Patent Application
20230121333
The present invention relates to a smart brake system, and a corresponding method for adjusting a brake clamping force to be applied to brake pads of a vehicle brake.
When a vehicle is braking, brake pads clamp or rub against a brake rotor (e.g. a brake disc or drum) thus slowing down the vehicle. However, as a result of heat dissipation processes during braking, the temperature of the rotor rises causing the rotor to expand. As the amount of clamping force exerted by the brake pads can vary depending on the amount of rotor expansion, for at least this reason there is a relation between brake clamping force needed to slow down or stop the vehicle and rotor temperature.
When the vehicle is stationary during parking or idling, a parking brake is typically engaged with the brake clamping force being automatically set by an on-board computer. The same is true for autonomous vehicles that need to decelerate or stop automatically without driver input. In these cases, the computer estimates the amount of brake clamping force needed to keep the vehicle from moving without applying a clamping force that is in excess of what is needed. If the rotor temperature is high when the brake is engaged, a higher clamping force may be required to compensate for subsequent cooling and contraction of the rotor. Additionally, any slope of the ground, mass of the vehicle, or other factors, such as environmental conditions, may need to be taken into account when determining the clamping force.
Some vehicles perform real-time estimation of the rotor temperature in order to determine a precise brake clamping force. Thus, one option is to install physical temperature sensors in the vehicle on or near the rotors or brake pads. However, for mass production units, the cost of installing such sensors is high. Alternatively, some vehicles estimate the rotor temperature using a virtual temperature sensor. Such a sensor comprises a computer implemented thermal model which calculates the heat energy loss and gain of the rotor when the vehicle is on a journey. For example, US 2017/080909 proposes a vehicle control apparatus which compensates an applied clamping force according to an estimated rotor temperature. In US 2017/080909 the rotor temperature is estimated by calculating the heat energy lost or gained during braking. This calculation entails complex thermal modelling of the brake systems and the heat transfer characteristics of the rotors. For example, the heat energy calculations of US 2017/080909 require parameters including the dimensions and weight of the brake apparatus, a thermal coefficient of the rotor material, and a friction coefficient of the brake friction material.
Similarly, US 2003/0081650 proposes an apparatus for estimating a vehicle brake rotor temperature. In US 2003/0081650 the rotor temperature is estimated by calculating the heat energy loss or gain of the rotor using analytical equations. The heat energy loss is added to a previously estimated rotor temperature to estimate a new rotor temperature. The new rotor temperature is then used in subsequent temperature estimations and this calculation is repeated throughout a journey.
Systems such as those proposed in US 2017/080909 and US 2003/0081650 require detailed knowledge of the physical and thermal properties of the rotors and brake systems on which to base their complex thermal models. Moreover, errors in the estimated rotor temperature can cumulatively add up throughout a journey adversely affecting the reliability and safety assurance of the vehicle brake system.
JP660262B2 proposes a brake control system that estimates a vehicle brake rotor temperature by calculating the heat energy loss or gain of the rotor using analytical equations. The heat energy gain/loss is added to a previously estimated rotor temperature to estimate a new rotor temperature.
Therefore, a smart brake system is required which can more reliably estimate vehicle brake temperature from more readily available data.
Accordingly, the present invention provides a smart brake system for adjusting a brake clamping force to be applied to brake pads of a vehicle brake in which the brake pads rub against a rotor to slow the vehicle, the system comprising:
Advantageously, the rotor temperature can be estimated using data that is readily available in most vehicles. The temperature prediction model does not require complex thermal modelling of the rotor or surrounding systems. Moreover, the inclusion of elapsed time since a previous brake event and the duration of a previous brake event in the temperature prediction model allows appropriate weighting to be given to the heating effects of the previous brake event depending on how recently it happened. As a result, the smart brake system is able to more accurately estimate the current rotor temperature and perform finer adjustments to the applied brake clamping force.
