Technical Field

[0001] This disclosure relates to regeneration control in electric vehicles.

Background

[0002] One drawback of electric vehicles (EVs) compared to conventional internal combustion engines is that the range of EVs is limited by their battery capacity and recharging the battery takes significantly longer than refuelling a fuel tank. A larger battery would increase the range but would have a significant impact on the cost of the EV. As a result, EV developers have focussed significant effort on increasing the range of EVs without making batteries bigger, heavier and to reduce cost.

[0003] One such effort is regenerative braking with the aim of recuperating energy to recharge the battery. More specifically, the kinetic energy available by the moving mass of the vehicle can be converted back into electric energy and stored in the battery. This is typically achieved by using the electric motor that drives the EV as a generator to slow down the EV and thereby recuperate the kinetic energy. A motor controller applies a negative torque to the electric motor which slows down the vehicle and also results in a net current that the controller uses to charge the battery.

[0004] Any discussion of documents, acts, materials, devices, articles or the like which has been included in the present specification is not to be taken as an admission that any or all of these matters form part of the prior art base or were common general knowledge in the field relevant to the present disclosure as it existed before the priority date of each of the appended claims.

[0005] Throughout this specification the word "comprise", or variations such as "comprises" or "comprising", will be understood to imply the inclusion of a stated element, integer or step, or group of elements, integers or steps, but not the exclusion of any other element, integer or step, or group of elements, integers or steps.

Summary

[0006] A method for controlling a motor of an electric vehicle, the method comprising: receiving first sensor data indicative of an inclination of the vehicle; receiving second sensor data indicative of a mass of the vehicle; receiving, from a driver of the electric vehicle, an indication of a desired behaviour of the electric vehicle; calculating, based on the first sensor data and the second data, an amount of torque required to achieve an actual behaviour of the vehicle that matches the desired behaviour of the vehicle independently from the mass of the vehicle and the inclination of the vehicle; and controlling the motor to apply the calculated amount of torque.

[0007] In some embodiments, calculating the amount of torque comprises evaluating a function with the first sensor data and the second sensor data being an input parameter to the function.

[0008] In some embodiments, the function comprises a mechanical downhill force acting on the electrical vehicle.

[0009] In some embodiments, the downhill force is in a direction of travel of the electric vehicle.

[0010] In some embodiments, the function comprises a mass value of the electric vehicle as a parameter. [0011] In some embodiments, the function is configured to calculate the mechanical downhill force F as F = m -9.81 -sin(a)~ , where m is the mass value, and a is the measured inclination.

[0012] In some embodiments, the method further comprises measuring, by the inclination sensor, a physical effect that is directly related to the inclination of the vehicle.

[0013] In some embodiments, the method further comprises measuring, by the inclination sensor, the inclination of the vehicle relative to a gravity vector.

[0014] In some embodiments, the method comprises using a rotational accelerometer as part of the inclination sensor.

[0015] In some embodiments, the method comprises using a gyroscope as part of the inclination sensor.

[0016] In some embodiments, the desired behaviour of the electric vehicle comprises a set speed provided by cruise control.

[0017] In some embodiments, the method further comprises performing a control method to correct an error between the set speed and a current speed.

[0018] In some embodiments, the calculated amount of torque is an input to the control method.

[0019] In some embodiments, the control method comprises a proportional element that calculates an output torque that is proportional to the error.

[0020] In some embodiments, the control method comprises an integral element that calculates the output torque as an integral over time of the error and/or a differential component that calculates the output torque as a time differential of the error. [0021] In some embodiments, the controller is further configured to add the calculated amount of torque to the output torque.

[0022] In some embodiments, the desired behaviour comprises maintaining the vehicle stationary on an incline.

[0023] In some embodiments, the method further comprises releasing a brake of the vehicle and the desired behaviour comprises maintaining the vehicle stationary upon releasing the brake.

