Method and system for controlling the movement of an electric vehicle (EV)

Abstract

An embodiment of the present disclosure discloses a method and system for controlling motion of an electric vehicle (EV). The method includes determining a speed profile for moving the EV from an initial speed over a period of time by minimizing energy dissipation according to an energy loss function. The energy loss function maps acceleration and speed values ​​of the EV to energy dissipation of the EV, and the energy dissipation occurs by controlling one or more electric motors of the EV to move the EV at the corresponding acceleration and speed values. The speed profile is a function of time. The method further includes controlling one or more electric motors of the EV to generate torque to move the EV according to the speed profile.

Description

Technical Field

[0001] The present disclosure generally relates to optimizing the energy efficiency of electric vehicles, and more specifically, to methods and systems for controlling the movement of electric vehicles (EVs).

Background Art

[0002] An EV is a vehicle powered by electrical energy, which is a renewable energy source. Improving energy efficiency increases the driving range of the EV, makes the EV more cost-effective to operate, more environmentally friendly, and more attractive to a wider range of users. An EV may use a battery pack and one or more electric motors for propulsion instead of an internal combustion engine. The battery pack stores electrical energy that powers one or more motors to drive the EV. Recently, EVs have gained popularity because they have the potential to reduce the operating cost of automobiles and the pollutant emissions of automobiles. In some cases, an EV may correspond to a hybrid electric vehicle (HEV) that uses an internal combustion engine in combination with one or more electric motors. The dynamics of the internal combustion engine and the dynamics of the electric motor may be different. For example, an electric motor can generate high torque, while an internal combustion engine can reach that high torque after increasing its speed. Since the internal combustion engine can be slower than the electric motor, it may limit the control operation of the EV.

[0003] An EV may use a transmission such as a continuously variable transmission (CVT) that converts electrical power supplied by one or more electric motors into momentum for driving the wheels of the EV. A CVT is a type of transmission that seamlessly changes gears regardless of the speed of the EV. However, such a transmission in an EV may affect the efficiency of the electric motor. The efficiency of an electric motor corresponds to the difference between the mechanical output of the electric motor and the electrical energy consumed by the electric motor. Usually, energy is lost during the conversion from electrical energy to mechanical energy, so the mechanical output is lower than the electrical energy consumed. This energy loss increases when a transmission such as a CVT is used for gear changes, which can thereby reduce the efficiency of the electric motor. Furthermore, the efficiency of the electric motor may be affected by differences in the type of electric motor and changes in torque requirements for moving the EV. For example, one electric motor may correspond to a powerful electric motor, and the other electric motor may correspond to a weak electric motor, and their efficiencies may vary based on torque requirements. A weak electric motor may require low torque, and a powerful electric motor may require high torque.

[0004] Therefore, it is necessary to overcome the above-mentioned problems. More specifically, it is necessary to develop a method and a system for controlling an electric vehicle in an efficient and feasible manner. SUMMARY OF THE INVENTION

[0005] An EV such as a battery-powered EV may use one or more electric motors to operate the EV. The electric motor may have an efficiency that is a function of the torque and rotational speed of the electric motor. When the electric motor operates to achieve the desired torque and rotational speed for moving the EV, there may be energy losses that affect the efficiency of the electric motor. The impact on electric motor efficiency may further affect the high-efficiency control of the EV's energy efficiency.

[0006] Therefore, an object of some embodiments is to use an energy loss function to minimize the energy loss of an EV. The energy loss function is a cost function that can be derived from tabular data of the efficiency map of an electric motor. The efficiency map may include a contour plot that determines the maximum efficiency of the electric motor for any combination of the torque and rotational speed of the electric motor. The cost function may include a quasiconvex loss function that can be optimized for high-efficiency control of the EV's energy efficiency.

[0007] In particular, the quasiconvex loss function may include the energy loss of the EV that can correspond to the speed and acceleration of the EV. The energy loss includes the amount of dissipated energy that may not be recovered. For example, the energy loss due to the efficiency of one or more electric motors is dissipated and cannot be recovered. The energy loss of the EV may also include the drive loss and the electric motor loss of the EV. The drive loss may correspond to the energy loss caused by rolling resistance or aerodynamic drag while driving the EV. The electric motor loss corresponds to the energy loss caused by the operation of one or more electric motors due to the efficiency of the electric motor. In some cases, the energy used to accelerate the EV may be recovered when the EV decelerates by operating one or more electric motors in the generator mode. The generator mode corresponds to regenerating energy by the deceleration of the EV.

[0008] Also, an object of some embodiments is to optimize the efficiency by using the degrees of freedom of the EV. In particular, the efficiency of one or more electric motors may be optimized by using the degrees of freedom. The degrees of freedom herein correspond to parameters of the EV, such as the torque split ratio, the gear ratio, and / or the speed profile of the EV. The torque split ratio corresponds to the allocation of the corresponding torque among one or more electric motors. For example, an EV having four in-wheel electric motors may allocate or divide the total torque requirement among the four electric motors of the EV. The gear ratio corresponds to the gear ratio for controlling the transmission of the EV. The speed profile corresponds to the braking operation and the acceleration operation tracked by the EV.

[0009] Some embodiments are based on the recognition that the speed profile tracked by an EV may affect energy losses. Therefore, some embodiments use an energy loss function to calculate a speed profile for minimizing the energy losses of the EV. The speed profile may include the initial speed (i.e., the current speed) of the EV and the target speed of the EV. The current speed may be determined using one or more sensors of the EV. The target speed may be determined based on the motion plan of the EV and the period assigned to the speed profile. In some embodiments, the speed profile is determined based on at least one or more of a constraint on the average speed of the speed profile, the current position and current speed of a leading vehicle traveling ahead of the EV, and the current position, the current speed, and the predicted speed over a certain period of the leading vehicle traveling ahead of the EV. The speed profile and the initial speed may be recalculated for each sampling time step for realizing high energy efficiency control of the EV.

[0010] In some embodiments, the gear ratio may be determined using a predetermined motor efficiency function. The predetermined motor efficiency function may be learned using kernel-based regression that fits the data of an efficiency map of one electric motor that associates the efficiency of the electric motor with the motor speed and motor torque. Also, the kernel-based regression may adapt the data of one or more efficiency maps of one or more electric motors to an energy loss function based on a model of the drive system losses of one or more electric motors and a model of the drive losses caused by the aerodynamics and rolling resistance of the EV. The model of the drive system losses includes losses corresponding to the energy usage and the efficiency of energy regeneration due to the acceleration and deceleration of the EV. More specifically, the kernel-based regression learns a quasiconvex loss function having the minimum value, i.e., the smallest value overall, in the data of one or more efficiency maps. In some embodiments, the minimum value may be determined using the gradient descent method. The gradient descent method is an iterative optimization approach.