The temperature prediction model may be a machine learning model which is trained using historical and/or simulated vehicle operation data and brake event data. For instance, the memory device may store a file of pretrained mathematical weights which can be applied to a function for predicting temperature. The model may be a regression model trained using K-nearest neighbours or a similar technique. The machine learning model may be a neural network such as an MLP (Multi Layer Perceptron). By using machine learning to develop the temperature prediction model, many different input parameters can easily be included which can increase the accuracy of the model. Moreover, such a system is able to estimate an absolute rotor temperature value as opposed to simply estimating a heat energy gain or loss. This reduces or avoids the production of cumulative errors, which can be a problem in conventional thermal models applied to brake systems.
The memory device may also store vehicle specification data, and the controller may be further configured to estimate the current temperature of the rotor from the vehicle specification data using the temperature prediction model; the vehicle specification data including one or more selected from the list comprising: dimensions of the rotor, mass of the rotor, brake pad contact surface area of the rotor, specific heat capacity of the rotor, dimensions of braked wheels of the vehicle, and vehicle mass. The use of such vehicle specification data can improve the accuracy of the current temperature estimation.
The memory device may also store a previous estimated temperature of the rotor which is the most recently estimated temperature of the rotor. The controller may be further configured to estimate the current temperature of the rotor from the previous estimated temperature using the temperature prediction model. Using the most recently estimated temperature in this way can also improve the accuracy of the current temperature estimation.
The vehicle operation data may also include one or more selected from the list comprising: current vehicle acceleration, current travel distance of a brake pedal, an estimate of kinetic energy of the rotor converted to heat during braking, the product of brake clamping force and vehicle speed, and the product of brake clamping force, vehicle speed and duration of the current brake event.
The estimate of the kinetic energy of the rotor converted to heat during braking (Disc_heating_KE_1) may be calculated as:
Disc_heating_KE_1=½mcv12−½mcv22
where mc is the mass of the vehicle, v1 is the vehicle speed at braking start, and v2 is the current vehicle speed during braking. Alternatively, the estimate of the kinetic energy of the rotor converted to heat during braking (Disc_heating_KE_2) may be calculated as:
Disc_heating_KE_2=(pv2fr)/(mdcd)
where p is brake clamping force, v2 is the current vehicle speed during braking, fr is brake pad contact surface area of the rotor, md is the mass of the brake disc, and cd is the specific heat capacity of the rotor. Indeed, the vehicle operation data may include two or more different estimates of the kinetic energy of the rotor converted to heat during braking (e.g. Disc_heating_KE_1 and Disc_heating_KE_2).
The data about the current brake event may also include one or both of: the vehicle speed at the beginning of the current brake event and the vehicle speed at the end of the current brake event.
The data about the previous brake event may also include the vehicle speed at the start of the previous brake event and the vehicle speed at the end of the previous brake event. These data further help to determine the appropriate weighting to be given to the heating effects of the previous brake event as they relate to how much heat was generated in the previous brake event. The data about the previous brake event may further include the average deceleration the previous brake event and/or the distance travelled by the vehicle since the end of the previous brake event.
The controller may be configured to adjust one or more brake fluid pressures to adjust the brake clamping force. The one or more brake fluid pressures may include a master cylinder pressure which controls a total brake clamping force applied to the brake pads of plural vehicle brakes of the vehicle and/or local brake cylinder pressures which each control a local brake clamping force applied to the brake pads of a respective one of the plural vehicle brakes of the vehicle.
Conveniently, the current brake clamping force of the vehicle operation data may be in the form of or derived from one or more currently-measured brake fluid pressures. As there is typically a direct relationship between brake fluid pressure and brake clamping force, using one or more brake fluid pressures in the temperature prediction model is generally equivalent to using an actual brake clamping force. Brake fluid pressures are typically more convenient to measure by sensors than actual brake clamping forces. Alternatively or additionally, the current brake clamping force may be estimated or measured from one or more other measurements such as a direct force (e.g. load cell) measurement, measurement of distance travelled by, or force applied to, a brake actuator (e.g. brake pedal), and measurement of deceleration of the vehicle.
