Traction control method and control system

Abstract

A control system (300) for controlling a traction electric machine 210 of a vehicle. Comprising obtaining a traction control target for controlling wheel speed and obtaining sensed feedback indicating

Description

TECHNICAL FIELD The present disclosure relates to a traction control method and control system. Aspects of the invention relate to a control system, to a system, to a vehicle, to a method, and to computer readable instructions. BACKGROUND During attempted vehicle acceleration, an anti-lock braking system (ABS) controller may activate a traction control function in response to a traction loss event when at least one driven vehicle wheel is slipping. When active, the traction control function causes the ABS controller to control the friction brake of the slipping wheel, via a hydraulic system, to apply braking until traction of the wheel is regained. Friction brakes are more responsive than traditional internal combustion engines. It is an aim of the present invention to address one or more of the disadvantages associated with the prior art. SUMMARY OF THE INVENTION Aspects and embodiments of the invention provide a control system, a system, a vehicle, a method, and computer readable instructions as claimed in the appended claims. According to an aspect of the present invention, there is provided a control system for controlling a traction electric machine of a vehicle, the control system comprising one or more processors collectively configured to: obtain a traction control target for controlling wheel speed; obtain sensed feedback indicating one or more state parameters of the traction electric machine; determine a torque restriction in dependence on the traction control target and the sensed feedback, using an optimal control scheme; and output a control signal to control output torque of the traction electric machine in dependence on the torque restriction. The term ‘robust control scheme’, or ‘control scheme configured to optimise an objective function’, can be used instead. An advantage is an improved traction control method. Traction electric machines can control vehicle wheel speed considerably faster than friction brakes. However, their use introduces control challenges. Optimal control schemes are robust for traction loss events on a wide range of different surfaces, without requiring prohibitive levels of calibration effort. This robustness is due to the ability of the control scheme to handle a plurality of state inputs, and due to its natural stability. The use of an optimal control scheme to control the traction electric machine, as part of a traction control function, can therefore significantly improve vehicle acceleration on complex surfaces. An optimal control scheme is a consistent and objective term for a class of process controllers configured to find a control for a dynamical system over time such that one or more objective functions are optimised. Optimising an objective function can comprise minimising a cost function, for example. The term ‘robust control scheme’, or ‘control scheme configured to optimise an objective function’, can be used instead. 1 Optionally, the optimal control scheme is hosted by an inverter controller of the control system. Optionally, the inverter controller is a controller of an inverter of the traction electric machine. Optionally, the control signal is a gate control signal. Optionally, the control signal is a gate control signal to control a gate control section of an inverter for the traction electric machine. Optionally, the gate control signal is transmitted over a local communication network of the traction electric machine to the gate control section. Optionally, the gate control signal is transmitted end-to-end to the gate control section over the local communication network. Optionally, the gate control signal is transmitted end-to-end within the inverter to the gate control section over the local communication network. Optionally, the gate control signal is sent by the inverter controller within the inverter and received by the gate control section within the inverter. In examples, the gate control signal is not transmitted over a vehicle communication network. An advantage is a reduced magnitude of initial ‘flare’ of wheel speed at the beginning of a traction loss event, due to the very low latency of the connection between the inverter controller and the gate control section. The latency may be a fraction of that which is encountered when sending a message over a vehicle communication network. Optionally, the traction control target is obtained from an anti-lock braking system (ABS) controller. Optionally, the traction control target is obtained from a traction control target generator hosted by the ABS controller. Optionally, the traction control target comprises a motor speed target and the traction control target generator Is a target speed generator. Optionally, the ABS controller is comprised in an anti-lock braking system. Optionally, the ABS controller is operably coupled to the inverter controller via a vehicle communication network. Optionally, the ABS controller is configured to arbitrate between the traction electric machine and friction brakes when determining the traction control target. An advantage of the ABS controller generating / sending the traction control target is that the ABS controller can arbitrate between the EDU and friction brakes of the vehicle, to provide a target level of wheel slip. Another advantage is that the ABS controller can coordinate slip control with other ABS algorithms such as vehicle stability control. Although the latency of sending the target from the ABS controller to the inverter controller over a vehicle communication network is higher, the inverter controller can still operate within a fast control loop within the inverter controller, so there is no latency penalty in overall system performance. Optionally, the control system comprises the ABS controller and the inverter controller, wherein the ABS controller hosts an outer control loop and the inverter controller hosts an inner control loop, wherein the ABS controller in the outer control loop is configured to determine the traction control target based on sensed vehicle speed and sensed wheel speed, and wherein the inverter controller in the inner control loop is configured to determine the torque restriction based on the traction control target and the sensed feedback. Optionally, the optimal control scheme comprises a linear quadratic regulator (LQR) control scheme. An advantage of the LQR control scheme is the prevention of initial wheel flare, and very accurate target tracking compared to proportional-integral-derivative (PID) schemes, over a wide range of different surfaces. This is achieved with a fraction of the calibration effort compared to PID schemes, and with greater computational efficiency than more complex model predictive control schemes. Greater computational efficiency allows the LQR scheme to run at a high frequency. Optionally, the traction control target comprises a motor speed target of the traction electric machine. An advantage is that the traction electric motor is used to help a slipping accelerating wheel to regain traction. Optionally, the one or more state parameters indicated by the sensed feedback indicate at least one of: a speed of the traction electric machine; a position of the traction electric machine; or torque generated by the traction electric machine. Optionally, the optimal control scheme is configured to optimise one or more tracking accuracy parameters and one or more energy consumption parameters. Optionally, the one or more tracking accuracy parameters comprise at least one of: a parameter indicative of speed of the traction electric machine; a parameter indicative of rotor position of the traction electric machine; or a parameter dependent on torque of the traction electric machine. An advantage is that the optimal control scheme is optimising many parameters rather than just one. Optionally, the one or more energy consumption parameters comprise a parameter indicative of torque of the traction electric machine. Optionally, the optimal control scheme is biased towards tracking accuracy over energy consumption. An advantage is that the optimal control scheme provides responsive and accurate traction control. The additional energy consumption is not a