[0024] In some embodiments, the method further comprises releasing a brake system and controlling the motor to apply the required torque in response to a requested torque requested by the driver surpassing the required torque.

[0025] In some embodiments, the desired behaviour comprises an acceleration or deceleration as indicated by a pedal position set by the driver.

[0026] In some embodiments, the pedal has a first position for maintaining a constant speed and a change in pedal position in a first direction indicates an acceleration and a change in pedal in a second direction indicates a deceleration.

[0027] In some embodiments, the pedal position in the second direction, indicates a braking distance as the desired behaviour and calculating the amount of torque comprises calculating the amount of torque required to achieve the indicated braking distance independently from the mass of the vehicle and the inclination of the vehicle.

[0028] An electric vehicle comprises a controller configured to perform the above method.

[0029] An electric vehicle controller is configured to perform the above method.

[0030] A non-transitory computer readable medium has program code stored thereon that, when executed by a processor, causes the processor to perform the above method. [0031] An electric vehicle comprises: a battery configured to provide energy to an electric motor; a controller configured to control the motor to generate torque by consuming electrical energy from the batter or by generating electrical energy and providing the generated electrical energy to the battery to charge the battery; an inclination sensor to provide first sensor data indicative of an inclination of the vehicle; a load sensor to provide second sensor data indicative of a mass of the vehicle; wherein the controller is configured to: receive the first sensor data indicative of the inclination and the second sensor data indicative of the mass, calculate an amount of torque based on the first sensor data and the second sensor data to maintain a set speed under the measured inclination, and control the motor to apply the calculated amount of torque to maintain the set speed under the measured inclination and to thereby consume the electrical energy provided by the battery or generate the electrical energy provided to the battery.

[0032] In some embodiments, the controller is further configured to perform a control method to correct an error between the set speed and a current speed.

[0033] In some embodiments, the calculated amount of torque is an input to the control method.

[0034] In some embodiments, the control method comprises a proportional element that calculates an output torque that is proportional to the error.

[0035] In some embodiments, the control method comprises an integral element that calculates the output torque as an integral over time of the error and/or a differential component that calculates the output torque as a time differential of the error.

[0036] In some embodiments, the controller is further configured to add the calculated amount of torque to the output torque. [0037] In some embodiments, the set speed is set by a driver of the electric vehicle by way of a user input interface associated with a cruise control.

[0038] In some embodiments, calculating the amount of torque comprises evaluating a function with the sensor data being an input parameter to the function.

[0039] In some embodiments, the function comprises a mechanical downhill force acting on the electrical vehicle.

[0040] In some embodiments, the downhill force is in a direction of travel of the electric vehicle.

[0041] In some embodiments, the function comprises a mass value of the electric vehicle as a parameter.

[0042] In some embodiments, the function is configured to calculate the mechanical downhill force F as F = m -9.81 -sin(a)~ , where m is the mass value, and a is the measured inclination.

[0043] In some embodiments, the inclination sensor is configured to measure a physical effect that is directly related to the inclination of the vehicle.

[0044] In some embodiments, the inclination sensor is configured to measure the inclination of the vehicle relative to a gravity vector.

[0045] In some embodiments, the inclination sensor comprises a rotational accelerometer.

[0046] In some embodiments, the inclination sensor comprises a gyroscope.

[0047] In some embodiments, the controller is further configured to: determine an amount of required torque for holding the vehicle stationary based on the measured inclination and the measured mass; releasing a brake system and controlling the motor to apply the required torque in response to a requested torque requested by a driver surpassing the required torque.

[0048] In some embodiments, the controller is further configured to: determine an amount of required torque for achieving a positive or negative acceleration as indicated by a driver input device based on the measured inclination or the measured mass or both; control the motor to apply the amount of required torque

[0049] A method for controlling a motor of an electric vehicle comprises: receiving first sensor data indicative of an inclination of the vehicle; receiving second sensor data indicative of a mass of the vehicle; calculating an amount of torque based on the first sensor data and the second sensor data to maintain a set speed under the measured inclination and the measured mass; and controlling the motor to apply the calculated amount of torque to maintain the set speed under the measured inclination and the measured mass and to thereby consume the electrical energy provided by the battery or generate the electrical energy provided to the battery.