The smart brake system of the first aspect may further comprise the vehicle sensors for measuring the vehicle operation data.
The controller may be configured: to estimate the current temperature of each of the rotors of plural vehicle brakes of the vehicle, and to adjust the brake clamping force applied to the brake pads of the vehicle brakes to compensate for the estimated current temperatures.
In a second aspect a method is provided of adjusting a brake clamping force to be applied to brake pads of a vehicle brake in which the brake pads rub against a rotor to slow the vehicle, the method comprising:
Thus, the method of the second aspect corresponds to the smart brake system of the first aspect. Accordingly, optional features of the smart brake system of the first aspect discussed above apply also to the method of the second aspect.
In a third aspect, there is provided a computer program comprising code which, when the code is executed on a computer-based controller, causes the controller to perform the method of the second aspect.
In a fourth aspect, there is provided a computer-readable data carrier storing thereon the computer program of the third aspect.
In a fifth aspect, there is provided a vehicle having one or more vehicle brakes and fitted with the smart brake system of claims the first aspect.
The invention includes the combination of the aspects and preferred features described except where such a combination is clearly impermissible or expressly avoided.
Embodiments and experiments illustrating the principles of the invention will now be discussed with reference to the accompanying figures in which:
Aspects and embodiments of the present invention will now be discussed with reference to the accompanying figures. Further aspects and embodiments will be apparent to those skilled in the art. All documents mentioned in this text are incorporated herein by reference.
The smart brake system 100 includes an interface 102 for receiving vehicle operation data from on-board vehicle sensors 101. A memory device 103 is provided for storing data about current 113 and previous 111 brake events. In addition, the memory device stores a temperature prediction model 109. A controller 105 is provided which implements a temperature prediction routine 113 to estimate a rotor temperature of the vehicle brake. The temperature prediction routine inputs the vehicle operation data and incoming data from the memory device as parameters to the temperature prediction model 109 in order to estimate the rotor temperature. The controller then uses a brake clamping force adjustment routine 115 to adjust the clamping force applied by the brake pads to the rotor according to the estimated rotor temperature.
If, as is usual, the vehicle has plural brakes, a similar process may be repeated in parallel to estimate the rotor temperature of multiple brakes on the same vehicle, and to adjust their brake clamping forces.
The controller 103 for the smart brake system 100 may be included as a software function in an ECU (electronic control unit) of the vehicle. Alternatively, the controller for the smart brake system may be separate to any such ECU. In this case, it may receive the vehicle operation data and data about current and previous brake events from the ECU, or some or all of the relevant vehicle sensors may communicate directly with the controller to provide these data.
Typically, the brake system is a hydraulic braking system in which the brakes are applied by increasing or decreasing one or more brake fluid pressures. This in-turn causes the brake clamping force applied by the brake pads to the rotor(s) to increase or decrease. The one or more brake fluid pressures may include a master cylinder pressure which controls a total brake clamping force applied to the brake pads of plural vehicle brakes of the vehicle. Alternatively or additionally, the brake fluid pressures may include local brake cylinder pressures which each control a local brake clamping force applied to the brake pads of a respective vehicle brake. Accordingly, the application of the brake clamping force in the smart brakes system 100 is typically controlled indirectly by controlling the one or more brake fluid pressures. Similarly, the brake clamping force is typically measured indirectly by measuring the one or more brake fluid pressures. However, other approaches may be used to apply and/or measure the brake clamping force. These may be particularly relevant, e.g. in the case of mechanically or electronically actuated brakes, which may not use hydraulics at all.
The temperature prediction model 109 is typically a machine learning model which is trained on previously collected brake event and vehicle operation data. Throughout a journey and when applying a parking brake at the end of a journey, the vehicle operation data and data about a current 113 and previous 111 brake event are continuously updated. The rotor temperature can then be re-estimated by the controller 105 using the temperature prediction model 109 and the updated data at regular intervals. The last estimated temperature of the brake discs can also be used as an input parameter to the temperature prediction model. On start-up or after a long idling time, the last estimated temperature (or initial temperature) can be assumed to be the ambient temperature as measured by the vehicle sensors 101.