[0011] In addition to or instead of this, some embodiments may use a predetermined torque splitting function to allocate the total torque among one or more motors. In some embodiments, the predetermined torque splitting function may be learned using a predetermined motor efficiency function of one or more electric motors.

[0012] In some embodiments, a torque profile of one or more electric motors may be determined to move the EV according to a speed profile. The torque profile may be determined using a model corresponding to the longitudinal motion of the EV. The model of the longitudinal motion associates the acceleration and speed of the EV with the total torque of the EV. The total torque may be generated according to the torque profile by controlling one or more electric motors. In some cases, the torque profile may be split for each of one or more electric motors. For example, the torque profile may be split into a first torque profile of a first electric motor among one or more electric motors and a second torque profile of a second electric motor among one or more electric motors.

[0013] In addition to or instead of this, the energy loss function may be derived at least from a predetermined motor efficiency function, a predetermined torque splitting function, and a model of the longitudinal motion of the EV.

[0014] In addition to or instead of this, the energy loss function may be optimized using gradient descent. Gradient descent is an iterative optimization approach for determining the minimum value, i.e., the quasi-optimal value, of the energy loss function. In some embodiments, the energy efficiency of the EV may be controlled based on a sequence of control inputs of the EV and an automatic feedback control mechanism. The sequence of control inputs may correspond to torque and gear ratio commands for high-efficiency control of the EV's energy. The automatic feedback control may include using a feedback signal that includes a sequence of measurements indicative of the state corresponding to the sequence of control inputs. In some exemplary embodiments, a feedback controller may be used to determine, at each control step, the current control input of a sequence of control inputs for controlling the energy efficiency of the EV based on the feedback signal. The feedback signal may include the current measurement of the current state of the corresponding sequence of control inputs. The feedback controller may apply a control policy for converting the current measurement into the current control input based on the current value of a control parameter within a set of control parameters of the feedback controller. Further, the state of the feedback controller defined by the control parameter may be iteratively updated by a Kalman filter that uses a prediction model for predicting the value of the control parameter that is subject to process noise. The predicted value may be updated based on a measurement model based on the sequence of measurements. The updated predicted value is used to determine the current value of the control parameter that explains the sequence of measurements according to the performance goal.

[0015] In addition to or instead of this, in some embodiments, a Kalman filter is used to update the speed profile using an energy loss function. Specifically, in response to receiving a feedback signal indicative of the current state of a vehicle tracking the speed profile, some embodiments execute a Kalman filter to update the speed profile of the current state of the vehicle and improve the likelihood of energy efficiency according to a probabilistic measurement model including an energy loss function. This is advantageous because the energy loss function is deterministic while vehicle control is inherently probabilistic. For example, by using a probabilistic measurement model including a deterministic energy loss function modified by probabilistic noise, the nature of the control can be reflected, the efficiency of determining or updating the speed profile can be improved, and the incompleteness of the estimation of the energy loss function can be taken into account.

[0016] Accordingly, one embodiment discloses a computer-implemented method for controlling the movement of an electric vehicle. The method uses a processor coupled to a memory storing an energy loss function that maps values of acceleration and speed of the EV to energy dissipation of the EV, the energy dissipation being caused by controlling one or more electric motors and / or one or more gearboxes of the EV to move the EV with corresponding acceleration and speed values. The processor is configured to execute instructions stored in the memory for implementing the method. The instructions, when executed by the processor, perform the steps of the method, the method including determining a speed profile that moves the EV from an initial speed over a period of time by minimizing the energy dissipation according to the energy loss function, the speed profile being a function of time, and the method further including controlling the one or more electric motors and / or the one or more gearboxes of the EV to generate torque for moving the EV according to the speed profile.

[0017] Accordingly, another embodiment discloses a system for controlling the movement of an electric vehicle. The system includes a processor coupled to a memory storing an energy loss function that maps values of the EV's acceleration and speed to the EV's energy dissipation, where the energy dissipation results from controlling one or more electric motors of the EV to move the EV with corresponding acceleration and speed values. The processor is configured to determine a speed profile for moving the EV from an initial speed over a period of time by minimizing the energy dissipation according to the energy loss function, where the speed profile is a function of time, and further configured to control the one or more electric motors of the EV to generate torque for moving the EV according to the speed profile. BRIEF DESCRIPTION OF THE DRAWINGS

[0018]

Figure 1A

Figure 1B

Figure 2

Figure 3

Figure 4

Figure 5A

Figure 5B

Figure 5C

Figure 5D

Figure 5E

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Embodiments for Carrying Out the Invention

[0019] In the following description, for the sake of explanation, numerous specific details are set forth in order to provide a thorough understanding of the present disclosure. However, it will be apparent to one of ordinary skill in the art that the present disclosure may be practiced without these specific details. In other instances, devices and methods are shown in block diagram form only to avoid obscuring the present disclosure.

[0020] As used in this specification and the claims, the terms “for example,” “as an example,” and “such as,” and each of the verbs “comprising,” “having,” “including,” and other verb forms thereof, when used in conjunction with a listing of one or more components or other items, are to be construed as open-ended, meaning that the listing is not to be considered as excluding further components or items. The term “based on” means at least partially based on. Further, it should be understood that the style and terminology used in this specification are for the purpose of description and should not be considered limiting. Any headings used in this specification are for convenience only and have no legal or limiting effect.

[0021] FIG. 1A shows an environment 100A of a system 100 for controlling the movement of an electric vehicle (EV) 102 according to an embodiment of the present disclosure. The EV 102 can correspond to a battery-driven EV such as a four-wheel electric vehicle 102A (e.g., an electric car) or a two-wheel electric vehicle 102B (e.g., an electric bicycle). The EV 102 can be operated by one or more electric motors. Each of the one or more electric motors can exhibit a corresponding efficiency that varies as a function of the torque and rotational speed of the corresponding one or more electric motors. Therefore, by using the system 100 for high-energy-efficiency control of the EV 102, the efficiency of the corresponding one or more electric motors can be optimized.