[0050] An electric vehicle controller is configured to: receive first sensor data indicative of an inclination of the vehicle; receive second sensor data indicative of a mass of the vehicle; calculate an amount of torque based on the first sensor data and the second sensor data to maintain a set speed under the measured inclination and the measured mass, and control the motor to apply the calculated amount of torque to maintain the set speed under the measured inclination and the measured mass and to thereby consume the electrical energy provided by the battery or generate the electrical energy provided to the battery. [0051] In some embodiments, the controller comprises a processor configured to execute program code stored on a non-volatile computer-readable medium.

[0052] A method for maintaining an electric vehicle stationary on an inclined road comprises: receiving first sensor data indicative of an inclination of the vehicle on the inclined road; receiving second sensor data indicative of a mass of the vehicle; calculating an amount of torque based on the first sensor data and the second sensor data to maintain the electric vehicle stationary under the measured inclination and the measured mass; and controlling the motor to apply the calculated amount of torque to maintain the vehicle stationary under the measured inclination and the measured mass.

[0053] In some embodiments, the method further comprises releasing a brake system upon controlling the motor to apply the required torque

[0054] In some embodiments, releasing the brake system is in response to a requested torque requested by a driver equals the required torque.

[0055] A method for controlling a motor of an electric vehicle comprises: receiving sensor data indicative of a mass of the vehicle; receiving, from a driver of the electric vehicle, an indication of an acceleration or deceleration as indicated by a pedal position set by a driver of the electric vehicle; calculating, based on the sensor data, an amount of torque required to achieve the indicated acceleration or deceleration of the vehicle independently from the mass of the vehicle; and controlling the motor to apply the calculated amount of torque. Brief Description of Drawings

[0056] Fig. 1 illustrates a control circuit to control an electric vehicle (EV) according to an embodiment.

[0057] Fig. 2 illustrates a modified control circuit according to an embodiment.

[0058] Fig. 3 illustrates the complete EV 300 according to an embodiment.

[0059] Fig. 4 illustrates a method for controlling downhill regenerative braking of electric vehicle according to an embodiment.

[0060] Fig. 5 illustrates a plot of the applied torque in kNm over the inclination in degrees according to an embodiment.

Description of Embodiments

[0061] This disclosure relates to motor control, such as the application of a positive torque to hold an electric vehicle on an incline or accelerate, or a negative torque for generative braking, such as in downhill operation of an electric vehicle and in particular, during use of cruise control to maintain a constant speed. More specifically, the electric vehicle uses an inclination sensor and a load sensor. With the inclination and the load (which is indicative of the mass of the vehicle), the vehicle controller can calculate the exact amount of torque that is required for the desired vehicle behaviour, such as downhill cruise control, uphill hill assist, and single pedal operation.

Cruise control

[0062] In general, cruise control involves a set speed that is typically set by the driver, such as by pushing a ‘Set’ button at the desired speed or by adjusting the set speed by pushing ‘+’ and buttons or other means. A motor controller then measures the current speed and compares the current speed to the set speed, such as by calculating a speed difference. The controller then adjusts vehicle parameters, namely the voltage applied to the motor, to reduce that speed difference to zero.

