Radar triggered road friction estimation

Through polarization radar technology and auxiliary physical friction estimation system, the problem of inaccurate road friction estimation in heavy vehicles is solved, the accuracy and safety of vehicle motion management are improved, and component wear is reduced.

CN118876984AInactive Publication Date: 2024-11-01VOLVO TRUCK CORP
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410478122.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2023-12-11
Filing Date
2024-04-19
Publication Date
2024-11-01
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The prior art is difficult to accurately and universally estimate road friction, especially in the motion management of heavy vehicles, resulting in insufficient precision and safety in vehicle control.

Method used

The polarized radar transceiver is used to transmit radar signals to the vehicle surface, determine the road friction condition by processing the backscattering signal, and trigger the auxiliary physical friction estimation system when the friction parameter changes are detected, including generating steering pulses or propulsion torque pulses, to Accurately estimate road friction.

Benefits of technology

Accurate estimation of the wheel force generation capacity of heavy-duty vehicles is achieved, the accuracy and safety of vehicle motion management is improved, the excessive actuation of the mechanical friction estimation system is reduced, and the service life of the components is extended.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118876984A_ABST
    Figure CN118876984A_ABST
Patent Text Reader

Abstract

The invention relates to radar triggered road friction estimation, in particular to a computer-implemented method for determining a road friction condition associated with at least one wheel (150, 160, 170, 430) on a heavy vehicle (100), the method comprises: transmitting (S1), by at least one polarized radar transceiver (180), a radar signal (185) to a surface (101) supporting the vehicle (100) and receiving backscatter from the transmitted radar signal, wherein the radar signal (185) comprises a first polarized component and a second polarized component different from the first polarized component; processing (S2) the received backscatter by a processing device (130) to determine a friction parameter relating to a road friction condition of the surface (101); monitoring (S3) the friction parameter over time to detect a change in the friction parameter; and triggering (S4) the friction force estimation by the assisted physical friction force estimation system if a change in the friction parameter is detected.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present disclosure generally relates to road friction estimation and also to vehicle motion management based at least in part on estimated road friction. In a particular aspect, the present disclosure relates to a radar-triggered road friction estimation method. The present disclosure may be applicable to heavy vehicles such as trucks, buses, and construction equipment, among other vehicle types. Although the present disclosure may be described with respect to a particular vehicle, the present disclosure is not limited to any particular vehicle. Background Art

[0002] Road friction plays an important role in the behavior of most vehicles, primarily because it limits the wheel forces that a vehicle can generate at any given point in time. For example, the road friction coefficient μ is related to the normal force F acting on the tire. z Together, the achievable longitudinal tire force F x Limited to F x ≤μF z Road friction also determines how much lateral wheel force F the tire can support without excessive lateral wheel slip. y , and therefore determines how quickly the vehicle can enter a curve without leaving the road.

[0003] The driver of a manually driven vehicle typically has at least a rough idea of ​​the current road conditions, for example, whether the road is slippery or icy, and can adjust vehicle motion (such as speed) accordingly. However, conditions can change rapidly, and not all drivers have sufficient experience to accurately determine road friction. Furthermore, for autonomous vehicles without a driver, it would certainly be helpful to have at least a rough idea of ​​the road friction conditions before executing control operations. Therefore, methods for determining road friction conditions are needed.

[0004] A great deal of work has been put into devising methods for automatically determining road friction based on applying controlled amounts of torque through different vehicle propulsion devices. Some such examples include DE102012220238, DE102019202772 and EP3562702.

[0005] JPH10288559 and JP2008170237 also disclose methods for automatically determining road friction. These methods instead estimate road friction based on applying steering.

[0006] However, despite the availability of automatic road friction determination techniques, there remains a need for more versatile and accurate methods to estimate road friction. Summary of the Invention

[0007] A first aspect of the present disclosure relates to a computer-implemented method for determining a road friction condition associated with at least one wheel on a heavy vehicle. The method includes transmitting a radar signal to a surface supporting the vehicle via at least one polarization radar transceiver and receiving backscatter from the transmitted radar signal. The radar signal includes a first polarization component and a second polarization component different from the first polarization component, such as a vertical polarization component and a horizontal polarization component or a left circular polarization component and a right circular polarization component. The method also includes processing the received backscatter via a processing device to determine a friction parameter associated with the road friction condition of the surface. The method also includes monitoring the friction parameter over time to detect a change in the friction parameter, and if a change in the friction parameter is detected, triggering friction force estimation via an auxiliary physical friction force estimation system.

[0008] This means that known mechanical friction estimation mechanisms only need to be activated when needed (i.e., when road friction conditions change). This is advantageous because too frequent activation of mechanical friction estimation systems can cause discomfort to occupants of the vehicle and can also lead to increased wear on components, such as tires.

[0009] Any method known in the art can be used to detect changes in friction parameters. For example, a test metric determined based on the polarization characteristics of received backscatter can be calculated and compared to a nominal value. If the test metric deviates from the nominal value by more than a predetermined amount, a change is detected. After a change is detected, the nominal value can then be updated to a new nominal value, and monitoring can continue using the new nominal value.

[0010] According to some aspects, the assisted physical friction estimation system includes generating steering pulses, ie, rapid steering changes by the steering axle, the amplitude of the steering pulses being generally small, in order to detect the road surface.

[0011] The assisted physical friction force estimation system may also include generating a propulsion torque pulse, such as a braking torque pulse or a propulsion torque pulse.

[0012] According to some aspects, the friction parameters monitored over time include two-port scattering parameters of the road surface at a first polarization and a second polarization. Prior art has shown that these two-port scattering parameters vary with friction. Therefore, they can be used to detect changes in road friction. The friction parameters can also be related to the scatter coherence matrix of the received backscatter, target entropy, and / or the polarization basis of the backscattered radar signal, or a combination of several different indices and quantities that can be determined based on processing the received backscatter.

[0013] The computer-implemented method optionally further includes monitoring a normal force associated with one or more wheels on the vehicle over time to detect a change in the normal force, and triggering a friction force estimation by an auxiliary physical friction force estimation system if a change in the normal force is detected. The normal force can be measured or estimated in a variety of ways, such as by monitoring bellows pressure in a suspension system, by monitoring vertical displacement of the vehicle chassis relative to a road surface, by monitoring tire pressure or effective rolling radius, etc. If a change in the normal force is detected, such as a change in the weight of the vehicle's cargo, it may be recommended to update the friction force estimate by triggering a friction force estimation by an auxiliary physical friction force estimation system.

[0014] Aspects of the method disclosed herein also relate to a computer-implemented method that includes determining a wheel force generation capability of one or more wheels on a heavy vehicle based on a friction force estimation performed by an auxiliary physical friction force estimation system. This wheel force generation capability is useful in performing various vehicle motion management operations, as discussed in detail below.

[0015] Aspects of the methods disclosed herein also relate to a computer-implemented method that includes updating a tire model based on a friction estimate performed by an auxiliary physical friction estimation system. For example, the method may include updating an inverse tire model associated with one or more wheels on a vehicle based on an estimated current road friction. The inverse tire model describes the relationship between wheel slip and the resulting tire force, and this relationship is dependent on road friction. By updating the tire model based on the friction estimate performed by the auxiliary physical friction estimation system, a more accurate tire model is obtained.

[0016] The method described herein may also include transmitting radar signals toward a surface supporting the vehicle via multiple polarimetric radar transceivers, and receiving backscatter from the transmitted radar signals, wherein each transmitted radar signal is associated with a corresponding wheel or group of wheels on the vehicle. By operating several polarimetric radar transceivers in this manner, road friction estimation for a specific wheel or group of wheels can be performed. This advantageously increases the amount of information available for advanced vehicle control.

[0017] Aspects of the present disclosure also relate to hardware devices, such as control units, computer systems, and processing circuits.

[0018] The disclosed aspects, examples (including any preferred examples) and / or the appended claims may be appropriately combined with each other, as will be apparent to anyone skilled in the art. Additional features and advantages are disclosed in the detailed description, claims, and drawings, and in part will be apparent to those skilled in the art or recognized by practicing the disclosure as described herein.

[0019] Also disclosed herein are computer systems, control units, code modules, computer-implemented methods, computer-readable media, and computer program products associated with the technical advantages discussed above. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Examples are described in more detail below with reference to the accompanying drawings.

[0021] Figure 1 is an exemplary illustration of a heavy vehicle.

[0022] Figure 2 The wheel is shown being subjected to the tire forces.

[0023] Figure 3 A graph showing the applied steering pulse is shown.

[0024] Figure 4 A motion support device for a steering axle is shown schematically.

[0025] Figure 5 The architecture of vehicle motion management is shown.

[0026] Figure 6 An exemplary heavy vehicle is shown.

[0027] Figure 7 is a graph showing exemplary tire forces as a function of wheel slip.

[0028] Figure 8 is a flow chart illustrating the method.

[0029] Figure 9 is a schematic diagram of an exemplary computer system for implementing the examples disclosed herein. DETAILED DESCRIPTION

[0030] The detailed description set forth below provides information and examples of the disclosed technology in sufficient detail to enable those skilled in the art to practice the disclosure.

[0031] The technology of the present disclosure will now be described more fully below with reference to the accompanying drawings, in which certain aspects of the invention are shown. However, the invention may be embodied in many different forms and should not be construed as limited to the embodiments and aspects set forth herein; rather, these embodiments are provided by way of example so that this disclosure will be thorough and complete and will fully convey the scope of the invention to those skilled in the art. Throughout this specification, like reference numerals refer to like elements.

[0032] It should be understood that the present invention is not limited to the embodiments described herein and shown in the accompanying drawings; rather, those skilled in the art will recognize that many changes and modifications may be made within the scope of the appended claims. Therefore, many (if not most) of the technical features discussed below are not inextricably linked to each other, but can be used independently to advantage.

[0033] Figure 1 A heavy vehicle 100 is shown. This particular example includes a tractor unit 110 arranged to tow a trailer unit 120 in a known manner. The tractor 110 includes a vehicle control unit (VCU) 130 arranged to control various functions of the vehicle 100. For example, the VCU may be arranged to perform vehicle motion management (VMM) functions, including control of wheel slip, vehicle unit stability, etc. The trailer unit 120 optionally also includes a VCU, which then controls one or more functions on the trailer 120. One or more VCUs may be communicatively coupled to a remote server 140, for example via a wireless link 135. The remote server may be arranged to perform various VCU configurations and provide various forms of data to the VCU 130, such as data about the brand and type of tires mounted on the vehicle 100, as will be discussed in more detail below.

[0034] A heavy vehicle may be understood as a motor vehicle with a gross vehicle weight rating (GVWR) exceeding 8,500 pounds (approximately 3855 kg). A heavy vehicle may also be understood as a motor vehicle with a basic vehicle frontal area exceeding 45 square feet (approximately 4.2 m2). Examples of heavy vehicles include trucks, semi-trailers, and the like. It will be appreciated that the methods and control units disclosed herein may also be advantageously applied to other types of heavy vehicles, such as trucks with a drawbar connection, construction equipment, buses, and the like.

[0035] The vehicle combination 100 may of course also include additional vehicle units, such as one or more dolly units and more than one trailer unit.The teachings herein may also be applied to unibody vehicles, such as unibody trucks and passenger cars.

[0036] Vehicle 100 is supported by wheels, each of which includes a tire. Tractor unit 110 has front wheels 150, which are typically steered, and rear wheels 160, at least one pair of which are driven wheels. Trailer unit 120 is supported on trailer wheels 170. Trailers with driven wheels and even one or more steered axles are also possible. The wheels of vehicle 100 support the vehicle on a surface 101, which can be paved, graveled, or other surfaces.

[0037] Vehicle 100 is supported on surface 101 by wheels 150, 160, 170, and as mentioned above, friction between the road and the tires plays a significant role in determining how vehicle 100 responds to various control commands. For example, under low friction conditions, applying a certain amount of torque may cause excessive wheel slip, whereas applying the same amount of torque under more favorable friction conditions will result in smooth acceleration of the vehicle.

[0038] Road friction can be characterized by the coefficient of road friction, typically denoted as μ. This is a dimensionless scalar value that describes the ratio of the friction between two objects to the force pressing them together. The coefficient of friction depends on the material used; for example, rubber on ice has a relatively low coefficient of friction, while rubber on a dry road has a higher coefficient of friction.

[0039] Figure 2 A vehicle tire 210 is schematically shown, such as a tire on one of the wheels 150, 160, 170 on the vehicle 100. A positive or negative torque T may be applied about its axis of rotation to generate a longitudinal force F x , to accelerate or brake the vehicle 100. However, this longitudinal force cannot exceed the normal force F z The product of μF and the road friction coefficient μ z If the wheel is a steerable wheel, a steering angle δ may also be applied, which then typically generates a lateral tire force F y .

[0040] Figure 7 Graph 700 is shown. In this document, longitudinal ground velocity can be determined relative to the vehicle, in which case the velocity direction refers to the vehicle's forward direction, or relative to the wheels, in which case the velocity direction refers to the wheel's forward direction or rolling direction. The same is true for lateral ground velocity, which can be the vehicle's lateral velocity or the wheel's lateral ground velocity relative to its rolling direction. The meaning is clear from the context, and it should be understood that a direct transformation can be applied to convert ground velocity between the vehicle's coordinate system and the wheel's coordinate system, and vice versa. For example, Thomas Gillespie discusses vehicle coordinate systems and wheel coordinate systems in Fundamentals of Vehicle Dynamics, published by the Society of Automotive Engineers in Warrendale, Pennsylvania, in 1992.

[0041] Longitudinal wheel slip λ x It can be defined according to SAE J370 (SAE Vehicle Dynamics Standards Committee, January 24, 2008) as

[0042]

[0043] Where R is the effective wheel radius in meters, ω i is the angular velocity of the i-th wheel on vehicle 100, and v x is the longitudinal velocity of the wheel (in the wheel's coordinate system). Therefore, λ x The value ranges from -1 to 1 and quantifies the degree of wheel slip relative to the road surface. The terms "road surface" and "surface" are used interchangeably herein. Both terms refer to the surface 101 supporting the vehicle 100 and are a reference to the ground velocity in both the lateral and longitudinal directions. Wheel slip is essentially the speed difference measured between the wheel and the vehicle. Therefore, the techniques disclosed herein are applicable to any type of wheel slip definition. It should also be understood that, in the wheel's coordinate system, the wheel slip value is equivalent to the wheel speed value given the wheel's velocity on the road surface.

[0044] The slip angle α (also called the side slip angle) is the angle between the direction the wheel is pointing and the actual direction of travel (i.e., the longitudinal velocity component v x With the wheel forward speed v x With the lateral velocity v y The slip angle generates a force, namely, a steering force, that lies in the plane of the contact patch and is perpendicular to the intersection of the contact patch and the midplane of the wheel. The steering force increases approximately linearly within the first few degrees of slip angle, then increases nonlinearly to a maximum value, and then begins to decrease.

[0045] The wheel slip angle α is usually defined as

[0046]

[0047] where v x is the longitudinal velocity of the wheel, and v y is the lateral velocity of the wheel (in the wheel's coordinate system). It will be appreciated that, depending on the motion of the vehicle 100, v x 、v y Therefore, it may be advantageous to deploy the wheel end modules 140 in conjunction with different wheels on the vehicle 100 or in conjunction with different axles on the vehicle 100 in order to obtain accurate information about the vehicle's motion.

[0048] In order for a wheel (or tire) to generate wheel forces that affect the motion of a heavy vehicle, such as acceleration or deceleration (negative acceleration), wheel slip must occur. For small slip values, the relationship between slip and generated force is approximately linear, with the proportionality constant usually expressed as the tire's slip stiffness C. x The tire is subjected to a longitudinal force F x , lateral force F y and the normal force F z Normal force F zIt is the key to determining some important vehicle properties. For example, the normal force largely determines the longitudinal tire force F that can be achieved by the wheel. x , because usually F x ≤μF z , where μ is a coefficient of friction associated with road friction conditions. The technology disclosed herein is intended to provide a method for at least approximately determining μ and F for at least some of the wheels 102 on the vehicle 100 and / or one or more axles of the vehicle 100. z Both wheel end modules 140 , systems and methods.

[0049] Graph 700 is a graph showing the longitudinal wheel slip λ x Example of achievable tire forces as a function of F. x is the longitudinal tire force, and F y is the maximum lateral wheel force achievable for a given wheel slip. This relationship between wheel slip and the resulting tire force is often referred to as the inverse tire model. Figure 7 The example in represents a positive wheel force, i.e., acceleration. A similar relationship exists between wheel slip and negative wheel force (i.e., braking). The maximum available lateral force for a given lateral slip can be described using the so-called magic formula as described in "Tyre and vehicle dynamics," by Hans Pacejka, published by Elsevier Ltd. in 2012, ISBN 978-0-08-097016-5, where wheel slip and tire forces are also discussed in detail.

[0050] The inverse tire model can be used to calculate the desired longitudinal tire force F x and longitudinal wheel slip λ x Convert between. Figure 7 The illustrated tire models can be used to coordinate motion actuators (such as brakes and propulsion) and steering on a vehicle to achieve the desired vehicle motion. However, if the wrong road friction is assumed, the tire model will not be very accurate. By using a tire model parameterized by road friction, the accuracy of the mapping between wheel slip and tire force is greatly improved. Some technologies disclosed herein relate to methods and systems that maintain a database of several tire models that can be selected based on the current road friction. The system maintains an up-to-date estimate of road friction and uses the corresponding tire model, i.e., a tire model that provides the correct mapping between wheel slip and tire force given the current road friction conditions. A lookup table can be used to map road friction to a tire model. The tire model can also be defined as a mathematical function with road friction as a parameter. These road friction-dependent tire models can be predetermined through actual experimentation, perhaps through computer simulation.

[0051] Reference again Figure 7, exemplary longitudinal tire force F x A nearly linearly increasing portion 710 is shown for small wheel slips, followed by a portion 720 with more nonlinear behavior for larger wheel slips. It is desirable to maintain vehicle operation in the linear region 710, where the longitudinal forces available in response to applied braking commands are more predictable and where sufficient lateral tire forces can be generated if needed. To ensure operation in this region, a wheel slip limit λ of approximately, for example, 0.1 can be applied to a given wheel. lim Therefore, by accurately knowing the current wheel slip, it is possible to ensure operation within the linear region, which greatly simplifies vehicle motion control to improve safety, efficiency and driver comfort. Knowing the current longitudinal wheel slip λ x and understanding the current wheel force generation capability (i.e., μF z ) also provides important information about a given wheel's ability to generate lateral force. Advantageously, the wheel slip limit is configured as a function of the current road friction. Therefore, according to some aspects, the tire model also includes control parameters such as the wheel slip limit. This enables vehicle control to better adapt to current road friction conditions, thereby improving vehicle safety and efficiency.

[0052] Vehicle 100 includes multiple sensors, such as torque sensors capable of accurately detecting small changes in torque about a wheel axle (such as the steering axle), inertial measurement units (IMUs) capable of accurately measuring acceleration and yaw motion across various vehicle units, and wheel speed sensors capable of very quickly acquiring small changes in wheel speed in response to certain actuator controls. It is now recognized that these vehicle sensors are becoming so sensitive and fast that an onboard vehicle control unit can measure and analyze the vehicle's response to very small amplitude and short duration vehicle control actions. This can be used to estimate road friction while the vehicle is in motion without significantly affecting vehicle behavior, as only very small control actions are required to generate vehicle response data, which can then be analyzed to infer current road friction conditions.

[0053] Specifically, road friction can be estimated based on the vehicle's response to small-amplitude and short-duration steering pulses—pulses that are small enough and / or short enough in duration to not significantly affect the vehicle's path. Thus, road friction can be estimated independently of the propulsion system. This is advantageous because the propulsion system includes components such as the transmission that typically complicate the estimation of road friction. Furthermore, by allowing steering-based road friction estimation, a redundant method is provided that can be used to supplement propulsion-based road friction estimation methods.

[0054] Figure 3A diagram 300 illustrating the general concept is shown. A steering pulse d(t) 310 is generated at time T0. This steering pulse 310 comprises a temporary, slight change in the steering angle relative to some nominal value, preferably, but not necessarily, followed by a more or less immediate reversal of the steering angle back to the nominal value. The vehicle's response r(t) 320 to this steering pulse is then recorded and analyzed to discern the current road friction conditions. Test data can be collected by applying the steering pulse while the vehicle is traveling on a road surface with a known coefficient of friction and stored in memory. The vehicle control unit can then repeat the measurement on a road surface with unknown friction and investigate which pulse response in the stored data is most similar to the current measurement.

[0055] The vehicle's response r(t) will have amplitude characteristics, such as peak amplitude M and duration T. In low-complexity implementations, these simple measurements can be used to determine road friction. Of course, more complex implementations are possible and may even be preferred due to the increased accuracy. Such more complex implementations may involve frequency analysis of the impulse response, possibly enhanced by machine learning techniques. Such estimation methods may be parameterized and typically configured by a service technician performing a vehicle software update, or periodically by a remote server 140.

[0056] Because the steering pulses are very short and small, the vehicle does not exhibit any noticeable motion deviations during the friction estimation process and does not stray from the road while performing the friction estimation. Instead, a short-duration sound may be heard and some vibration may be felt as the measurement is repeated multiple times to improve the quality of the estimated signal. Therefore, the methods discussed herein can be applied while the vehicle is in motion, even when traveling relatively fast (such as approximately 80 km / h or even faster).

[0057] A similar method involves generating a propulsion or braking pulse at one or more wheels, i.e., a temporary and slight change in the applied torque T relative to some nominal value, followed preferably, but not necessarily, by a more or less immediate reversal of the applied torque T back to the nominal value. The vehicle's response to this wheel torque pulse is then recorded and analyzed to identify the current road friction conditions.

[0058] However, friction estimation based on controlling the steering and / or propulsion devices on a vehicle to "sense" road conditions cannot be performed continuously, at least because it causes excessive inconvenience to the driver or passengers of the vehicle due to the noise and vibration generated. Furthermore, such friction sensing actuation cannot be performed continuously due to the undesirable large amount of energy consumed. This type of mechanical friction estimation also leads to increased component wear (e.g., tire wear and wear of the steering linkage on the vehicle).

[0059] The present disclosure aims to provide a method and apparatus capable of detecting changes in road friction and subsequently reliably estimating friction using applied torque and / or steering pulses (as examples of a mechanical friction estimation system). This eliminates the need to continuously perform mechanical friction estimation, but rather only requires it when a change in road friction is suspected. Alternatively, the mechanical friction estimation system on a vehicle can be triggered by a detected change in the normal force acting on one or more wheels of the vehicle.

[0060] To this end, the vehicle 100 includes a polarimetric radar sensor system comprising one or more radar transceivers 180. The radar sensor system is directed toward the ground and can also be used to determine the vehicle's ground speed. The ground speed radar module can be configured to determine a two-dimensional velocity vector [v x , v y ], or only determine a one-dimensional velocity vector, such as the longitudinal velocity of vehicle 100.

[0061] The principle of determining ground speed utilized herein is based on the disclosure of US 2004 / 0138802. In addition to the disclosure of US 2004 / 0138802, ground speed radar systems have been proposed previously in a number of disclosures.

[0062] The technology disclosed herein also builds upon previous work in which dual-polarized high-carrier frequency radar signals have been used to estimate road surface properties, from which the coefficient of friction can be at least approximately determined. In “Road Surface Recognition at mm-Wavelengths Using a Polarimetric Radar,” IEEE Transactions on Intelligent Transportation Systems, Vol. 23, No. 7, July 2022, Vessen Vassilev describes a system for estimating road surface conditions using polarimetric radar. For example, the system can be used to detect when a thin layer of ice forms on the road surface. At least in some cases, the system can also be used to detect when the road surface is wet and when it is dry. This disclosure also cites a number of references [1] to

[10] , which contain similar work that can be applied in conjunction with the teachings of this disclosure. A drawback of previous work using polarimetric radar to estimate road friction is that it is difficult to determine the absolute value of friction, i.e., the actual friction. This is because some high-friction surfaces have similar scattering properties to low-friction surfaces. However, it has been recognized that polarimetric radar sensor systems can more reliably detect changes in road friction. The present disclosure exploits this knowledge and uses a polarimetric radar friction estimation system to primarily detect friction changes. It then uses a mechanical road friction estimation device (such as one that includes steering pulses or wheel torque pulses) to reliably determine the actual friction value, i.e., the road friction condition after the friction change.

[0063] A polarimetric radar transceiver is a dual-polarization radar transceiver arranged to receive and detect radar backscatter in at least two different polarizations. In the system disclosed herein, two linear polarizations (such as horizontal and vertical) or two circular polarizations (such as right-hand circular polarization and left-hand circular polarization) can be used.

[0064] It will be appreciated that known systems for determining road surface properties, such as the road friction coefficient, can sometimes be adequately calibrated to provide the actual friction coefficient μ. However, accurately measuring the road friction coefficient based solely on polarimetric radar is not sufficient to obtain a reliable estimate of road friction conditions.

[0065] It should be understood that the tires on the wheels of vehicle 100 play a significant role in determining the behavior and capabilities of vehicle 100. A well-designed set of tires will provide both good traction and fuel economy, while a poorly designed set of tires or excessively worn tires may reduce both traction and fuel economy, and may even cause instability in the vehicle combination, which is certainly undesirable. The present disclosure also relates to software tire models that model tire parameters and tire behavior for given vehicle conditions, such as vehicle speed, normal load, etc. Tire models can be helpful in estimating road friction, as the properties of the tires installed on vehicle 100 may affect the vehicle's response 320 to steering pulses 310. For example, the tire models disclosed herein can be used to model the relationship between generated steering pulses 310 and / or generated torque pulses, vehicle response 320, and road friction conditions. For example, worn tires may cause the vehicle to respond in a manner similar to that of a vehicle with better tires but traveling on a road with a lower coefficient of friction, and vice versa.

[0066] The tire models discussed herein are preferably configured to be dynamically updated as the vehicle operates. Thus, preferably, the tire model is dynamic rather than static, meaning that the tire model will be able to better and more rigorously model tire parameters based on, for example, the overall driving scenario and the changing state of the tire over time due to changes in tire characteristics due to wear and changes in inflation pressure and temperature.

[0067] Figure 4 The functionality 400 for controlling wheels 430 via some exemplary motion support devices (MSDs) is schematically illustrated, here including a power steering device 420 and a propulsion device 410 (such as an electric motor or an internal combustion engine). The power steering device 420 and the propulsion device 410 are examples of actuators that can be controlled by one or more motion support device control units 440, for example, to generate control pulses to the wheels 430 in order to induce a vehicle response, which can then be recorded and analyzed by vehicle sensors to estimate current road friction conditions.

[0068] The MSD control unit 440 may be arranged to control one or more actuators. For example, it is not uncommon for the MSD control unit to be arranged to control the MSDs for both wheels of an axle. By estimating the vehicle unit motion using, for example, a global positioning system, vision-based sensors, wheel speed sensors, radar sensors, and / or lidar sensors and converting the vehicle unit motion into a local coordinate system for a given wheel (in terms of, for example, a longitudinal velocity component and a lateral velocity component), wheel slip may be accurately estimated in real time by comparing the vehicle unit motion in the wheel reference coordinate system with data obtained from wheel speed sensors arranged in conjunction with the wheel.

[0069] As discussed above, functionality 400 also includes a friction-sensing polarimetric radar device 450. The radar device is configured to detect changes in road friction. VMM functionality 460 and / or MSD control unit 440 can be configured to use applied steering pulses and / or applied torque pulses to trigger friction estimation in order to detect new friction conditions.

[0070] Both the steering device 420 and the propulsion device 410 interact with the road surface via the tires of the wheels. Therefore, as mentioned above, the tire properties generally have a significant impact on how different control actions of the power steering device 420 and the propulsion device 410 generate vehicle motion. A software-based tire model 480 is therefore optionally included in the system 400. The tire model provides information about the tires currently mounted on the wheels 430, their properties and behavioral characteristics. The tire model can also include a mapping between wheel slip and generated tire forces, as described above in conjunction with Figure 7 As discussed above. The VMM control unit 460 and / or the MSD control unit 440 use the information provided by the tire model to predict the results of different control allocations and interpret the vehicle's response to, for example, steering pulses in order to infer the current road conditions based primarily on friction. As discussed above, the tire model can be implemented as a lookup table or other type of function. The tire model is parameterized, i.e., defined by one or more tire parameters. This means that the function itself varies depending on the properties of the tire. The tire model can be used to model the various relationships illustrated above, such as the relationship or mapping between steering pulses and generated wheel forces, and / or the mapping between tire wear rate and the vehicle's response to steering pulses applied for different road friction conditions. It should be understood that the present disclosure is not limited to any particular form of tire model structure. Rather, it should be understood that many different types of mathematical and / or experimentally based functions and mappings can be used as tire models.

[0071] Also refer to Figure 5 The Traffic Situation Management (TSM) function 470 plans driving maneuvers in a time range of, for example, about 1 to 10 seconds. This time range corresponds to, for example, the time it takes for the vehicle 100 to negotiate a curve. The vehicle maneuvers planned and executed by the TSM may be associated with acceleration profiles and curvature profiles that describe the desired vehicle speed and turning for a given maneuver. The TSM continuously requests the desired acceleration profile a from the VMM function 460. req and the curvature curve c req , the VMM function performs force distribution to satisfy requests from the TSM in a safe and robust manner. The TSM function 470 can also optionally determine vehicle handling based on the tire model 480, such as Figure 4 Instructions.

[0072] The VMM function 460 operates in a time range of about 0.1 to 1.5 seconds and continuously adjusts the acceleration curve a req and the curvature curve c req Converted into control commands for controlling vehicle motion functions actuated by the different MSDs of the vehicle 100, which report back capabilities to the VMM, which in turn are used as constraints in vehicle control. This control is advantageously based on current road friction data obtained by the method disclosed herein. Figure 5 , the VMM function 460 performs vehicle state or motion estimation 510, that is, the VMM function 460 continuously determines the vehicle state (usually a vector variable) including the position, velocity, acceleration, yaw motion, normal force and articulation angle of different units in the vehicle combination by monitoring the vehicle state and behavior using various sensors 450 arranged on the vehicle 100, which are usually but not always associated with the MSD.

[0073] The friction estimation method is executed in the friction estimation module 560, which triggers the generation of steering pulses through control signals to the MSD coordination function 530, thereby generating steering angle commands to the power steering device on the vehicle 100 and / or generating torque pulses through friction brakes or one or more electric motors on the vehicle 100. The vehicle's response to this actuation is measured by sensors 450 (including torque sensors for the power steering device and / or propulsion device), and this response is sent back to the friction estimation module 560, where friction is estimated. This friction estimation system is an example of what is referred to herein as a mechanical friction estimation system.

[0074] The results of the motion estimation 510 (i.e., the estimated vehicle state) are input to the global force generation module 520, which determines the required global forces that need to be generated on the vehicle units to meet the motion request from the TSM 470. The MSD coordination function 530 distributes, for example, wheel forces and coordinates other MSDs, such as steering and suspension. The coordinated MSDs then work together to provide the desired lateral force Fy and longitudinal force Fx on the vehicle units, as well as the required moment Mz, to achieve the desired motion of the vehicle combination 100. Figure 5 Indicated in the figure, the MSD coordination function 530 can shift the wheel slip to λ i , wheel speed ω, torque T i and / or steering angle δ i Any of them are output to different MSDs.

[0075] The MSD coordination function 530 is optionally supported by a tire model function 480, which can continuously update a software-based model of the tires on the vehicle based, at least in part, on current road friction conditions. The MSD coordination function 530 can, for example, use the tire model to determine the relationship between wheel slip and generated wheel forces. The tire model is parameterized by one or more tire parameters, such as tire wear, tire normal load, tire slip stiffness, and current road friction conditions. These tire parameters can, of course, be preconfigured. However, additional advantages can be achieved if the tire parameters are estimated or otherwise determined based on tire data obtained from memory 540 or based on tire data obtained from one or more sensors 450.

[0076] It will be appreciated that different types of vehicles will respond differently to a given stimulus, and very heavy vehicles will likely require a stronger steering pulse in order for the vehicle to exhibit a sufficiently strong measurable response to accurately estimate road friction. Thus, according to some aspects, the duration and / or amplitude of the steering pulse is limited based on one or more dynamic characteristics of the heavy vehicle 100. For example, appropriate settings for the steering pulse and / or torque pulse in terms of steering angle and / or torque amplitude and duration may be configured at the factory when the vehicle is assembled, or these appropriate settings may be updated periodically, for example, by a remote server 140. The steering pulse and / or torque pulse may be generated as a pulse that includes a suddenly applied control change followed by an immediate reversal of the control command to its nominal value. The response of the vehicle 100 will then be an impulse response, which may be analyzed using the methods discussed in more detail below. The above in conjunction with Figure 3 An example of such a steering pulse 310 and the resulting vehicle response 320 is discussed. However, other forms of input steering stimulus may also be used. For example, the applied steering pulse may be generated as a step function, including a sudden applied steering angle starting from a nominal steering angle. However, this type of step function steering pulse is more likely to affect the vehicle path than a steering pulse that includes a reversal back to the nominal value.

[0077] Reference again Figure 5The MSD coordination function 530 can also be used to at least partially compensate for the steering pulse. For example, assume a large steering pulse is applied, which, if no other action is taken, would cause the vehicle 100 to veer left. According to some aspects, the MSD coordination function 530 is configured to perform motion prediction, thereby predicting the consequences of applying the steering pulse. The MSD coordination function 530 can then perform MSD coordination to compensate for the steering pulse. In other words, the method optionally includes compensating for the steering pulse ( S14 ) via one or more other motion-supporting devices of the vehicle 100. For example, a left turn can be compensated by applying a short-duration braking torque on the right side of the vehicle 100 or a short-duration positive torque pulse on the right side of the vehicle to counteract the yaw motion caused by the steering pulse. However, it should be understood that such vehicle motion compensation will invalidate any vehicle response measurements involving yaw motion, lateral force, etc. Nevertheless, such vehicle motion compensation may still be useful if the primary source of information for assessing road friction is the torque applied by the power steering device to generate the steering pulse.

[0078] This may cause some vehicles to become more sensitive to applied steering pulses and / or torque pulses at higher speeds than at lower speeds. Therefore, the method optionally includes configuring the amplitude and / or duration of the steering pulses and / or torque pulses based on the current speed of the heavy-duty vehicle 100. Generally speaking, the faster a vehicle is traveling, the greater its stored kinetic energy, and the smaller the steering pulses required to generate a sufficiently large response to be measured and used to estimate road friction. Of course, it may also be advantageous to configure the amplitude and / or duration of the steering pulses and / or torque pulses based on the operating environment of the heavy-duty vehicle 100. For example, different steering pulse amplitudes and / or durations may be more appropriate for estimating road friction on a low-speed gravel road than on a highway. The vehicle's effective wheelbase may also affect the amplitude and / or duration of the steering pulses. Longer wheelbases generally require a larger steering angle to achieve the same curvature. Therefore, a vehicle with a longer wheelbase can tolerate larger steering pulses than a vehicle with a shorter wheelbase without significantly disrupting its motion.

[0079] As discussed above, what constitutes a finite duration and a finite amplitude should be interpreted broadly and take into account factors such as vehicle type, operating environment, and vehicle speed. However, for most heavy vehicles, the method may include generating a steering pulse having an amplitude between 1 and 5 degrees, and preferably between 2 and 3 degrees, and a duration between 0.1 and 1.0 seconds, and preferably around 0.2 to 0.5 seconds.

[0080] The response of the vehicle to the applied steering pulses and / or torque pulses can be measured in different ways using a single sensor or a combination of sensors. For example, the method can include measuring the response of the vehicle 100 as the torque applied by the power steering system of the vehicle 100. This is one of the preferred methods of measuring the response of the vehicle because it has been shown to provide relevant measurements even for small steering pulses of limited duration and amplitude. Basically, referring to Figure 5 A steering angle command for a given steering angle change is transmitted from the VMM 460 to the MSD control unit 440, which then controls the power steering system to apply the steering angle change. This is followed immediately by a command to reverse the steering angle. The power steering system then measures the applied torque required to produce the steering angle change and records this data as a vector of torque values, sampled, for example, at a sufficiently high rate (e.g., approximately several kHz). This data is then sent back to the friction estimation function 560 in the VMM 460, which performs an analysis to estimate the current road friction. This analysis can be performed at varying levels of complexity. For example, a simple method for determining road friction based on this torque data is to evaluate the torque required to cause the steering angle change. If the wheels turn very easily, the measured torque value will be relatively small, indicating low friction. On the other hand, if turning the wheels requires more torque, the potential road friction will be higher. This torque can also be estimated based on the current vehicle ground speed and vehicle load, as a stationary or slower-moving vehicle typically requires more torque to change the steering angle than a faster-moving vehicle.

[0081] If torque pulses are used as an alternative to or in addition to steering pulses, similar action is applied. If little resistance is encountered when applying the torque pulses, i.e., if the wheel speed is changing rapidly, then the road friction is likely low compared to the wheel's strong resistance to the applied torque pulses.

[0082] The method may also include measuring the response of vehicle 100 as a yaw motion of the vehicle caused by the applied steering pulse and / or torque pulse, at least if the torque pulse is applied to one side of the vehicle and / or symmetrically to both sides. Of course, if the friction between the vehicle and the road is low, a given steering pulse and / or torque pulse is also expected to produce little yaw motion. On the other hand, if the friction is high, the same steering pulse and / or torque pulse may result in more yaw motion, at least if the vehicle is traveling at the same speed. Friction estimation functionality 560 may be configured with lookup tables, etc., that allow the recorded yaw motion caused by a given steering pulse to be converted into an estimated road friction coefficient. These lookup tables are preferably also functions of vehicle speed and / or vehicle load. Similarly, the method may include measuring the response of vehicle 100 as a lateral force and / or lateral acceleration generated by the applied steering pulse. The lateral acceleration is preferably measured at the vehicle's steering axle.

[0083] Road friction conditions can be estimated based on the vehicle's measured response to applied steering pulses in a variety of different ways, which can be applied independently or in combination. One preferred method involves estimating road friction values ​​based on frequency analysis of the measured response. For example, on slippery roads, oscillations in the vehicle's response to steering pulses have been observed. On the other hand, a firmer response and less oscillatory behavior generally indicate greater road friction.

[0084] Alternatively or in addition to and with reference to Figure 3 , the method comprises estimating a road friction value based on the magnitude M and / or duration T of the measured response.

[0085] Methods involving machine learning may also be applied to the measured vehicle responses to estimate the current road friction conditions. Thus, the method may comprise estimating the road friction value using a machine learning method applied to the measured responses. Of course, such methods may involve conventional neural networks, which may be trained using measurement data obtained on a road section with a known friction coefficient. The neural network may be fed with raw response data and / or pre-processed measurement data, such as a frequency domain representation of the measurement data. The detection mechanism is advantageously based on machine learning techniques. Different types of machine learning techniques have been successfully applied, but algorithms based on random forest techniques have been found to be particularly effective and provide robust road friction estimates.

[0086] Random forest or random decision forest represents an ensemble learning method for classification, regression and other tasks, which operates by constructing many decision trees during training and outputting the class (classification) mode or mean / average prediction (regression) of each tree. Random decision forest is associated with the advantage of being able to correct the habit of decision trees to overfit with their training set. Random forest is generally superior to decision tree-based algorithms. As an alternative to the random forest classification method, a less complex decision tree algorithm can be used, commonly referred to as the regression tree algorithm, which is basically a single-tree random forest algorithm.

[0087] As mentioned above, the tires on the vehicle also affect the vehicle's response to applied steering impulses. Less worn tires will naturally provide better grip than nearly completely worn tires, which will be reflected in the vehicle's response to a given steering impulse. Therefore, advantageously, the method optionally includes software-based models of the vehicle's 100 tires to estimate road friction values. Examples of such software-based tire models are discussed above. Of course, these software-based tire models can also be updated using data obtained from the friction estimation algorithms discussed herein, i.e., the method can include updating the software-based models of the vehicle's 100 tires based on measured responses.

[0088] The driver or passengers of heavy vehicle 100 may notice the actions of the control unit as described herein, for example, through increased noise levels or slightly jerky movements of the vehicle. To inform the driver about the cause of the vehicle's noise and behavior, the method may include triggering a notification to the vehicle driver in conjunction with generating the steering pulses. This notification may include, for example, a symbol on the instrument panel, similar to when an anti-lock braking system (ABS) becomes active or an anti-skid system is engaged.

[0089] The methods disclosed herein are advantageously used in conjunction with autonomous or semi-autonomous vehicles without a driver to infer road conditions based on the overall feel of the vehicle and visual inspection of the road ahead of the vehicle. However, a driver manually operating a vehicle may also wish to trigger these methods to determine road friction. This functionality may be used, for example, by a driver who is unsure about road conditions and wishes to obtain a "second opinion" as to whether the road conditions are slippery or whether they simply appear to be slippery. This may occur, for example, if the road is covered in snow and may be wet (very slippery) or dry (less slippery). To this end, the methods disclosed herein may include generating steering pulses and / or torque pulses in response to a request from the driver of the vehicle 100. The request may be generated, for example, by some form of input control, such as a button or option on a touch screen.

[0090] Figure 6An exemplary heavy vehicle is shown in which a plurality of polarization radar transceivers 610 have been mounted in conjunction with respective wheels or groups of wheels. On this exemplary vehicle, each set of wheels (i.e., the front steering wheels, the rear traction wheels, and the trailer wheels) is associated with a radar transceiver on the right and left sides of the vehicle. Each such radar transceiver monitors polarization backscatter from an area of ​​road surface 620 in front of the respective wheel or group of wheels, such as Figure 6 As shown, a mechanical friction estimation system is used to detect changes in road friction. Once a change in road friction is detected ahead of a given wheel or wheel group, an auxiliary mechanical friction estimation system is triggered for that specific wheel or wheel group to more reliably determine the actual road friction conditions for that specific wheel or wheel group. Advantageously, the onboard road friction estimation system then obtains reliable wheel-specific road friction condition data. This road friction data can also be used to estimate the tire force generation capability of a specific wheel, which can be used to improve vehicle motion management.

[0091] Figure 8 A flowchart summarizing some of the teachings herein as a method is shown. It should be understood that the method can be implemented in a computer system for execution by processing circuitry (e.g., processing circuitry on a vehicle control unit 130). The flowchart illustrates a computer-implemented method for determining road friction conditions associated with at least one wheel 150, 160, 170, 430 on a heavy vehicle 100. The road friction condition can be an approximate road friction coefficient, or some other indicator, such as a unitless value within some predetermined range from low friction to high friction. The method includes transmitting, via at least one polarization radar transceiver 180, an S1 radar signal 185 toward a surface 101 supporting the vehicle 100 and receiving backscatter from the transmitted radar signal. The radar signal 185 includes a first polarization component and a second polarization component different from the first polarization component. According to some aspects, the polarization radar signal is also used to determine the ground speed of the vehicle 100, but it should be understood that the friction estimation function and the ground speed estimation function are not inseparable. The method further comprises: processing S2 the backscatter received by the processing device 130 to determine a friction parameter associated with the road friction condition of the surface 101; monitoring S3 the friction parameter over time to detect a change in the friction parameter; and triggering S4 to perform friction estimation by an auxiliary physical friction estimation system if a change in the friction parameter is detected.

[0092] The change in the friction parameter may, for example, include comparing the difference between the friction parameter and a nominal value with a threshold value. Of course, more advanced detection criteria may also be implemented. The change detection criteria may be determined through actual experiments and / or computer simulations.

[0093] To detect changes in a friction parameter, test statistics can be determined as a predetermined function of the friction parameter. Such test statistics can be monitored over time to detect changes. The change in the friction parameter can be measured relative to a nominal value of the friction parameter. Once the change in the friction parameter has been determined, a new nominal friction parameter value can be configured, and monitoring can then be performed based on the new nominal friction parameter value. Those skilled in the art will appreciate that changes in the friction parameter can be implemented in several different ways, and the present disclosure is not limited to any particular way of detecting changes in the friction parameter.

[0094] The auxiliary physical friction force estimation system may include, for example, generating steering pulses S41, generating propulsion torque pulses S42, and / or generating braking torque pulses S43, as discussed above. More than one mechanical actuation (e.g., a combination of steering pulses and torque pulses) may be used to determine road friction via the auxiliary physical friction force estimation system. Such a combination may be used to compensate for the effects of the auxiliary friction force estimation system on vehicle motion. For example, the steering pulse may be combined with differential propulsion or braking on the sides of the vehicle to compensate for yaw motion caused by the steering pulse. The propulsion torque pulse may be offset in a similar manner by the braking torque pulse. For example, the above combined Figure 5 The discussed MSD coordination function 530 may be used to implement these aspects of the method.

[0095] The torque pulses are advantageously generated S44 by an electric motor of the heavy vehicle 100. The use of electric motors in this application is advantageous because they can typically operate at a high bandwidth, i.e., they can be used to generate torque pulses of short, significant duration. The electric motors can also be used to monitor the wheel's response to the applied torque pulses. Many vehicles already have electric motors available, which can be advantageously retrofitted with polarimetric radar systems and updated software to implement the advanced friction estimation methods discussed herein.

[0096] Braking torque pulses may also be generated S45 by an electromechanical braking device, which may also typically be actuated at a high bandwidth.

[0097] Aspects of the computer-implemented method discussed herein relate to friction estimation, where the monitored friction parameters include two-port scattering parameters S21 of the road surface 101 at a first polarization and a second polarization, as discussed in "Road Surface Recognition at mm-Wavelengths Using a Polarimetric Radar," IEEE Transactions on Intelligent Transportation Systems, Vol. 23, No. 7, July 2022. The friction parameters may also be correlated with a scattering coherence matrix of received backscatter S22, with target entropy, and / or with a polarization basis of the backscattered radar signal S23. Other friction parameters may also be defined and used to similar effect.

[0098] The friction parameter monitored over time to detect changes in the friction parameter can be any of the following: a two-port scattering parameter of the road surface 101 at the first and second polarizations, a scattering coherence matrix of received backscatter, target entropy, and / or a polarization basis of the backscattered radar signal. Functions of these indicators can also be used to detect changes. Suitable functions can be determined based on actual experiments and / or computer simulations of the interaction between polarimetric radar signals and various road surfaces.

[0099] In general, those skilled in the art understand that the backscatter from the transmitted polarized radar signal is first processed to determine the appropriate friction parameter. Changes in the friction parameter are then detected by comparing the friction parameter or a function thereof with a certain nominal value. If the value deviation exceeds a set amount, or other more advanced detection criteria are met, a change is detected. The article "Road Surface Recognition at mm-Wavelengths Using aPolarimetricRadar" in Volume 23, Issue 7 of "IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS" in July 2022 discusses various indicators that can be used to determine road friction conditions. Any of these indicators can be monitored over time and compared with the corresponding nominal value. If the deviation of the monitored indicator exceeds a predetermined amount, a change is detected.

[0100] According to some aspects, the computer-implemented method further comprises monitoring S5 a normal force associated with one or more wheels on the vehicle 100 over time to detect a change in the normal force, and if a change in the normal force is detected, triggering S6 a friction force estimation by an auxiliary physical friction force estimation system. The normal force can be measured in a variety of ways, such as by monitoring bellows pressure in a suspension system of the vehicle 100, by monitoring the distance of a heavy vehicle chassis from the ground, and / or by measuring the linear displacement of the suspension of the vehicle 100. The change in normal force is related to the ability to generate tire force, as described above in conjunction with Figure 7 As discussed, changes in normal force may produce similar effects to changes in road friction and are therefore also relevant to VMM functionality and MSD control.

[0101] The computer-implemented methods discussed herein may also include determining S7 wheel force generation capabilities based on friction estimation performed by an auxiliary physical friction estimation system (eg, based on a tire model parameterized by road friction).

[0102] The computer-implemented methods discussed herein may also include updating the S8 tire model based on the friction estimation performed by the auxiliary physical friction estimation system.

[0103] The computer-implemented method discussed herein may also include transmitting S11 radar signals 185 to the surface 101 supporting the vehicle 100 via a plurality of polarization radar transceivers 180, and receiving backscatter from the transmitted radar signals, wherein each transmitted radar signal is associated with a corresponding wheel or group of wheels on the vehicle 100. This is described above in conjunction with Figure 6 By using multiple polarimetric radar transceivers 180 in this manner, wheel-specific road friction conditions, or at least wheel group-specific road friction conditions, can be determined. Advantageously, vehicle motion management and actuator control can then be customized for each wheel, or at least each wheel group, taking into account the current road friction for that particular wheel or wheel group.

[0104] Figure 9is a schematic diagram of a computing system 900 for implementing the examples disclosed herein. The computer system 900 is adapted to execute instructions from a computer-readable medium to perform these and / or any functions or processes described herein. The computer system 900 can be connected (e.g., networked) to other machines in a LAN (local area network), a LIN (local interconnect network), an automotive network communication protocol (e.g., FlexRay), an intranet, an extranet, or the Internet. Although only a single device is shown, the computer system 900 may include any collection of devices that individually or collectively execute an instruction set (or multiple instruction sets) to perform any one or more of the methods discussed herein. Therefore, any reference in this disclosure and / or claims to a computer system, computing system, computer device, computing device, control system, control unit, electronic control unit (ECU), processor device, processing circuit, etc. includes reference to one or more such devices that individually or collectively execute an instruction set (or multiple instruction sets) to perform any one or more of the methods discussed herein. For example, the control system may include a single control unit or multiple control units that are connected or otherwise communicatively coupled to each other, such that any performed functions can be distributed among the control units as needed. Furthermore, such devices may communicate with each other and other devices through various system architectures, such as directly or via a Controller Area Network (CAN) bus.

[0105] Computer system 900 may include at least one computing device or electronic device capable of including firmware, hardware, and / or executing software instructions to implement the functionality described herein. Computer system 900 may include processing circuitry 902 (e.g., processing circuitry including one or more processor devices or control units), memory 904, and a system bus 906. Computer system 900 may include at least one computing device having processing circuitry 902. System bus 906 provides an interface for system components including, but not limited to, memory 904 and processing circuitry 902. Processing circuitry 902 may include any number of hardware components for performing data or signal processing or for executing computer code stored in memory 904. Processing circuitry 902 may include, for example, a general-purpose processor, a special-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a circuit containing processing components, a group of distributed processing components, a group of distributed computers configured for processing, or other programmable logic devices, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functionality described herein. Processing circuitry 902 may also include computer-executable code that controls the operation of the programmable device.

[0106] The system bus 906 can be any of several types of bus structures, which can be further interconnected to a memory bus (with or without a memory controller), a peripheral bus, and / or a local bus using any of a variety of bus architectures. The memory 904 can be one or more devices for storing data and / or computer code to complete or facilitate the methods described herein. The memory 904 can include a database component, an object code component, a script component, or any type of information structure for supporting various activities herein. Any distributed or local memory device can be utilized together with the system and method of this specification. The memory 904 can be communicatively connected to the processing circuit 902 (e.g., via a circuit or any other wired, wireless, or network connection) and can include computer code for executing one or more processes described herein. The memory 904 may include nonvolatile memory 908 (e.g., read-only memory (ROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), etc.) and volatile memory 910 (e.g., random access memory (RAM)), or any other medium that can be used to carry or store desired program code in the form of machine-executable instructions or data structures and that can be accessed by a computer or other machine having the processing circuit 902. A basic input / output system (BIOS) 912 may be stored in the nonvolatile memory 908 and may include the basic routines that help to transfer information between elements within the computer system 900.

[0107] The computer system 900 may also include or be coupled to non-transitory computer-readable storage media such as storage device 914, which may include, for example, an internal or external hard disk drive (HDD) (e.g., Enhanced Integrated Drive Electronics (EIDE) or Serial Advanced Technology Attachment (SATA)), a HDD for storage (e.g., EIDE or SATA), flash memory, etc. The storage device 914 and other drives associated with computer-readable and computer-usable media may provide non-volatile storage of data, data structures, computer-executable instructions, etc.

[0108] The computer code of hard coding or soft coding can be provided in the form of one or more modules. The module can be implemented as software and / or hard-coded in the circuit to implement the functionality described herein in whole or in part. The module can be stored in a storage device 914 and / or volatile memory 910 that may include an operating system 916 and / or one or more program modules 918. All or a portion of the examples disclosed herein can be implemented as a computer program 920 stored on a temporary or non-temporary computer-usable or computer-readable storage medium (e.g., a single medium or multiple media) such as a storage device 914, which includes complex programming instructions (e.g., complex computer-readable program code) that cause the processing circuit 902 to perform the actions described herein. Therefore, the computer-readable program code of the computer program 920 may include software instructions for implementing the functionality of the examples described herein when executed by the processing circuit 902. In some examples, storage device 914 may be a computer program product (e.g., a readable storage medium) having a computer program 920 stored thereon, wherein at least a portion of computer program 920 may be loadable (e.g., loaded into a processor) for implementing the functionality of the examples described herein when executed by processing circuitry 902. Processing circuitry 902 may serve as a controller or control system for computer system 900 for implementing the functionality described herein.

[0109] The computer system 900 may include an input device interface 922 configured to receive input and selections to be transmitted to the computer system 900 when executing instructions, such as from a keyboard, mouse, touch-sensitive surface, etc. Such input devices can be connected to the processor circuit 902 via the input device interface 922 coupled to the system bus 906, but can be connected via other interfaces (such as a parallel port, an Institute of Electrical and Electronics Engineers (IEEE) 1394 serial port, a universal serial bus (USB) port, an IR interface, etc.). The computer system 900 may include an output device interface 924 configured to forward output to a display, a video display unit (e.g., a liquid crystal display (LCD) or a cathode ray tube (CRT)). The computer system 900 may include a communication interface 926 suitable for communicating with a network as appropriate or as needed.

[0110] The operational actions described in any of the exemplary aspects herein are described to provide examples and discussion. These actions may be performed by hardware components, may be embodied in machine-executable instructions to cause a processor to perform these actions, or may be performed by a combination of hardware and software. Although a specific order of method actions may be shown or described, the order of the actions may vary. In addition, two or more actions may be performed simultaneously or partially simultaneously.

[0111] The terms used herein are for the purpose of describing particular aspects only and are not intended to limit the present disclosure. As used herein, unless the context clearly indicates otherwise, the singular forms "a", "an", and "the" are intended to include the plural forms as well. As used herein, the term "and / or" includes any and all combinations of one or more of the relevant listed items. It should also be understood that when used herein, the terms "include" and / or "comprising" indicate the presence of the described features, integers, actions, steps, operations, elements and / or parts, but do not exclude the presence or addition of one or more other features, integers, actions, steps, operations, elements, parts and / or groups thereof.

[0112] It should be understood that although the terms first, second, etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, a first element may be referred to as a second element, and similarly, a second element may be referred to as a first element without departing from the scope of this disclosure.

[0113] Relative terms such as "below" or "above", or "upper" or "lower", or "horizontal" or "vertical" may be used herein to describe the relationship of one element to another, as shown in the figures. It should be understood that these terms and those discussed above are intended to encompass different device orientations in addition to the orientations depicted in the figures. It should be understood that when an element is referred to as being "connected" or "coupled" to another element, the element can be directly connected or directly coupled to the other element, or intervening elements may be present. In contrast, when an element is referred to as being "directly connected" or "directly coupled" to another element, there are no intervening elements.

[0114] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs. It should also be understood that, unless otherwise expressly defined herein, the terms used herein should be interpreted as having the same meaning as in the context of this specification and the relevant art, and should not be interpreted in an idealized or overly formal sense.

[0115] It should be understood that the present disclosure is not limited to the aspects described above and shown in the accompanying drawings; rather, those skilled in the art will recognize that many changes and modifications may be made within the scope of the present disclosure and the appended claims. In the drawings and description, various aspects have been disclosed for illustrative purposes only and not for purposes of limitation, the scope of the disclosure being set forth in the appended claims.

Claims

1. A computer-implemented method for determining road friction conditions associated with at least one wheel (150, 160, 170, 430) on a heavy vehicle (100), the method comprising: transmitting (S1) a radar signal (185) by at least one polarization radar transceiver (180) to a surface (101) supporting the vehicle (100) and receiving backscatter from the transmitted radar signal, wherein the radar signal (185) includes a first polarization component and a second polarization component different from the first polarization component, processing (S2) the received backscatter by means of a processing device (130) to determine a friction parameter associated with the road friction condition of the surface (101), monitoring (S3) the friction parameter over time to detect changes in the friction parameter, and If a change in the friction parameter is detected, Then, it is triggered (S4) to perform friction estimation through the auxiliary physical friction estimation system.

2. The computer-implemented method of claim 1, wherein the auxiliary physical friction force estimation system includes generating a steering pulse (S41).

3. The computer-implemented method of claim 1 or 2, wherein the auxiliary physical friction force estimation system comprises generating a propulsion torque pulse (S42).

4. The computer-implemented method of any preceding claim, wherein the auxiliary physical friction force estimation system comprises generating a braking torque pulse (S43).

5. The computer-implemented method of claim 3 or 4, wherein the torque pulses are generated by an electric machine (S44).

6. The computer-implemented method of claim 4, wherein the braking torque pulse is generated by an electromechanical braking device (S45).

7. A computer-implemented method according to any preceding claim, wherein the friction parameters comprise 2-port scattering parameters (S21) of the road surface (101) at the first polarization and the second polarization.

8. A computer-implemented method according to any preceding claim, wherein the friction parameter is correlated to a scatter coherence matrix of the received backscatter (S22).

9. A computer-implemented method according to any preceding claim, wherein the friction parameter is related to target entropy and / or a polarisation basis of the backscattered radar signal (S23).

10. A computer-implemented method according to any preceding claim, comprising monitoring (S5) a normal force associated with one or more wheels on the vehicle (100) over time to detect changes in the normal force, and If a change in the normal force is detected, Then trigger (S6) to perform friction estimation through the auxiliary physical friction estimation system.

11. A computer implemented method according to any preceding claim, comprising determining (S7) a wheel force generation capability based on the friction force estimation performed by the auxiliary physical friction force estimation system.

12. A computer implemented method according to any preceding claim, comprising updating (S8) a tire model based on the friction estimation performed by the auxiliary physical friction estimation system.

13. A computer-implemented method according to any preceding claim, comprising transmitting (S11) radar signals (185) to the surface (101) supporting the vehicle (100) via a plurality of polarization radar transceivers (180), and receiving backscatter from the transmitted radar signals, wherein each transmitted radar signal is associated with a corresponding wheel or group of wheels on the vehicle (100).

14. A computer program product comprising program code which, when executed by a processing circuit, performs the method according to any one of claims 1 to 13.

15. A non-transitory computer-readable storage medium comprising instructions which, when executed by a processing circuit, cause the processing circuit to perform the method of any one of claims 1 to 13.

16. A road friction estimation system for a heavy vehicle (100), comprising at least one polarimetric radar transceiver (180) and a processing device (130) configured to determine road friction conditions associated with at least one wheel (150, 160, 170, 430) on the heavy vehicle (100), wherein the radar transceiver (180) is a polarimetric radar transceiver (180) arranged to transmit a corresponding radar signal (185) toward a surface (101) supporting the vehicle (100) and to receive backscatter from the transmitted radar signal, wherein the radar signal (185) comprises corresponding different first and second polarization components, wherein the processing means (130) is arranged to process the received backscatter to determine at least one friction parameter related to the road friction condition of the surface (101), wherein the processing means (130) is arranged to monitor the friction parameter over time to detect changes in the friction parameter, and If a change in the friction parameter is detected, This triggers friction estimation through the auxiliary physical friction estimation system.

17. A heavy vehicle (100) comprising a road friction estimation system according to claim 16.