[0022] In some exemplary embodiments, EV102 may be connected to system 100 via network 104. Network 104 may include a wired network, a wireless network, and the like. In some other exemplary embodiments, system 100 may be integrated into a controller or control unit (not shown) of EV102. System 100 will be further described with reference to FIG. 1B below.

[0023] FIG. 1B shows a block diagram 100B of a system 100 for controlling the movement of EV102 according to some embodiments of the present disclosure.

[0024] System 100 includes a processor 106 and a memory 108 that stores an energy loss function 110. The energy loss function 110 maps the acceleration and velocity values of EV102 to the energy dissipation of EV102. The energy dissipation occurs by controlling one or more electric motors of EV102 to move EV102 with the corresponding acceleration and velocity values. The processor 106 is configured to cause the system 100 to determine a velocity profile for moving EV102 from an initial velocity over a period of time by executing instructions stored in the memory 108. The velocity profile may be a function of time determined by minimizing the energy dissipation according to the energy loss function 110.

[0025] In some exemplary embodiments, the initial speed is determined as the current speed of EV102. The current speed may be estimated using one or more sensors of EV102. The one or more sensors may include an accelerometer, a gyroscope, a global positioning system (GPS), a speed sensor, a wheel speed sensor, and the like. Some embodiments are based on the recognition that recalculation using sensor data can provide energy-efficient motion control to the EV. Therefore, the initial speed and the speed profile may be recalculated for each sampling time. The sampling time may correspond to the duration for which the initial speed and the speed profile are sampled for energy-efficient control of EV102. Such recalculated initial speed and speed profile can improve the accuracy for controlling the motion of EV102 with efficient energy consumption.

[0026] The processor 106 is further configured to control one or more electric motors of EV102 to generate torque for moving EV102 according to the speed profile. EV102 can be moved from the initial speed to the target speed of the speed profile. The target speed may be determined based on the motion plan of EV102 and the period assigned to the speed profile.

[0027] The procedure for controlling the motion of EV102 using the system 100 will be further described with reference to FIG. 2.

[0028] FIG. 2 shows a schematic diagram depicting a procedure 200 for controlling the motion of EV102 according to some embodiments of the present disclosure.

[0029] In some exemplary embodiments, system 100 may obtain motor efficiency data 206 of one or more electric motors of EV102. The motor efficiency data 206 may be obtained from dynamometer tests 202 of one or more electric motors of EV102. The dynamometer tests 202 may correspond to a series of experiments that test one or more electric motors at a specific constant speed and a specific constant torque, a high acceleration aggressive driving schedule (US06) test, an urban dynamometer driving schedule (UDDS) test, and the like. The motor efficiency data 206 may be used to learn a quasiconvex loss function 210. The quasiconvex loss function 210 is a cost function that may be obtained from drive loss data 208. The drive loss data 208 may be obtained from a vehicle model 204 corresponding to the drive losses of EV102, or may be obtained from testing EV102 on a test bed. The drive losses include energy losses due to aerodynamic drag and rolling resistance of EV102.

[0030] In some embodiments, the quasiconvex loss function 210 includes both a drive mode that uses energy to accelerate EV102 and a drive mode that regenerates energy while decelerating EV102. The quasiconvex loss function 210 includes a quasiconvex shape that enables differentiability of the energy losses in the motor efficiency data 206, whereby the system 100 becomes executable for real-time energy optimal control 212 of EV102.

[0031] Therefore, the system 100 may use a speed profile to minimize the energy losses of EV102 and enable energy optimal control 212. The minimization of energy losses based on the speed profile will be further described with reference to FIG. 3.

[0032] Figure 3 shows a schematic diagram depicting a procedure 300 for controlling the movement of EV102 according to some other embodiments of the present disclosure. Procedure 300 is executed by system 100 to minimize the efficiency of one or more electric motors using the degrees of freedom of EV102. The degrees of freedom include a speed profile 302, a total torque profile 306, and a gear ratio 312. The speed profile 302 is determined using the energy loss function 110. The speed profile 302 is a function of time corresponding to braking operations and acceleration operations tracked by EV102. The acceleration is the time derivative of the speed of EV102 and represents the value of the acceleration of EV102.

[0033] The total torque profile 306 is determined using a model 304 of the vertical movement of EV102 (hereinafter referred to as the "vertical movement model 304"). The vertical movement model 304 associates the acceleration and speed of EV102 with the total torque of EV102.

[0034]

Number

[0035]

Number

[0036]

Number

[0037] In some exemplary embodiments, the total torque profile 306 is assigned to one or more electric motors to move the EV 102 according to the speed profile 302. The total torque profile 306 is allocated among one or more electric motors based on a predefined torque splitting function 308. The total torque profile 306 is split into motor torques 1, motor torque N, etc. for the "N" multiple electric motors of the EV 102. For example, the EV 102 may use a first electric motor 316A and a second electric motor 316B. The total torque profile 306 may be assigned to a first torque profile of the first electric motor 316A and a second torque profile of the second electric motor 316B.

[0038]

Number

[0039]

Number

[0040]

Number

[0041]

Number

[0042]

Number

[0043] Such a total torque profile 306 is shown as a graph plot in FIG. 5A.

[0044]

Number

[0045] In some exemplary embodiments, system 100 may control the vertical movement of EV 102 by using feedback signal 318. Feedback signal 318 may include information corresponding to EV 102 and a vehicle traveling in front of EV 102. This will be further described with reference to FIG. 4 in conjunction with FIG. 3.

[0046] FIG. 4 shows an exemplary scenario 400 depicting EV 102 on road 404 and a vehicle 402 traveling in front of EV 102, according to some embodiments of the present disclosure. EV 102 and vehicle 402 are traveling on road 404. Vehicle 402 may include a fully electric vehicle, a fuel-powered vehicle, etc. The vehicle may be autonomous, semi-autonomous, or manually operated.

[0047] In some embodiments, speed profile 302 may be determined based on one or more of a constraint on the average speed of speed profile 302, the current position of vehicle 402, and / or the current speed of vehicle 402. In some other embodiments, speed profile 302 may be determined based on the current position, current speed, and predicted speed of vehicle 402 over a period of time. Thus, the efficiency of one or more electric motors may be optimized using automatic feedback control. The automatic feedback control corresponds to feedback signal 318 from EV 102, which may include the current speed of EV 102, the current position of EV 102, the current speed of vehicle 402, the position of vehicle 402 during travel, etc.