Control circuit

[0063] Fig. 1 illustrates a control circuit 100 to control an electric vehicle (EV) 101 according to an embodiment. EVs described in this disclosure can be fully electric where the electric motor is the only means of providing torque to the wheels and the battery is the only source of electric power. In other examples, however, EVs may be hybrid electric vehicles including an internal combustion engine for charging the battery or for applying additional torque to the wheels or both, or hydrogen fuel cell vehicles where the hydrogen fuel cell provides electric energy for the electric motor, potentially also including a battery. EV 101 comprises a speed sensor that provides sensor data as the current speed V_cur 102. The driver of EV 101 sets the desired speed which is provided to circuit 100 as set speed V_set 103. Circuit 100 substrates the current speed 102 from the set speed 103 to calculate an error 104. It is clear that error 104 is zero if the current speed 102 exactly matches the set speed 103.

[0064] The error is an input to computational modules comprising a proportional module 105 that multiplies the error with a first constant, an integral module 106 that integrates the error over time and multiplies the result with a second constant. Further, there is a differential module 107 that calculates the derivative of the error with respect to time (i.e. rate of change) and multiplies the result with a third constant. It is noted that not all modules 105, 106, 107 may be present. In particular, the differential module

107 may often be omitted due to difficulties regarding stability and convergence.

[0065] The output values of the computational modules 105, 106 and 107 are added at

108 to provide a torque output 109. The torque output 109 may be in the form of a controlled voltage or other parameters. When in use, if there is a difference between the current speed 102 and set speed 103, the proportional module 105 outputs a value that is proportional to the error. However, that may result in a stable state with constant error over time. To counteract this constant error, the integral module 106 provides an output value that increases over time if the error remains constant in order to reduce that error over time. Finally, the output of the differential module 107 depends on the rate of change of the input. So if the input changes quickly, the differential module 107 will dampen that change in order to ensure smoother operation.

[0066] It is noted that the circuit 100 including modules 105, 106, and 107 typically requires tuning by determining the first, second and third constants, that is, the weights applied to the proportional, integral and differential calculations. This tuning can become difficult for EVs resulting in sub-optimal control including oscillations and overshoot problems.

Inclination sensor

[0067] Fig. 2 illustrates a modified control circuit 200, according to an embodiment, which shares many of the components from Fig. 1. It is noted that control circuit 200 may be implemented integrally with a vehicle controller or as a separate component. Control circuit 200 may be implemented as a field programmable gate array, application specific integrated circuit, or a microprocessor executing program code.

[0068] Compared to Fig. 1, there is now an inclination sensor 201 (or tilt or slope sensor) added to the circuit. Inclination sensor 201 provides sensor data indicative of the pitch of the EV 101, that is, a back to front tilt. In other words, the tilt sensor 201 indicates the incline on which the EV moves. That is, inclination sensor 201 indicates whether EV 101 moves uphill or downhill. More specifically, inclination sensor 201 provides the degrees of inclination (or similar measure) of the road surface on which EV 101 travels. It is noted that, while some examples herein relate to downhill applications, the disclosed solutions can equally be applied to uphill applications. More specifically, the acceleration torque required to maintain a constant speed when driving uphill can be calculated using the same equation as the regeneration torque required for constant speed downhill.

[0069] In some examples, the inclination sensor 201 measures a physical effect that is directly related to the rotation angle of the sensor. This angle may be with reference to the gravity vector. For example, inclination sensor 201 may comprise a rotatable proof weight with connected electrodes. As the sensor tilts, the proof weight rotates relative to the sensor and so the capacitance between the electrodes of the proof weight and static electrodes of the sensor changes. This change in capacitance is measured and can be converted into a rotation angle. Inclination sensor 201 may be a microelectromechanical system (MEMS). Further, inclination sensor 201 may be a static inclinometer or a dynamic inclinometer. A static inclinometer is a sensor with a proof weight that is designed to align with the Earth’s gravity vector. A dynamic inclinometer provides changes in rotation angle. In other words, the dynamic inclinometer is a rotational accelerometer. The actual inclination angle can then be obtained by integrating the rotational acceleration twice. In yet a further example, the inclination sensor 201 comprises gyroscopes to measure an inclination angle.

