TDE 1360 may be located in the vehicle 1300, or they may be remote from the vehicle 1300. The hardware, logic, and functionality may each be split between local and remote. Local hardware may communicate with remote hardware over a communication system of any type capable of electronically transmitting and/or receiving information. Logic may be embodied in hardware, or in software instructions accessible from hardware devices including tangible storage or communication hardware.
The trip dynamics kernel 1362 uses a model of the vehicle 1300, such as shown in
A trip dynamics kernel 1362 that has available a physical dynamics model as illustrated by
Once the required data is obtained from the base sources, the relevant equations, which have already been identified in traversing the tree from root to leaf nodes, can be applied to obtain the target variable. In effect, the above discussion demonstrates that all the processes listed in
The above method for obtaining a process whereby any target variable in the variables tables can be sourced or calculated is summarized by
The method of
In
If a variable is not a base source variable, it may be computed from an equation. Equation numbers that correspond to
Accordingly, score 113 (in this particular embodiment) is found in equation (5) to depend directly on four variables, namely, fuel economy score 104, drivability score 110, fuel economy weight factor 106, and drivability weight factor 112. As taught by
In fact, recursion through this particular tree may involve nearly all variables and equations in the model. Triangle 1810 indicates that the process compute fuel economy 1608 to compute instantaneous fuel economy at steady state 303 uses equation (23), the tree expansion of which is omitted from
A few closing remarks about
The types of models described above can be applied to many useful purposes in addition to guiding the operation of a vehicle 1300 by a driver 1350. Also, data collected by a trip dynamics logger 1361, whether or not in a context of driver guidance, can be accumulated and analyzed for such other purposes.
In particular, information collected about drivers can be collected and analyzed to build a driver model 2902. The driver model 2902 may predict what a driver 1350 will do under a given set of circumstances. Various statistical methods can be used to make such predictions based on observations such as those collected by the trip dynamics logger 1361 regarding state of vehicles 1300 and their components, route, and environment. These data are often in the form of time series. Examples of such prediction methods include regression and time series analysis. Such a driver model 2902 might be used to predict how a driver 1350 will drive a particular vehicle 1300 over a particular route under particular environment 1351 conditions. The driver 1350 may, for example, engage the clutch, depress the throttle, shift gears, or apply the brake. Using a model of the vehicle 1300, or a virtual vehicle 1300, and a particular virtual route, one or more indicia of goodness, of the types already described, may be calculated.
A vehicle 1300 may include a set of components such as those shown in
A route model 2903 may include elements that change in space are static in time over a particular mission, such as grade, minimum and maximum speed (dictated both by law and by safety), rolling resistance coefficient, friction coefficient, and elevation. Other elements, such as the influence of weather 2904 (e.g., wind speed, air temperature, and road icing), may be treated as static or time dependent. A route model 2903 may range in complexity, depending on how realistic it is required to be for some purpose. Clearly, there are significant differences between the range and frequency of environmental conditions typically encountered in different locales. Compare, for example, Canada and the southern United States in winter with respect to wind, precipitation, and road conditions.
In general, if we fix any two of the factors/models of
For a pair of a given driver model 2902 and a given vehicle model 2901 and a given set of weather conditions—a driver/vehicle/weather (DWT) triple—during a mission, we can find a route specification that optimizes a goodness score. In other words, we can compare route models to see which achieves the best score for that DWT triple. The weather conditions (e.g., wind, precipitation, and road conditions) might be static, or might vary over space and time during the trip. A comparison might be done for a single driver or for a set of drivers, such as a suite of drivers typical for a fleet of a given type. The route might be selected for a single vehicle, or for a set of vehicles, such as a suite of vehicles typical for a fleet. The route might be simple, requiring the driver to get from point A to point B. Or the route might consist of a sequence of segments, such as A to B, then B to C, with the roads to be taken for one or more segments being chosen by the process. An overall average (or weighted average) of best scores over a suite of DWT driver/vehicle pairs might be computed.
A process for find an optimum configuration, relative to some scoring criterion, is illustrated by
Selection of routes, or route models, for comparison might be done through a user interface, managed by a processor, on an electronic device, managed by a processor. Selection of vehicles or vehicle components, driver models, and weather conditions might also be done through such an interface.
Note that in the above flowcharts, the order may be varied, some steps might be eliminated, or some additional ones may be added. Some more obvious steps are not shown for clarity.
Throughout this document and claims, the word “or” is used in the inclusive sense unless otherwise specified. Of course, many variations of the above method are possible within the scope of the invention. The present invention is, therefore, not limited to all the above details, as modifications and variations may be made without departing from the intent or scope of the invention. Consequently, the invention should be limited only by the following claims and equivalent constructions.
This application is a continuation-in-part of U.S. utility application Ser. No. 13/251,711 filed Oct. 3, 2011, and entitled “Fuel Optimization Display”, which is incorporated in its entirety by this reference. This application claims the benefit of U.S. Provisional Application No. 61/524,832, filed Aug. 18, 2011, and entitled “Fuel Optimization Display”, which is incorporated in its entirety by this reference. This application is related to U.S. utility application No. ______ filed Oct. 31, 2011, and entitled “Selecting a Vehicle to Optimize Fuel Efficiency for a Given Route and a Given Driver”, which is incorporated in its entirety by this reference.
| Number | Date | Country | |
|---|---|---|---|
| 61524832 | Aug 2011 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 13251711 | Oct 2011 | US |
| Child | 13285340 | US |
