Method and system for controlling electric heavy commercial vehicle
By using a secondary planning method to optimize the control of the motion support system in heavy commercial vehicles, and coordinating the electric motor and driving brakes, the problem of energy loss and operating characteristics of the vehicle between different axles is solved, and higher operation is achieved Economicality and user satisfaction.
Patent Information
- Application Number
- CN202280094100.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2022-04-06
- Publication Date
- 2024-11-08
AI Technical Summary
The prior art is difficult to effectively coordinate the power transmission system in heavy commercial vehicles, especially in the differences in motor operating characteristics and energy loss between different axles, making it difficult for vehicles to achieve energy-saving operations within a given actuator limit range. .
Using a method based on the secondary planning (QP) problem, select the current driving mode according to the predefined driving mode, combine the current status of the vehicle and the actuator limit, optimize the control of the motion support system, and use numerical solvers such as the inner point method, The alternating direction multiplier method and other methods independently control the coordination of the electric motor and the driving brake.
It has achieved improved vehicle operation economy and production efficiency under different uses and road conditions. It is suitable for multi-function vehicles, reducing the optimization problems between power loss and torque and improving user satisfaction.
Smart Images

Figure CN118922325A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the field of heavy commercial vehicles, in particular vehicles equipped with a motion support system comprising at least two electric motors at respective axles and a set of service brakes. The present disclosure proposes an optimization-based method for controlling the motion support system depending on the currently active driving mode. Background Art
[0002] In a vehicle such as a battery electric vehicle or a fuel cell vehicle, there are several actuators in the powertrain that can be used to perform actions requested by the driver or an automated driving system (ADS). The powertrain is typically an over-actuated system (more control inputs than control outputs), so its control represents an underdetermined problem (fewer equations than unknowns) and requires coordination of the motions of the actuators. This coordination may be referred to as control allocation. Attempts have been made in this area to use optimization-assisted automatic control techniques, and more accurately the control allocation problem is treated and solved as an optimization problem. In this form, the control allocation problem may consist of minimizing a cost function that penalizes errors in global forces and moments (torques) as well as errors in the driver's motion requests.
[0003] Electric heavy-duty commercial vehicles often contain multiple electric motors to accelerate and (regeneratively) brake the vehicle. Depending on the layout of the drivetrain, the electric motors can be centrally mounted on the chassis, centrally mounted on each driven axle or mounted as wheel-end drives. Figure 3 4 are top views of the three types of chassis, wherein reference numeral 220 indicates a service brake, 250, 250A, 250B indicate an electric motor, 251, 251A, 251B indicate a gearbox, 252 indicates a drive shaft, 253, 253A, 253B indicate a drive axle, 254 indicates an air spring, 255 indicates a non-drive axle, 256 indicates a vehicle frame, and a hollow arrow indicates the direction in which the vehicle travels forward. Figure 4A and Figure 4B In the latter two layouts, there is usually a transmission or gearbox for each axle or wheel. Depending on the gearbox configuration, wheel diameter, and other transmission-related characteristics, the gear ratios of the driven axles can be the same or different.
[0004] The operating characteristics of the electric motors vary from axle to axle due to variations in normal loads, differences in the type and size of electric motors installed, and differences in gear ratios. Specifically, each motor may behave differently in terms of energy lost to heat dissipation, flux leakage, etc. The responsibility for ensuring energy-efficient operation of the vehicle within given actuator limits is typically assigned to a centralized Vehicle Motion Management (VMM) system that receives status signals and understands the capabilities of the electric motors and determines coordinated actuator requests based on requests from the human driver, driver assistance units such as Traffic Situation Management (TSM) systems or ADS. The general architecture described in detail below is as follows: Figure 2 , where the VMM is indicated by reference numeral 260 and the TSM is indicated by reference numeral 270. The VMM system may specifically implement the distribution of torque to the wheels based on the user's preferences for energy economy and performance (expressed through configuration settings).