[0070] The inclination sensor 201 may process the measurement, such as by applying a cosine function and multiplying the result by a fourth constant. In other examples, the vehicle controller receives the raw sensor data and makes those calculations.

[0071] The advantage of including inclination sensor 201 is that a slope in the road surface now directly leads to a change in torque 109. That is, in response to measuring uphill motion, a positive torque 109 is applied. Conversely, in response to measuring downhill motion, a negative torque 109 is applied, , to keep the speed constant. In some examples, there is a motor controller that receives as input the applied torque and controls the output power, such as by controlling the voltage applied to the motor windings. Especially in downhill motion, the negative torque leads to regeneration (i.e. recuperation) of kinetic energy that would have otherwise been dissipated as heat by the brakes. Further, the output values of the computational modules 105, 106, 107 are significantly reduced because the inclination sensor 201 already provides most of the correction. This may be an at least tenfold reduction, such as from 100 Nm to only 10 Nm that is to be controlled by the computational modules 105, 106, 107. As a result, oscillations and overshoots are significantly reduced, leading to reduced wear and tear, and increased driver comfort. Ultimately, the disclosed method increases range compared to less efficient control without tilt sensor 201. [0072] In one example, the downward force on a hill with a slope (i.e. inclination) of a degrees is: F = m • 9.81 • sin(a)~ , where m is the mass of the EV in kg. With wheel radius r, the wheel torque T is then T = F - r = m -9.81 - sin(a)~ ■ r . The motor torque is given by Tm = T / DR, where DR is the vehicle differential ratio. The mass of the EV can be set at manufacture or deployment as the empty mass or the fully laden mass. In either case, the controller may need to correct an error but that error is much smaller than without the inclination sensor. Further, the wheel radius can be set at a typical wheel radius for that particular type of EV.

Mass sensor

[0073] In yet a further example, the EV comprises a further sensor that provides sensor data indicative of the mass of the vehicle. This sensor may be a load sensor that senses the mass of the load. The total mass of the vehicle is then a known tare mass of the empty vehicle plus the measured load. Examples of load sensors include airbag sensors or deflector sensors, which may be based on springs or other compression elements. In other example, the sensor may be a tire pressure sensor that is calibrated using the tire pressure while the vehicle is empty. Including the actual measured mass of the vehicle, and not only an estimate, in addition to the inclination, provides a significantly more accurate calculation of the torque required for constant speed. Therefore, the control, such as PID control becomes less error prone and quicker to settle.

[0074] In some examples, the EV controller determines whether the EV is coasting or braking. In coasting, the EV maximises the available kinetic energy to move a maximum distance before stopping. So the motor of the EV would apply zero torque, i.e. consume zero energy and regenerate zero energy. In contrast, when the EV is braking, the aim is to maximise the amount of energy regenerated until the EV has stopped or reached the desired lower speed. It is often difficult for an EV controller to determine whether the driver intends to coast or to brake. Therefore, some examples disclosed herein make use of cruise control where the driver provides a set speed. On an even road (zero incline) the torque required to maintain that speed is almost constant so that only small errors need to be corrected. However, when the inclination changes frequently, control becomes difficult. More particularly, when the inclination changes abruptly, such as from flat to uphill, the speed drops relatively quickly. If a proportional-integral controller is used, the proportional component will immediately apply a higher torque due to the error between the set speed and the reduced speed due to the inclination. However, the higher torque is required to lift the vehicle mass uphill and does therefore not lead to an increase in speed and therefore not to a correction of the error. It is the integral component that then increases the torque to compensate for the remaining error. While the error can eventually be reduced to zero, the integral part is typically relatively slow because it integrates the error overtime. Further, it often leads to an overshoot that then needs to be corrected. The current disclosure provides a solution involving an inclination sensor that provides input to the control method. As a result, the torque correction can be immediate - similar to the proportional par - without the delay of the integral part. Therefore, settling times of frequent changes are reduced and oscillations and overshoot are suppressed. This applies to both to the inclination and the mass sensors. Each of those sensors individually provide an advantage and in combination that advantage is most advantageous.