disadvantage because a traction loss event typically has a short duration, so the overall energy cost is low. According to another aspect of the present invention, there is provided a system comprising an electric drive unit, and the control system, wherein the electric drive unit comprises the traction electric machine, an inverter, and a first controller of the control system which hosts the optimal control scheme, and wherein the first controller is configured to obtain the traction control target and the sensed feedback, determine the torque restriction, and output the control signal to the inverter. Optionally, the first controller is an inverter controller as defined above. Optionally, the system further comprises an anti-lock braking system, wherein the anti-lock braking system comprises a second controller of the control system, and wherein the second controller is configured to determine the traction control target, and send the traction control target to the first controller over a vehicle communication network. Optionally, the second controller is an ABS controller as defined above. According to a further aspect of the present invention, there is provided a vehicle comprising the control system or the system. According to a further aspect of the present invention, there is provided a method of controlling a traction electric machine of a vehicle, the method comprising: obtaining a traction control target for controlling wheel speed; obtaining sensed feedback indicating one or more state parameters of the traction electric machine; determining a torque restriction in dependence on the traction control target and the sensed feedback, using an optimal control scheme; and outputting a control signal to control output torque of the traction electric machine in dependence on the torque restriction. According to a further aspect of the present invention, there is provided a method of controlling a traction electric machine of a vehicle, the method comprising: obtaining a traction control target for controlling wheel speed; obtaining sensed feedback indicating one or more state parameters of the traction electric machine; determining a torque restriction in dependence on the traction control target and the sensed feedback, using a control scheme, wherein the control scheme is hosted by an inverter controller of an inverter of the traction electric machine; and outputting a control signal to control output torque of the traction electric machine in dependence on the torque restriction, and wherein the control signal is a gate control signal to control a gate control section of an inverter for the traction electric machine. An advantage of hosting the control scheme in the inverter controller is the latency benefit as mentioned earlier. Based on the above statement, the control scheme in the inverter controller may either be an optimal control scheme or a different control scheme. According to a further aspect of the present invention, there is provided a method of controlling a traction electric machine of a vehicle, the method comprising: obtaining a target for controlling wheel speed; obtaining sensed feedback indicating one or more state parameters of the traction electric machine; determining a parameter in dependence on the target and the sensed feedback, using a control scheme; and outputting a control signal to control output torque of the traction electric machine in dependence on the parameter. According to a further aspect of the present invention, there is provided a control system for controlling a traction electric machine of a vehicle, the control system comprising one or more processors collectively configured to perform the steps of any one of the above methods. The control system comprises one or more controllers collectively comprising at least one electronic processor having an electrical input for receiving an input signal; and at least one memory device electrically coupled to the at least one electronic processor and having instructions stored therein; and wherein the at least one electronic processor is configured to access the at least one memory device and execute the instructions thereon so as to execute the steps of any one or more of the above methods. According to a further aspect of the present invention, there is provided computer readable instructions which, when executed by a computer, are arranged to perform any one or more of the above methods. According to a further aspect of the invention there is provided a non-transitory computer readable medium comprising computer readable instructions that, when executed by one or more electronic processors, causes the one or more electronic processors to carry out any one or more of the methods described herein. Within the scope of this application it is expressly intended that the various aspects, embodiments, examples and alternatives set out in the preceding paragraphs, in the claims and / or in the following description and drawings, and in particular the individual features thereof, may be taken independently or in any combination that falls within the scope of the appended claims. That is, all embodiments and / or features of any embodiment can be combined in any way and / or combination that falls within the scope of the appended claims, unless such features are incompatible. The applicant reserves the right to change any originally filed claim or file any new claim accordingly, including the right to amend any originally filed claim to depend from and / or incorporate any feature of any other claim although not originally claimed in that manner. BRIEF DESCRIPTION OF THE DRAWINGS One or more embodiments of the invention will now be described, by way of example only, with reference to the accompanying drawings, in which: FIG. 1 illustrates a perspective view of an example vehicle; FIG. 2 illustrates a schematic view of example components of a vehicle; FIG. 3 illustrates a schematic view of example components of a system and control system; FIG. 4 illustrates a schematic view of an example non-transitory computer-readable storage-medium; FIGS. 5A, 5B illustrate control block diagrams of an example control scheme; FIG. 6 illustrates a control block diagram of an example Linear-Quadratic Regulator (LQR); FIG. 7 illustrates an example method; and FIGS. 8A-8B illustrate graphs comparing wheel speeds controlled by LQR and PID-based speed regulators, respectively. DETAILED DESCRIPTION A vehicle 1 in accordance with an embodiment of the present invention is described herein with reference to the accompanying FIG. 1. In some, but not necessarily all examples, the vehicle 1 is a passenger vehicle, also referred to as a passenger car or as an automobile. In other examples, embodiments of the invention can be implemented for other applications, such as commercial vehicles. Various components of the vehicle 1 are now described with reference to FIG. 2. The vehicle 1 comprises a front left wheel FL, a front right wheel FR, a rear left wheel RL, and a rear right wheel RR. The number of wheels and their layout may vary depending on the type of vehicle 1. Each wheel is provided with a friction brake 226 such as a disc brake or drum brake. FIG. 2 also shows a torque source 210 comprising an electric drive unit 210 (EDU) for driving one or more of the vehicle wheels. In the illustration, the EDU 210 is configured to drive the rear wheels RL, RR. Alternatively, or additionally, the EDU 210 is configured to drive the front wheels FL, RR. Alternatively, the EDU 210 is configured to drive one of the wheels FL, FR, RL, RR. In examples, the vehicle 1 may comprise a plurality of torque sources, such as a plurality of EDUs. The illustrated vehicle 1 is a battery electric vehicle (BEV). Alternatively, the vehicle 1 is a hybrid electric vehicle (HEV) further comprising an internal combustion engine (not shown) as a further torque source. The EDU 210 comprises a housing arrangement 212. For example, the housing arrangement 212 comprises one or more enclosures containing various components of the EDU 210. If there are multiple enclosures, the enclosures may be connected to each other, for example by mechanical fixings / sealant, to form a module. The components of the EDU 210 comprise: - an electric machine 216 operable as a motor and optionally also as a generator (e.g., for regenerative braking); - a transmission 218 providing a fixed or changeable gear ratio; - power