[0005] The present disclosure is intended to further develop available VMM technology to enable new capabilities and reduce drawbacks. Summary of the invention
[0006] An object of the present disclosure is to provide a method for controlling a motion support system (MSS) in an electric heavy commercial vehicle. A further object is to propose an MSS control method suitable for vehicles where regenerative braking alone is insufficient to achieve a prescribed minimum braking force. A further object is to propose an MSS control method that can be adapted to different uses of the vehicle. A further object is to propose a controller for these uses.
[0007] At least some of these objects are achieved by the invention as defined by the independent claims. The dependent claims relate to advantageous embodiments of the invention.
[0008] In a first aspect of the present invention, a method for controlling the motion coordination of an MSS in a heavy commercial vehicle and / or a VMM in such a vehicle is provided. The method comprises the steps of: repeatedly selecting a current driving mode from a plurality of predefined driving modes; solving a quadratic programming (QP) problem associated with optimal control of independently controlled actuators in the MSS based on the current state of the vehicle (e.g., including device states, motor temperature, motor angular velocity and linear velocity, etc.) and subject to constraints representing actuator limits; and controlling the motion support system using the solution to the QP problem. The independently controlled actuators include at least two electric motors located at respective axles and a set of service brakes. The predefined driving modes include: (i) a first driving mode, wherein the QP problem represents a control allocation problem for the MSS, and (ii) a second driving mode, wherein the QP problem represents a minimum power loss in the motion support system.
[0009] Recall that all of the described motion support devices may be considered to constitute a MSS.
[0010] According to the first aspect, since the MSS control problem is formulated in the form of a QP problem, a range of numerical solvers with excellent performance and stability can be used, such as the interior point (IP) method, the active set (AS) method, and the alternating direction method of multipliers (ADMM) method. In addition, since the method has a first driving mode and a second driving mode and possibly additional driving modes available, the method can adapt to different uses of the vehicle. This makes the method suitable for multi-purpose or multi-purpose vehicles, such as trucks and buses designed for urban and highway use, agricultural and forestry equipment that performs tasks over a wide geographical area, and construction equipment that performs multiple material handling tasks (loading, transporting, etc.). In these applications, the method can contribute to good operating economy, production efficiency, and high user satisfaction.
[0011] In some embodiments, the constraints in the QP problem representing the actuator limits vary depending on the current state of the vehicle. Such variations can be used, for example, to simulate a global power maximum for the vehicle; Figure 6 In FIG. 6 , it can be seen that the maximum total propulsion force of the cruise actuator 601A and the start actuator 601B gradually decays at high speeds. As another example, when the battery of the vehicle is fully loaded, regenerative braking may not be available.
[0012] In some embodiments, the QP problem in the first driving mode (performance-oriented mode) is not related to power loss. More precisely, power loss is not explicitly (or separately) included in the QP problem and is therefore not particularly penalized.
[0013] In some embodiments, the QP problem in the second driving mode (minimum power loss mode) includes a relationship between the power loss of at least two of the actuators and the torque. This relationship can more fully reflect the energy loss of the vehicle and allow for more detailed optimization. In some embodiments, these relationships have a linear dependence on torque (service brakes), or a quadratic dependence on torque (electric drive), or a combination of these.
[0014] In a further development of these embodiments, the relationship between power loss and torque is provided as a smooth approximation of the measurement or simulation. The smoothness of these relationships may lead to better conditioned QP problems that can be solved while maintaining accuracy and reducing computational cost.