[0048]

Number

[0049]

Number

[0050]

Number

[0051]

Number

[0052]

Number

[0053] The speed profile 302, the total torque profile 306, and the gear ratio 312 are shown in FIGS. 5A, 5B, 5C, and 5D.

[0054] FIG. 5A shows a graph display 500A depicting a graph plot 502 corresponding to a speed profile (e.g., speed profile 302) of the EV102 and a graph plot 504 corresponding to a total torque profile (e.g., total torque profile 306) of the EV102, according to some embodiments of the present disclosure.

[0055] As described with reference to FIG. 3, the total torque profile 306 may be determined from the speed profile 302 using the longitudinal motion model 304. Such a total torque profile 306 is depicted in the graph plot 504.

[0056] When the EV102 is traveling on the road 404, the torque requirements of the plurality of electric motors of the EV102 may vary. For example, when the EV102 is traveling and trying to get ahead of the vehicle 402, the torque requirement may be high. When the EV102 is decelerating while trying to turn, the torque requirement may be low. In some cases, the torque requirement may peak when the EV102 is traveling uphill. To control the plurality of electric motors of the EV102, the total torque profile 306 may be allocated among the plurality of electric motors using a torque splitting function. The plurality of electric motors may correspond to two electric motors such as a first electric motor 316A and a second electric motor 316B.

[0057]

Number

[0058] The assignment of the total torque profile 306 among a plurality of electric motors is shown in FIGS. 5B and 5C.

[0059] FIG. 5B shows an exemplary graphical display 500B depicting the splitting of the total torque profile 305 according to some embodiments of the present disclosure. The graphical display 500B includes a graph plot 506 and a graph plot 508. The graph plot 506 corresponds to the speed trajectory of the EV 102 when the total torque profile 306 is split between two electric motors of the EV 102, such as the first electric motor 316A and the second electric motor 316B. The graph plot 508 corresponds to splitting the torque profile 306 into a first torque profile 508A (τ1) of the first electric motor 316A and a second torque profile 508B (τ2) of the second electric motor 316B. For example, as shown in FIG. 5A, the first torque profile 508A includes a maximum power of 150 kW, and the second torque profile 508B includes a maximum power of 50 kW.

[0060] The first electric motor 316A is controlled according to the first torque profile 508A, and the second electric motor 316B is controlled according to the second torque profile 508B. Each of the first electric motor 316A and the second electric motor 316B may be used for different torque requirements. For example, the first electric motor 316A may be used for high torque requirements when the EV 102 is traveling forward in front of the vehicle 402. The second electric motor 316B may be used for low torque requirements when the EV 102 is decelerating while trying to turn on the road 404. In some cases, both the first electric motor 316A and the second electric motor 316B may be used for peak torque requirements when the EV 102 is traveling uphill.

[0061] In some cases, the plurality of electric motors may correspond to the four electric motors of the EV102. The total torque profile 306 is allocated among the four electric motors of the EV102. This is shown in FIG. 5C.

[0062] FIG. 5C shows an exemplary graphical display 500C depicting the splitting of the total torque profile 306 according to some other embodiments of the present disclosure. The graphical display 500C includes a graph plot 510 corresponding to the total torque profile 306. The total torque profile 306 is allocated among the four electric motors of the EV102 using a predefined torque splitting function 308. As shown in FIG. 5C, the total torque profile 306 is split into a first torque profile of the first electric motor depicted in graph plot 512, a second torque profile of the second torque electric motor depicted in graph plot 514, a third torque profile of the third electric motor depicted in graph plot 516, and a fourth torque profile of the fourth electric motor depicted in graph plot 518.

[0063] Furthermore, the torque profile and the speed profile are used to calculate the gear ratio of the transmission of one or more electric motors. The calculated gear ratio is represented in a graph plot. This is shown in FIG. 5D.

[0064] FIG. 5D shows an exemplary graphical display 500D depicting the gear ratio (e.g., gear ratio 312) of the transmission of one or more electric motors of the EV102 according to some embodiments of the present disclosure. The graphical display 500D includes a graph plot 520 corresponding to the gear ratio 312 calculated from the first torque profile depicted in graph plot 512 and the speed profile 302 depicted in graph plot 502 using a predefined motor efficiency function 310.

[0065] In an exemplary scenario, the four electric motors of EV102 may include a first electric motor of EV102 that can be a powerful electric motor and a second electric motor of EV102 that can be a weak electric motor. The difference in the output of the electric motors can affect energy consumption. Accelerating the first electric motor and the second electric motor can result in higher energy consumption. In such a scenario, the acceleration of the electric motors is adjusted based on the speed profile 302. By adjusting the acceleration, energy is saved. This is shown in the graph display of FIG. 5E.

[0066] FIG. 5E shows an exemplary graph display 500E depicting the speed profile 302 of EV102 according to some embodiments of the present disclosure. The graph display 500E includes a graph plot 522 and a graph plot 524 that describe an energy-optimal speed profile for minimizing the energy consumption by one or more electric motors of EV102. The graph plot 522 includes a curve 522A corresponding to the speed profile of EV102 without a transmission for one or more electric motors of EV102, a curve 522B corresponding to the speed profile when EV102 uses one electric motor such as the first electric motor 316A of EV102 with a transmission such as a continuously variable transmission (CVT), a curve 522C corresponding to the speed profile when EV102 uses the second electric motor 316B with a CVT transmission, and a curve 522D corresponding to the speed profile when EV102 uses both the first electric motor 316A and the second electric motor 316B of EV102 with a CVT transmission to change the gears of EV102. Similarly, the graph plot 524 includes a curve 524A, a curve 524B, a curve 524C, and a curve 524D that describe different speed profiles of EV102 having the different configurations described above. As shown in FIG. 5E, the energy-optimal speed profile depends on the configuration of EV102, such as the use of electric motors with transmissions, the use of multiple electric motors in EV102 with or without transmissions.

[0067] In one exemplary scenario, the first electric motor 316A may correspond to the powerful electric motor of the EV102, and the second electric motor 316B may correspond to the weak electric motor of the EV102. In some cases, the EV102 may not use a CVT to change the gears of the EV102. In such a case, as shown in graph plot 522 (e.g., curve 522A on the time axis from 0 to 8 s) and graph plot 524 (e.g., curve 524A on the time axis from 0 to 9 s), first the first electric motor 316A may be accelerated. Next, as shown in graph plot 522 (e.g., curve 522A on the time axis from 8 to 26 s) and graph plot 524 (e.g., curve 524A on the time axis from 9 to 22 s), the acceleration may switch to the second electric motor 316B. Further, as shown in graph plot 522 (e.g., curve 522A on the time axis from 26 to 30 s) and graph plot 524 (e.g., curve 524A on the time axis from 22 to 30 s), the acceleration may switch back to the first electric motor 316A again.

[0068] Using the speed profile 302, the EV102 can save energy, such as 1.5 - 13% energy savings. For example, the EV102 accelerates from initial speeds of 10 km / h and 20 km / h to reach target speeds of 60 km / h and 100 km / h, respectively, within a period such as 30 seconds. The average speed of the speed profile 302 may be constrained such that the moving distance in the period assigned to the speed profile 302 is the same.

[0069] The derivation of the energy loss function 110 of the plurality of electric motors of the EV102 will be further described with reference to FIG. 6.

[0070] FIG. 6 shows a schematic block diagram 600 corresponding to a procedure for learning an energy loss function 110 for controlling the movement of the EV102 according to some embodiments of the present disclosure.

[0071] In some embodiments, the energy loss function 110 may be learned using kernel-based regression. The kernel-based regression adapts data of one or more electric motor efficiencies 602 of one or more electric motors to the energy loss function 110 based on a model of drive system losses of one or more electric motors and a model of drive losses caused by aerodynamic resistance and rolling resistance of the EV 102. The model of drive system losses may be determined from one or more efficiencies resulting from energy use and energy regeneration due to acceleration and deceleration of the EV 102.

[0072] In some exemplary embodiments, the one or more electric motor efficiencies 602 may be obtained from corresponding electric motor efficiency data of one or more electric motors of the EV 102. For example, the one or more electric motor efficiencies 602 may be obtained from first electric motor efficiency data 602A, second electric motor efficiency data 602B, and Nth electric motor efficiency data 602N of one or more electric motors of the EV 102.

[0073] Each of the first electric motor efficiency data 602A, the second electric motor efficiency data 602B, and the Nth electric motor efficiency data 602N may be used to determine a motor efficiency function 604 of the corresponding one or more electric motors. The motor efficiency function 604 is an example of a predefined motor efficiency function 310. For example, the first electric motor efficiency data 602A is used to determine a first electric motor efficiency function 604A of the first electric motor of the EV 102, the second electric motor efficiency data 602B is used to determine a second electric motor efficiency function 604B of the second electric motor of the EV 102, and the Nth electric motor efficiency data 602N is used to determine an Nth electric motor efficiency function 604N of the Nth electric motor of the EV 102.

[0074] Furthermore, the motor efficiency function 604 of one or more electric motors of the EV102 may be used together with the vertical motion model 606 of the EV102 to determine the torque split function 608. The vertical motion model 606 corresponds to the vertical motion model 304, and the torque split function 608 corresponds to the predetermined torque split function 310. In some embodiments, all of the motor efficiency function 604, the vertical motion model 606, and the torque split function 608 may be used to determine the energy loss function 110.

[0075] In some exemplary embodiments, kernel regression learns a model corresponding to the motor efficiency function 604, the vertical motion model 606, the torque split function 608, and the energy loss function 110. Kernel regression will be further described with reference to FIG. 7.

[0076] FIG. 7 shows an exemplary graphical display depicting a graph plot 700 corresponding to the kernel function 702 of kernel regression for learning a model of the energy loss function 110 according to some embodiments of the present disclosure. Kernel regression is a data-based model fitting method that uses the kernel function 702. The kernel function 702 may use a set of training data (X) corresponding to one of the one or more electric motor efficiencies 602 of FIG. 6 to learn a continuous function from a set of training data. The training data may be obtained using a grid of data regarding the speed of the EV102 and the acceleration of the EV102. This will be further described with reference to FIG. 9.

[0077] In particular, the kernel function 702 associates two data points within the training data set (X) to determine the energy loss function 110.

[0078]

Number

[0079]

Number

[0080]

Number

[0081] More specifically, the kernel-based function 702 may be used as a quasiconvex loss function having a minimum value in the data of one or more efficiencies 602, i.e., the smallest value overall. Therefore, Equation (19b) may be used to determine the global optimum value (x * ), and the set X may include the operating range of one or more electric motors of the EV102. For example, the motor efficiency function 604(η) may be a function of the motor speed (ω) and the motor torque (τ). The motor efficiency function 604 can be defined as η = k([ω,τ] T ,X)α + α0.

[0082]

Number

[0083]

Number

[0084]

Number

[0085]

Number

[0086]

Number

[0087] The kernel regression of (25) may be used to learn the motor efficiency function 604 (described in FIG. 6) from the training data set. This will be further described with reference to FIG. 8.

[0088]

Number

[0089] FIG. 9 shows an exemplary graphical representation depicting an interpolation function 900 corresponding to the torque split function 608 according to some embodiments of the present disclosure. The interpolation function 900 may use a grid 902 of sample data regarding the speed of the EV 102 and the total torque of the EV 102. For each sample grid point within the grid 902, an optimal torque split ratio is determined using a predefined motor efficiency function 802 for all electric motors of the EV 102 and / or an optimal gear ratio for any or all of the electric motors of the EV 102. The optimal torque split ratios for all sample grid points within the grid 902 may be stored in a memory such as the memory 108 of the system 100. The torque split function may use the stored grid 902 of the speed and total torque of the EV 102, together with the corresponding torque split ratios of the electric motors of the EV 102.

[0090]

Number

[0091]

Number

[0092]

Number

[0093]

Number

[0094]

Number

[0095]

Number

[0096] An exemplary graphical representation of the energy loss function 110 will be further described with reference to FIG. 10.

[0097] FIG. 10 shows an exemplary graphical representation 1000 depicting an energy loss function 1002 according to some embodiments of the present disclosure. The energy loss function 1002 is an example of the energy loss function 110. The energy loss function 1002 is expressed as a function of the speed and acceleration of the EV 102. This is because the drive losses of the EV 102 such as air resistance and rolling resistance depend on the speed, and the motor losses due to motor efficiency (e.g., one or more motor efficiencies 602) depend on the motor torque related to the speed and acceleration of the EV 102.

[0098] As shown in FIG. 10, the graphical representation 1000 includes energy loss lines separated by 300 watts. For example, the energy loss line 1004 in the graphical representation 1000 indicates an energy loss of 300 watts.

[0099] FIG. 11 shows a flowchart of a method 1100 for controlling the movement of an EV (such as EV 102) according to some embodiments of the present disclosure. The method 1100 can be executed by the system 100.

[0100] In step 1102, method 1100 includes determining a speed profile that moves the EV from an initial speed over a period of time by minimizing energy dissipation according to an energy loss function, where the speed profile is a function of time. In some embodiments, the initial speed may be estimated as the current speed using one or more sensors of the EV. Also, the speed profile may include a target speed of the EV that can be determined based on the motion plan of the EV and the period assigned to the speed profile. The target speed may be at the end of the speed profile. In some embodiments, the speed profile may be determined based on at least one or more of a constraint on the average speed of the speed profile, the current position and current speed of a leading vehicle traveling ahead of the EV, and the predicted speed of the leading vehicle traveling ahead of the EV and the current position and the current speed.

[0101] In step 1104, method 1100 includes controlling one or more electric motors of the EV to generate torque to move the EV according to the speed profile.

[0102] In some embodiments, controlling one or more electric motors includes determining a total torque profile of one or more electric motors to move the EV according to the speed profile using a model of the longitudinal motion of the EV related to vehicle acceleration and vehicle speed. In some cases, the EV may use a first electric motor (e.g., the first electric motor 316A) and a second electric motor (e.g., the second electric motor 316B). In such a case, the total torque profile is divided into a first torque profile and a second torque profile based on a predetermined torque splitting function. The first electric motor and the second electric motor are controlled according to the first torque profile and the second torque profile, respectively.

[0103] FIG. 12 shows a block diagram 1200 of an apparatus 1202 for controlling the operation of the system 100 according to some embodiments of the present disclosure. In some embodiments, the apparatus 1202 may be integrated with the system 100. In some other embodiments, the apparatus 1202 may be connected to the system 100 via the network 104. The apparatus 1202 may control the system 100 for controlling the movement of the EV 102.

[0104] The apparatus 1202 may include at least one processor 1204 and a memory 1206 storing instructions, including executable modules executed by the at least one processor 1204 during the control of the system 100. The executable modules may include a transceiver 1208, a feedback controller 1210, and a Kalman filter 1212.

[0105] In an exemplary embodiment, the memory 1206 may be configured to store an energy loss function 1214, a torque splitting function 1216, and a motor efficiency function 1218 for highly efficient control of the EV 102. The energy loss function 1214 corresponds to the energy loss function 110, the torque splitting function 1216 corresponds to the torque splitting function 608, and the motor efficiency 1218 corresponds to the motor efficiency function 604. In an alternative exemplary embodiment, any or all of the energy loss function 1214, the torque splitting function 1216, or the motor efficiency function 1218 (referred to herein as functions 1214, 1216, and 1218) may be accessed from the system 100 via the network 104.

[0106] Memory 1206 may be embodied as a storage medium such as a RAM (Random Access Memory), ROM (Read Only Memory), hard disk, or any combination thereof. For example, memory 1206 may store instructions executable by at least one processor 1204. The at least one processor 1204 may be embodied as a single-core processor, a multi-core processor, a computing cluster, or any number of other configurations. The at least one processor 1204 may be operably connected to memory 1206 and / or transceiver 1208 via bus 1220. In one embodiment, the at least one processor 1204 may be configured to calculate a speed profile (e.g., speed profile 302) of EV102 using energy loss function 1214 and torque split function 1216, and / or may be configured to calculate a gear ratio of the transmission of EV102 using motor efficiency function 1218.

[0107] Using the calculated speed profile and / or gear ratio, a sequence of control inputs for controlling a movement such as a longitudinal movement of EV102 is presented to system 100 via transceiver 1208. The sequence of control inputs changes the initial speed of EV102 to track a specific speed reference, i.e., a target speed in the speed profile. In some exemplary embodiments, the sequence of control inputs may be transmitted to EV102 via transceiver 1208 as torque commands and gear ratio commands 1222. Transceiver 1208 may be, for example, a radio frequency (RF) transceiver.

[0108] In some cases, EV102 may be traveling behind another vehicle such as vehicle 402. In such a case, the apparatus 1202 may control the longitudinal movement of EV102 by using a feedback signal 1224 that may include the speed of EV102, the position of EV102, the speed of vehicle 402, the position of vehicle 402, and the like. The feedback signal 1224 may be received by a feedback controller 1210. The feedback controller 1210 is configured to determine a current control input for controlling the system 100 based on a feedback signal 1222 that includes current measurements of the current state of the system 100 by applying a control policy at each control step. The control policy may correspond to a method of converting current measurements into a current control input based on the current values of the control parameters within a set of control parameters of the feedback controller 1210. The apparatus 1202 may determine an optimized speed profile based on the feedback signal 1220 and enable the system 100 to control EV102 using the optimized speed profile.

[0109] In some embodiments, the Kalman filter 1212 is configured to repeatedly update the state of the feedback controller 1210. This state may be defined by control parameters that use a prediction model that predicts values of control parameters subject to process noise and a measurement model that updates predicted values of the control parameters based on a sequence of measurements subject to measurement noise, thereby generating current values of the control parameters that explain the sequence of measurements according to performance objectives of the system 100 such as energy-efficient motion control of EV102.

[0110] In addition to or instead of this, in some embodiments, a Kalman filter is used to update the speed profile using an energy loss function. Specifically, in response to receiving a feedback signal indicating the current state of the vehicle tracking the speed profile, some embodiments execute a Kalman filter to update the speed profile of the current state of the vehicle and improve the likelihood of energy efficiency according to a probabilistic measurement model including an energy loss function. This is advantageous because the energy loss function is deterministic while vehicle control is inherently probabilistic. For example, by using a probabilistic measurement model including a deterministic energy loss function modified by probabilistic noise, the nature of the control can be reflected, the efficiency of determining or updating the speed profile can be improved, and the incompleteness of the estimation of the energy loss function can be considered.

[0111] In some exemplary embodiments, apparatus 1202 may be embodied within the control system of EV102. This will be further described with reference to FIG. 13.

[0112] FIG. 13 shows a schematic block diagram 1300 of an EV such as EV102 having a control system 1302 for controlling the movement of EV102, according to some other exemplary embodiments of the present disclosure. The control system 1302 is embodied within EV102. In some exemplary embodiments, the control system 1302 may be connected to system 100 and apparatus 1202. In some other exemplary embodiments, the control system 1302 may be embodied together with system 100 and apparatus 1202.

[0113] EV102 includes a battery 1304, a gearbox 1306, a first electric motor such as electric motor 1308A, and a second electric motor such as electric motor 1308B. Electric motor 1308A is connected to gearbox 1306 to drive the front axle (not shown) of EV102. Electric motor 1308B may use a fixed gear to drive the rear axle of EV102.

[0114] Furthermore, EV102 may include one or more sensors collectively referred to as sensor 1310. Sensor 1310 may be any combination of a positioning sensor such as a global positioning system (GPS), an acceleration sensor such as an inertial measurement unit (IMU), a gyroscope sensor, a radar sensor, a camera, etc. In an exemplary embodiment, sensor 1310 may be used to measure the speed and position of EV102. In another exemplary embodiment, sensor 1310 may be used to measure the speed and position of EV102, as well as the speed and position of a leading vehicle 402 traveling ahead of EV102.

[0115] The control system 1302 may calculate a speed profile of EV102 using sensor measurements from sensor 1310. The control system 1302 may output the motor torque of electric motor 1308A, the motor torque of electric motor 1308B, and the gear ratio of the gearbox 1306 using the calculated speed profile. In particular, the control system 1302 determines the total torque profile of electric motor 1308A and electric motor 1308B. The total torque profile is divided between electric motor 1308A and electric motor 1308B. For example, the total torque profile is divided into a first torque and a second torque for the corresponding wheels of electric motor 1308A and electric motor 1308B. The first torque and the second torque are transmitted as torque commands to electric motor 1308A and electric motor 1308B.

[0116] FIG. 14 shows a use case 1400 for controlling the movement of EV1402 using system 100 according to some embodiments of the present disclosure. In an exemplary scenario for illustration, EV1402 and vehicle 1404 are traveling on a road. Vehicle 1404 is traveling ahead of EV1402 as shown in FIG. 14. EV1402 is an example of EV102, and vehicle 1404 is an example of vehicle 402.

[0117] System 100 determines the speed profile of EV1402 to move EV1402 from an initial speed of, for example, 10 km / h to a target speed of, for example, 60 km / h within a period such as 30 seconds. To maintain a minimum distance as a safety distance from vehicle 1404, System 100 may determine the speed profile of EV1402 based on one or more constraints (such as inequality constraints) on the average speed of the speed profile, the current position and current speed of vehicle 1404, and / or the current position, current speed, and predicted speed within a period such as 30 seconds. For example, the current speed of vehicle 1404 may be 20 km / h and the predicted speed of vehicle 1404 may be 100 km / h. In some exemplary embodiments, the current position and current speed of vehicle 1404 may be obtained from a feedback signal (such as feedback signal 1222) of EV102.

[0118] Next, System 100 controls one or more electric motors of EV1402 to generate a total torque profile for moving EV1402 based on the calculated speed profile. The total torque profile may be divided for the corresponding one or more electric motors of EV1402 using a torque splitting function (such as a predefined torque splitting function 308). Also, using the calculated speed profile, the gear ratio of the transmission of one or more electric motors of EV1402 may be determined using a predefined motor efficiency function (such as a predefined motor efficiency function 310). Next, System 100 minimizes the energy consumption of EV1402 while following vehicle 1404 and maintaining a safety distance from vehicle 1404.

[0119] Such an optimal control procedure of System 100 is efficient for an EV including a plurality of electric motors with different specifications and can improve the overall energy efficiency of the EV by taking advantage of the different electric motors.

[0120] Also, individual embodiments may be described as processes shown as flowcharts, flow diagrams, data flow diagrams, structure diagrams, or block diagrams. A flowchart may describe the operations as a sequential process, but many of the operations can be executed in parallel or simultaneously. Additionally, the order of the operations may be rearranged. A process may end when its operations are completed, but may have additional steps not discussed or included in the figure. Further, all of the operations in any specifically described process may not occur in all embodiments. A process may correspond to a method, function, procedure, subroutine, subprogram, etc. When a process corresponds to a function, the end of the function may correspond to returning the function to the calling function or the main function.

[0121] Furthermore, embodiments of the disclosed subject matter may be implemented, at least in part, either manually or automatically. Manual or automatic implementation may be executed or at least assisted through a machine, hardware, software, firmware, middleware, microcode, hardware description language, or any combination thereof. When implemented in software, firmware, middleware or microcode, the program code or code segments for performing the necessary tasks may be stored on a machine-readable medium. A processor (or processors) may perform the necessary tasks.

[0122] The embodiments of the present disclosure can be realized in any of a number of ways. For example, these embodiments may be realized using hardware, software, or a combination thereof. When realized in software, the software code can be executed in any suitable processor or collection of processors, whether provided on a single computer or distributed among multiple computers. Such processors may be realized as integrated circuits with one or more processors within an integrated circuit component. That being said, the processors may be realized using circuitry in any suitable format.

[0123] The various methods or processes outlined herein may be encoded as software executable on one or more processors employing any one of a variety of operating systems or platforms. Additionally, such software may be described using any of a plurality of suitable programming languages and / or programming or scripting tools, and may also be compiled as executable machine language code or intermediate code to be executed on a framework or virtual machine. Typically, the functionality of program modules may be combined or distributed as desired in various embodiments.

[0124] Also, the embodiments of the present disclosure may be implemented as a method, and an example thereof is provided. The order of operations performed as part of this method may be determined in any suitable way. Accordingly, the embodiments may be configured such that the operations are performed in an order different from the illustrated order, which may include performing some operations simultaneously, although they are shown as a series of operations in the exemplary embodiments. Accordingly, it is the object of the appended claims to cover all such variations and modifications that fall within the true spirit and scope of the present disclosure.

[0125] Although the present disclosure has been described with reference to several preferred embodiments, it should be understood that various other adaptations and modifications can be made within the spirit and scope of the present disclosure. Accordingly, it is the aspect of the appended claims to cover all such variations and modifications that fall within the true spirit and scope of the present disclosure.

