WO2023087153A1 - 一种标定方法、装置及系统 - Google Patents

一种标定方法、装置及系统 Download PDF

Info

Publication number
WO2023087153A1
WO2023087153A1 PCT/CN2021/131000 CN2021131000W WO2023087153A1 WO 2023087153 A1 WO2023087153 A1 WO 2023087153A1 CN 2021131000 W CN2021131000 W CN 2021131000W WO 2023087153 A1 WO2023087153 A1 WO 2023087153A1
Authority
WO
WIPO (PCT)
Prior art keywords
calibration
objective function
objective
parameters
calibration parameters
Prior art date
Application number
PCT/CN2021/131000
Other languages
English (en)
French (fr)
Inventor
罗杰
刘栋豪
古强
庄雨铮
张永生
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to PCT/CN2021/131000 priority Critical patent/WO2023087153A1/zh
Priority to EP21964319.4A priority patent/EP4425110A1/en
Priority to CN202180007894.5A priority patent/CN116457636A/zh
Publication of WO2023087153A1 publication Critical patent/WO2023087153A1/zh
Priority to US18/664,682 priority patent/US20240295420A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00—Programme-control systems
    • G05B19/02—Programme-control systems electric
    • G05B19/04—Programme control other than numerical control, i.e. in sequence controllers or logic controllers
    • G05B19/042—Programme control other than numerical control, i.e. in sequence controllers or logic controllers using digital processors
    • G05B19/0426—Programming the control sequence
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
    • G01D18/00—Testing or calibrating apparatus or arrangements provided for in groups G01D1/00 - G01D15/00
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60K—ARRANGEMENT OR MOUNTING OF PROPULSION UNITS OR OF TRANSMISSIONS IN VEHICLES; ARRANGEMENT OR MOUNTING OF PLURAL DIVERSE PRIME-MOVERS IN VEHICLES; AUXILIARY DRIVES FOR VEHICLES; INSTRUMENTATION OR DASHBOARDS FOR VEHICLES; ARRANGEMENTS IN CONNECTION WITH COOLING, AIR INTAKE, GAS EXHAUST OR FUEL SUPPLY OF PROPULSION UNITS IN VEHICLES
    • B60K35/00—Instruments specially adapted for vehicles; Arrangement of instruments in or on vehicles
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
    • G01M17/00—Testing of vehicles
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/08—Configuration management of networks or network elements
    • H04L41/0803—Configuration setting
    • H04L41/0813—Configuration setting characterised by the conditions triggering a change of settings
    • H04L41/082—Configuration setting characterised by the conditions triggering a change of settings the condition being updates or upgrades of network functionality
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/08—Configuration management of networks or network elements
    • H04L41/0803—Configuration setting
    • H04L41/0823—Configuration setting characterised by the purposes of a change of settings, e.g. optimising configuration for enhancing reliability
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/50—Testing arrangements
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/20—Pc systems
    • G05B2219/26—Pc applications
    • G05B2219/2637—Vehicle, car, auto, wheelchair
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/16—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks using machine learning or artificial intelligence
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/06—Generation of reports