[0015] As a primary use case for the present invention, heavy commercial vehicles can be envisaged with relatively limited electric propulsion power per unit of vehicle mass (and therefore also limited electric regenerative braking power). The power-to-weight ratio may be 20 kW / ton or less, 10 kW / ton or less or 5 kW / ton or less. Even lower power-to-weight ratios are expected for special road transport, oversized loads, etc. Propulsion power limitations are particularly important for electric vehicles, since regulatory minimums apply to the braking efficiency of each vehicle in traffic (e.g., ≥3 kN / ton, ≥4 kN / ton, ≥5 kN / ton). Already at a moderate speed of 50 km / h, an instantaneous braking force of 3 kN / ton converts approximately 42 kW of kinetic energy per ton of vehicle mass into battery energy or heat. A power of 42 kW / ton is generally much higher than what can be achieved by pure regenerative braking (i.e., converted into battery energy), so the service brakes must generally be engaged when the vehicle is driven at street or highway speeds. Moderate use of the service brakes, purposefully combined with regenerative braking on one or more axles, may benefit the operating economy of the vehicle. If different relationships between power loss and torque apply to different axles, then balancing and combining all of these potential brake actuators is an important optimization problem.
[0016] In a second aspect of the invention, there is provided a controller configured to control an MSS in a heavy commercial vehicle. The controller may be suitable for deployment as a VMM controller in a vehicle. It comprises an input interface configured to receive data indicating a current vehicle state, a processing circuit configured to perform the method of the first aspect, and an output interface configured to feed a control signal to the MSS.
[0017] The controller having the technical features according to the second aspect of the present invention generally has the same effects and advantages as the first aspect, and it can be embodied in technical variations to the same extent.
[0018] The present invention also relates to a computer program comprising instructions for causing a computer or in particular a controller to perform the above-mentioned method of the first aspect. The computer program may be stored or distributed on a data carrier. As used herein, a "data carrier" may be a temporary data carrier, such as a modulated electromagnetic wave or light wave, or a non-temporary data carrier. Non-temporary data carriers include volatile and non-volatile memories, such as permanent and non-permanent storage media of magnetic, optical or solid-state type. Still within the scope of a "data carrier", such memory may be fixedly mounted or portable.
[0019] Generally, unless otherwise explicitly defined herein, all terms used in the claims should be interpreted according to their ordinary meaning in the art. Unless otherwise explicitly stated, all references to "a / an / the element, device, component, means, step, etc." will be interpreted openly as referring to at least one instance of the element, device, component, means, step, etc. Unless explicitly stated otherwise, the steps of any method disclosed herein do not have to be performed in the exact order described. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] Various aspects and embodiments will now be described by way of example with reference to the accompanying drawings, in which:
[0021] Figure 1 is a side view of a commercial vehicle;
[0022] Figure 2 is a block diagram illustrating a control architecture of an electric vehicle including a VMM system;
[0023] Figure 3 , Figure 4A and Figure 4B is a top view of a vehicle chassis with different powertrain layouts;
[0024] Figure 5 A top view of the vehicle chassis, where quantities related to motion control are introduced;
[0025] Figure 6 is a graph of maximum positive and negative propulsive forces achievable on two different axles of the vehicle as a function of speed versus speed;
[0026] Fig. 7A and Figure 7B Graphs showing the dependence of the efficiency coefficient η of the cruising axle and the starting axle of the vehicle on the torque.
[0027] Figure 8 is the corresponding relationship between the power loss and torque of the cruising axle and the starting axle; and
[0028] Fig. 9 is a flow chart of a MSS control method according to an embodiment of the present invention. DETAILED DESCRIPTION
[0029] Various aspects of the present disclosure will now be described more fully below with reference to the accompanying drawings, in which certain embodiments of the present invention are shown. However, these aspects may be embodied in many different forms and should not be construed as limiting; rather, these embodiments are provided by way of example so that the present disclosure will be thorough and complete and the scope of all aspects of the present invention will be fully conveyed to those skilled in the art. Throughout this specification, the same reference numerals refer to the same elements.
[0030] Figure 1 1 is a side view of a commercial vehicle 100 to which the technology disclosed herein may be applied. The vehicle 100 includes a tractor or towing vehicle 110 supported on front wheels 150 and rear wheels 160, at least some of which are driven wheels. Typically, but not necessarily, all wheels on the tractor 110 are brake wheels. The tractor 110 is configured to tow a first trailer unit 120 supported on trailer wheels 170 and a fifth wheel connection in a known manner. The trailer wheels 170 are typically brake wheels, but may also include driven wheels on one or more axles. It should be understood that the technology disclosed herein may also be advantageously applied to other types of heavy commercial vehicles, such as single unit vehicles, tractor-trailer combinations connected by a drawbar or another rigid coupling, construction equipment, buses, etc.
[0031] Specifically, the tractor 110 includes a vehicle unit computer (VUC) 130 or control unit for controlling various functions, including propulsion, braking, and steering. Some trailer units 120 also include a VUC or control unit 140 for controlling various functions of the trailer 120, such as braking of trailer wheels 170, and sometimes steering and propulsion of these wheels. The VUCs 130, 140 may be implemented in a centralized manner, or distributed over several processing components. Portions of the vehicle control functions may be performed remotely, for example on a remote server 190 connected to the vehicle 100 via a wireless link 180 extending from a wireless access network 185 to a transceiver (not shown) in the vehicle 100.
[0032] The VUC 130 on the tractor 110 (and possibly the VUC 140 on the trailer 120 ) may be configured to execute a vehicle control method organized according to a layered functional architecture, where some functions may be included in a traffic situation management (TSM) domain in a higher layer and some other functions may be included in a vehicle movement management (VMM) domain residing in a lower functional layer.
[0033] Now refer to Figure 2The motion control architecture of the vehicle 100 is described. The block diagram schematically shows the functionality 200 of controlling wheels 210 by some exemplary motion support devices (MSD) or actuators, including service brakes 220 (such as disc brakes or drum brakes) and propulsion devices 250. The term "service brakes 220" may be any type of non-regenerative brake designed to convert kinetic energy more or less completely into heat energy. The service brakes 220 may be implemented as friction brakes (e.g., electromechanical or pneumatic disc or drum brakes) or non-contact brakes (e.g., eddy current brakes), for example. The service brakes 220 and the propulsion devices 250 are examples of wheel torque generating devices, which may also be referred to as actuators and may be controlled by one or more MSD control units 230. All motion support devices may be considered to constitute a motion support system (MSS). The propulsion device 250 may be an electric motor, which may be operated to provide acceleration torque in forward and reverse directions, and preferably to provide regenerative braking in forward and reverse directions. The propulsion device 250 is powered via appropriate drive circuitry from an electrical energy source (not shown), which may be an on-board battery, an on-board generator or fuel cell, or a current collector in contact with an overhead line or third rail.
[0034] The technology disclosed herein is particularly applicable to vehicles 100 in which the propulsion device 250 is configured to have a total electric propulsion power (and approximately equal electric regenerative braking power) of 20 kW / ton or less, 10 kW / ton or less, or 5 kW / ton or less per unit vehicle mass. The vehicle mass refers to the fully loaded state with a trailer. The total mass of a vehicle or vehicle combination with a maximum design load may be on the order of tens of tons, such as ≥10 tons, ≥20 tons, ≥30 tons, or ≥40 tons. The technology is also applicable to lower power-to-weight ratios, for example as low as 2.2 kW / ton, which is the minimum motorization standard set by some jurisdictions. In addition, the vehicle 100 is also configured to have a braking force of at least 3 kN / ton, such as at least 4 kN / ton, such as at least 5 kN / ton.
[0035] The control of the actuators is based on, for example, measurement data obtained from wheel speed sensors 240 and from other vehicle state sensors 280, such as radar sensors, lidar sensors, and vision-based sensors such as camera sensors and infrared detectors. Other exemplary wheel torque-generating motion support devices that can be controlled according to the principles discussed herein are engine retardation devices and power steering devices. Each MSD control unit 230 can be arranged to control one or more actuators; for example, one MSD control unit 230 can be configured to control two wheels on an axle.