Claims

**Claim 1** A method for controlling the movement of an electric vehicle (EV), the method using a processor coupled to a memory storing an energy loss function, the energy loss function mapping values of acceleration and velocity of the EV to energy dissipation of the EV, the energy dissipation resulting from controlling one or more electric motors of the EV to move the EV with corresponding acceleration and velocity values, the processor being configured to execute instructions stored in the memory for implementing the method, the instructions, when executed by the processor, performing steps of the method, the method comprising: determining a speed profile for moving the EV from an initial speed over a period by minimizing the energy dissipation according to the energy loss function, the speed profile being a function of time, the method further comprising: controlling the one or more electric motors of the EV to generate torque for moving the EV according to the speed profile, the energy loss function corresponding to a quasiconvex function having a minimum value, the energy loss function being determined based at least on a model of longitudinal motion related to vehicle acceleration and vehicle speed, a predetermined torque splitting function, and a predetermined motor efficiency function, wherein the model of longitudinal motion is a model used to determine a total torque profile of the one or more electric motors for moving the EV according to the speed profile, wherein the predetermined torque splitting function is a function used to allocate the total torque profile among the one or more electric motors, and wherein the predetermined motor efficiency function is determined from efficiency data of the one or more electric motors. **Claim 2** The step of controlling the one or more electric motors comprises: determining the total torque profile using the model of longitudinal motion; and controlling the one or more electric motors based on the determined total torque profile. The method according to claim 1. **Claim 3** The one or more electric motors include a first electric motor of the EV and a second electric motor of the EV, and further, Based on a predetermined torque splitting function, a step of allocating the total torque profile to a first torque profile and a second torque profile, the predetermined torque splitting function allocates the total torque profile between at least the first electric motor and the second electric motor, and further, The method according to claim 2, further comprising the step of controlling the first electric motor and the second electric motor according to the first torque profile and the second torque profile respectively.

4. Determining a gear ratio of a transmission of the one or more electric motors using a predetermined motor efficiency function; And a step of controlling the transmission based on the determined gear ratio, wherein the predetermined motor efficiency function is learned using a kernel-based regression that fits data of an efficiency map of the first electric motor that associates the efficiency of the first electric motor with the motor speed and motor torque of the first electric motor. The method according to claim 3.

5. The method according to claim 4, wherein the predetermined torque splitting function is learned using the predetermined motor efficiency function of the one or more electric motors.

6. The energy loss function is learned using a kernel-based regression that adapts data of one or more efficiency maps of the one or more electric motors to the energy loss function based on a model of drive system losses of the one or more electric motors and a model of drive losses caused by aerodynamics and rolling resistance of the EV. The model of drive system losses corresponds to the efficiency of the motor mode and the generator mode of the EV. The motor mode corresponds to the use of electrical energy for vehicle acceleration, and the generator mode corresponds to the regeneration of the electrical energy due to vehicle acceleration and deceleration of the EV. The method further comprises operating the one or more electric motors in at least the motor mode and the generator mode of the EV using the total torque profile. The method according to claim 1.

7. The method according to claim 1, further comprising the step of optimizing the energy loss function based on the gradient descent method.

8. Determining the initial velocity as the current velocity estimated using one or more sensors of the EV; Further comprising determining a target velocity based on the motion plan of the EV and the period assigned to the velocity profile, the target velocity being at the end of the velocity profile, and further comprising, for each sampling time step for controlling the motion of the EV, recalculating the velocity profile and the initial velocity, or the step of recalculating the velocity profile is Receiving a feedback signal indicative of the current state of the EV tracking the velocity profile; Executing a Kalman filter to update the velocity profile of the current state of the EV and improve the likelihood of energy efficiency according to a probabilistic measurement model including the deterministic energy loss function. The method according to claim 1.

9. Constraints on the average velocity of the velocity profile, The current position and current velocity of a leading vehicle traveling ahead of the EV, and The predicted velocity of the leading vehicle traveling ahead of the EV over a period of time, the current position and the current velocity The method according to claim 6, further comprising determining the velocity profile based on at least one or more of the foregoing.

10. A system for controlling the motion of an electric vehicle (EV), comprising A processor coupled to a memory storing an energy loss function, the energy loss function mapping values of the acceleration and velocity of the EV to energy dissipation of the EV, the energy dissipation being caused by controlling one or more electric motors of the EV to move the EV with corresponding acceleration and velocity values, the processor, by executing instructions stored in the memory, causes the system to Determine a velocity profile for moving the EV from an initial velocity over a period of time by minimizing the energy dissipation according to the energy loss function, the velocity profile being a function of time, and further cause the system to configured to cause execution of a step of controlling the one or more electric motors of the EV to generate torque for moving the EV according to the speed profile, wherein the energy loss function corresponds to a quasiconvex function having a minimum value, and the energy loss function is determined based on at least a model of longitudinal motion related to vehicle acceleration and vehicle speed, a predetermined torque splitting function, and a predetermined motor efficiency function, The model of the longitudinal motion is a model used to determine a total torque profile of the one or more electric motors for moving the EV according to the speed profile, The predetermined torque splitting function is a function used to allocate the total torque profile among the one or more electric motors, The predetermined motor efficiency function is a system determined from efficiency data of the one or more electric motors. **Claim 11** To control the one or more electric motors, the processor causes the system to, by executing instructions stored in the memory, determine the total torque profile using a model of the longitudinal motion of the EV, and control the one or more electric motors based on the determined total torque profile. The system according to claim 10. **Claim 12** The one or more electric motors include a first electric motor of the EV and a second electric motor of the EV. The processor further causes the system to, by executing instructions stored in the memory, allocate the total torque profile into a first torque profile and a second torque profile based on a predetermined torque splitting function, the predetermined torque splitting function allocating the total torque profile between at least the first electric motor and the second electric motor, and further causes the system to control the first electric motor and the second electric motor according to the first torque profile and the second torque profile, respectively. The system according to claim 11. **Claim 13** The processor further causes the system to, by executing instructions stored in the memory, determine a gear ratio of a transmission of the one or more electric motors using a predetermined motor efficiency function; and control the transmission based on the determined gear ratio. The system according to claim 12. **Claim 14** The predetermined motor efficiency function is learned using a kernel-based regression that fits data of an efficiency map of the first electric motor that associates the efficiency of the first electric motor with the motor speed and motor torque of the first electric motor, and the predetermined torque splitting function is learned using the predetermined motor efficiency function of the one or more electric motors. The system according to claim 12. **Claim 15** The energy loss function is learned using a kernel-based regression that fits data of one or more efficiency maps of the one or more electric motors to the energy loss function based on a model of drive system losses of the one or more electric motors and a model of drive losses caused by aerodynamics and rolling resistance of the EV. The model of drive system losses corresponds to the efficiency of the motor mode and the generator mode of the EV. The motor mode corresponds to the use of electrical energy for vehicle acceleration, and the generator mode corresponds to the regeneration of the electrical energy due to vehicle acceleration and deceleration of the EV. The system according to claim 10.

Images