Definitions

  • the embodiments of the present application relate to the field of calibration, and in particular, to a calibration method, device, and system.
  • Vehicle calibration refers to the process of optimizing the parameters in the vehicle controller software in order to meet the vehicle factory requirements, relevant international standards and industry standards after the controller hardware, software and related sensors are determined. At present, vehicle calibration mainly adopts manual calibration.
  • Figure 1 is a schematic diagram of a manual calibration method, as shown in Figure 1, the method includes the following steps: In the first step, the calibration engineer sets the calibration parameters according to experience, and downloads the calibration parameters to the controller through the calibration tool. In the second step, professional drivers drive the vehicle for testing and collect relevant data. In the third step, the professional driver drives the vehicle back to the initial position. In the fourth step, the calibration engineer conducts data analysis through the collected data and gives suggestions for adjusting the calibration parameters.
  • Embodiments of the present application provide a calibration method, device, and system, which can improve calibration efficiency and reduce calibration costs.
  • a calibration method includes: firstly, acquiring observation signals of a system to be tested. Secondly, calculate the corresponding performance index according to the observed signal of the system under test. Then, based on the performance index, the constraint condition, and at least one objective function, an optimization algorithm is used to obtain updated calibration parameters, and the above objective function is used to indicate the functional relationship between the performance indexes. Finally, the updated calibration parameters are sent to the controller.
  • the system to be tested may be a system of an automobile or a system of other devices.
  • the embodiment of the present application does not limit the specific type of the system to be tested and the specific device to which the system to be tested belongs. That is, the calibration method provided by this application can be used to calibrate automobiles, and can also be used to calibrate other equipment.
  • the performance index is calculated with the observation signal, and based on the performance index, constraint conditions, and at least one objective function, an optimization algorithm is used to obtain the updated calibration parameters, and the entire calibration process does not need Compared with manual calibration, the participation of calibration engineers can effectively reduce the cost of vehicle calibration and improve the calibration effect and efficiency.
  • the method further includes: the above-mentioned performance indicators include objective performance indicators and subjective performance indicators, and at least one objective function includes at least one of objective objective functions or subjective objective functions , the objective objective function is the weighted sum of multiple objective performance indicators, the sum of the weighted coefficients of multiple objective performance indicators is 1, the subjective objective function is the weighted sum of multiple subjective performance indicators, and multiple subjective perceptions The sum of the weighting coefficients of class performance indicators is 1.
  • the performance of the system under test is evaluated by establishing at least one objective function, the at least one objective function includes at least one of the objective objective function or the subjective objective function, the objective objective function is calculated and weighted based on the observed signal, and the subjective The class objective function is obtained by weighting the scores of professional drivers, so the calibration parameters optimized based on this multi-objective function not only make the performance indicators of the vehicle meet the industry standards, but also improve the driving experience of users.
  • the method further includes: determining the type of the calibration parameter.
  • the types of the calibration parameters include at least the Map class and the Table class. Based on the type of the calibration parameter and the initial value of the calibration parameter, an initial coupling coefficient between the calibration parameters is determined.
  • the coupling relationship between the Map class calibration parameters and the coupling relationship between the Table class calibration parameters can be obtained, and the optimization algorithm is used When optimizing, the optimized coupling coefficient can be used instead of directly optimizing the calibration parameters. Since the number of coupling coefficients is less than the number of calibration parameters, the number of calibration parameters can be reduced, the optimization time can be reduced, and the optimization efficiency can be improved.
  • the method further includes: based on the performance index, constraint conditions, at least one objective function, and the coupling coefficient before updating, using an optimization algorithm to obtain the updated coupling coefficient.
  • the coupling coefficient before updating includes the initial coupling coefficient.
  • the method further includes: the number of initial coupling coefficients is smaller than the number of calibration parameters.
  • the number of coupling coefficients is smaller than the number of calibration parameters. Therefore, in the optimization process of determining the calibration parameters, the calculation amount of the optimization algorithm can be reduced, the optimization time can be reduced, and the optimization efficiency can be improved.
  • the method further includes: when the observed signal satisfies the first preset condition, performing data processing on the observed signal, and calculating a corresponding performance index according to the processed observed signal.
  • the first preset condition may be a set range associated with the observed signal.
  • the observed signal is within the set range, the observed signal is processed; when the observed signal is not within the set range, the observed signal is not processed. to process.
  • the first preset condition can also be a set value associated with the observed signal, when the observed signal is greater than or equal to the set value, the observed signal is processed, and when the observed signal is smaller than the set value, the observed signal is not to process.
  • the embodiment of the present application does not limit the specific content and form of the first preset condition.
  • the data satisfying the first preset condition is processed, and the data not satisfying the first preset condition is not processed, which can ensure that the observed signals that meet the conditions participate in the calibration parameters.
  • the accuracy of the calibration parameters obtained by using the optimization algorithm can be ensured.
  • the method further includes: generating abnormal alarm information when an abnormality occurs in the calibration process of the system under test, and the abnormal alarm information is used to prompt the user that the calibration is abnormal.
  • an abnormal alarm message is generated in time to remind the user of the abnormal calibration, which can avoid the long-term waiting of the operator because he does not know the abnormal calibration, and the operator can continue to calibrate after handling the relevant abnormality, which can ensure Calibration efficiency.
  • the method further includes: generating a calibration report when the value of at least one objective function satisfies a second preset condition.
  • the calibration report includes at least one of performance indicators, distribution of objective function values, optimal calibration parameters, or important observation graphics.
  • the second preset condition may be a range of preset objective function values, and when the value of at least one objective function is within the range of preset objective function values, a calibration report is generated.
  • the embodiment of the present application does not limit the specific content and form of the second preset condition.
  • a calibration report can be generated when the calibration parameters are determined to meet the second preset condition, and the operator can better summarize experience and understand the current performance of the vehicle according to the content of the calibration report.
  • the method further includes that the optimization algorithm includes at least one of the following: a Bayesian optimization algorithm, a particle swarm optimization algorithm, a genetic algorithm, and a machine learning algorithm.
  • the optimization algorithm includes at least one of the following: a Bayesian optimization algorithm, a particle swarm optimization algorithm, a genetic algorithm, and a machine learning algorithm.
  • the method further includes: constraints may include constraints on observed signals, constraints on performance indicators, constraints on the value of an objective function, or constraints on coupling coefficients between calibration parameters at least one of the conditions.
  • the constraints can be multiple constraints, through which the optimization time of the calibration parameters can be shortened, and when the calibration parameters determined through the multiple constraints are applied to the equipment or system, it can have more good performance.
  • the method further includes: sending prompt information to the controller; the prompt information is used to prompt the user to at least one of calibration start or calibration end.
  • prompt information is sent to the controller of the device to be calibrated, and the device to be calibrated can prompt the user according to the prompt information, and the prompt can include sound, light and vibration.
  • the embodiment of the present application does not limit the specific type of the device to be calibrated to the user.
  • the calibration process can be made more humanized by prompting the user to start or end the calibration.
  • a calibration device includes: a transceiver module and a processing module.
  • the transceiver module is used to obtain the observation signal of the system under test.
  • the processing module is used to calculate the corresponding performance index according to the observed signal of the system under test.
  • the processing module is further configured to use an optimization algorithm to obtain updated calibration parameters based on performance indicators, constraint conditions, and at least one objective function.
  • the above-mentioned objective function is used to indicate a functional relationship between performance indicators.
  • the transceiver module is also used to send the updated calibration parameters to the controller.
  • the above-mentioned performance indicators include objective performance indicators and subjective performance indicators
  • the at least one objective function includes at least one of objective objective functions or subjective objective functions
  • objective The class objective function is the weighted sum of multiple objective class performance indicators, and the sum of the weighted coefficients of multiple objective class performance indicators is 1.
  • the subjective class objective function is the weighted sum of multiple subjective perception performance indicators. The sum of the weighting coefficients of the indicators is 1.
  • the processing module is further configured to determine the type of the calibration parameter, and the type of the calibration parameter includes a Map class and a Table class.
  • the processing module is also used to determine the initial coupling coefficient between the calibration parameters based on the type of the calibration parameter and the initial value of the calibration parameter.
  • the number of initial coupling coefficients is smaller than the number of calibration parameters.
  • the processing module is specifically configured to use an optimization algorithm to obtain an updated coupling coefficient based on performance indicators, constraints, at least one objective function, and a coupling coefficient before updating.
  • the coupling coefficient of includes the initial coupling coefficient.
  • the processing module is specifically configured to perform data processing on the observed signal when the observed signal satisfies the first preset condition, and calculate a corresponding performance index according to the processed observed signal.
  • the processing module is further configured to generate abnormal alarm information when the calibration process of the system under test is abnormal, and the abnormal alarm information is used to prompt the user for abnormal calibration.
  • the processing module is further configured to generate a calibration report when the value of at least one objective function satisfies a second preset condition.
  • the calibration report includes at least one of performance indicators, distribution of objective function values, optimal calibration parameters, or important observation graphics.
  • the above optimization algorithm includes at least one of the following: Bayesian optimization algorithm, particle swarm optimization algorithm, genetic algorithm, and machine learning algorithm.
  • the above constraints include the constraints of the observed signal, the constraints of the performance index, the constraints of the value of the objective function, or the constraints of the coupling coefficient between the calibration parameters. at least one.
  • the transceiver module is further configured to send prompt information to the controller; the prompt information is used to prompt the user to at least one of calibration start or calibration end.
  • a calibration device including: a memory and a processor; the memory is coupled to the processor; the memory is used to store computer program codes, and the computer program codes include computer instructions; when When the processor executes the computer instructions, the calibration device executes the calibration method provided in the first aspect above.
  • a fourth aspect of the embodiments of the present application provides a computer-readable storage medium including instructions.
  • the computer is made to execute the calibration method provided in the first aspect above.
  • a fifth aspect of the embodiments of the present application provides a computer program product, which, when the computer program product is run on a computer, causes the computer to execute the calibration method provided in the first aspect above.
  • a sixth aspect of the embodiments of the present application provides a calibration system, the calibration system includes a controller, and the calibration device as described in the second aspect above, and the controller is coupled to the calibration device.
  • a seventh aspect of the embodiments of the present application provides a vehicle, the vehicle includes a controller, and the calibration device as described in the second aspect above, and the controller is coupled to the calibration device.
  • the vehicle may be a new energy vehicle or a smart vehicle.
  • FIG. 1 is a schematic diagram of a manual calibration method provided in the embodiment of the present application.
  • Fig. 2 is a schematic diagram of a calibration method provided in the embodiment of the present application.
  • FIG. 3 is a schematic structural diagram of a calibration system provided in an embodiment of the present application.
  • FIG. 4 is a schematic structural diagram of a calibration device provided in an embodiment of the present application.
  • FIG. 5 is a schematic structural diagram of another calibration system provided in the embodiment of the present application.
  • FIG. 6 is a schematic flow chart of a calibration method provided in the embodiment of the present application.
  • FIG. 7 is a schematic flowchart of another calibration method provided in the embodiment of the present application.
  • FIG. 8 is a schematic diagram of the application of a calibration method provided in the embodiment of the present application.
  • Fig. 9 is a schematic diagram of the application of another calibration method provided by the embodiment of the present application.
  • FIG. 10 is a schematic flowchart of another calibration method provided in the embodiment of the present application.
  • FIG. 11 is a schematic flowchart of another calibration method provided in the embodiment of the present application.
  • FIG. 12 is a schematic flowchart of another calibration method provided in the embodiment of the present application.
  • FIG. 13 is a schematic structural diagram of another calibration device provided in the embodiment of the present application.
  • At least one item (piece) of a, b or c can represent: a, b, c, a and b, a and c, b and c, or, a and b and c, wherein a, b and c can be single or multiple.
  • words such as “first” and “second” are used to distinguish the same or similar items with basically the same function and effect, Those skilled in the art can understand that words such as “first” and “second” do not limit the quantity and execution order.
  • first in the first preset condition in the embodiment of the present application and “second” in the second preset condition are only used to distinguish different preset conditions.
  • the first, second, etc. descriptions that appear in the embodiments of this application are only for illustration and to distinguish the description objects, and there is no order, nor does it represent a special limitation on the number of devices in the embodiments of this application, and cannot constitute a limitation on the number of devices in this application. Any limitations of the examples.
  • the embodiment of this application provides a calibration method, which does not require the participation of calibration engineers, can not only improve the calibration efficiency, reduce the calibration cost, but also adopt the method of the embodiment of the application
  • the calibration parameters determined by the calibration method can improve the driving experience of the user.
  • the calibration method provided by the embodiment of this application can be applied to the calibration of various controllers or systems in vehicles, and can also be applied to the calibration of other equipment other than vehicles.
  • the following embodiments illustrate the calibration method and device provided in the embodiments of the present application by taking the calibration method applied to vehicle calibration as an example.
  • this calibration method can be used for electronic stability control system (electronic stability control system, ESC), vehicle state estimation system (vehicle state estimation, VSE), brake anti-lock braking system (antilock brake system, ABS), traction force Control system (traction control system, TCS), vehicle control unit system (vehicle control unit, VCU), thermal management system (thermal management system, TMS), electric power steering system (electric power steering, EPS), value-added function system ( value added function, VAF) and other systems for calibration.
  • electronic stability control system electronic stability control system
  • VSE vehicle state estimation system
  • VSE brake anti-lock braking system
  • ABS traction force Control system
  • TCS traction control system
  • vehicle control unit system vehicle control unit system
  • thermal management system thermal management system
  • electric power steering system electric power steering, EPS
  • value-added function system value added function, VAF
  • Fig. 2 is a calibration method provided by the embodiment of the present application.
  • the method may include the following steps:
  • the calibration device sends the calibration parameters to the controller.
  • professional drivers drive the vehicle for testing and collect relevant data.