[0036] The Traffic Situation Management (TSM) function 270 plans driving maneuvers on a relatively short time horizon. Such planning corresponds to stabilization (operational decisions), which is a level below guidance (tactical decisions), which in turn is a level below navigation (strategic decisions). The TSM function can be configured to make decisions on the order of 10 seconds. The vehicle maneuvers planned and executed by the TSM function 270 can be associated with acceleration and curvature profiles that describe the expected vehicle speed and steering pattern for a given maneuver, that is, these quantities as a function of time within the time horizon of the TSM function 270. The TSM function 270 continuously provides the expected acceleration profiles to the VMM controller 260. and curvature curve , the VMM controller is configured to perform force and torque distribution to satisfy requests from TSM function 270 in a safe and robust manner. VMM controller 260 may be constructed as: an input interface 202 configured to receive data indicating current vehicle status from sensors 240, 280; a set of processing circuits 204; and an output interface 206 configured to feed control signals to the MSS, preferably through the intermediary of MSD controller 230.
[0037] The VMM controller 260 continuously feeds capability information to the TSM function 270, which details the vehicle's current capabilities in terms of the forces that can be generated, maximum speed, acceleration, etc. The TSM function 270 may act as (or include) an automated driving system (ADS), i.e., it accepts driving tasks from the user and autonomously generates acceleration profiles. and curvature curve to perform the task. The acceleration profile and curvature profile may alternatively be obtained from the heavy vehicle driver via conventional control input devices such as a steering wheel, accelerator pedal and brake pedal. Advanced VMM functionality may be combined with conventional manual driver control inputs to form so-called Advanced Driver Assistance Systems (ADAS).
[0038] The VMM controller 260 may operate on a time scale of the order of 1 second. The VMM controller 260 may be operable to continuously and curvature curve The VMM controller 260 converts the capability information into control commands for controlling vehicle motion functions actuated by the different MSDs 220, 250 of the vehicle 100, which report the capability information back to the VMM controller 260. The capability information can be used to formulate or update constraints in the vehicle control process. The VMM controller 260 performs vehicle state or motion estimation, that is, it continuously determines the vehicle state s including the position, velocity, acceleration and articulation angle of the different units in the vehicle combination by monitoring the operation using inputs from various sensors 280 arranged on the vehicle 100, which can be arranged in conjunction with the MSDs 220, 250, although this is not necessary. The result of the motion estimation (i.e. the estimated vehicle state s) is input to the force generation module, which determines the global force required for the two vehicle units. ,in Each of the has a longitudinal component and the lateral component (and optionally a yaw moment component) to move the vehicle 100 according to the requested acceleration and curvature profile. It is noteworthy that the satisfaction of the global force requirement is always subject to vehicle constraints; for example, due to combined slip, the total available road-wheel friction is divided between longitudinal and lateral components.
[0039] like Figure 2 As shown, VMM controller 260 and TSM function 270 may be implemented as functional blocks within VUCs 130 , 140 .
[0040] Required global force vector Input to the MSD coordination function in the VMM controller 260, which distributes wheel forces and coordinates other MSDs (such as steering and suspension). Control signals to the MSDs may include Figure 5 The signal shown is - Control signal to the service brake: ; - Control signal to the electric motor (cruise actuator) on the cruise axle 253A: ; - Control signal to the electric motor (starting actuator) on the starting axle 253B: .
[0041] The gear ratios between the cruise axle 253A and the starter axle 253B are different. Indicative gear ratio figures for heavy commercial vehicles are 15:1 and 23:1, respectively. Another difference may be that the cruise actuator and the starter actuator have different mechanical designs or different sizes. In addition, in Figure 5 middle, represents the wheelbase of the cruising axle 253A, and is the distance from the center of the vehicle to the cruise axle 253A. The values of these quantities affect the kinematics of the vehicle and can be included in the control allocation problem. Steering angle It can be controlled by the driver, or it can be controlled autonomously by the VMM controller 260. The MSDs coordinated by the VMM controller 260 ideally work together to apply lateral forces on each of the two vehicle units. and longitudinal force and yaw moment , to achieve the desired movement of the vehicle combination 100.
[0042] VMM controller 260 may be configured according to Fig. 9 The method 900 is shown performing the actions.