If the brake system is being fitted to an existing vehicle, the vehicle sensors 101 may include existing sensors of the vehicle. However, this does not exclude that vehicle sensors can be installed along with the brake system. The vehicle operation data may include data relating to the current journey or operation of the vehicle, and also data relating to conditions external to the vehicle. These data include current ambient temperature, current brake clamping force and current vehicle speed. For example, they may include, or be in the form of, any one or more of: current wheel speed, current data relating to vehicle speed, current acceleration, brake fluid pressure, current ambient temperature, other information about the weather conditions, the slope of the ground that the vehicle is parked or driving on, and any other external factors which may affect the temperature of the brake rotors. The vehicle operation data may also include the status of driver controls such as travel distance of a brake pedal or whether a parking brake has been applied.
The data stored in the memory device 103 about a current brake event 113 contain information relating to a current application of the brakes. A brake event may involve the application of the vehicle brakes by the driver during a journey or the application of a parking brake at the end of a journey. The data about a current brake event include a current braking time (or sliding time). This parameter is the elapsed time since the brake was first activated at the beginning of the current brake event. The current braking time may be deduced from changes in acceleration, wheel speed or brake fluid pressure, or driver controls may be monitored to signal that a brake event has started. The data about a current brake event may also include one or more of: estimated rotor temperature at the beginning of the brake event, vehicle distance travelled since the current brake event started, vehicle speed at the beginning of the brake event, and any other information about a current brake event which may be available from the vehicle sensors.
Similarly, the memory device 103 stores data about a previous brake event 111. These may include some or all of the data recorded for a current brake event, which are then retained in the memory device when that brake event finishes. The previous brake event is typically the most recent brake event where the vehicle brakes were applied and then released. However, these data may include data from multiple previous brake events, and/or a last significant brake event where the brakes were applied for a minimum duration or with a minimum force. The data about a previous brake event include: elapsed time since the previous brake event, and duration of the last brake event. They may also include one or more of: the vehicle speed at the start of the previous brake event, the vehicle speed at the end of the previous brake event, the average deceleration previous last brake event, distance travelled since the previous brake event, estimated temperature at the end of the previous brake event, and the number of brake events in a given journey. In this way, the temperature prediction model can account for residual heating in the brake discs as a result of recent braking.
By including an elapsed time since a previous brake event finished, the temperature prediction model 109 can apply an appropriate weighting to the heating effects of the last brake event. For instance, if a long time has passed since the previous brake event, the model may ignore the heating effects of the previous braking entirely in preference for assuming the current rotor temperature is close to ambient temperature. In this way, the model (described below) is able discriminate between the how much weight it gives the input data it receives. This improves the accuracy of the rotor temperature estimation since more sources of data may be used to inform the model.
The data from a previous brake event may also relate to braking from a previous journey where a vehicle has been driven to a destination and parked/turned off. Therefore, residual heating as an effect of braking during a different journey may be taken into account when the vehicle is restarted.
The temperature prediction model may also use vehicle specification data as input parameters. The vehicle specification data may be stored in the memory device 103 and may include any of: dimensions of the rotor or brake disc, mass of the rotor, brake pad contact surface area of the rotor, specific heat capacity of the rotor, dimensions of braked wheels of the vehicle, vehicle mass, vehicle type or model, position of the vehicle brake (for example at the front or rear of a vehicle), and/or other known specifications about the vehicle or brakes.