  • the professional driver drives the vehicle back to the initial position.
  • the calibration equipment uses the optimization algorithm to analyze the data based on the collected data, and gives suggestions for adjusting the calibration parameters.
  • the calibration device sends the calibration parameters to the controller for the first time, it can set the initial value for each calibration parameter according to the experience of the calibration engineer, or it can obtain each calibration parameter from the cloud. The initial value of the parameter, and then send the initial calibration parameter to the controller.
  • the calibration device subsequently delivers calibration parameters to the controller, it can deliver the calibration parameters optimized by using the optimization algorithm to the controller.
  • Calibration devices in Figure 2 include, but are not limited to, cell phones, tablet computers, desktops, laptops, handheld computers, notebook computers, ultra-mobile personal computers (UMPCs), netbooks, and cellular phones, personal digital Assistant (personal digital assistant, PDA), augmented reality (augmented reality, AR) ⁇ virtual reality (virtual reality, VR) equipment and other electronic equipment, the calibration equipment is used to execute software codes or computer programs to achieve the embodiment of the application provided calibration method.
  • PDA personal digital Assistant
  • augmented reality augmented reality, AR
  • VR virtual reality
  • the calibration method shown in Figure 2 uses an optimization algorithm to calibrate the vehicle through the calibration equipment, which does not require the participation of calibration engineers, can reduce vehicle calibration costs, and improve calibration efficiency.
  • Fig. 3 is a schematic structural diagram of a calibration system provided by an embodiment of the present application.
  • the calibration system includes a calibration device and a vehicle, and the calibration device can be connected to a controller through a calibration tool.
  • the calibration device is used to execute software codes or computer programs to implement the calibration method provided in the embodiment of the present application.
  • the calibration device sends the calibration parameters to the controller, and the controller sends control information to the simulator/bench/real vehicle according to the calibration parameters, and the calibration device sends the control information to the simulator/bench/real vehicle according to the response (the first observation signal, the second Two observation signals), using the optimization algorithm to update the calibration parameters through continuous optimization and iteration, and finally get the optimal combination of calibration parameters.
  • the calibration system shown in Figure 3 does not require the participation of calibration engineers, can effectively reduce the cost of vehicle calibration, and improve the calibration effect and efficiency.
  • Fig. 4 is a calibration device provided in the embodiment of the present application.
  • the calibration device can be connected with a controller (for example, electronic control unit, ECU) through a calibration tool, or can be directly connected with the controller.
  • the calibration device includes: manual calibration experience digitization module, observation signal input module, calibration working condition start-stop recognition module, calibration data processing module, multi-objective calculation module, calibration parameter automatic optimization module, calibration parameter download module, calibration fault alarm module, Calibration post-processing module. The specific functions of the nine modules are introduced below.
  • the digital module of manual calibration experience can digitally extract the calibration experience of experts and calibration engineers through digital strategies, and can also digitally extract the empirical values of calibration parameters through digital strategies to guide optimization algorithms or machines
  • the learning algorithm finds the optimal combination of calibration parameters faster and better.
  • the embodiment of the present application does not limit the specific digitization strategy.
  • the following embodiment uses the digitization strategy including digitization of a reasonable range of calibration parameters, digitization of coupling relationships of calibration parameters, and digitization of constraint conditions as examples for illustration.
  • the digitization of the reasonable range of calibration parameters refers to the digitization of the more important calibration parameters and adjustment ranges accumulated by the calibration engineer based on experience to obtain the reasonable range of the upper and lower limits of the calibration parameters.
  • the calibration parameters can be optimized and solved according to the reasonable range of the upper and lower limits of the calibration parameters, which can improve the calibration efficiency.
  • the digitalization of the coupling relationship of calibration parameters refers to obtaining the coupling relationship between multiple calibration parameters according to the type of calibration parameters.
  • the types of the calibration parameters include Table class, Map class and scalar class.
  • the optimization of the calibration parameters of the Table class can be converted into the optimization of the coupling coefficients p and q, so the number of calibration parameters can be reduced, the optimization time can be reduced, and the optimization efficiency can be improved.
  • Digitization of constraints refers to the digitization of at least one of the constraints of the observed signal, the constraints of the performance index, the constraints of the objective function value, or the constraints of the coupling coefficient between the calibration parameters, which is used to guide the optimization algorithm to update Find the optimal combination of calibration parameters faster and better.
  • the empirical values of the calibration parameters, the range of the calibration parameters, the coupling relationship of the calibration parameters, constraints, objective functions, performance indicators, etc. obtained by the manual calibration experience digitization module can be stored in the calibration device or in the cloud.
  • the observation signal input module receives the observation signal from the controller through the calibration tool.
  • the specific type of the observed signal is related to the type of the system to be calibrated, and the embodiment of the present application does not limit the specific type of the observed signal.
  • the observation signal can be used in the start-stop identification module of the calibration working condition to determine the entry and exit of the calibration function, and can also be used in the multi-objective calculation module to determine the value of the objective function.
  • the start-stop recognition module of the calibration working condition comprehensively judges the entry and exit of the calibration function through the state of the function flag and the value of the relevant observation signal.
  • the calibration data processing module performs data processing on the input observation signal.
  • the calibration data processing module does not perform data processing on the input observation signal.
  • the embodiment of the present application does not limit the specific conditions for the start-stop identification module of the calibration working condition to determine the entry and exit of the calibration function.
  • the calibration data processing module performs validity judgment and data processing on the observation signal input by the observation signal input module, and the processed data can be used for the calculation of the objective function by the multi-objective calculation module.
  • data processing may include, but is not limited to, data dimensionality reduction, data interception, and filtering processing.
  • data processing may include, but is not limited to, data dimensionality reduction, data interception, and filtering processing.
  • the embodiment of the present application does not limit the specific manner of data processing.
  • the multi-objective calculation module calculates multiple performance indicators of the system to be calibrated based on the processed observation signals output by the calibration data processing module, and calculates at least one objective function based on multiple performance indicators, and the objective function is used to indicate the function between performance indicators relation.
  • the at least one objective function includes a subjective objective function and an objective objective function
  • the performance index may include an objective performance index and a subjective performance index.
  • the subjective objective function is the weighted sum of multiple subjective performance indicators, and the sum of the weighted coefficients of the multiple subjective performance indicators is 1.
  • the objective function of the objective class is the weighted sum of multiple objective class performance indexes, and the sum of the weight coefficients of the multiple objective class performance indexes is 1.
  • the calibration parameter optimization module uses an optimization algorithm or a machine learning algorithm to obtain updated calibration parameters according to the objective function value output by the multi-objective calculation module, the observation constraint conditions and the calibration parameter coupling conditions.
  • the calibration parameter download module writes the updated calibration parameters obtained by the calibration parameter optimization module into the controller by calling an application programming interface (API) of the calibration tool.
  • API application programming interface
  • the calibration fault alarm module is used to identify and alarm various faults that occur in the above calibration process.
  • the module can be activated to issue a calibration fault alarm and remind the calibration personnel to intervene.
  • the post-calibration processing module disconnects the signal and calibration parameter interaction between the automation calibration software and the controller by calling relevant commands, exits the functional automation calibration process, and automatically generates a calibration report based on the test data of all test groups.
  • the content of the calibration report may include performance evaluation indicators, distribution of objective function values, optimal calibration parameter value combinations, and graphic display of important observation signals.
  • the calibration device provided in the embodiment of the present application does not require the participation of calibration engineers when calibrating the systems or controllers of equipment such as vehicles through the above nine modules, which can improve calibration efficiency and reduce calibration costs. Moreover, in the process of optimizing the calibration parameters, the calibration device combines the subjective objective function and the coupling relationship between the calibration parameters. Therefore, the determined calibration parameters can not only improve the driving experience of the user, but also reduce the number of calibration parameters and reduce the number of calibration parameters. Optimize time and improve optimization efficiency.
  • the embodiment of the present application also provides a calibration system, as shown in FIG. 5 , the calibration system includes the calibration device in FIG. 4 , and a controller, and the controller is coupled to the calibration device.
  • the controller is used for sending observation signals to the calibration device and receiving calibration parameters from the calibration device.
  • FIG. 6 shows a calibration method provided by the embodiment of the present application, which can be executed by a calibration device. As shown in FIG. 6 , the method includes the following steps S601-S604.
  • the system to be tested is a system that needs to be calibrated.
  • the system to be tested may be a system of an automobile or a system of other devices.
  • the specific type of the system to be tested and the specific device to which the system to be tested belong are not limited.
  • the system to be tested can be the vehicle’s electronic stability control system ESC, vehicle state estimation system VSE, brake anti-lock braking system ABS, traction control system TCS, vehicle control unit system VCU, thermal management At least one system in the system TMS, electric power steering system EPS, value-added function system VAF and other systems.
  • vehicle electronic stability control system ESC
  • vehicle state estimation system VSE vehicle state estimation system
  • brake anti-lock braking system ABS brake anti-lock braking system ABS
  • traction control system TCS vehicle control unit system VCU
  • thermal management At least one system in the system TMS, electric power steering system EPS, value-added function system VAF and other systems.
  • TCS electronic stability control system
  • the observed signal may be a signal detected by a sensor in the system under test.
  • the observation signal may be a signal detected by the vehicle sensor and related to vehicle operation, and the observation signal may include parameters such as left front axle speed, right front axle speed, driving speed peak, driving speed trough, yaw rate, etc.
  • the embodiment of the present application does not limit the specific type of the observation signal, and the observation signal may be the same or different for different systems to be tested.
  • the above step S601 may include receiving an observation signal from a vehicle controller.
  • the observation signal in the vehicle controller can be a parameter collected by the sensor when the vehicle is running, and the parameter can be stored in the controller, and the observation signal can be read from the controller when the vehicle is calibrated.
  • the observation signal of the system under test obtained in the above step S601 may be the observation signal of the system under test detected by the sensor after the calibration device sends the initial calibration parameters to the controller and the vehicle runs based on the initial calibration parameters.
  • the observation signal of the system under test obtained in the above step S601 may also be the observation signal detected by the sensor after the vehicle controller operates based on the updated calibration parameters obtained in the following step S603.
  • Performance indicators can be used to evaluate the performance of the system under test.
  • the performance index of the system under test can be one or more.
  • the performance indicators of the system to be tested may include objective performance indicators and subjective performance indicators.
  • Objective performance indicators are related to enterprise internal standards, industry standards, international standards, customer performance indicators requirements and regulatory requirements.
  • Subjective performance indicators are related to the driver's driving experience.
  • the objective performance index can be represented by the functional relationship between multiple observation signals, and the specific value of the objective performance index can be obtained by substituting the observation signal into the preset performance index function.
  • Subjective performance indicators can be scored by professional drivers.
  • the objective performance indicators of the system to be tested may include front axle shaft speed performance index, driving speed fluctuation performance index, yaw rate performance index, stable driving shaft speed performance index, etc.
  • the subjective performance indicators of the system under test may include control jitter, acceleration performance, acceleration noise, driver handling, etc.
  • axle speed performance index of the front axle can be expressed by the following function:
  • vWhlobj sum((vWhlFL+vWhlFR)/2-vTarKarAxlFA);
  • vWhlobj is the front axle speed performance index
  • vWhlFL is the left front axle speed
  • vWhlFR is the right front axle speed
  • vTarKarAxlFA is the target axle speed
  • the target axle speed is a preset value.
  • the performance index of driving speed fluctuation can be expressed by the following function:
  • ⁇ nDriveobj sum(nDrivepeak[i]-nDrivevalley[i]);
  • ⁇ nDriveobj is the driving speed fluctuation performance index
  • nDrivepeak[i] is the driving speed peak
  • nDrive valley[i] is the driving speed valley
  • the yaw rate performance index can be expressed by the following function:
  • YawRateobj is the yaw rate performance index
  • YawRate is the yaw rate
  • the performance index of the stable drive shaft speed can be expressed by the following function:
  • vAxlWhlobj sum(0.85*Vxref+4-(vWhlFL+vWhlFR)/2);
  • vAxlWhlobj is the performance index of the stable drive shaft speed
  • Vxref is the reference shaft speed, which is a preset value
  • vWhlFL is the left front shaft speed
  • vWhlFR is the right front shaft speed.
  • an optimization algorithm is used to obtain updated calibration parameters, and the objective function is used to indicate a functional relationship between the performance indexes.
  • the latest performance metrics, constraints, and objective functions can be obtained from the cloud.
  • the functionality of the system under test can be evaluated using at least one objective function.
  • the at least one objective function may include an objective objective function and a subjective objective function, the objective objective function may be a weighted sum of multiple objective performance indicators, the sum of the weighted coefficients of multiple objective performance indicators is 1, and the subjective objective function
  • the function may be a weighted sum of multiple subjective performance indicators, and the sum of the weighted coefficients of multiple subjective performance indicators is 1.
  • the system to be tested includes four objective performance indicators, which are the front axle shaft speed performance index, the driving speed fluctuation performance index, the yaw angular velocity performance index, and the stable driving shaft speed performance index as an example.
  • the objective function of the system under test can be the following formula:
  • f objective c1*vWhlobj+c2* ⁇ nDriveobj+c3*YawRateobj+c4*vAxlWhlobj;
  • the system under test includes 4 subjective performance indicators, which are control jitter score, acceleration performance score, acceleration noise score, and driver controllability score as an example.
  • the system under test The objective function of subjective feeling class can be the following formula:
  • score control jitter is the score of control jitter
  • score acceleration performance is the score of acceleration performance
  • score acceleration noise is the score of acceleration noise
  • score driver controllability is the score of driver controllability
  • Constraints are conditions that need to be met during the calibration process.
  • the constraints can be obtained by digitizing industry standards, international standards, and regulatory standards, or by digitizing the adjustment range of calibration parameters accumulated by engineers based on experience.
  • the constraints can be represented by upper and lower thresholds.
  • the constraints may include at least one of constraints on observed signals, constraints on performance indicators, constraints on values of objective functions, or constraints on coupling coefficients between calibration parameters.
  • the constraint conditions of the observation signal can be obtained by digitizing the yaw rate constraints, steering wheel rotation angle constraints, and slip rate constraints in regulations and standards.
  • the optimization algorithm may include Bayesian optimization algorithm, particle swarm optimization algorithm, genetic algorithm, or machine learning algorithm.
  • the above step S603 may include: calculating the objective objective function and subjective objective function corresponding to TCS, by calling the Bayesian optimization algorithm, according to the current
  • the values of the objective objective function and the current subjective objective function, the constraints of the objective function, and the constraints of the calibration parameters are used to solve the optimization problem and generate optimized calibration parameters (that is, updated calibration parameters).
  • the optimized calibration parameters can be used for the next round of testing.
  • the calibration parameters of the same device in different countries and regions may be the same or different, which is not limited in this embodiment of the present application.
  • the initial values of the calibration parameters, constraints, and weighting coefficients of the objective function can be adjusted in a targeted manner, so that the determined calibration parameters are more suitable for the vehicle, and the performance of the vehicle can be better utilized.
  • the vehicle when the vehicle is used in Germany, the vehicle can be calibrated using performance indicators, constraints, at least one objective function and weighting coefficients of the at least one objective function of the German standard.
  • the vehicle when the vehicle is used in China, the vehicle can be calibrated using Chinese standard performance indicators, constraint conditions, at least one objective function and weighting coefficients of at least one objective function.
  • the calibration method provided by this application solves the optimization problem through an optimization algorithm or a machine learning algorithm, and can quickly and accurately obtain optimal calibration parameters, which can improve calibration efficiency and reduce calibration costs.
  • the latest initial value of the calibration parameter may be obtained from the cloud, and the latest initial value of the calibration parameter may be sent to the controller.
  • the controller receives the updated calibration parameters. Since the calibration parameters will affect the performance of the vehicle, when the updated calibration parameters are obtained, the vehicle can run again based on the updated calibration parameters, and further obtain the corresponding Based on the observation signal of the system to be tested, the corresponding performance index and objective function are calculated, and the above optimization algorithm is used to obtain the re-updated calibration parameters, and the re-updated calibration parameters are sent to the controller.
  • the vehicle can run again based on the updated calibration parameters, and the calibration device repeats the above steps S601-S604 until the difference between the value of at least one objective function and the preset objective function value is the smallest, and when the constraint conditions are met, the calibration at this time
  • the parameters are determined as the optimal calibration parameters.
  • the calibration method provided in the embodiment of the present application does not require the participation of a calibration engineer during calibration, which can improve calibration efficiency and reduce calibration costs. Moreover, the calibration device combines subjective objective functions in the process of optimizing the calibration parameters, so the determined calibration parameters can improve the driving experience of the user.
  • the updated calibration parameters are sent to the controller of the equipment to be calibrated. After the equipment to be calibrated is restarted, the equipment to be calibrated will run with the updated calibration parameters, and then perform a new round of calibration.
  • the embodiment of the present application also provides a calibration method, as shown in FIG. 7 , which may further include steps S605-S606 before the above step S601.
  • S605. Determine the type of the calibration parameter, where the type of the calibration parameter includes a Map class and a Table class.
  • the type of the calibration parameter may be preset, and the preset type may be determined according to the experience of the calibration engineer.
  • Map classTable class there is a coupling relationship between multiple calibration parameters.
  • the type of the calibration parameter may also include a scalar type, where the calibration parameters of the scalar type are independent calibration parameters, and there is no coupling relationship between the calibration parameters of the scalar type.
  • the number of initial coupling coefficients is smaller than the number of calibration parameters.
  • the initial value of the calibration parameter may be a value determined according to the experience of the calibration engineer.
  • the type of the calibration parameter is a Map class or a Table class
  • the optimization coupling coefficient can be used to replace the directly optimized calibration parameters. Since the number of coupling coefficients is much smaller than the number of calibration parameters, optimizing the coupling coefficients can greatly reduce the number of calibration parameters, reduce optimization time, and improve optimization efficiency compared with directly optimizing calibration parameters.
  • an initial value can be set for each Map-type calibration parameter, and corresponding input data x, y and output data z can be obtained according to the initial value of the calibration parameter.
  • an optimization algorithm which can effectively reduce the number of parameters that need to be calibrated , to improve the calibration efficiency.
  • the embodiment of the present application does not limit what kind of polynomial is used to fit the calibration parameters of the Map class.
  • an initial value can be set for each Table-type calibration parameter, and corresponding input data x and output data y can be obtained according to the initial value of the calibration parameter.
  • the embodiment of the present application does not limit which polynomial is used to fit the calibration parameters of the Table class.
  • step S603 includes: based on the performance index, constraint conditions, at least one objective function, and the coupling coefficient before updating, using an optimization algorithm to obtain the updated coupling coefficient.
  • the aforementioned coupling coefficients before updating include initial coupling coefficients.
  • the coupling coefficient before updating may be the initial coupling coefficient in step S606.
  • the initial coupling coefficient can be updated by using the optimization algorithm, and the updated coupling coefficient can be obtained.
  • the updated coupling coefficient is the updated calibration parameter. That is to say, compared with directly optimizing the calibration parameters, this solution can reduce the number of parameters to be calibrated by optimizing the coupling coefficient.
  • the coupling coefficient before updating is the coupling coefficient determined in the previous optimization.
  • the coupling coefficient before updating is the coupling coefficient determined for the first optimization.
  • the coupling coefficient before updating is the coupling coefficient determined for the second optimization.
  • step S607 may also be included after step S601 and before step S602.
  • the observation signal satisfying the first preset condition and the observation signal used for calculation in step S602 above may be the same observation signal, or may be a different observation signal, that is, the same observation signal may be used to determine whether to perform data processing, or may be used as
  • the object of data processing is not limited in this embodiment of the present application.
  • the first preset condition may be a set range associated with the observed signal.
  • the observed signal is within the set range, the observed signal is processed; when the observed signal is not within the set range, the observed signal is not processed. to process.
  • the first preset condition can also be a set value associated with the observed signal, when the observed signal is greater than or equal to the set value, the observed signal is processed, and when the observed signal is smaller than the set value, the observed signal is not to process.
  • the embodiment of the present application does not limit the specific content and form of the first preset condition, and the following embodiments take the first preset condition that the preset left front axle speed is K as an example for illustration.
  • the speed of the left front axle keeps increasing when the professional driver presses the accelerator to the bottom, and the speed of the left front axle increases continuously when the professional driver releases the accelerator.
  • the speed of the middle left front axle keeps decreasing.
  • the embodiment of the present application also provides a calibration method, as shown in FIG. 11 , which may further include step S608 in addition to the steps S601 to S607 described above.
  • corresponding abnormal alarm information can be generated to promptly prompt the user that the abnormal calibration needs to be dealt with, which can reduce the waiting time of the user because they do not know about the alarm, and can improve optimization efficiency.
  • the embodiment of the present application does not limit the sequence of execution of step S608 and steps S601 to S607.
  • An abnormal alarm may occur during execution of any of steps S601 to S607.
  • Figure 11 takes step S608 executed after step S604 as an example Give an example.
  • the embodiment of the present application also provides a calibration method, as shown in FIG. 12 , which may further include step S609 after the above steps S601 to S607.
  • the calibration report can be generated by various data processing in the optimization process, and the content of the calibration report can include performance evaluation indicators, distribution of objective function values, optimal calibration parameter groups, and graphic display of important observations.
  • the second preset condition may be a range of preset objective function values, and when the value of at least one objective function is within the range of preset objective function values, a calibration report is generated.
  • the embodiment of the present application does not limit the specific content and form of the second preset condition, and the following embodiments take the second preset condition as a preset objective function value as an example for illustration.
  • the preset objective function value is M
  • the preset subjective function value is N.
  • prompt information may be sent to the controller; the prompt information is used to prompt the user to at least one of calibration start or calibration end.
  • prompt information is sent to the controller of the device to be calibrated, and the device to be calibrated can prompt the user, and the prompt can include sound, light and vibration.
  • the embodiment of the present application does not limit the specific type of the device to be calibrated to the user.
  • the calibration device can send a prompt message to the controller of the device to be calibrated, and the device to be calibrated can send a prompt sound of "Calibration Start" to the user according to the prompt message, and when the calibration ends, the calibration device can Send prompt information to the controller of the device to be calibrated, and the device to be calibrated can send out a prompt sound of "calibration complete" to the user according to the prompt information.
  • the calibration method provided in the embodiment of this application solves the optimization problem through an optimization algorithm or a machine learning algorithm, and can quickly and accurately obtain the optimal calibration parameters.
  • This method is compatible with various types of equipment or systems, and can be used without The participation of calibration engineers is required to improve calibration efficiency and reduce calibration costs.
  • the method combines the subjective objective function, so the determined calibration parameters can improve the driving experience of the user.
  • the embodiment of the present application further provides a calibration device 130 .
  • the calibration device 130 may include a transceiver module 131 and a processing module 132 .
  • the calibration device 130 may further include a storage module 133, and the storage module 133 may store computer program codes for implementing the calibration method shown in FIG. 6 , FIG. 7 , FIG. 10 , FIG. 11 or FIG. 12 .
  • the transceiver module 131 is configured to acquire observation signals of the system under test.
  • the processing module 132 is configured to calculate the corresponding performance index according to the observed signal of the system under test.
  • the processing module 132 is further configured to use an optimization algorithm to obtain updated calibration parameters based on performance indicators, constraint conditions, and at least one objective function, and the above objective function is used to indicate a functional relationship between performance indicators.
  • the constraint conditions include at least one of constraint conditions of observed signals, constraints of performance indicators, constraints of objective function values, or constraints of coupling coefficients between calibration parameters.
  • the performance indicators include objective performance indicators and subjective performance indicators
  • at least one objective function includes at least one of objective function or subjective objective function
  • the objective function is the weighted weight of multiple objective performance indicators and, the sum of the weighted coefficients of multiple objective performance indicators is 1
  • the subjective objective function is the weighted sum of multiple subjective performance indicators
  • the sum of the weighted coefficients of multiple subjective performance indicators is 1.
  • the processing module 132 is further configured to determine the type of the calibration parameter, and the type of the calibration parameter includes a Map class and a Table class.
  • the processing module 132 is further configured to determine an initial coupling coefficient between calibration parameters based on the type of the calibration parameter and the initial value of the calibration parameter.
  • the number of initial coupling coefficients is smaller than the number of calibration parameters.
  • the processing module 132 is specifically configured to obtain the updated coupling coefficient by using an optimization algorithm based on the performance index, the constraint condition, at least one objective function, and the coupling coefficient before the update.
  • the coupling coefficient before updating includes the initial coupling coefficient.
  • the transceiver module 131 is also configured to send the updated calibration parameters to the controller.
  • the processing module 132 is specifically configured to perform data processing on the observed signal when the observed signal satisfies the first preset condition, and calculate a corresponding performance index according to the processed observed signal.
  • the processing module 132 is further configured to generate abnormality warning information when the calibration process of the system under test is abnormal, and the abnormality warning information is used to prompt the user that the calibration is abnormal.
  • the processing module 132 is further configured to generate a calibration report when the value of at least one objective function satisfies a second preset condition.
  • the calibration report includes at least one of performance indicators, distribution of objective function values, optimal calibration parameters, or important observation graphics.
  • the optimization algorithm includes Bayesian optimization algorithm, particle swarm optimization algorithm, genetic algorithm, or machine learning algorithm.
  • the transceiver module 131 is further configured to send prompt information to the controller; the prompt information is used to prompt the user to at least one of calibration start or calibration end.
  • the calibration equipment provided in the embodiment of the present application solves the optimization problem through an optimization algorithm or a machine learning algorithm, and can quickly and accurately obtain the optimal calibration parameters.
  • This method is compatible with various equipment or systems, and can be used without The participation of calibration engineers is required to improve calibration efficiency and reduce calibration costs.
  • the method combines the subjective objective function, so the determined calibration parameters can improve the driving experience of the user.
  • the embodiment of the present application also provides a calibration device, which includes a memory and a processor; the memory and the processor are coupled; the memory is used to store computer program codes, and the computer program codes include computer instructions; when the processor executes the computer instructions, the calibration The device executes the calibration method shown in FIG. 6 , FIG. 7 , FIG. 10 , FIG. 11 or FIG. 12 .
  • the embodiment of the present application also provides a computer-readable storage medium, where computer program code is stored in the computer-readable storage medium.
  • the electronic device executes the steps shown in Fig. 6 , Fig. 7 , Fig. 10 , The calibration method shown in Figure 11 or Figure 12.
  • the embodiment of the present application also provides a computer program product.
  • the computer program product When the computer program product is run on a computer, the computer is made to execute the calibration method shown in FIG. 6 , FIG. 7 , FIG. 10 , FIG. 11 or FIG. 12 .
  • An embodiment of the present application further provides a vehicle, the vehicle includes a controller, and the calibration device as shown in FIG. 13 , the controller is coupled to the calibration device.
  • the vehicle may be a new energy vehicle or a smart vehicle.
  • the steps of the methods or algorithms described in connection with the disclosure of this application can be implemented in the form of hardware, or can be implemented in the form of a processor executing software instructions.
  • Software instructions can be composed of corresponding software modules, and software modules can be stored in random access memory (Random Access Memory, RAM), flash memory, erasable programmable read-only memory (Erasable Programmable ROM, EPROM), electrically erasable Programmable read-only memory (Electrically EPROM, EEPROM), registers, hard disk, removable hard disk, CD-ROM, or any other form of storage medium known in the art.
  • An exemplary storage medium is coupled to the processor such the processor can read information from, and write information to, the storage medium.
  • the storage medium can also be a component of the processor.
  • the processor and storage medium can be located in the ASIC.
  • the ASIC may be located in the core network interface device.
  • the processor and the storage medium may also exist in the core network interface device as discrete components.
  • the functions described in the present invention may be implemented by hardware, software, firmware or any combination thereof.
  • the functions may be stored on or transmitted over as one or more instructions or code on a computer-readable medium.
  • Computer-readable media includes both computer storage media and communication media including any medium that facilitates transfer of a computer program from one place to another.
  • a storage media may be any available media that can be accessed by a general purpose or special purpose computer.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Chemical & Material Sciences (AREA)
  • Combustion & Propulsion (AREA)
  • Transportation (AREA)
  • Mechanical Engineering (AREA)
  • Automation & Control Theory (AREA)
  • Testing Or Calibration Of Command Recording Devices (AREA)

Abstract

本申请实施例公开了一种标定方法、装置及系统，涉及标定领域，能够提高新能源汽车、智能汽车的标定效率，降低标定成本。具体方案为：首先，获取待测系统的观测信号。其次，根据待测系统的观测信号计算对应的性能指标。然后，基于性能指标、约束条件，以及至少一个目标函数，采用优化算法得到更新后的标定参数，其中目标函数用于指示性能指标之间的函数关系。最后，向控制器发送更新后的标定参数。

Description

一种标定方法、装置及系统 技术领域
本申请实施例涉及标定领域，尤其涉及一种标定方法、装置及系统。
背景技术
车辆标定是指在控制器硬件、软件和相关传感器等器件确定之后，为了满足车辆出厂要求、相关国际标准及行业标准，对车辆控制器软件中的参数进行优化的过程。目前，车辆标定主要采用人工标定方式。
图1为一种人工标定方法的示意图，如图1所示，该方法包括以下步骤：第一步，标定工程师根据经验设置标定参数，并通过标定工具将标定参数下载至控制器中。第二步，专业驾驶员驾驶车辆进行测试，并收集相关数据。第三步，专业驾驶员驾驶车辆回到初始位置。第四步，标定工程师通过收集的数据进行数据分析，并给出标定参数的调整建议。
由于人工标定时不同标定工程师的标定结果差异很大，一致性较差，因此对标定工程师的要求较高。而且人工标定很难对全局进行标定，只能保证局部性能最优，再者人工标定需要耗费大量的人力、物力和财力。因此，人工标定具有成本高、效果差、效率低的问题。
发明内容
本申请实施例提供一种标定方法、装置及系统，能够提高标定效率，降低标定成本。
本申请实施例采用如下技术方案：
本申请实施例的第一方面，提供一种标定方法，该标定方法包括：首先，获取待测系统的观测信号。其次，根据待测系统的观测信号计算对应的性能指标。然后，基于性能指标、约束条件，以及至少一个目标函数，采用优化算法得到更新后的标定参数，上述目标函数用于指示性能指标之间的函数关系。最后，向控制器发送更新后的标定参数。
可选的，待测系统可以为汽车的系统，也可以为其他设备的系统，本申请实施例对于待测系统的具体类型，待测系统所属的具体设备并不限定。即本申请提供的标定方法可以用于标定汽车，也可以用于标定其他设备。
基于本方案，通过获取待测系统的观测信号，用观测信号计算性能指标，并基于性能指标、约束条件，以及至少一个目标函数，采用优化算法得到更新后的标定参数，整个标定过程可以不需要标定工程师的参与，与人工标定相比，能够有效降低车辆标定成本，提高标定效果和标定效率。
结合第一方面，在一种实现方式中，该方法还包括：上述性能指标包括客观类性能指标和主观感受类性能指标，至少一个目标函数包括客观类目标函数或主观类目标函数中的至少一个，客观类目标函数为多个客观类性能指标的加权和，多个客观类性 能指标的加权系数之和为1，主观类目标函数为多个主观感受类性能指标的加权和，多个主观感受类性能指标的加权系数之和为1。
基于本方案，通过建立至少一个目标函数来评价待测系统的性能，该至少一个目标函数包括客观类目标函数或主观类目标函数中的至少一个，客观类目标函数基于观测信号计算加权得到，主观类目标函数通过专业驾驶员的打分加权得到，从而基于该多目标函数优化得到的标定参数，不仅使得车辆的性能指标能够满足行业标准，而且能够提升用户的驾驶体验。
结合第一方面，在一种实现方式中，该方法还包括：确定标定参数的类型。标定参数的类型至少包括Map类和Table类。基于标定参数的类型和标定参数的初始值，确定标定参数之间的初始耦合系数。
基于本方案，通过确定标定参数的类型，将Map类和Table类的标定参数分别进行拟合，可以得到Map类标定参数之间的耦合关系和Table类标定参数之间的耦合关系，采用优化算法优化时可以用优化耦合系数代替直接优化标定参数，由于耦合系数的数量少于标定参数的数量，因此能够降低标定参数的数量，减少优化时间，提升优化效率。
结合第一方面，在一种实现方式中，该方法还包括：基于性能指标、约束条件、至少一个目标函数，以及更新前的耦合系数，采用优化算法得到更新后的耦合系数。更新前的耦合系数包括初始耦合系数。
基于本方案，通过优化耦合系数代替直接优化标定参数，在确定标定参数的优化过程中，能够降低优化算法的运算量，可以减少优化时间，提升优化效率。
结合第一方面，在一种实现方式中，该方法还包括：初始耦合系数的数量小于标定参数的数量。
基于本方案，通过优化耦合系数代替直接优化标定参数，偶合系数的数量小于标定参数的数量。因此，在确定标定参数的优化过程中，能够降低优化算法的运算量，可以减少优化时间，提升优化效率。
结合第一方面，在一种实现方式中，该方法还包括：在观测信号满足第一预设条件时，对观测信号进行数据处理，并根据处理后的观测信号计算对应的性能指标。
可选的，第一预设条件可以是与观测信号关联的一个设定范围，当观测信号在设定范围内时，对观测信号进行处理，当观测信号不在设定范围内时，不对观测信号进行处理。
可选的，第一预设条件还可以是与观测信号关联的一个设定值，当观测信号大于等于设定值时，对观测信号进行处理，当观测信号小于设定值时，不对观测信号进行处理。本申请实施例对于第一预设条件的具体内容及形式并不限定。
基于本方案，通过设置第一预设条件，对满足第一预设条件的数据进行处理，对不满足第一预设条件的数据不进行处理，可以确保符合条件的观测信号参与到标定参数的优化过程中，能够确保采用优化算法得到的标定参数的准确性。
结合第一方面，在一种实现方式中，该方法还包括：当待测系统的标定过程出现异常时生成异常告警信息，异常告警信息用于提示用户标定异常。
基于本方案，在标定过程中出现异常时，及时生成异常告警信息，提示用户标定 异常，可以避免操作人员因为不知道标定异常导致长时间等待，操作人员处理相关异常后可以继续进行标定，能够确保标定效率。
结合第一方面，在一种实现方式中，该方法还包括：在至少一个目标函数的数值满足第二预设条件时，生成标定报告。该标定报告包括性能指标、目标函数值分布、最佳标定参数，或重要观测图形中的至少一种。
可选的，第二预设条件可以为预设目标函数值的范围，当至少一个目标函数的数值在预设目标函数值的范围内时，生成标定报告。本申请实施例对于第二预设条件的具体内容及形式并不限定。
基于本方案，通过设置第二预设条件，当满足第二预设条件标定参数确定后，可以生成标定报告，操作人员根据标定报告的内容，可以更好的总结经验，了解车辆当前的性能。
结合第一方面，在一种实现方式中，该方法还包括，优化算法包括以下至少一种：贝叶斯优化算法、粒子群优化算法、遗传算法、机器学习算法。
基于本方案，通过采用贝叶斯优化算法、粒子群优化算法、遗传算法、或机器学习算法来优化标定参数，相较于人工标定，不但可以对更多的标定参数进行优化，而且优化速度更快，采用优化算法确定标定参数时还不需要标定工程师参与，能够有效降低车辆标定成本，提高标定效果和标定效率。
结合第一方面，在一种实现方式中，该方法还包括：约束条件可以包括观测信号的约束条件、性能指标的约束条件、目标函数数值的约束条件、或标定参数之间的耦合系数的约束条件中的至少一种。
基于本方案，约束条件可以为多个方面的约束条件，通过该多个约束条件可以缩短标定参数的优化时间，并且通过该多个约束条件确定的标定参数应用于设备或系统时，可以拥有更好的性能。
结合第一方面，在一种实现方式中，该方法还包括：向控制器发送提示信息；该提示信息用于提示用户标定开始或标定结束中的至少一项。
可选的，向待标定设备的控制器发送提示信息，待标定的设备根据提示信息可以对用户进行提示，该提示可以包括声、光和振动。本申请实施例对于待标定设备对用户提示的具体类型并不限定。
基于本方案，通过提示用户标定开始或标定结束，可以让标定过程更人性化。
本申请实施例的第二方面，提供一种标定装置，该标定装置包括：收发模块和处理模块。收发模块，用于获取待测系统的观测信号。处理模块，用于根据待测系统的观测信号计算对应的性能指标。处理模块还用于基于性能指标、约束条件，以及至少一个目标函数，采用优化算法得到更新后的标定参数，上述目标函数用于指示性能指标之间的函数关系。收发模块还用于向控制器发送更新后的标定参数。
结合第二方面，在一种可能的实现方式中，上述性能指标包括客观类性能指标和主观感受类性能指标，该至少一个目标函数包括客观类目标函数或主观类目标函数中的至少一个，客观类目标函数为多个客观类性能指标的加权和，多个客观类性能指标的加权系数之和为1，主观类目标函数为多个主观感受类性能指标的加权和，多个主观感受类性能指标的加权系数之和为1。
结合第二方面，在一种可能的实现方式中，处理模块还用于确定标定参数的类型，标定参数的类型包括Map类和Table类。处理模块还用于基于标定参数的类型和标定参数的初始值，确定标定参数之间的初始耦合系数。
结合第二方面，在一种可能的实现方式中，初始耦合系数的数量小于标定参数的数量。
结合第二方面，在一种可能的实现方式中，处理模块具体用于基于性能指标、约束条件、至少一个目标函数，以及更新前的耦合系数，采用优化算法得到更新后的耦合系数，更新前的耦合系数包括初始耦合系数。
结合第二方面，在一种可能的实现方式中，处理模块具体用于在观测信号满足第一预设条件时，对观测信号进行数据处理，并根据处理后的观测信号计算对应的性能指标。
结合第二方面，在一种可能的实现方式中，处理模块还用于当待测系统的标定过程出现异常时生成异常告警信息，异常告警信息用于提示用户标定异常。
结合第二方面，在一种可能的实现方式中，处理模块还用于在至少一个目标函数的数值满足第二预设条件时，生成标定报告。该标定报告包括性能指标、目标函数值分布、最佳标定参数，或重要观测图形中的至少一种。
结合第二方面，在一种可能的实现方式中，上述优化算法包括以下至少一种：贝叶斯优化算法、粒子群优化算法、遗传算法、机器学习算法。
结合第二方面，在一种可能的实现方式中，上述约束条件包括观测信号的约束条件、性能指标的约束条件、目标函数数值的约束条件、或标定参数之间的耦合系数的约束条件中的至少一种。
结合第二方面，在一种可能的实现方式中，收发模块还用向控制器发送提示信息；提示信息用于提示用户标定开始或标定结束中的至少一项。
本申请实施例的第三方面，提供一种标定设备，包括：存储器和处理器；存储器和所述处理器耦合；所述存储器用于存储计算机程序代码，所述计算机程序代码包括计算机指令；当所述处理器执行所述计算机指令时，所述标定设备执行如上述第一方面提供的标定方法。
本申请实施例的第四方面，提供一种计算机可读存储介质，包括指令。当指令在计算机上运行时，使得计算机执行如上述第一方面提供的标定方法。
本申请实施例的第五方面，提供一种计算机程序产品，当计算机程序产品在计算机上运行时，使得计算机执行如上述第一方面提供的标定方法。
本申请实施例的第六方面，提供一种标定系统，该标定系统包括控制器，以及如上述第二方面所述的标定装置，所述控制器与所述标定装置之间耦合。
本申请实施例的第七方面，提供一种车辆，该车辆包括控制器，以及如上述第二方面所述的标定装置，所述控制器与所述标定装置之间耦合。可选的，该车辆可以为新能源汽车或智能汽车等。
本申请中第二方面、第三方面、第四方面、第五方面、第六方面和第七方面的描述，可以参考第一方面的详细描述；并且，第二方面、第三方面、第四方面、第五方面、第六方面和第七方面的有益效果，可以参考第一方面的有益效果分析，此处不再 赘述。
附图说明
图1为本申请实施例提供的一种人工标定方法的示意图；
图2为本申请实施例提供的一种标定方法的示意图；
图3为本申请实施例提供的一种标定系统的结构示意图；
图4为本申请实施例提供的一种标定装置的结构示意图；
图5为本申请实施例提供的另一种标定系统的结构示意图；
图6为本申请实施例提供的一种标定方法的流程示意图；
图7为本申请实施例提供的另一种标定方法的流程示意图；
图8为本申请实施例提供的一种标定方法的应用示意图；
图9为本申请实施例提供的另一种标定方法的应用示意图；
图10为本申请实施例提供的又一种标定方法的流程示意图；
图11为本申请实施例提供的再一种标定方法的流程示意图；
图12为本申请实施例提供的再一种标定方法的流程示意图；
图13为本申请实施例提供的另一种标定装置的结构示意图。
具体实施方式
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行描述。在本申请中，“至少一个”是指一个或者多个，“多个”是指两个或两个以上。“和/或”，描述关联对象的关联关系，表示可以存在三种关系，例如，A和/或B，可以表示：单独存在A，同时存在A和B，单独存在B的情况，其中A，B可以是单数或者复数。字符“/”一般表示前后关联对象是一种“或”的关系。“以下至少一项(个)”或其类似表达，是指的这些项中的任意组合，包括单项(个)或复数项(个)的任意组合。例如，a，b或c中的至少一项(个)，可以表示：a，b，c，a和b，a和c，b和c，或，a和b和c，其中a、b和c可以是单个，也可以是多个。另外，为了便于清楚描述本申请实施例的技术方案，在本申请的实施例中，采用了“第一”、“第二”等字样对功能和作用基本相同的相同项或相似项进行区分，本领域技术人员可以理解“第一”、“第二”等字样并不对数量和执行次序进行限定。比如，本申请实施例中的第一预设条件的“第一”和第二预设条件中的“第二”仅用于区分不同的预设条件。本申请实施例中出现的第一、第二等描述，仅作示意与区分描述对象之用，没有次序之分，也不表示本申请实施例中对设备个数的特别限定，不能构成对本申请实施例的任何限制。
需要说明的是，本申请中，“示例性的”或者“例如”等词用于表示作例子、例证或说明。本申请中被描述为“示例性的”或者“例如”的任何实施例或设计方案不应被解释为比其他实施例或设计方案更优选或更具优势。确切而言，使用“示例性的”或者“例如”等词旨在以具体方式呈现相关概念。
为了解决人工标定成本高、效果差、效率低的问题，本申请实施例提供一种标定方法，该方法不需要标定工程师参与，不仅能够提高标定效率，降低标定成本，而且采用本申请实施例的标定方法确定的标定参数，能够提升用户的驾驶体验。
本申请实施例提供的标定方法可以适用于车辆中各种控制器或系统的标定，也可以适用于除车辆以外的其他设备的标定，本申请实施例对于该标定方法标定的具体设 备的类型并不限定，下述实施例以该标定方法应用于车辆标定为例对本申请实施例提供的标定方法及装置进行说明。例如，该标定方法可以对车辆中的电子稳定控制系统(electronic stability control system，ESC)、车辆状态估计系统(vehicle state estimation，VSE)、制动防抱死系统(antilock brake system，ABS)、牵引力控制系统(traction control system，TCS)、整车控制单元系统(vehicle control unit，VCU)、热管理系统(thermal management system，TMS)、电动助力转向系统(electric power steering，EPS)、增值功能系统(value added function，VAF)等系统进行标定。
图2为本申请实施例提供的一种标定方法，如图2所示，该方法可以包括以下步骤：第一步，标定设备将标定参数下发至控制器中。第二步，专业驾驶员驾驶车辆进行测试，并收集相关数据。第三步，专业驾驶员驾驶车辆回到初始位置。第四步，标定设备根据收集的数据，采用优化算法进行数据分析，并给出标定参数的调整建议。为了更快更好的找到最优的标定参数，标定设备首次向控制器下发标定参数时，可以根据标定工程师的经验给每个标定参数设定初始值，或者，可以从云端获取每个标定参数的初始值，然后将该初始标定参数下发至控制器中。标定设备在后续向控制器下发标定参数时，可以将采用优化算法优化后的标定参数下发至控制器。
图2中的标定设备包括但不限于手机、平板电脑、桌面型、膝上型、手持计算机、笔记本电脑、超级移动个人计算机(ultra-mobile personal computer，UMPC)、上网本，以及蜂窝电话、个人数字助理(personal digital assistant，PDA)、增强现实(augmented reality，AR)\虚拟现实(virtual reality，VR)设备等电子设备，该标定设备用于执行软件代码或计算机程序以实现本申请实施例提供的标定方法。
相较于图1所示的采用人工标定的方式，图2所示的标定方法通过标定设备采用优化算法对车辆进行标定，可以不需要标定工程师的参与，能够降低车辆标定成本，提高标定效率。
图3为本申请实施例提供的一种标定系统的结构示意图，如图3所示，该标定系统包括标定设备和车辆，标定设备可以通过标定工具与控制器连接。该标定设备用于执行软件代码或计算机程序以实现本申请实施例提供的标定方法。标定设备将标定参数下发至控制器，控制器根据标定参数向模拟器/台架/实车下发控制信息，标定设备根据模拟器/台架/实车的响应(第一观测信号、第二观测信号)，采用优化算法通过不断优化迭代更新标定参数，最终得到标定参数的最优组合。图3所示的标定系统相较于人工标定而言，不需要标定工程师的参与，能够有效降低车辆标定成本，提高标定效果和标定效率。
图4为本申请实施例提供的一种标定装置。如图4所示，该标定装置可以通过标定工具与控制器(例如，electronic control unit，ECU)连接，也可以直接与控制器连接。该标定装置包括：人工标定经验数字化模块、观测信号输入模块、标定工况启停识别模块、标定数据处理模块、多目标计算模块、标定参数自动优化模块、标定参数下载模块、标定故障告警模块、标定后处理模块。下面对该九个模块的具体功能进行介绍。
人工标定经验数字化模块作为标定装置的输入模块，可以通过数字化策略将专家及标定工程师的标定经验进行数字化提取，还可以通过数字化策略将标定参数的经验 值进行数字化提取，用于指导优化算法或机器学习算法更快更好的找到标定参数的最优组合。本申请实施例对于具体的数字化策略并不限定，下述实施例以数字化策略包括标定参数合理范围数字化、标定参数耦合关系数字化、约束条件数字化为例进行示例性说明。
标定参数合理范围数字化是指将标定工程师根据经验积累的比较重要的标定参数及调节范围进行数字化，得到标定参数的上下限合理范围。在对标定参数优化过程中，可以依据该标定参数的上下限合理范围对标定参数进行优化求解，能够提高标定效率。
标定参数耦合关系数字化是指根据标定参数的类型得到多个标定参数之间的耦合关系。该标定参数的类型包括Table类、Map类和标量类。
在一些示例中，由于Table类标定参数之间存在耦合关系，因此可以将Table类标定参数以多项式函数(如y＝ax2+bx+c)的方式进行拟合，得到耦合系数a，b，c，在对标定参数进行优化时，可以将Table类的标定参数的优化转换为对耦合系数a，b，c的优化，因此能够减降低标定参数的数量，减少优化时间，提升优化效率。
在另一些示例中，由于Map类标定参数之间存在耦合关系，因此可以将Map类标定对象以多项式函数(如z＝x2/2p+y2/2q)的方式进行拟合，得到耦合系数p、q，在对标定参数进行优化时，可以将Table类的标定参数的优化转换为对耦合系数p、q的优化，因此能够减降低标定参数的数量，减少优化时间，提升优化效率。
约束条件数字化是指将观测信号的约束条件、性能指标的约束条件、目标函数数值的约束条件、或标定参数之间的耦合系数的约束条件中的至少一种进行数字化，用于指导优化算法更快更好的找到标定参数的最优组合。
可选的，人工标定经验数字化模块得到的标定参数的经验值、标定参数的范围、标定参数的耦合关系、约束条件、目标函数、性能指标等可以存储在标定装置中，也可以存储在云端。
观测信号输入模块通过标定工具接收来自控制器的观测信号。该观测信号的具体类型与待标定的系统的类型有关，本申请实施例对于观测信号的具体类型并不限定。该观测信号可以用于标定工况启停识别模块确定标定功能的进入与退出，也可以用于多目标计算模块确定目标函数的数值。
标定工况启停识别模块通过功能标志位的状态及相关观测信号的数值综合判断标定功能的进入与退出。当标定工况启停识别模块确定进入标定功能时，标定数据处理模块对输入的观测信号进行数据处理。当标定工况启停识别模块确定退出标定功能时，标定数据处理模块对输入的观测信号不进行数据处理。本申请实施例对于标定工况启停识别模块确定标定功能的进入与退出的具体条件并不限定。
标定数据处理模块对观测信号输入模块输入的观测信号进行有效性判断及数据处理，处理后的数据可以用于多目标计算模块进行目标函数的计算。
可选的，数据处理可以包括但不限于数据降维、数据截取、滤波处理，本申请实施例对于数据处理的具体方式并不限定。
多目标计算模块基于标定数据处理模块输出的处理后的观测信号，计算待标定系统的多个性能指标，并根据多个性能指标计算至少一个目标函数，目标函数用于指示性能指标之间的函数关系。
可选的，本申请实施例中的目标函数可以为至少一个，该至少一个目标函数包括主观类目标函数和客观类目标函数，性能指标可以包括客观类性能指标和主观感受类性能指标。主观类目标函数为多个主观感受类性能指标的加权和，该多个主观感受类性能指标的加权系数之和为1。客观类目标函数为多个客观类性能指标的加权和，该多个客观类性能指标的加权系数之和为1。
标定参数优化模块根据多目标计算模块输出的目标函数值、观测量约束条件和标定参数耦合条件，采用优化算法或机器学习算法，得到更新后的标定参数。
标定参数下载模块通过调用标定工具的应用程序接口(application programming interface，API)将标定参数优化模块得到的更新后的标定参数写入到控制器中。
标定故障告警模块用于对上述标定过程中出现的各种故障进行识别及告警。当标定功能在自动化标定过程中遇到异常场景时，可以激活该模块进行标定故障告警，提醒标定人员介入。
标定后处理模块通过调用相关命令断开自动化标定软件与控制器间的信号及标定参数交互，退出功能自动化标定流程，并根据所有测试组别的测试数据自动生成标定报告。
可选的，标定报告内容可以包括性能评价指标、目标函数值分布、最佳标定参数值组合、重要观测信号图形显示。
本申请实施例提供的标定装置，通过上述九个模块对车辆等设备的系统或控制器进行标定时，不需要标定工程师的参与，能够提高标定效率，降低标定成本。而且该标定装置在对标定参数优化过程中，结合了主观类目标函数以及标定参数之间的耦合关系，因此确定的标定参数不仅能够提升用户的驾驶体验，而且能够减降低标定参数的数量，减少优化时间，提升优化效率。
本申请实施例还提供一种标定系统，如图5所示，该标定系统包括图4中的标定装置，以及控制器，控制器与标定装置之间耦合连接。控制器用于向标定装置发送观测信号，并接收来自标定装置的标定参数。
图6为本申请实施例提供的一种标定方法，该方法可以由标定设备执行，如图6所示，该方法包括以下步骤S601～S604。
S601、获取待测系统的观测信号。
待测系统可以为一个，也可以为多个。该待测系统为需要标定的系统。该待测系统可以为汽车的系统，也可以为其他设备的系统，本申请实施例对于待测系统的具体类型，待测系统所属的具体设备并不限定。
例如，以标定车辆为例，该待测系统可以为车辆的电子稳定控制系统ESC、车辆状态估计系统VSE、制动防抱死系统ABS、牵引力控制系统TCS、整车控制单元系统VCU、热管理系统TMS、电动助力转向系统EPS、增值功能系统VAF等系统中的至少一个系统。下述实施例以待测系统为TCS为例进行示例性说明。
可选的，观测信号可以为待测系统中的传感器检测到的信号。例如，观测信号可以为车辆的传感器检测到的与车辆运行相关的信号，该观测信号可以包括左前轴速、右前轴速、驱动转速波峰、驱动转速波谷、横摆角速度等参数。本申请实施例对于观测信号的具体类型并不限定，对于不同的待测系统，观测信号可以相同也可以不同。
在一些示例中，上述步骤S601可以包括接收来自车辆控制器的观测信号。该车辆控制器中的观测信号可以为车辆运行时传感器采集的参数，该参数可以存储到控制器中，并在对车辆标定时从控制器读出观测信号。
由于标定参数是影响车辆性能的重要指标，当标定参数不同时，获取的待测系统的观测信号也会随之改变。上述步骤S601中获取的待测系统的观测信号可以是标定设备将初始标定参数下发至控制器以后，车辆基于该初始标定参数运行后，传感器检测到的待测系统的观测信号。上述步骤S601中获取的待测系统的观测信号也可以是车辆控制器基于下述步骤S603得到的更新后的标定参数运行后，传感器检测到的观测信号。
S602、根据待测系统的观测信号计算对应的性能指标。
性能指标可以用来评价待测系统的性能。待测系统的性能指标可以为一个或多个。
待测系统的性能指标可以包括客观类性能指标和主观感受类性能指标。客观类性能指标与企业内部标准、行业标准、国际标准、客户性能指标要求及法规要求等有关。主观感受类性能指标与驾驶员的驾驶感受有关。
客观类性能指标可以通过多个观测信号之间的函数关系表示，将观测信号代入预设的性能指标函数可以得到该客观类性能指标的具体数值。主观感受类性能指标可以通过专业的驾驶人员打分得到。
示例性的，以待测系统为TCS为例，该待测系统的客观类性能指标可以包括前轴轴速性能指标、驱动转速波动性能指标、横摆角速度性能指标、稳定驱动轴速性能指标等。该待测系统的主观感受类性能指标可以包括控制抖动、加速性能、加速噪声、驾驶员操控性等。
例如，前轴轴速性能指标可以用如下函数表示：
vWhlobj＝sum((vWhlFL+vWhlFR)/2-vTarKarAxlFA)；
其中，vWhlobj为前轴轴速性能指标，vWhlFL为左前轴速，vWhlFR为右前轴速，vTarKarAxlFA为目标轴速，该目标轴速为预设值。根据步骤S601获取的观测信号左前轴速、右前轴速，结合前轴轴速性能指标函数，可以得到前轴轴速性能指标的数值。
再例如，驱动转速波动性能指标可以用如下函数表示：
ΔnDriveobj＝sum(nDrivepeak[i]-nDrive valley[i])；
其中，ΔnDriveobj为驱动转速波动性能指标，nDrivepeak[i]为驱动转速波峰，nDrive valley[i]为驱动转速波谷。根据步骤S601获取的观测信号驱动转速波峰、驱动转速波谷，结合驱动转速波动性能指标函数，可以得到驱动转速波动性能指标的数值。
再例如，横摆角速度性能指标可以用如下函数表示：
YawRateobj＝sum(abs(YawRate))；
其中，YawRateobj为横摆角速度性能指标，YawRate为横摆角速度。根据步骤S601获取的观测信号横摆角速度，结合横摆角速度性能指标函数，可以得到横摆角速度性能指标的数值。
再例如，稳定驱动轴速性能指标指标可以用如下函数表示：
vAxlWhlobj＝sum(0.85*Vxref+4-(vWhlFL+vWhlFR)/2)；
其中，vAxlWhlobj为稳定驱动轴速性能指标，Vxref为参考轴速，该参考轴速为预设值，vWhlFL为左前轴速，vWhlFR为右前轴速。根据步骤S601获取的观测信号 左前轴速、右前轴速，结合稳定驱动轴速性能指标函数，可以得到稳定驱动轴速性能指标的数值。
S603、基于性能指标、约束条件，以及至少一个目标函数，采用优化算法得到更新后的标定参数，该目标函数用于指示性能指标之间的函数关系。
可选的，可以从云端获取最新的性能指标、约束条件，以及目标函数。
待测系统的功能可以用至少一个目标函数来评价。该至少一个目标函数可以包括客观类目标函数和主观类目标函数，客观类目标函数可以为多个客观类性能指标的加权和，多个客观类性能指标的加权系数之和为1，主观类目标函数可以为多个主观感受类性能指标的加权和，多个主观感受类性能指标的加权系数之和为1。
例如，以待测系统为TCS，该待测系统包括4个客观类性能指标，分别为前轴轴速性能指标、驱动转速波动性能指标、横摆角速度性能指标、稳定驱动轴速性能指标为例，该待测系统的客观类目标函数可以为下式：
f客观＝c1*vWhlobj+c2*ΔnDriveobj+c3*YawRateobj+c4*vAxlWhlobj；
其中，c1、c2、c3、c4为加权系数，c1+c2+c3+c4＝1。
再例如，以待测系统为TCS，该待测系统包括4个主观感受类性能指标，分别为控制抖动得分、加速性能得分、加速噪声得分、驾驶员操控性得分为例，该待测系统的主观感受类目标函数可以为下式：
f主观＝d1*score控制抖动+d2*score加速性能+d3*score加速噪声+d4*score驾驶员操控性；
其中，score控制抖动为控制抖动得分，score加速性能为加速性能得分，score加速噪声为加速噪声得分，score驾驶员操控性为驾驶员操控性得分，d1、d2、d3、d4为加权系数，d1+d2+d3+d4＝1。
约束条件为标定过程中需要满足的条件。该约束条件可以是将行业标准、国际标准及法规标准进行数字化处理得到约束条件，也可以是将工程师根据经验积累的标定参数的调节范围进行数字化处理得到约束条件。该约束条件可以用上下限阈值表示。
在一些示例中，约束条件可以包括观测信号的约束条件、性能指标的约束条件、目标函数数值的约束条件、或标定参数之间的耦合系数的约束条件中的至少一种。例如，以待测系统为TCS为例，观测信号的约束条件可以是将法规标准中的横摆角速度约束、方向盘转角约束、滑移率约束等数字化处理得到约束条件。
示例性的，优化算法可以包括贝叶斯优化算法、粒子群优化算法、遗传算法、或机器学习算法。
例如，以待测系统为TCS，优化算法为贝叶斯优化算法为例，上述步骤S603可以包括：计算TCS对应的客观类目标函数和主观类目标函数，通过调用贝叶斯优化算法，根据当前客观类目标函数和当前主观类目标函数的数值、目标函数的约束条件，以及标定参数的约束条件，求解优化问题并生成优化后的标定参数(即更新后的标定参数)。该优化后的标定参数可以用于下一轮测试。
可选的，同一设备，在不同国家、地区标定参数可以相同，也可以不同，本申请实施例对此并不限定。
示例性的，以对车辆进行标定为例，当车辆在不同国家使用时，由于各个国家的 标准、道路风格不同，可以对车辆标定不同的参数。在对车辆进行标定的过程中，可以有针对性的调整标定参数的初始值、约束条件以及目标函数的加权系数，使得确定的标定参数与车辆更适合，可以更好的发挥车辆的性能。
例如，当该车辆在德国使用时，可以使用德国标准的性能指标、约束条件、至少一个目标函数以及至少一个目标函数的加权系数，对车辆进行标定。当该车辆在中国使用时，可以使用中国标准的性能指标、约束条件、至少一个目标函数以及至少一个目标函数的加权系数，对车辆进行标定。
本申请提供的标定方法，通过优化算法或机器学习算法对优化问题进行求解，可以快速准确的得到最优的标定参数，能够提高标定效率，降低标定成本。
S604、向控制器发送更新后的标定参数。
可选的，在对设备进行标定之前，可以先从云端获取最新的标定参数的初始值，向控制器发送该最新的标定参数的初始值。
控制器接收该更新后的标定参数，由于标定参数会影响车辆性能，因此在得到更新后的标定参数时，车辆可以基于该更新后的标定参数再次运行，进一步获取该更新后的标定参数对应的待测系统的观测信号，基于该观测信号计算对应的性能指标和目标函数，并采用上述优化算法得到再次更新后的标定参数，将该再次更新后的标定参数下发至控制器。车辆可以基于该再次更新后的标定参数再次运行，标定设备一直重复上述步骤S601-S604，直至至少一个目标函数的数值与预设目标函数值差距最小，且满足约束条件时，将此时的标定参数确定为最优标定参数。
本申请实施例提供的标定方法在进行标定时，不需要标定工程师的参与，能够提高标定效率，降低标定成本。而且该标定装置在对标定参数优化过程中，结合了主观类目标函数，因此确定的标定参数能够提升用户的驾驶体验。
可选的，向待标定设备的控制器发送更新的标定参数，当待标定的设备重新启动后，待标定的设备将使用更新的标定参数运行，然后进行新一轮的标定。
本申请实施例还提供一种标定方法，如图7所示，在上述步骤S601之前还可以包括步骤S605～S606。
S605、确定标定参数的类型，该标定参数的类型包括Map类和Table类。标定参数的类型可以是预先设置好的，该预先设置的类型可以是根据标定工程师的经验确定的。当标定参数的类型为Map类或Table类时，多个标定参数之间存在耦合关系。Map类Table类。
可选的，标定参数的类型还可以包括标量类，标量类的标定参数为独立的标定参数，标量类的标定参数不存在耦合关系。
S606、基于标定参数的类型和标定参数的初始值，确定标定参数之间的初始耦合系数。
初始耦合系数的数量小于所述标定参数的数量。
标定参数的初始值可以是根据标定工程师的经验确定的数值。当标定参数的类型为Map类或Table类时，多个标定参数之间存在耦合关系。由于标定参数的数量较多，如果步骤S603中采用优化算法直接优化标定参数，将导致优化的时间较长，优化的效率得不到大幅提升。因此可以通过将Map类或Table类的标定参数以多项式函数的方 式进行拟合，得到耦合系数，步骤S603中采用优化算法优化时可以用优化耦合系数代替直接优化标定参数。由于耦合系数的数量较标定参数的数量小很多，故优化耦合系数相较于直接优化标定参数能够大大降低标定参数的数量，减少优化时间，提升优化效率。
示例性的，如图8所示，根据标定工程师的经验可以给每个Map类标定参数设定初始值，根据标定参数的初始值可以得到对应的输入数据x、y及输出数据z。将每个标定参数对应的点标注在如图8所示的三维坐标系中，这些离散的点可以用多项式函数进行拟合。例如使用多项式函数z＝x2/2p+y2/2q进行拟合，得到初始耦合系数系数p和q，然后再结合上述步骤S603采用优化算法优化耦合系数p和q，能够有效减少需要标定参数的数量，提高标定效率。本申请实施例对于使用何种多项式来拟合Map类标定参数并不限定。
例如，以标定车辆需要标定40个Table类标定参数为例，根据标定工程师的经验可以给每个Map类标定参数设定初始值，根据标定参数的初始值可以得到对应的输入数据x、y及输出数据z。将每个标定参数对应的点标注在如图8所示的三维坐标系中，这些离散的点可以用多项式函数进行拟合。例如使用多项式函数z＝x2/2p+y2/2q进行拟合，得到初始耦合系数p、q，然后再结合上述步骤S603采用优化算法优化2个耦合系数p、q，而不是上述40个Map类标定参数，通过计算40个Map类标定参数之间的耦合关系，可以将需要优化的标定参数的数量由40个减少为2个，能够有效减少需要标定的参数的数量，提高标定效率。
示例性的，如图9所示，根据标定工程师的经验可以给每个Table类标定参数设定初始值，根据标定参数的初始值可以得到对应的输入数据x及输出数据y。将每个标定参数对应的点标注在如图9所示的二维坐标系中，这些离散的点可以用多项式函数进行拟合。例如使用多项式函数y＝ax2+bx+c进行拟合，得到初始耦合系数a，b，c，然后再结合上述步骤S603采用优化算法优化耦合系数a、b、c，能够有效减少需要标定参数的数量，提高标定效率。本申请实施例对于使用何种多项式来拟合Table类标定参数并不限定。
例如，以标定车辆需要标定20个Table类标定参数为例，根据标定工程师的经验可以给每个Table类标定参数设定初始值，根据标定参数的初始值可以得到对应的输入数据x及输出数据y。将每个标定参数对应的点标注在如图9所示的二维坐标系中，这些离散的点可以用多项式函数进行拟合。例如使用多项式函数y＝ax2+bx+c进行拟合，得到初始耦合系数a，b，c，然后再结合上述步骤S603采用优化算法优化3个耦合系数a、b、c，而不是上述20个Table类标定参数，通过计算20个Table类标定参数之间的耦合关系，可以将需要优化的标定参数的数量由20个减少为3个，能够有效减少需要标定的参数的数量，提高标定效率。
相应的，上述步骤S603包括：基于性能指标、约束条件、至少一个目标函数，以及更新前的耦合系数，采用优化算法得到更新后的耦合系数。上述更新前的耦合系数包括初始耦合系数。
示例性的，在第一次执行步骤S603进行优化时，该更新前的耦合系数可以为步骤S606中的初始耦合系数。基于性能指标、约束条件和至少一个目标函数的数值，采用 优化算法可以对初始耦合系数进行更新，得到更新后的耦合系数。该更新后的耦合系数即为更新后的标定参数。也就是说，与直接优化标定参数相比，本方案通过优化耦合系数，能够将需要标定的参数的数量减少。
示例性的，在第二次或者第二次以后执行步骤S603进行优化时，该更新前的耦合系数为前一次优化确定的耦合系数。例如，第二次执行步骤S603进行优化时，该更新前的耦合系数为第一次优化确定的耦合系数。再例如，第三次执行步骤S603进行优化时，该更新前的耦合系数为第二次优化确定的耦合系数。
本申请实施例还提供一种标定方法，如图10所示，在上述步骤S601之后步骤S602之前还可以包括步骤S607。
S607、在观测信号满足第一预设条件时，对观测信号进行数据处理，并根据处理后的观测信号计算对应的性能指标。
满足第一预设条件的观测信号与上述步骤S602用于计算的观测信号可以是相同的观测信号，也可以是不同的观测信号，即同一观测信号可以用于判断是否执行数据处理，还可以作为数据处理的对象，本申请实施例对此并不限定。
可选的，第一预设条件可以是与观测信号关联的一个设定范围，当观测信号在设定范围内时，对观测信号进行处理，当观测信号不在设定范围内时，不对观测信号进行处理。
可选的，第一预设条件还可以是与观测信号关联的一个设定值，当观测信号大于等于设定值时，对观测信号进行处理，当观测信号小于设定值时，不对观测信号进行处理。本申请实施例对于第一预设条件的具体内容及形式并不限定，下述实施例以第一预设条件为预设左前轴速为K为例进行实例性说明。
例如，以第一预设条件为左前轴速大于或等于预设左前轴速K为例，在专业驾驶员将油门踩到底的过程中左前轴速不断增加，在专业驾驶员松开油门的过程中左前轴速不断减小，当左前轴速大于或等于预设左前轴速K时，对观测信号进行数据处理，当左前轴速小于预设左前轴速K时，不对观测信号进行数据处理。
本申请实施例还提供一种标定方法，如图11所示，除包括上述步骤S601～S607之外，还可以包括步骤S608。
S608、当待测系统的标定过程出现异常时生成异常告警信息，异常告警信息用于提示用户标定异常。
示例性的，当标定工具连接失败或标定软件初始化失败时，可以生成相应的异常告警信息，及时提示用户标定异常需要处理，能够减少用户因为不知道报警而等待的时间，可以提高优化效率。
本申请实施例对于步骤S608和步骤S601～S607的先后执行顺序并不限定，步骤S601～S607中的任一个步骤在执行时都可能出现异常告警，图11以步骤S608在步骤S604之后执行为例进行示例性示意。
本申请实施例还提供一种标定方法，如图12所示，在上述步骤S601～S607之后还可以包括步骤S609。
S609、在至少一个目标函数的数值满足第二预设条件时，生成标定报告。
标定报告可以由优化过程中的各项数据处理生成，标定报告的内容可以包括性能 评价指标、目标函数值分布、最佳标定参数组、重要观测图形显示。
可选的，第二预设条件可以为预设目标函数值的范围，当至少一个目标函数的数值在预设目标函数值的范围内时，生成标定报告。本申请实施例对于第二预设条件的具体内容及形式并不限定，下述实施例以第二预设条件为预设目标函数值为例行进示例性说明。
示例性的，以TCS功能为例，预设客观类目标函数值为M，预设主观类目标函数值为N，当客观类目标函数值等于M，主观类目标函数值等于N时，生成标定报告。
在标定开始或结束中的任一阶段，可以向控制器发送提示信息；该提示信息用于提示用户标定开始或标定结束中的至少一项。
可选的，向待标定设备的控制器发送提示信息，待标定的设备可以对用户进行提示，该提示可以包括声、光和振动。本申请实施例对于待标定设备对用户提示的具体类型并不限定。
示例性的，在标定开始时，标定设备可以向待标定设备的控制器发送提示信息，待标定的设备根据提示信息可以对用户发出“标定开始”的提示声音，当标定结束时，标定设备可以向待标定设备的控制器发送提示信息，待标定的设备根据提示信息可以对用户发出“标定完成”的提示声音。
本申请实施例提供的标定方法，通过优化算法或机器学习算法对优化问题进行求解，可以快速准确的得到最优的标定参数，该方法可以兼容各类设备或系统，而且在进行标定时可以不需要标定工程师的参与，能够提高标定效率，降低标定成本。而且该方法在对标定参数优化过程中，结合了主观类目标函数，因此确定的标定参数能够提升用户的驾驶体验。
如图13所示，本申请实施例还提供一种标定装置130。该标定装置130可以包括收发模块131和处理模块132。可选的，标定装置130还可以包括存储模块133，该存储模块133可以存储用于实现图6、图7、图10、图11或图12所示的标定方法的计算机程序代码。
收发模块131，用于获取待测系统的观测信号。
处理模块132，用于根据待测系统的观测信号计算对应的性能指标。
处理模块132，还用于基于性能指标、约束条件，以及至少一个目标函数，采用优化算法得到更新后的标定参数，上述目标函数用于指示性能指标之间的函数关系。
可选的，约束条件包括观测信号的约束条件、性能指标的约束条件、目标函数数值的约束条件、或标定参数之间的耦合系数的约束条件中的至少一种。
可选的，性能指标包括客观类性能指标和主观感受类性能指标，至少一个目标函数包括客观类目标函数或主观类目标函数中的至少一个，客观类目标函数为多个客观类性能指标的加权和，多个客观类性能指标的加权系数之和为1，主观类目标函数为多个主观感受类性能指标的加权和，多个主观感受类性能指标的加权系数之和为1。
处理模块132，还用于确定标定参数的类型，标定参数的类型包括Map类和Table类。
处理模块132，还用于基于标定参数的类型和标定参数的初始值，确定标定参数之间的初始耦合系数。初始耦合系数的数量小于标定参数的数量。
处理模块132，具体用于基于性能指标、约束条件、至少一个目标函数，以及更新前的耦合系数，采用优化算法得到所述更新后的耦合系数。该更新前的耦合系数包括初始耦合系数。
收发模块131，还用于向控制器发送更新后的标定参数。
处理模块132，具体用于在观测信号满足第一预设条件时，对观测信号进行数据处理，并根据处理后的观测信号计算对应的性能指标。
处理模块132，还用于当待测系统的标定过程出现异常时生成异常告警信息，异常告警信息用于提示用户标定异常。
处理模块132，还用于在至少一个目标函数的数值满足第二预设条件时，生成标定报告。标定报告包括性能指标、目标函数值分布、最佳标定参数，或重要观测图形中的至少一种。
可选的，优化算法包括贝叶斯优化算法、粒子群优化算法、遗传算法、或机器学习算法。
收发模块131，还用于向控制器发送提示信息；该提示信息用于提示用户标定开始或标定结束中的至少一项。
本申请实施例提供的标定设备，通过优化算法或机器学习算法对优化问题进行求解，可以快速准确的得到最优的标定参数，该方法可以兼容各类设备或系统，而且在进行标定时可以不需要标定工程师的参与，能够提高标定效率，降低标定成本。而且该方法在对标定参数优化过程中，结合了主观类目标函数，因此确定的标定参数能够提升用户的驾驶体验。
本申请实施例还提供一种标定设备，该标定设备包括存储器和处理器；存储器和处理器耦合；存储器用于存储计算机程序代码，计算机程序代码包括计算机指令；当处理器执行计算机指令时，标定设备执行图6、图7、图10、图11或图12所示的标定方法。
本申请实施例还提供一种计算机可读存储介质，该计算机可读存储介质中存储有计算机程序代码，当上述处理器执行该计算机程序代码时，电子设备执行图6、图7、图10、图11或图12所示的标定方法。
本申请实施例还提供了一种计算机程序产品，当该计算机程序产品在计算机上运行时，使得计算机执行图6、图7、图10、图11或图12所示的标定方法。
本申请实施例还提供一种车辆，该车辆包括控制器，以及如图13所述的标定装置，所述控制器与所述标定装置之间耦合。可选的，该车辆可以为新能源汽车或智能汽车等。
结合本申请公开内容所描述的方法或者算法的步骤可以硬件的方式来实现，也可以是由处理器执行软件指令的方式来实现。软件指令可以由相应的软件模块组成，软件模块可以被存放于随机存取存储器(Random Access Memory，RAM)、闪存、可擦除可编程只读存储器(Erasable Programmable ROM，EPROM)、电可擦可编程只读存储器(Electrically EPROM，EEPROM)、寄存器、硬盘、移动硬盘、只读光盘(CD-ROM)或者本领域熟知的任何其它形式的存储介质中。一种示例性的存储介质耦合至处理器，从而使处理器能够从该存储介质读取信息，且可向该存储介质写入信息。当然，存储 介质也可以是处理器的组成部分。处理器和存储介质可以位于ASIC中。另外，该ASIC可以位于核心网接口设备中。当然，处理器和存储介质也可以作为分立组件存在于核心网接口设备中。
本领域技术人员应该可以意识到，在上述一个或多个示例中，本发明所描述的功能可以用硬件、软件、固件或它们的任意组合来实现。当使用软件实现时，可以将这些功能存储在计算机可读介质中或者作为计算机可读介质上的一个或多个指令或代码进行传输。计算机可读介质包括计算机存储介质和通信介质，其中通信介质包括便于从一个地方向另一个地方传送计算机程序的任何介质。存储介质可以是通用或专用计算机能够存取的任何可用介质。
以上所述的具体实施方式，对本发明的目的、技术方案和有益效果进行了进一步详细说明，所应理解的是，以上所述仅为本发明的具体实施方式而已，并不用于限定本发明的保护范围，凡在本发明的技术方案的基础之上，所做的任何修改、等同替换、改进等，均应包括在本发明的保护范围之内。

Claims (25)

  1. 一种标定方法，其特征在于，所述方法包括：
    获取待测系统的观测信号；
    根据所述待测系统的观测信号计算对应的性能指标；
    基于所述性能指标、约束条件，以及至少一个目标函数，采用优化算法得到更新后的标定参数；所述目标函数用于指示所述性能指标之间的函数关系；
    向控制器发送所述更新后的标定参数。
  2. 根据权利要求1所述的方法，其特征在于，所述性能指标包括客观类性能指标和主观感受类性能指标，所述至少一个目标函数包括客观类目标函数或主观类目标函数中的至少一个，所述客观类目标函数为多个客观类性能指标的加权和，所述多个客观类性能指标的加权系数之和为1，所述主观类目标函数为多个主观感受类性能指标的加权和，所述多个主观感受类性能指标的加权系数之和为1。
  3. 根据权利要求1或2所述的方法，其特征在于，所述方法还包括：
    确定所述标定参数的类型；所述标定参数的类型包括Map类和Table类；
    基于所述标定参数的类型和所述标定参数的初始值，确定所述标定参数之间的初始耦合系数。
  4. 根据权利要求3所述的方法，其特征在于，所述初始耦合系数的数量小于所述标定参数的数量。
  5. 根据权利要求1-4中任一项所述的方法，其特征在于，所述基于所述性能指标、约束条件，以及至少一个目标函数，采用优化算法得到更新后的标定参数，包括：
    基于所述性能指标、所述约束条件、所述至少一个目标函数，以及更新前的耦合系数，采用优化算法得到所述更新后的耦合系数；所述更新前的耦合系数包括所述初始耦合系数。
  6. 根据权利要求1-5中任一项所述的方法，其特征在于，所述根据所述待测系统的观测信号计算对应的性能指标，包括：
    在所述观测信号满足第一预设条件时，对所述观测信号进行数据处理，并根据处理后的所述观测信号计算对应的性能指标。
  7. 根据权利要求1-6中任一项所述的方法，其特征在于，所述方法还包括：
    当所述待测系统的标定过程出现异常时生成异常告警信息，所述异常告警信息用于提示用户标定异常。
  8. 根据权利要求1-7中任一项所述的方法，其特征在于，所述方法还包括：
    在所述至少一个目标函数的数值满足第二预设条件时，生成标定报告；所述标定报告包括性能指标、目标函数值分布、最佳标定参数，或重要观测图形中的至少一种。
  9. 根据权利要求1-8中任一项所述的方法，其特征在于，所述优化算法包括以下至少一种：贝叶斯优化算法、粒子群优化算法、遗传算法、机器学习算法。
  10. 根据权利要求1-9中任一项所述的方法，其特征在于，所述约束条件包括所述观测信号的约束条件、所述性能指标的约束条件、所述目标函数数值的约束条件、或所述标定参数之间的耦合系数的约束条件中的至少一种。
  11. 根据权利要求1-10中任一项所述的方法，其特征在于，所述方法还包括：
    向所述控制器发送提示信息；所述提示信息用于提示用户标定开始或标定结束中的至少一项。
  12. 一种标定装置，其特征在于，所述装置包括：收发模块和处理模块；
    所述收发模块，用于获取待测系统的观测信号；
    所述处理模块，用于根据所述待测系统的观测信号计算对应的性能指标；
    所述处理模块，还用于基于所述性能指标、约束条件，以及至少一个目标函数，采用优化算法得到更新后的标定参数；所述目标函数用于指示所述性能指标之间的函数关系；
    所述收发模块，还用于向控制器发送所述更新后的标定参数。
  13. 根据权利要求12所述的标定装置，其特征在于，所述性能指标包括客观类性能指标和主观感受类性能指标，所述至少一个目标函数包括客观类目标函数或主观类目标函数中的至少一个，所述客观类目标函数为多个客观类性能指标的加权和，所述多个客观类性能指标的加权系数之和为1，所述主观类目标函数为多个主观感受类性能指标的加权和，所述多个主观感受类性能指标的加权系数之和为1。
  14. 根据权利要求12或13所述的标定装置，其特征在于，所述处理模块还用于：
    确定所述标定参数的类型；所述标定参数的类型包括Map类和Table类；
    基于所述标定参数的类型和所述标定参数的初始值，确定所述标定参数之间的初始耦合系数。
  15. 根据权利要求14所述的标定装置，其特征在于，所述初始耦合系数的数量小于所述标定参数的数量。
  16. 根据权利要求12-15中任一项所述的标定装置，其特征在于，所述处理模块，具体用于基于所述性能指标、所述约束条件、所述至少一个目标函数，以及更新前的耦合系数，采用优化算法得到所述更新后的耦合系数；所述更新前的耦合系数包括所述初始耦合系数。
  17. 根据权利要求12-16中任一项所述的标定装置，其特征在于，所述处理模块，具体用于在所述观测信号满足第一预设条件时，对所述观测信号进行数据处理，并根据处理后的所述观测信号计算对应的性能指标。
  18. 根据权利要求12-17中任一项所述的标定装置，其特征在于，所述处理模块，还用于当所述待测系统的标定过程出现异常时生成异常告警信息，所述异常告警信息用于提示用户标定异常。
  19. 根据权利要求12-18中任一项所述的标定装置，其特征在于，所述处理模块，还用于在所述至少一个目标函数的数值满足第二预设条件时，生成标定报告；所述标定报告包括性能指标、目标函数值分布、最佳标定参数，或重要观测图形中的至少一种。
  20. 根据权利要求12-19中任一项所述的标定装置，其特征在于，所述优化算法包括以下至少一种：贝叶斯优化算法、粒子群优化算法、遗传算法、机器学习算法。
  21. 根据权利要求12-20中任一项所述的标定装置，其特征在于，所述约束条件包括所述观测信号的约束条件、所述性能指标的约束条件、所述目标函数数值的约束条件、或所述标定参数之间的耦合系数的约束条件中的至少一种。
  22. 根据权利要求12-21中任一项所述的标定装置，其特征在于，所述收发模块还用向所述控制器发送提示信息；所述提示信息用于提示用户标定开始或标定结束中的至少一项。
  23. 一种标定设备，其特征在于，所述标定设备包括存储器和处理器；所述存储器和所述处理器耦合；所述存储器用于存储计算机程序代码，所述计算机程序代码包括计算机指令；当所述处理器执行所述计算机指令时，所述标定设备执行如权利要求1-11中任意一项所述的标定方法。
  24. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质存储有计算机程序，当所述计算机程序在标定设备上运行时，使得所述标定设备执行如权利要求1-11中任意一项所述的标定方法。
  25. 一种标定系统，其特征在于，所述标定系统包括控制器，以及如权利要求12-22中任一项所述的标定装置，所述控制器与所述标定装置之间耦合。
PCT/CN2021/131000 2021-11-16 2021-11-16 一种标定方法、装置及系统 WO2023087153A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
PCT/CN2021/131000 WO2023087153A1 (zh) 2021-11-16 2021-11-16 一种标定方法、装置及系统
EP21964319.4A EP4425110A1 (en) 2021-11-16 2021-11-16 Calibration method, device and system
CN202180007894.5A CN116457636A (zh) 2021-11-16 2021-11-16 一种标定方法、装置及系统
US18/664,682 US20240295420A1 (en) 2021-11-16 2024-05-15 Calibration method, apparatus, and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2021/131000 WO2023087153A1 (zh) 2021-11-16 2021-11-16 一种标定方法、装置及系统

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US18/664,682 Continuation US20240295420A1 (en) 2021-11-16 2024-05-15 Calibration method, apparatus, and system

Publications (1)

Publication Number Publication Date
WO2023087153A1 true WO2023087153A1 (zh) 2023-05-25

Family

ID=86396176

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/131000 WO2023087153A1 (zh) 2021-11-16 2021-11-16 一种标定方法、装置及系统

Country Status (4)

Country Link
US (1) US20240295420A1 (zh)
EP (1) EP4425110A1 (zh)
CN (1) CN116457636A (zh)
WO (1) WO2023087153A1 (zh)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113340334A (zh) * 2021-07-29 2021-09-03 新石器慧通（北京）科技有限公司 用于无人车的传感器标定方法、装置及电子设备
CN113479186A (zh) * 2021-07-02 2021-10-08 中汽研（天津）汽车工程研究院有限公司 一种混合动力汽车能量管理策略优化方法

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113479186A (zh) * 2021-07-02 2021-10-08 中汽研（天津）汽车工程研究院有限公司 一种混合动力汽车能量管理策略优化方法
CN113340334A (zh) * 2021-07-29 2021-09-03 新石器慧通（北京）科技有限公司 用于无人车的传感器标定方法、装置及电子设备

Also Published As

Publication number Publication date
EP4425110A1 (en) 2024-09-04
US20240295420A1 (en) 2024-09-05
CN116457636A (zh) 2023-07-18

Similar Documents

Publication Publication Date Title
CN109591811B (zh) 车辆制动方法、装置及存储介质
CN109543245B (zh) 无人车应对能力边界信息确定方法、装置和电子设备
CN109522673B (zh) 一种测试方法、装置、设备和存储介质
US10471967B2 (en) Optimization of a vehicle to compensate for water contamination of a fluid of a vehicle component
CN112744120B (zh) 估算电动车辆续驶里程的方法、云端服务器、车辆和介质
CN113219962B (zh) 一种面向混行队列跟驰安全的控制方法、系统及存储介质
CN112131655B (zh) 制动踏板感性能确定方法、装置及计算机可读存储介质
WO2023087153A1 (zh) 一种标定方法、装置及系统
CN109752952B (zh) 一种获取多维随机分布及强化控制器的方法和装置
CN116910552A (zh) 一种车辆故障的检测方法、装置、车载终端及存储介质
CN103994745A (zh) 一种导向轮偏转角度测量方法及装置
CN111123730B (zh) 无人驾驶故障仿真方法、装置、设备及计算机可读介质
CN114548608B (zh) 模型处理方法、装置、目标交通设备及存储介质
CN114537421A (zh) 调整系统故障的方法、装置、设备和计算机可读介质
CN114676781A (zh) 车辆故障诊断预测方法、装置、设备及车辆
US20210146945A1 (en) Method of calculating stopping distance and related devices
CN113204742A (zh) 车辆控制参数的标定方法、装置及车辆
CN115092136B (zh) 车速规划方法、装置、车辆及存储介质
CN116030632B (zh) 一种面向混合交通流的性能指标计算方法及系统
CN111806431B (zh) 一种泊车控制方法、装置、计算机设备和存储介质
CN112683216B (zh) 用于生成车辆长度信息的方法、装置、路侧设备和云控平台
CN114228722B (zh) 驾驶风格划分方法、装置、设备、存储介质和程序产品
WO2024174172A1 (zh) 一种车辆动力学模型的建模方法、装置及计算设备
US11760368B2 (en) System and method of same-loop adaptive simulation for autonomous driving
CN114913714A (zh) 确定车辆安全距离的方法

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 202180007894.5

Country of ref document: CN

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21964319

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 2021964319

Country of ref document: EP

ENP Entry into the national phase

Ref document number: 2021964319

Country of ref document: EP

Effective date: 20240528