[0043] The method begins at step 910, i.e., selecting a current driving mode from at least a first driving mode and a second driving mode. In the case of an autonomous vehicle, the selection can be based on the assigned task, such as the time urgency of the task, the terrain of the environment in which the task is to be completed, etc. For vehicles driven by humans, the driving mode can be selected by observing the driver's behavior, such as the regularity (smoothness) of acceleration and braking over time, peaks in acceleration and braking, terrain, manual input indicating the driver's desired mode, etc. Pattern recognition technology can be used to map the observed driving behavior to a suitable driving mode. For both autonomous driving and human driving, one of the energy-optimized driving modes can be implemented when the battery charge of the vehicle 100 is below a predetermined threshold (such as 20%).
[0044] Step 910 may be repeated as needed during operation of the vehicle 100, such as based on an event trigger or periodically. Events that may trigger a driving mode update include driver commands. The method 900 may also be implemented as follows: when a sensor report is received indicating that the performance of the vehicle 100 has significantly degraded in the recent past, step 190 is repeated; this may indicate that the vehicle 100 can be operated more economically in a different driving mode, or with greater user satisfaction. If a new execution (i.e., repetition) of step 910 results in a change in driving mode, steps 912 and 914 are repeated for the new driving mode.
[0045] The predefined driving modes generally include a first driving mode that is performance-oriented and a second driving mode that prioritizes energy economy. The degree of specialization of the driving modes can vary between different implementations, in particular by defining a cost function to be optimized, where different weighting coefficients are applied to power loss terms and terms representing driving errors (e.g., the difference between actual vehicle behavior and a control request from the driver or ADS). In an extreme example, the first driving mode is a pure control allocation problem, completely ignoring the power loss issue. This can be achieved by defining a cost function that is independent of power losses, for example, a cost function that lacks a term that explicitly captures power losses; this focuses more effort on satisfying control requests. The first driving mode and the second driving mode may optionally include cost terms or constraints for ensuring the stability of the vehicle (in particular yaw stability).
[0046] In a second step 912, a QP problem is solved. The QP problem is related to the control of multiple independently controlled actuators (or MSDs) in the MSS, including two or more electric motors 250 located at respective axles 253A, 253B and the service brakes 220. The QP problem is so relevant because it reflects the static characteristics (e.g., geometry, kinematics, actuator limits) and current state of the vehicle 100, and the control signals for the actuators can be derived from the solution to the QP problem. The QP problem can be updated at runtime (e.g., by quantitative adjustments) to reflect the changing vehicle state , such as when the vehicle enters a higher or lower speed range. The updating of the QP problem may occur independently of the reselection of the driving mode, and may occur more frequently, or may occur synchronously with the reselection of the driving mode.
[0047] In the third step 914 of method 900, the solution μ to the QP problem is used to control the motion support system. Depending on how the QP problem is formulated, the solution μ can be used directly as a control signal or must be processed to be suitable for this purpose. As described above, execution of method 900 can be restarted from step 910 on a periodic or event-triggered basis.
[0048] To formulate the actuator coordination optimization problem for the motion control problem, the inputs are: vehicle parameters such as mass, inertia, tire radius, tire cornering stiffness; and desired control targets such as desired longitudinal and lateral forces, target service brake power, and target powertrain power. It is assumed that the MSS can be modeled as a linear dynamic system near the operating point of the current vehicle state, with a state vector Controlled by the following equation in is the observation signal, and is a constant matrix. This can represent the linearization of the nonlinear vehicle model valid at the operating point. Let represents the desired actuator request, Denotes the global forces and moments, the optimization problem can be written as follows:
[0049] Here, the cost function is the so-called control effectiveness matrix, which is basically a static model of the global forces as a function of the actuator signals, and which does not need to be related to the linear dynamic system equations. Consistent. Constant represents the actuator limit, which can be a function of the current vehicle state. Figure 6 As shown in FIG. 1 , due to power limitation, the maximum propulsion force 601A (positive and negative, corresponding to acceleration and regenerative braking) of the cruise actuator begins to decrease with speed at about 60 km / h. In the same figure, it can be seen that when the speed is above 25 km / h, the maximum positive and negative propulsion force 601B of the start actuator is power limited. In addition, Figure 6 , the total propulsive force is plotted as curve 601A+B, and the dashed line represents the desired (or ideal) behavior approximated by the combination of the starting actuator and the cruise actuator.