Electric vehicle

[0075] Fig. 3 illustrates the complete EV 300 according to an embodiment. EV 300 comprises a battery 301 to provide energy to an electric motor 302. EV 300 further comprises an inclination sensor 303 and a load sensor 307. EV 300 further comprises a controller 304 (effectively implementing circuit 200 in Fig. 2) that receives sensor data indicative of the downhill inclination and the load, calculates an amount of torque based on the sensor data as described above. The torque is calculated so it maintains a set speed, as set by the driver, under (or in light of) the inclination measured by the inclination sensor 303 and the load measured by load sensor 307. Controller 304 further controls the motor to apply the calculated amount of torque to maintain the set speed under the measured inclination and to thereby generate the electrical energy provided to the battery. [0076] Controller 304 may be implemented in a variety of different forms, such as a programmable logic device (PLD) based on read only memory, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC). In yet another example, the controller 304 is implemented as a processor, such as a microcontroller, that executes program code stored on non-volatile computer-readable medium (ROM). Other controller architectures are equally possible.

Control method

[0077] Fig. 4 illustrates a method 400 for controlling a motor of electric vehicle 300 according to an embodiment. Method 400 may be performed by controller 304 such that controller 304 receives 401 sensor data indicative of the inclination such as by receiving sensor data from inclination sensor 303. Controller 304 further receives 402 sensor data indicative of the mass of the electric vehicle 300, such as by receiving sensor data from load sensor 307. Controller 304 also receives 403, from a driver of the electric vehicle 300, an indication of a desired behaviour of the electric vehicle, such as keeping the electric vehicle 300 stationary uphill, a desired acceleration or a desired deceleration, which may be indicated by a desired braking distance as set by a brake pedal or the acceleration pedal in single-pedal operation.

[0078] Then, controller 304 calculates 404 an amount of motor torque based on the sensor data to achieve an actual behaviour of the vehicle that matches the desired behaviour. The actual behaviour is then independent from the mass of the vehicle and the inclination of the vehicle. This means that the driver can set the desired behaviour, such as by cruise control or by operating the acceleration pedal, and the vehicle behaves substantially identical whether it is fully loaded or empty. For example, the vehicle maintain a set speed, remain stationary uphill, or brake within a set distance given the measured inclination and mass . The mechanical equations are provided above for this calculation. That is, controller 302 calculates the amount of torque by evaluating a function with the sensor data being an input parameter to the function and controller 302 calculates a number that is indicative of the torque. The controller 302 can then controls 405 the motor accordingly by providing that number, or corresponding control signals to a regulator to provide the appropriate amount of energy to the motor to achieve the calculated amount of torque. For example, controller 304 may output a pulse width modulation (PWM) signal that has a duty cycle to reflect the calculated amount of torque.

[0079] More particularly, during acceleration, the controller 304 provides a current to the motor, which creates a magnetic field and in cooperation with another magnet (permanent or electric), creates a moment that is applied to the wheels. As the motor spins, the motor creates a reverse induced voltage, referred to as back electromotive force (back EMF) that opposes the current which induced it. The back EMF increases with motor speed so as the motor accelerates, eventually the back EMF equals the applied voltage. This means the voltage is increased to provide further acceleration. When the applied voltage and back EMF are equal, the current is almost zero and the EV is coasting, that is, maintains an almost constant speed. In downhill scenarios however, the motor accelerates and the back EMF increases. This results in a negative current that can be used by controller 304 to charge the battery. The amount of current used depends on the voltage that is applied to the motor.