electronics including an inverter 214; and - an inverter controller 301B to control the inverter 214 to regulate torque output of the EDU 210. The electric machine 216 comprises a rotor-stator pair. The electric machine 216 can comprise a Permanent Magnet Synchronous Motor (PMSM), or alternatively an Induction Motor (IM) or Switched Reluctance Motor (SRM) or axial flux motor. The transmission 218 converts a speed of the rotor to a wheel speed of the driven wheel ordriven wheels of the vehicle 1. In other examples, the EDU 210 is a direct drive unit with the transmission 218 omitted. The inverter 214, when controlled by the inverter controller 301B, is configured to control the frequency and magnitude of electrical signals supplied to the EDU 210. The inverter controller 301B regulates the torque of an electric motor primarily by controlling the frequency and magnitude of the electrical signals supplied to the EDU 210. For example, the inverter controller 301B may be configured to adjust the magnitude of the voltage supplied by the inverter 214 to the EDU 210 to control the EDU's torque output. The inverter controller 301B may be configured to adjust the frequency of the electrical signals supplied to the EDU 210, to control the EDU's speed, thus indirectly controlling torque output. The inverter controller 301B may be configured to control a gate control section 318 (shown in FIG. 3) of the inverter 214 to adjust the magnitude and frequency by pulse-width modulation (PWM), which comprises adjusting the duty cycle of transistor gates in the gate control section 318 such as MOSFETs (Metal-Oxide-6 Semiconductor Field-Effect Transistors) or IGBTs (Insulated Gate Bipolar Transistors). The inverter controller 301B transmits gate control signals to the transistor gates, to control the torque output of the EDU 210. The gate control signals are transmitted over a local communication network / bus of the EDU 210 or of the inverter 214, to the gate control section 318. The inverter controller 301B has a low-latency connection to the gate control section 318, and for example may comprise a microcontroller mounted to a same circuit board as the gate control section 318. The EDU 210 further comprises a sensor arrangement including a motor electrical current sensor 220 and a motor position sensor 222 (rotor position sensor). The inverter controller 301B is configured to receive feedback from the sensor arrangement to determine the EDU's rotor position, speed, current, and torque. This feedback is used to adjust the gate control signals, ensuring that it operates at the desired torque level. Not all EDUs will require all the above sensors and signals. The motor electrical current sensor 220 is configured to measure the electrical current flowing through the motor windings of the EDU 210, and output feedback indicating the electrical current. The inverter controller 301B is configured to convert the feedback to a parameter indicative of actual torque currently being produced by the EDU 210, in dependence on a current:torque conversion. Optionally, voltage sensing may be included. Alternatively, direct torque measurement may be used. The motor position sensor 222 can comprise a Hall effect sensor, an electrical resolver, or the like. The motor position sensor 222 is configured to output a parameter indicative of a position of the EDU 210, such as the rotor of the EDU 210. The position parameter can be differentiated to determine the speed of the EDU 210. The motor position sensor 222 may be located either before or after the transmission 218. In examples, the parameter from the motor position sensor 222 may indicate the current angular position of a shaft of the EDU 210, wherein the shaft is mechanically coupled to the rotor. The inverter controller 301B is configured to process the signal to determine the current actual speed of the rotor of the EDU 210. Alternatively, the motor position sensor 222 may calculate and output the current actual speed of the rotor. FIG. 2 also illustrates an anti-lock braking system 228 (ABS) comprising an ABS controller 301A and an actuation means (not shown) to control the friction brakes 226 of the vehicle 1. The actuation means (actuator) can comprise a hydraulic circuit and solenoid valves, for example. The ABS controller 301A receives signals from a set of sensors and outputs one or more control signals in dependence on the signals to control the actuation means. The set of sensors for the ABS controller 301A comprises: - wheel speed sensors 206, for each wheel FL, FR, RL, RR of the vehicle 1; and - at least one inertial measurement unit (IMU) 202. The set of sensors 202, 206 are distributed around the vehicle 1. At least some of the sensors 202, 206 are operably coupled to the ABS 228 over a vehicle communication network 224, such as a Controller Area Network (CAN) bus or Flexray(TM) bus. Furthermore, the ABS 228 may be operably coupled to the EDU 210 7 via the vehicle communication network 224, enabling the ABS controller 301A of the ABS 228 to communicate with the inverter controller 301B of the EDU 210. The vehicle communication network 224 is a whole-vehicle network to which multiple different subsystems are operably connected, such as one or more of: Engine control modules (ECMs); Transmission control modules (TCMs); Anti-lock brake systems (ABS); Airbag control modules (ACMs); Instrument clusters; Powertrain control modules (PCMs); Body control modules (BCMs); Climate control systems; Infotainment systems; Door control modules; Sensor modules (e.g., for temperature, pressure, speed); or Actuator modules (e.g., for controlling motors, solenoids). The ABS controller 301A hosts various functions which depend on different ones of the set of sensors. One function of the ABS controller 301A comprises anti-lock braking. During braking, the ABS controller 301A may obtain wheel speed signals from the wheel speed sensors 206 and may obtain vehicle speed signal dependent on the wheel speed signals. The ABS controller 301A is configured to detect a wheel speed a threshold amount lower than the vehicle speed and / or than the other wheel speeds. In response, the ABS controller 301A is configured to control the friction brake 226 of the wheel to reduce braking effort of the wheel until traction is regained. The ABS controller 301A may further be configured to output a signal to the inverter controller 301B to cause the inverter controller 301B to reduce torque output of the EDU 210. Another function of the ABS controller 301A may comprise vehicle stability control. The ABS controller 301A may receive signals from one or more of the IMUs 202, the wheel speed sensors 206, or a steering angle sensor. The ABS controller 301A may be configured to determine a stability parameter configured to detect understeer and oversteer, in dependence on the signals. The ABS controller 301A may be configured to determine that the understeer or oversteer exceeds a threshold. In response, the ABS controller 301A may be configured to control friction brakes 226 of the vehicle 1 to apply differential friction braking to reduce the detected understeer or oversteer of the vehicle 1. The ABS controller 301A may further be configured to output a signal to the inverter controller 301B to cause the inverter controller 301B to decrease or perhaps increase torque output of the EDU 210 to reduce the detected understeer or oversteer. A further function comprises traction control. During acceleration, the ABS controller 301A may obtain wheel speed signals from the wheel speed sensors 206 and a vehicle speed signal. The vehicle speed signal is dependent on one or more of the wheel speed signals, accelerations from the IMU 202, ora satellite positioning signal (e.g., Global Positioning System, GPS). The ABS controller 301A may be configured to detect a traction loss event and activate a traction control function. Detecting a traction loss event can comprise detecting a wheel speed which is a threshold amount greater than the vehicle speed, for example. Optionally, activation of the traction control function may further depend on IMU-based signals for example indicative of vehicle yaw. When active, the traction control function causes the ABS controller 301A to control the friction brake 226 of the wheel to apply braking until traction of the wheel is regained. However, friction brakes 226 are too slow to arrest wheel acceleration during the initial stages