The vehicle operation data may include an estimate the amount of kinetic energy (KE) of the brake rotor that is converted to heat during braking. This estimate can then also be input to the temperature prediction model. The estimate of kinetic energy of the rotor converted to heat during braking (Disc_heating_KE_1) may be calculated as:
Disc_heating_KE_1=½mcv12−½mcv22
where mc is the mass of the vehicle, v1 is the vehicle speed at braking start, and v2 is the current vehicle speed during braking. Alternatively or additionally, the estimate of kinetic energy of the rotor converted to heat during braking (Disc_heating_KE_2) may be calculated as:
Disc_heating_KE_2=(pv2fr)/(mdcd)
where p is brake clamping force, v2 is the current vehicle speed during braking, fr is brake pad contact surface area of the rotor, ma is the mass of the brake disc, and cd is the specific heat capacity of the rotor. The machine learning model may be a regression model trained using analytical regression methods. For instance the regression model may be: K-nearest neighbours regression, Lasso regression (Least Absolute Shrinkage and Selection Operator), simple regression, ARD regression (Automatic Relevance Determination), Ridge regression, gradient boosted decision trees (such as XGBoost https://xgboost.readthedocs.io/en/latest/#), random forest regression, or random forest regression with extra tree regression (for example as described in https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.ExtraTreesRegressor). Alternatively, the machine learning model may be a neural network such as an MLP regressor.
The machine learning model may be trained and tested using a test vehicle equipped with suitable sensors or using a test bench brake system, of the type shown schematically in
A large amount of data can be collected to train and test the temperature prediction model 109 by simulating many different test scenarios using the test bench brake system. The test scenarios can be varied by adjusting parameters such as the number of brake events performed, the length of sliding time (brake duration), the vehicle speed, the vehicle mass, and the simulated slope of the ground.
In this way, the temperature prediction model 109 may be trained easily and adapted for new brake systems without the need to perform a complex thermal analysis of each brake rotor. Thus, developing updated temperature prediction models for new brake systems is more straightforward than for conventional thermal heat loss models. In particular, many of the above-mentioned input parameters are readily available from existing vehicle specifications and sensors. Additional, more complex parameters, for example the thermal properties of the rotor used to estimate the kinetic energy lost to heat during braking, may be measured and added to the temperature prediction model if desired. These optional additional parameters may increase the fidelity of the temperature prediction and improve the accuracy of the resulting adjustments of brake clamping force. Moreover, cumulative errors in temperature estimation can be reduced because a machine learning model can be trained to recognise how much weighting to give to previous temperature estimates. Therefore, the smart brake system can be more reliable and safer than conventional brake systems.
The features disclosed in the foregoing description, or in the following claims, or in the accompanying drawings, expressed in their specific forms or in terms of a means for performing the disclosed function, or a method or process for obtaining the disclosed results, as appropriate, may, separately, or in any combination of such features, be utilised for realising the invention in diverse forms thereof.
While the invention has been described in conjunction with the exemplary embodiments described above, many equivalent modifications and variations will be apparent to those skilled in the art when given this disclosure. Accordingly, the exemplary embodiments of the invention set forth above are considered to be illustrative and not limiting. Various changes to the described embodiments may be made without departing from the spirit and scope of the invention.
For the avoidance of any doubt, any theoretical explanations provided herein are provided for the purposes of improving the understanding of a reader. The inventors do not wish to be bound by any of these theoretical explanations.
Any section headings used herein are for organizational purposes only and are not to be construed as limiting the subject matter described.
Throughout this specification, including the claims which follow, unless the context requires otherwise, the word “comprise” and “include”, and variations such as “comprises”, “comprising”, and “including” will be understood to imply the inclusion of a stated integer or step or group of integers or steps but not the exclusion of any other integer or step or group of integers or steps.
It must be noted that, as used in the specification and the appended claims, the singular forms “a,” “an,” and “the” include plural referents unless the context clearly dictates otherwise. Ranges may be expressed herein as from “about” one particular value, and/or to “about” another particular value. When such a range is expressed, another embodiment includes from the one particular value and/or to the other particular value. Similarly, when values are expressed as approximations, by the use of the antecedent “about,” it will be understood that the particular value forms another embodiment. The term “about” in relation to a numerical value is optional and means for example +/−10%.
| Number | Date | Country | Kind |
|---|---|---|---|
| 21202677.7 | Oct 2021 | EP | regional |