[0080] For example, the EV is coasting with cruise control on a horizontal road and the controller calculated an amount of torque to overcome air resistance and other losses to maintain a constant speed at 200 Nm for example, which may equate to a voltage of 100 V applied to the motor windings. Then the EV enters a downhill slope and the controller 304 will need to reduce the torque to 150 Nm, by reducing the voltage to 80 V, for example, to maintain speed. This sudden reduction, however, may lead to oscillations and overshoot in existing controllers. Therefore, this disclosure provides an improved controller that uses the data from an inclination sensor to calculate the amount of torque and then control the voltage applied to the motor. So in this case, the inclination sensor would provide a reduction of 50 Nm. It is noted that the motor voltage itself can be referred to as a “torque value” because the motor voltage is directly related to the mechanical torque generated by the motor. The reduction in voltage as calculated based on the inclination sensor data results in a negative current that the controller 304 uses to charge the battery and reduces oscillations and overshoot. Thereby, controller 304 controls motor 302 to apply the calculated amount of torque to maintain the set speed under the measured inclination and to thereby generate the electrical energy provided to the battery 301.

[0081] Fig. 5 illustrates a plot of the applied torque in kNm over the inclination in degrees according to an embodiment. The torque in kNm can simply be transformed to a voltage value also indicating torque, by multiplying the torque in kNm by a constant or by a parameter that depends on the motor speed. It can be seen that over this narrow range of inclination angle, the actual curve is relatively close to a linear relationship. Therefore, it is possible to simply multiply the reading of the inclination sensor by 9.81 • m • r /DR to obtain a good approximation of the torque to be applied. The remaining error can be suppressed by the computational modules.

[0082] It is noted that the inclination sensor 201 is relatively sensitive to small changes in inclination and provides the measurement relatively quickly to the control circuit 208. As a result, the torque is adjusted quickly to reflect the current inclination. For example, the torque is adjusted within 1 s or within 100 ms, which is less that what would be noticeable considering a relatively inert EV. Again, having only the circuit in Fig. 1 without the inclination sensor would result in a significantly longer settling time for the controller to reduce the error to zero resulting in a loss of recuperated energy. Therefore, adding the inclination sensor according to this disclosure increases range and reduces wear of the EV.

State of charge

[0083] In a further example, the controller monitors the state of charge of the battery, such as a charging level, which may range from 0% to 100%. If the state of charge is full, or close to full, the controller inhibits regenerative braking because there is nowhere to store the recuperated energy. Instead, the controller uses the friction brakes to keep the vehicle at the desired speed downhill.

[0084] While the above description relates to constant speed control (i.e. “cruise control”), the disclosed methods can be used in other applications. Hill assist

[0085] For example, the mass of the vehicle can be used in hill assist. Hill assist refers to the control of the EV from standstill to moving forward while the vehicle is on an uphill inclination. The motor of the EV is able to keep the EV stationary by applying the right amount of torque that would prevent the vehicle from rolling downhill.

However, applying this torque uses electrical energy continuously, which is why most EVs use conventional (i.e. friction) brakes while the vehicle is in standstill and the driver does not activate the acceleration pedal. In commercial vehicles, that conventional brake may be an air pressure brake where the brake pads are pressed against the brake discs by compressed air. As the driver depresses the acceleration pedal to control torque, there is a point where the acceleration force created by the torque requested by the driver is equal to the downhill force of the mass of the vehicle. At that point, the vehicle releases the brakes to allow the vehicle to move forward as the torque increases.

[0086] Hill assist works well for cars and other vehicles of known mass, which may be assisted by an inclination sensor so that the downhill force and thus the standstill torque can be calculated. However, in vehicles where the mass can change significantly between empty and loaded state, it is difficult to implement hill assist. In particular, if the EV is empty, the calculated torque is too high, which means the vehicle moves forward as soon as the brakes are released. Conversely, if the EV is fully loaded, the EV rolls downhill as soon as the brakes are released. It is clear that both scenarios are undesirable and present safety risks.

[0087] To avoid unintended forward movement or downhill roll, the present disclosure provides an improvement involving a load sensor or other sensor that provides data indicative of the mass of the vehicle. With the vehicle mass at hand, the EV now calculates a more accurate torque value at which to release the brakes.