of a traction loss event. This can lead to an initial ‘flare’ of 8 wheel speed peaking at tens of kilometres per hour greater than the vehicle speed for hundreds of milliseconds, which is subsequently reduced to the vehicle speed by the friction brakes 226. The EDU 210 by contrast has a very fast response time, with dynamics approximately ten times faster than friction brakes 226. The EDU 210 also offers a fast torque rise rate of typically hundreds of newton metres per second. Therefore, according to aspects of the present invention, the EDU 210 is controlled in addition to, or even instead of, the friction brakes 226. The fast response of the EDU 210 can be utilised to minimise the wheel flare. The ABS controller 301A is therefore configured to output a signal to the inverter controller 301B to cause the inverter controller 301B to decrease a speed of the EDU 210 to reduce the speed of the wheel until traction of that wheel is regained. If the traction control function implements simultaneous friction braking and EDU-based speed reduction of the wheel, the ABS controller 301A may further comprise an arbitrator to control braking strength of the friction brake 226 relative to a speed reduction of the EDU 210. With reference to FIG. 3, there is illustrated a control system 300 for a vehicle 1. The control system 300 comprises one or more controllers 301. The illustrated control system 300 comprises the ABS controller 301A in the ABS 228 and the inverter controller 301B in the inverter 214 of the EDU 210. The ABS controller 301A and inverter controller 301B are communicably connected to each other via the vehicle communication network 224. The control system 300 is configured to receive data from the sensors 202, 206, 220, 222 and determine a motor torque demand in dependence on the data. The control system 300 may then output a gate control signal to control the gate control section 318 of the inverter 214 to cause the electric machine 216 of the EDU 21 Oto generate the torque demanded by the motor torque demand. The control system 300 as illustrated in FIG. 3 comprises two controllers 301 A, 301B, although it will be appreciated that this is merely illustrative. The ABS controller 301A comprises processing means 304A and memory means 306A. The processing means 304A may be one or more electronic processing device 304A which operably execute computer-readable instructions 308A. The memory means 306A may be one or more memory device 306A. The memory means 306A is electrically coupled to the processing means 304A. The memory means 306A is configured to store instructions 308A, and the processing means 304A is configured to access the memory means 306A and execute the instructions 308A stored thereon. The ABS controller 301A comprises an input means 310A and an output means 312A. The input means 310A may comprise an electrical input 310A of the ABS controller 301A. The output means 312A may comprise an electrical output 312A of the ABS controller 301A. The ABS controller 301A may have an interface 302A comprising an electrical input / output I / O 310A, 312A, or an electrical input 310A, or an electrical output 312A, for receiving information and interacting with external components. The input 310A is arranged to receive: wheel speed signals from the wheel speed sensors 206, and optionally a yaw signal from the IMUs 202. The wheel speed signals are electrical signal which indicate individual wheel speeds, and which collectively enable 9 the ABS controller 301A or an upstream controller to determine a vehicle speed signal indicative of a speed of the vehicle 1. Alternatively, or additionally, the vehicle speed may be dependent on accelerations or satellite position coordinates. The yaw signal is an electrical signal indicative of a yaw rate or other yaw parameter of the vehicle 1. In response to activation of the traction control function, the output 312A of the ABS controller 301A is arranged to output a motor speed target signal (motor speed demand), indicative of a target or setpoint speed of the electric machine 216 of the EDU 210 for controlling a wheel speed of a set of one or more driven wheels of the vehicle 1. The motor speed target signal may be referenced to a target axle speed or a target motor speed. More broadly, the motor speed target signal can be regarded as a traction control target signal. The traction control target signal is transmitted to the vehicle communication network 224. The output 312A may also be arranged to output traction-control friction-braking signals to control an amount of brake pressure of the friction brakes 226 of each wheel of the set. The inverter controller 301B comprises processing means 304B and memory means 306B. The processing means 304B may be one or more electronic processing device 304B which operably execute computer-readable instructions 308B. The memory means 306B may be one or more memory device 306B. The memory means 306B is electrically coupled to the processing means 304B. The memory means 306B is configured to store instructions 308B, and the processing means 304B is configured to access the memory means 306B and execute the instructions 308B stored thereon. The inverter controller 301B comprises an input means 310B and an output means 312B. The input means 310B may comprise an electrical input 310B of the inverter controller 301B. The output means 312B may comprise an electrical output 312B of the inverter controller 301B. The inverter controller 301B may have an interface 302B comprising an electrical input / output I / O 310B, 312B, or an electrical input 310B, or an electrical output 312B, for receiving information and interacting with external components. When the traction control function is active in the ABS controller 301 A, the input 310B is arranged to obtain the traction control target signal from the vehicle communication network 224, and sensed feedback from sensors 220, 222, indicating one or more state parameters of the traction electric machine 216. The sensed feedback comprises a state parameter indicative of a torque of the EDU 210 from the motor electrical current sensor 220, and a state parameter indicative of a speed of the EDU 210 from the motor position sensor 222. The parameters indicate the current torque (actual current torque) produced by the EDU 210, and the current speed (actual current speed) of the EDU 210. The output 312B is arranged to output a gate control signal indicative of a motor torque demand. The gate control signal is transmitted to the gate control section 318 of the inverter 214 via a local communication bus or network of the EDU 210, causing the gate control section 318 of the inverter 214 to control an amount of torque generated by the electric machine 216 of the EDU 210. FIG. 4 illustrates a non-transitory computer-readable storage medium 400 comprising the instructions 308 (computer software). FIGS. 5A and 5B each illustrate a control block diagram comprising a plurality of process controllers 502, 506, 510 of a traction control function 500. The process controllers 502, 506, 510 are hosted by the ABS controller 301A and by the inverter controller 301B. The process controllers 502, 506, 510 are collectively arranged in an outer control loop 522 and inner control loops 524, 526. The control system is configured to execute the process controllers 502, 506, 510. The process controllers 502, 506, 510 comprise a target speed generator 502, a torque restriction generator 506, and a torque demand generator 510. Additional or intervening process controllers may be included but are not shown. The outer control loop 522 is a longitudinal vehicle dynamics control loop providing sensed feedback 520 indicative of current (actual) longitudinal vehicle dynamics. One of the inner control loops 524 is a motor speed control loop 524 providing sensed feedback 516 indicative of current (actual) speed of the EDU 210. The innermost control loop 526 is a motor torque control loop providing sensed feedback 517 indicative of current (actual) torque of the EDU 210. The target speed generator 502 is a closed loop process controller within the longitudinal vehicle dynamics control loop 522 and outside the motor speed control loop 524. The torque restriction generator 506 is a closed loop process controller within the motor speed control loop 524 and outside the motor torque control loop 526. The torque demand generator 510 is a closed loop process controller within the motor torque control loop 526. The target speed generator 502 is configured to obtain the wheel speed signals, the vehicle speed signal, and optionally the yaw signal, as feedback signals 520 of the longitudinal vehicle dynamics control loop 522, and determine a motor speed target 504 in dependence on the signals 520. In examples, the signals 520 indicate the current (actual) wheel speed of the or each driven wheel driven by the EDU 210, and the vehicle speed and yaw rate. The target speed generator 502 is configured to determine and output the motor speed target 504 to the torque restriction generator 506. When the traction control function 500 is active, the target speed generator 502 is configured (e.g., calibrated) to determine a value of the motor speed target 504 that provides a desired / target level of wheel slip at the wheel-road interface. Generally, the value will be lower than the current speed of the EDU 210. The target level of wheel slip may depend on the ‘sportiness’ of the vehicle, and for example may depend on which one of a plurality of driving modes has been selected via a user interface. The motor speed target 504 allows the slipping driven wheel or wheels to decelerate until traction is regained and / or the target level of wheel slip is reached. The motor speed target 504 is just one example of a traction control target of the traction control function 500. Another example would be a motor torque target output by a target torque generator. The torque restriction generator 506 is configured to obtain the motor speed target 504 from the target speed generator 502. The torque restriction generator 506 is also configured to obtain sensed feedback 516 from the 11 motor speed control loop 524, which is indicative of a state parameter of the EDU 210, wherein the state parameter indicates the current position and / or speed of the EDU 210. For example, the sensed feedback 516 may comprise a signal from the motor position sensor 222, which is indicative of the current detected motor speed and / or current detected position of the rotor of the electric machine 216 of the EDU 210. The torque restriction generator 506 may differentiate the current position of the rotor to determine the current speed of the EDU 210. The torque restriction generator 506 is configured to determine and output a torque restriction 508 in dependence on the motor speed target 504 and the sensed feedback 516. In examples, the torque restriction 508 is in the form of a motor torque limit. The motor torque limit 508 restricts the maximum allowable torque which the EDU 210 can output. When the traction control function 500 is active, the target speed generator 502 and the torque restriction generator 506 are configured (e.g., calibrated) to set the motor torque limit 508 to less than the current torque of the EDU 210. Therefore, the speed of the spinning wheel will decrease until traction is regained. The specific control scheme used by the torque restriction generator 506 is described later in relation to FIG. 6. The torque demand generator 510 is configured to obtain the motor torque limit 508 from the torque restriction generator 506. The torque demand generator 510 is also configured to obtain a variable torque request from a higher-level controller (not shown) than the controller which is executing the torque demand generator 510. The torque demand generator 510 is configured to apply the motor torque limit 508 to the variable torque request. The variable torque request is dependent on either accelerator pedal depression or an automated longitudinal control algorithm, depending on whether longitudinal vehicle control is under manual control or automatic control. The torque demand generator 510 is also configured to obtain sensed feedback 517 from the motor torque control loop 526, which is indicative of a state parameter of the EDU 210, wherein the state parameter indicates the current torque of the EDU 210. For example, the sensed feedback 517 may comprise a signal from the motor electrical current sensor 220, which is indicative of the current torque of the EDU 210. The torque demand generator 510 is configured to determine and output a control signal 512 to control output torque of the EDU 210 in dependence on the torque restriction. The control signal may be in the form of a motor torque demand 512. The motor torque demand 512 is dependent on the motor torque limit 508, the variable torque request, and the sensed feedback 517. The motor torque demand 512 may be a gate control signal which is outputtothe gate control section 318 of the inverter214 of the EDU 210. In response, the duty cycle of one or more transistor gates of the gate control section 318 of the inverter 214 changes to modify the torque output of the electric machine 216 of the EDU 210 to track the motor torque demand 512. An example of determining the motor torque demand 512 comprises first determining a torque error between the variable torque request from a higher-level controller and the current torque of the EDU 210 from the 12 sensed feedback, and determining a torque offset in dependence on the torque error. The torque offset has a magnitude dependent on the magnitude of the error, and a sign (positive or negative) dependent on the sign of the torque error. After summing the variable torque request and the torque offset, the torque demand generator 510 compares the summation with the motor torque limit 508. If the motor torque limit 508 is lower than the summation, the motor torque demand 512 may be dependent on (e.g., equal to / capped at) the motor torque limit 508. If the motor torque limit 508 is not lower than the variable torque request, the motor torque demand 512 may be allowed to vary in dependence on the variable torque request and the torque offset without being dependent on the motor torque limit 508. The dashed box 514 represents motor dynamics which change in response to the motor torque demand 512, for illustrative purposes. The motor dynamics comprise the current speed and current torque of the electric machine 216, for example. Therefore, the electrical current sensor 220 and motor position sensor 222 are shown within the motor dynamics box 514. These sensors 220, 222 are respectively connected to the motor torque control loop 526 and motor speed control loop 524 which provide their signals to the torque demand generator 510 and torque restriction generator 506, respectively. The next dashed box 518 represents vehicle longitudinal dynamics which change in response to the change in motor dynamics (box 514). The vehicle longitudinal dynamics comprise the wheel speeds, vehicle speed, and yaw rate, for example. Therefore, the IMUs 202 and wheel speed sensors 206 are shown within the vehicle longitudinal dynamics box 518. These sensors 202, 206 are connected to the longitudinal vehicle dynamics control loop 522 which provides their signals to the target speed generator 502. In FIG. 5A, the ABS controller 301A hosts the target speed generator 502, and the inverter controller 301B hosts the torque restriction generator 506 and torque demand generator 510. FIG. 5B illustrates a variant in which the inverter controller 301B hosts the target speed generator 502, the torque restriction generator 506, and the torque demand generator 510. In a further variant (not shown), the ABS controller 301A hosts the target speed generator 502 and the torque restriction generator 506, and the inverter controller 301B hosts the torque demand generator 510. The topology of FIG. 5A is advantageous because the motor speed control loop 524 is executed within the EDU 210, which has a fast local communication bus / network, therefore the motor speed control loop 524 can be executed at a significantly faster frequency than the vehicle communication network 224 which has higher latency. The transport delay expected over the vehicle communication network 224 may be greater than 100 milliseconds. For example, the control system may be configured to execute the motor speed control loop 524 at greater than twice the frequency of the longitudinal vehicle dynamics control loop 522. This enables improved tracking accuracy. In both FIGS. 5A and 5B, the target speed generator 502 of the traction control function 500 is hosted by the ABS controller 301 A. One advantage is that the ABS controller 301A may optionally be configured to arbitrate 13 between the EDU 210 and the friction brakes 226 as part of the traction control function 500. Another advantage is that the ABS controller 301A may optionally be configured to coordinate the traction control function 500 with other ABS functions such as vehicle stability control. In another embodiment shown in FIG. 5B, the target speed generator 502 may be hosted by the inverter controller 301B. In both FIGS. 5A and 5B, the torque demand generator 510 is hosted by the inverter controller 301B. An advantage is that the motor torque control loop 526 can be executed at a high frequency with low latency, to minimise torque error. Although FIGS. 5A and 5B show three process controllers, it would be possible to implement aspects of the invention with fewer process controllers. For example, two or more of the target speed generator 502, torque restriction generator 506, and torque demand generator 510, may be merged to a single process controller. The specific control scheme of the torque restriction generator 506 is now described with reference to FIG. 6. The torque restriction generator 506 executes an optimal control scheme 600. The optimal control scheme 600 may optimise the objective function offline. Alternatively, some optimal control schemes optimise the function online, such as Model Predictive Control (MPC) or Generalized Predictive Control (GPC). The optimal control scheme 600 depicted in FIG. 6 is an LQR control scheme (LQR process controller). FIG. 6 is just one example of an LQR control scheme 600. In other embodiments, another optimal control scheme may be used such as MPC or GPC. An advantage of an LQR control scheme 600 is that less optimisation is required during runtime than MPC, and therefore is less computationally intensive. Input terminal block 602 comprises a Speed Target, ‘s’. In examples, the Speed Target is the motor speed target 504 obtained by the torque restriction generator 506 from the target speed generator 502. Block 606 receives the Speed Target from terminal block 602 and integrates (1 / s) the speed target to determine a Rotor Position Target. The LQR control scheme 600 comprises two stages 608, 620 dependent on the state of the process. Input terminal block 604 is connected to the first stage 608, and comprises a State Vector ‘x’. The State Vector comprises Motor Speed (feedback 516, FIGS. 5A-5B), Motor Position (from sensor 222), and Motor Actual Torque (from sensor 222: the parameter indicative of actual torque currently being produced by the EDU 210). The Motor Speed and Motor Position may be state parameters indicated by the sensed feedback of the motor speed control loop 524, and in dependence on information sensed by the motor position sensor 222. In other implementations, a MotorTorque Error Integral may be provided instead of Motor Actual Torque, by integrating the torque error. At the first stage 608, the State Vector is split out into its three component signals. The first stage 608 is connected to the second stage 620 by three signal paths 610, 612, 614. 14 The first signal path 610 comprises a summation block 616 which calculates a Speed Error, in dependence on the Motor Speed and Speed Target. The second signal path 612 comprises a summation block 618 which calculates a Position Error, in dependence on Motor Position / speed and the Rotor Position Target. The third signal path 614 passes the Motor Actual Torque directly to the second stage 620 without any operations. At the second stage 620, the Speed Error, Position Error, and Motor Actual Torque, are reconfigured into a vector signal. The second stage 620 is connected to the third stage 626 by a signal path 622 providing the vector signal. The signal path 622 comprises a matrix gain stage 624 which multiplies the vector signal ‘u’ by a matrix gain K according to the relationship u=-Kx. If an LQR control scheme is used, the gain matrix K may be determined offline through the solution of the Riccati equation, which comprises minimising a quadratic cost function, and then using this gain matrix to compute the control input in dependence on the current State Vector ‘x’ of the system. Alternatively, in MPC or GPC, the matrix gain K could be calculated online. The cost function for LQR comprises a quadratic objective function which comprises the general form: J = f (xtQx + uTRu)dt Jto where ‘x’ is the State Vector, u is the control input vector, Q is the tracking accuracy weight matrix (state cost matrix), and R is the energy consumption weight matrix (control cost matrix). The exact numerical weights are not specified in this disclosure. Therefore, the Motor Speed (x1), Motor Position (x2), and Motor Actual Torque (x3) of the State Vector ‘x’ are tracking accuracy parameters. Online, these are multiplied by the matrix gain K which was calculated offline based on weights matrices Q and R. The matrix gain K may be based on variable tracking accuracy weights: Motor Speed Weight (x1costA2), Motor Position Weight (x2costA2), and Motor Actual Torque Weight (x3costA2). Therefore, there are increasing costs when deviating from the Motor Speed, Position, and Motor Actual Torque. The rate of cost increase depends on the relative weights of the tracking accuracy parameters during calibration. In examples, the cost function of the LQR process controller is biased towards tracking accuracy over energy consumption. In other words, the tracking accuracy weights may impose a cost penalty on the State Vector at a greater rate than the energy consumption weight matrix imposes on the energy consumption parameter. The additional energy consumption is not a disadvantage because a traction loss event typically has a short duration, so the overall energy cost is low. When the torque restriction generator 506 implementing the LQR control scheme 600 is hosted in the inverter controller 301B, this provides the opportunity to run at a fast frequency with minimal delay, to improve control stability. Therefore, a frequency of the motor speed control loop 524 may be configured to provide a response delay of less than five milliseconds. This would not be possible if the torque restriction generator 506 was implemented outside the EDU 210, because the transport delay expected over the vehicle communication network 224 may be greater than 100 milliseconds. FIG. 7 illustrates a method 700 according to an embodiment of the invention. The method 700 is a method of controlling an EDU 210 of a vehicle 1, such as the vehicle 1 illustrated in FIG. 7. In particular, the method 700 is a method of controlling the EDU 210 of the vehicle 1 to implement a traction control function 500. The method 700 may be performed by the system or control system illustrated in FIG. 3. In particular, the memory 306A may comprise computer-readable instructions 308 which, when executed by the processor 304A, perform the method 700. During execution, the traction control function 500 may be active. In the example of FIG. 7, blocks 702, 704, 706 are executed by the ABS controller 301A of the ABS 228, and blocks 708, 710, 712, 714 are executed by the inverter controller 301B of the EDU 210. As described earlier, some operations may be hosted by different controllers, and some or all of the operations may be executed by just one of the controllers. Blocks 702, 704, 706 may be executed by the target speed generator 502 hosted in the ABS controller 301 A. Block 702 comprises obtaining wheel speed signals from the wheel speed sensors 206 and obtaining a vehicle speed signal as described above. The yaw signal may also be received. The signals may be received via the longitudinal vehicle dynamics control loop 522. Although not shown, the ABS controller 301A may check in dependence on the signals whether to deactivate the traction control function 500, before progressing further. Block 704 comprises determining the traction control target such as the motor speed target 504, in dependence on the signals obtained at block 702. Block 706 comprises outputting the traction control target to the vehicle communication network 224, as schematically represented by the dashed vertical line separating the controllers 301 A, 301B. Blocks 708, 710, 712, 714 may be executed by the torque restriction generator 506 and torque demand generator 510, hosted in the inverter controller 301B of the EDU 210. Block 708 comprises obtaining the traction control target 504 for controlling wheel speed from the vehicle communication network 224, that was output by block 706. Block 710 comprises obtaining sensed feedback 516 indicating one or more state parameters of the electric machine 216 of the EDU 210, such as the current speed of the EDU 210 calculated in dependence on the current position of the EDU 210 sensed by the motor position sensor 222. Block 712 comprises determining a torque restriction, such as the motor torque limit 508, in dependence on the traction control target and the sensed feedback. The determination is dependent on an optimal control scheme such as the LQR control scheme 600 described above. Block 712 may further comprise determining the motor torque demand 512 in dependence on the torque restriction, the variable torque request described above, and sensed feedback indicative of torque error as described above. Block 714 comprises outputting a control signal, such as the motor torque demand 512, to control output torque of the electric machine 216 of the EDU 210 in dependence on the torque restriction. For example, the motor torque demand 512 may be in the form of a gate control signal sent to the gate control section 318 of the inverter 214, in dependence on the motor torque demand 512. The gate control signal may be sent to the gate control section 318 over the local communication network / bus of the inverter 214 of the EDU 210. FIGS. 2-3 illustrate a system 30 for implementing at least blocks 708, 710, 712, 714 of the method 700. The system 30 comprises the EDU 210 and the control system 300, wherein the EDU 210 comprises a traction electric machine 216, an inverter 214, and a first controller 301B of the control system 300, wherein the first controller 301B hosts the optimal control scheme 600, and wherein the first controller is configured to obtain the traction control target (block 708) and the sensed feedback (block 710), determine the torque restriction (block 712), and output the control signal (block 714) to the inverter 214. Optionally, the system 30 further comprises an ABS 228, wherein the ABS 228 comprises a second controller 301A of the control system 300, and wherein the second controller 301A is configured to determine the traction control target (block 704), and send the traction control target (block 706) to the first controller 301A over a vehicle communication network 224. FIGS. 8A and 8B illustrate actual wheel speed (y-axis) overtime (x-axis) of a driven wheel of a vehicle 1 which is accelerating on a low-traction surface. In FIG. 8A, the torque restriction generator 506 implements an LQR control scheme 600, and is hosted in the inverter controller 301B as shown in FIG. 5A, to minimise latency. In FIG. 8B, the torque restriction generator 506 implements a PID (proportional-integral-derivative) control scheme instead. As shown, the PID control scheme remains unstable throughout the period of acceleration, with poor tracking accuracy compared to the LQR control scheme 600. The LQR control scheme 600 offers excellent tracking accuracy without the need for significant calibration effort. Although the instability of the PID control scheme could eventually have been resolved, the calibration effort required for an accurate PID control scheme can be prohibitive because a very large number of weights would be needed to tune the PID control scheme for different surfaces. For example, traction may vary under each wheel, which can affect dynamics. Different surface types can affect dynamics. Road wetness also affects dynamics. These scenarios would have required additional desktop and test track tuning. The LQR approach is robust for traction loss events fora wide range of surface types due to its ability to handle a plurality of state inputs and is naturally stable. This can greatly improve vehicle acceleration on complex surfaces. It is to be understood that the or each controller 301 can comprise a control unit or computational device having one or more electronic processors (e.g., a microprocessor, a microcontroller, an application specific integrated circuit (ASIC), etc.), and may comprise a single control unit or computational device, or alternatively different functions of the or each controller 301 maybe embodied in, or hosted in, different control units or computational devices. As used herein, the term “controller,” “control unit,” or “computational device” will be understood to include a single controller, control unit, or computational device, and a plurality of controllers, control units, or computational devices collectively operating to provide the required control functionality. A set of instructions could be provided which, when executed, cause the controller 301 to implement the control techniques described herein (including some or all of the functionality required for the method(s) described herein). The set of instructions 308 could be embedded in said one or more electronic processors 304A of the controller 301; or alternatively, the set of instructions 308 could be provided as software to be executed in the controller 301. A first controller or control unit may be implemented in software run on one or more processors. One or more other controllers or control units may be implemented in software run on one or more processors, optionally the same one or more processors as the first controller or control unit. Other arrangements are also useful. The, or each, electronic processor 304A may comprise any suitable electronic processor (e.g., a microprocessor, a microcontroller, an ASIC, etc.) that is configured to execute electronic instructions 308. The, or each, electronic memory device 306A may comprise any suitable memory device and may store a variety of data, information, threshold value(s), lookup tables or other data structures, and / or instructions therein or thereon. In an embodiment, the memory device 306A has information and instructions for software, firmware, programs, algorithms, scripts, applications, etc. stored therein or thereon that may govern all or part of the methodology described herein. The processor, or each, electronic processor 304A may access the memory device 306A and execute and / or use that or those instructions and information to carry out or perform some or all of the functionality and methodology described herein. The at least one memory device 306A may comprise a computer-readable storage medium (e.g. a non-transitory or non-transient storage medium) that may comprise any mechanism for storing information in a form readable by a machine or electronic processors / computational devices. Examples of the form include, without limitation: a magnetic storage medium (e.g. floppy diskette); optical storage medium (e.g. CD-ROM); magneto optical storage medium; read only memory (ROM); random access memory (RAM); erasable programmable memory (e.g. EPROM ad EEPROM); flash memory; or electrical or other types of medium for storing such information / instructions. It will be appreciated that embodiments of the present invention can be realised in any suitable form of hardware, software or a combination of hardware and software. For example, it is contemplated that the present invention is not limited to being implemented by way of programmable processing devices, and that at least some of, and in some embodiments all of, the functionality and or method steps of the present invention may equally be implemented by way of non-programmable hardware, such as by way of non-programmable ASIC, Boolean logic circuitry, etc. It will be appreciated that various changes and modifications can be made to the present invention without departing from the scope of the present application. 5 The blocks illustrated in the FIGS. 5A, 5B, 6, 7 may represent steps in a method and / or sections of code in the computer program 308. The illustration of a particular order to the blocks does not necessarily imply that there is a required or preferred order for the blocks and the order and arrangement of the block may be varied. Furthermore, it may be possible for some steps to be omitted. 10 Features described in the preceding description may be used in combinations other than the combinations explicitly described. Although functions have been described with reference to certain features, those functions may be performable by other features whether described or not. Although features have been described with reference to certain embodiments, those features may also be present in other embodiments whether described or not. 15

Claims

1. A control system for controlling a traction electric machine of a vehicle, the control system comprising one or more processors collectively configured to:obtain a traction control target for controlling wheel speed;obtain sensed feedback indicating one or more state parameters of the traction electric machine;determine a torque restriction in dependence on the traction control target and the sensed feedback, using an optimal control scheme; andoutput a control signal to control output torque of the traction electric machine in dependence on the torque restriction.

2. The control system of claim 1, wherein the optimal control scheme is hosted by an inverter controller of the control system, and wherein the control signal is a gate control signal to control a gate control section of an inverter for the traction electric machine.

3. The control system of claim 1 or 2, wherein the traction control target is obtained from an anti-lock braking system controller.

4. The control system of claim 1,2 or 3, wherein the optimal control scheme comprises a linear quadratic regulator control scheme.

5. The control system of any preceding claim, wherein the traction control target comprises a motor speed target of the traction electric machine.

6. The control system of any preceding claim, wherein the one or more state parameters indicated by the sensed feedback indicate at least one of: a speed of the traction electric machine; a position of the traction electric machine; or torque generated by the traction electric machine.

7. The control system of any preceding claim, wherein the optimal control scheme is configured to optimise one or more tracking accuracy parameters and one or more energy consumption parameters.

8. The control system of claim 7, wherein the one or more tracking accuracy parameters comprise at least one of:a parameter indicative of speed of the traction electric machine;a parameter indicative of rotor position of the traction electric machine; ora parameter dependent on torque of the traction electric machine.

9. The control system of claim 7 or 8, wherein the one or more energy consumption parameters comprise a parameter indicative of torque of the traction electric machine.

10. The control system of claim 7, 8 or 9, wherein the optimal control scheme is biased towards tracking accuracy over energy consumption.

11. A system comprising an electric drive unit, and the control system of any one of the preceding claims, wherein the electric drive unit comprises the traction electric machine, an inverter, and a first controller of the control system which hosts the optimal control scheme, and wherein the first controller is configured to obtain the traction control target and the sensed feedback, determine the torque restriction, and output the control signal to the inverter.

12. The system of claim 11, further comprising an anti-lock braking system, wherein the anti-lock braking system comprises a second controller of the control system, and wherein the second controller is configured to determine the traction control target, and send the traction control target to the first controller over a vehicle communication network.13, A vehicle comprising the control system of any one of claims 1 to 10, or the system of claim 11 or 12.

14. A method of controlling a traction electric machine of a vehicle, the method comprising: obtaining a traction control target for controlling wheel speed;obtaining sensed feedback indicating one or more state parameters of the traction electric machine; determining a torque restriction in dependence on the traction control target and the sensed feedback, using an optimal control scheme; andoutputting a control signal to control output torque of the traction electric machine in dependence on the torque restriction.

15. Computer readable instructions which, when executed by a computer, are arranged to perform a method according to claim 14.

Images