[0050] In the equality and inequality constraints of the optimization problem, is a constant matrix that absorbs the matrix The constraints may include equality constraints on the global longitudinal force, such as , which ensures that the control input is achieved. Another equality constraint may reflect the desired braking blend, i.e., the sum of the individual service braking forces should equal the total requested braking force.
[0051] In the cost function to be minimized, the matrix Set the desired actuator request vector The relative priority of the corresponding components of , and the matrix Indicates the relative priority of global forces and moments. The importance of the first and second terms in the cost function is balanced. Whether the cost function penalizes energy loss can be determined by the matrix Assigning zero or non-zero values to specific elements of is clear from the following calculations. Figure 5 shown In the case of a tractor chassis, which has four motors and four service brakes, the optimization variables There are eight components. Assuming the steering angle is controlled by the driver, and therefore its current value constitutes the vehicle state part of.
[0052] By expanding the term in the cost function and ignore the constant , the above optimization problem can be written as a standard QP problem: in
[0053] As mentioned above, several solvers are available in the literature or in commercial software that can efficiently and accurately solve the QP problem numerically, including the interior point (IP) method, the active set (AS) method, and the alternating direction method of multipliers (ADMM) method. Different optimization methods have different properties, and for some problems, there may be differences in execution time, convergence speed, accuracy of the solution, etc. These properties are particularly relevant for control optimizers running in embedded systems in vehicles, but there are limitations such as the accuracy of the floating point arithmetic used. The output of the solver can be an approximate solution , which is - or can be converted into - an actuator-signal vector, including the brake torque and the motor torque. It is worth noting that the approximate solution does not need to correspond to the actual optimal solution; the required accuracy and / or convergence may depend on the expected accuracy of the sensor signals, the expected accuracy of the actuators, and the tolerance for minor control imperfections in the use case.
[0054] As mentioned above, the equality condition ,in represents the global forces and moments, which can be included in the QP problem (1). In some embodiments, only those that satisfy the following conditions are accepted : This helps ensure that the solution to the QP problem (1) is feasible.
[0055] The matrix in the cost function may have different values in the first driving mode and the second driving mode, e.g. or In the second driving mode, the matrix in the quadratic term The relationship between power loss and torque is included, and in particular may include this relationship for at least two of the actuators.
[0056] Fig. 7A and Figure 7B This relationship is shown in FIG. The upper curves 701A, 701B represent the efficiency coefficients of the cruise actuator and the starter actuator, respectively, as a function of the torque recorded when the vehicle is traveling at a constant speed of 2.8 m / s. . Efficiency coefficient It can be interpreted as the ratio of useful output mechanical power to the supplied input electrical power. The lower curves 702A and 702B represent the fractional energy loss under the same operating conditions ( ).
[0057] Figure 8 is a graph of power loss in units of energy dissipated per unit time (1 kW). Consistent with the reference numerals used previously, curves 801A, 802A refer to the cruise actuator, and curves 801B, 802B refer to the cruise actuator. Here, the irregular curves 801A, 801B represent measured or simulated power loss values as a function of torque, while the parabolas 802A, 802B represent smooth approximations of these functions, and are in fact quadratic approximations. By using the matrix Configuration and optimization variables representing the electric motor torque The corresponding values of these elements of the components of can be easily integrated into the QP formula (1) for the quadratic problem. For simplicity, assume Including the cruise actuator torque, the start actuator torque and the brake torque in this order, then:
[0058] constant and May depend on gear ratio, wheel radius and similar parameters. In addition, can be set to a small positive value so that is positive definite. In a more realistic formula, the torque Each of has multiple components corresponding to different wheels of vehicle 100.
[0059] Matrix in Linear Terms can be used to model the power loss of the service brakes 220, i.e. the rate at which these brakes convert kinetic energy. A relationship of the form: in is the wheel speed A linear function of Again, this represents the service brake torque. Therefore, the value is calculated based on the current wheel speed , as part of the current vehicle state reported by sensors 240, 280, and assigned to the matrix The corresponding elements of:
[0060] It is important to note that the power loss of the service brake is a linear function, unlike the power loss of the electric motor which is a quadratic function.
[0061] For more background information on the modeling of MSS and the formulation of the optimization problem, see S. Janardhanan, “Concept design of electric cruise and startability axles for long haul heavy vehicles to maximise driving range,” Proceedings of the 2021 IEEE Vehicle Power and Propulsion Conference (VPPC), ISBN 978-1-6654-0528-7.
[0062] Aspects of the disclosure have mainly been described above with reference to a few embodiments. However, as is readily appreciated by a person skilled in the art, other embodiments than the ones disclosed above are equally possible within the scope of the invention, as defined by the appended patent claims.
Claims
1. A method (900) of controlling a motion support system (220, 250) in a heavy commercial vehicle (100), the method comprising: repeatedly ( 910 ) selecting a current driving mode from a plurality of predefined driving modes; solving (912) a quadratic programming (QP) problem associated with optimal control of independently controlled actuators in the motion support system based on a current state of the vehicle and subject to constraints representing actuator limits, wherein the independently controlled actuators include at least two electric motors (250) located at respective axles and a set of service brakes (220); as well as controlling the motion support system using the solution of the QP problem in (914), The predefined driving modes include: (i) a first driving mode, wherein the QP problem represents a control allocation problem for the motion support system, and (ii) A second driving mode, wherein the QP problem represents a minimum power loss in the motion support system. 2 . The method of claim 1 , wherein the constraints representing actuator limits vary depending on the current state of the vehicle. 3 . The method according to claim 1 , wherein the QP problem in the second driving mode comprises a relationship between power loss and torque of at least two of the actuators.
4. A method according to any one of the preceding claims, wherein the QP problem in the second driving mode includes a relationship between power loss and torque in quadratic form to at least one of the actuators and a relationship between power loss and torque in linear form to at least another of the actuators.
5. A method according to claim 3 or 4, wherein the relationship between power loss and torque is a measured or simulated smooth approximation.
6. A method according to any preceding claim, wherein the two independent electric motors comprise a cruise actuator and a starter actuator.
7. A method according to any one of the preceding claims, wherein the two types of independent electric motors include an electric axle with a cruise actuator and an electric axle with a starter actuator. 8 . The method of claim 7 , wherein the cruise actuator and the start actuator differ in gear ratio.
9. The method according to any of the preceding claims, wherein the electric propulsion power of the motion support system is at most 20 kW / tonne, such as at most 10 kW / tonne, such as at most 5 kW / tonne.
10. A method according to any preceding claim, wherein the motion support system is operable to provide a deceleration force of at least 3 kN / tonne, such as at least 4 kN / tonne, such as at least 5 kN / tonne.
11. The method according to any of the preceding claims, wherein the QP problem in the first driving mode is not related to power losses.
12. A controller (260) configured to control a motion support system (220, 250) in a heavy commercial vehicle (100), the controller comprising: an input interface (202) configured to receive data indicative of a current vehicle status; Processing circuitry (204) configured to perform the method according to any one of the preceding claims; as well as An output interface (206) is configured to feed a control signal to the motion support system.
13. A computer program comprising instructions for causing a controller (260) according to claim 12 to perform the steps of the method (900) according to any one of claims 1 to 11.
Citation Information
Cited By
A method for cooperative control of a commercial vehicle electric drive axle and a bridge end drive of an EMB
CN122876896A