Therefore, unintended forward movement and downhill rollback is reduced. Further, the energy consumption is reduced because less energy is waisted for generating torque while the brakes hold the vehicle. This improves the range of the vehicle in cases where hill assist is used frequently. As described above, the equation

T = F • r = m -9.81 - sin(a)~ • r can be used to calculate the torque required based on the mass m and the inclination a .

Single pedal operation

[0088] In a further application, the disclosed methods can be used to improve the driver experience under varying load conditions. In particular, the disclosed methods can be used to provide a consistent response by the acceleration pedal. The position of the acceleration pedal can be given in percent of the total depression, which means the position of the pedal is between 0% and 100%. In some examples, a neutral position is defined for the acceleration pedal, which may be at 50%, for example. The acceleration pedal provides for acceleration when it is depressed from the neutral position (e.g., more than 50%) and provides for regeneration (i.e. deceleration under energy recuperation) when the pedal is released from the neutral positon (e.g., less than 50%). In other words, for constant speed coasting, the driver keeps the pedal at the neutral position. Lifting the foot results in regeneration while pressing the foot down results in acceleration. This behaviour is also referred to as single pedal operation.

[0089] The difficulty with single pedal operation is that the behaviour of the vehicle changes with a change of vehicle mass. That is, at a certain regeneration position (e.g. 20%), a full vehicle will decelerate slower (travel further before coming to a halt) than an empty vehicle. This means a driver who is used to an empty vehicle tends to apply insufficient regeneration when driving a full vehicle. So the driver driving a full vehicle depresses the pedal by 20%, which is the require position for an empty vehicle to stop within the available distance. However, the required position for the full vehicle to stop within the available distance is 10%. This means that the driver needs to use the conventional brakes to bring the vehicle to a halt towards the end of the distance (e.g. at a red light). The result is a loss in recuperated energy. It is therefore desirable that the behaviour of the pedal is independent from the current loading of the vehicle. [0090] The distance that the vehicle travels before it comes to a halt is also referred to as the braking distance. The braking distance is also directly related to a (negative) acceleration a in m is ² , which means both terms can be used interchangeably. It is desirable that the braking distance (or acceleration) for a particular position of the pedal remains constant regardless of the loading of the vehicle. To achieve this aim, the EV uses the sensor data from the load sensor to calculate the torque that is required for a given braking distance. That is, the pedal position determines the braking distance (negative acceleration a) and the EV calculates the required torque, which now depends on the current loading of the EV. As a result, the braking distance for a particular pedal position does not depend on the loading and remains constant. This means the driver recuperates more energy because the use of conventional friction brakes is reduced.

[0091] It is noted that the same applies to a positive acceleration a in the sense that is desirable to accelerate the vehicle by the same rate for the same pedal position regardless of the mass of the vehicle or inclination of the road surface. For a positive acceleration, however, the advantage of increased energy recuperation may be less than for negative acceleration (i.e. deceleration).

[0092] It is noted that the inclination sensor can equally be used to calculate the required torque to decelerate the vehicle within the distance set by the pedal. So inclination and mass can be used separately or together. Essentially, inclination sensor and load sensor provide the data for calibrating the pedal behaviour so that the handling of the vehicle is independent from the loading and the inclination. The equation provided above can be adapted to incorporate the desired acceleration by

T = F - r = (F _dmvnhill + F _acceleration ) = m • 9.81 • V«(a ) ■ r + m - a , where a is now the desired acceleration/deceleration as indicated by the pedal position. It can be seen that this equation now compensates for both the inclination (from the inclination sensor) as well as the vehicle mass (from the load sensor).

[0093] It will be appreciated by persons skilled in the art that numerous variations and/or modifications may be made to the above-described embodiments, without departing from the broad general scope of the present disclosure. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive.