WO2023051065A1 - 参数标定的方法及装置 - Google Patents

参数标定的方法及装置 Download PDF

Info

Publication number
WO2023051065A1
WO2023051065A1 PCT/CN2022/112810 CN2022112810W WO2023051065A1 WO 2023051065 A1 WO2023051065 A1 WO 2023051065A1 CN 2022112810 W CN2022112810 W CN 2022112810W WO 2023051065 A1 WO2023051065 A1 WO 2023051065A1
Authority
WO
WIPO (PCT)
Prior art keywords
parameter
calibration
multiple devices
devices
evaluation results
Prior art date
Application number
PCT/CN2022/112810
Other languages
English (en)
French (fr)
Inventor
古强
庄雨铮
刘栋豪
罗杰
王滨
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 EP22874460.3A priority Critical patent/EP4386612A1/en
Publication of WO2023051065A1 publication Critical patent/WO2023051065A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04Q—SELECTING
    • H04Q9/00—Arrangements in telecontrol or telemetry systems for selectively calling a substation from a main station, in which substation desired apparatus is selected for applying a control signal thereto or for obtaining measured values therefrom
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/10—Geometric CAD
    • G06F30/15—Vehicle, aircraft or watercraft design
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B17/00—Monitoring; Testing
    • H04B17/20—Monitoring; Testing of receivers
    • H04B17/21—Monitoring; Testing of receivers for calibration; for correcting measurements
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04Q—SELECTING
    • H04Q2209/00—Arrangements in telecontrol or telemetry systems
    • H04Q2209/80—Arrangements in the sub-station, i.e. sensing device
    • H04Q2209/86—Performing a diagnostic of the sensing device

Definitions

  • the present application relates to the field of data processing, and more specifically, to a method and device for parameter calibration.
  • Parameter calibration refers to the process of finding a set of parameters that are applicable under various conditions, for example, under different hardware conditions or different environmental conditions.
  • vehicle calibration refers to the control of the components in the control software in order to obtain satisfactory vehicle performance, meet the requirements of the vehicle before leaving the factory, and meet relevant national standards after the controller hardware, control software, and related sensors are determined. The process of parameter optimization.
  • Vehicle calibration results are an important factor affecting vehicle-related performance.
  • it is usually calibrated by an experienced calibration engineer based on the sample vehicle.
  • it is necessary to conduct calibration experiments under different working conditions to find a suitable parameter combination.
  • the calibration process takes a long time and the calibration cost is high. Taking the braking system in a vehicle as an example, it usually takes a long calibration process of "two summers and one winter" to complete the parameter calibration in the braking system, which seriously affects the calibration efficiency.
  • the present application provides a parameter calibration method and device, which can improve the efficiency of parameter calibration.
  • a method for parameter calibration including: obtaining the evaluation results of the parameter combinations of multiple devices, and the evaluation results of the parameter combinations of the multiple devices are based on the parameter combinations of the multiple devices at the It is obtained by performing a calibration test within a period of time, and the first period of time is less than or equal to the first threshold; at least one adjusted parameter combination is obtained according to the evaluation results of the parameter combinations of multiple devices.
  • the multiple devices perform the calibration test within the same period of time, which can improve the efficiency of data collection, and is conducive to real-time adjustment of the parameters of the devices based on the evaluation results of the parameter combinations of multiple devices, thereby improving the calibration efficiency. Reduce the calibration cycle and reduce the calibration cost.
  • Multiple devices in the solution of the embodiment of the present application can perform calibration tests simultaneously under different working conditions, which improves calibration efficiency and reduces time and economic costs.
  • Multiple devices in the solution of the embodiment of the present application can be in different working conditions at the same time, which can realize real-time summary of data of multiple devices and parameter optimization, which is conducive to finding compatible devices under multiple working conditions at the same time.
  • the combination of parameters can improve the stability of parameters and improve the calibration efficiency.
  • parameter calibration is performed based on the evaluation results of test data of multiple devices, which reduces the cognitive bias caused by different calibration engineers, improves the calibration quality, and helps to avoid the recall of equipment after leaving the factory .
  • the solution of the embodiment of the present application can also use an automatic optimization algorithm to complete the calibration, reduce the number of calibration engineers, reduce labor costs, and further reduce the influence of the calibration engineer's subjective preference on the calibration result.
  • the multiple devices performing the calibration test within the first period of time may also be understood as the multiple devices may perform the calibration test within the same period of time, for example, the multiple devices may perform the calibration test at the same time.
  • the parameters in the parameter combination are the parameters to be calibrated.
  • a parameter combination includes at least one parameter.
  • the apparatus for parameter calibration may belong to the multiple devices, or may not belong to the multiple devices.
  • the parameter combinations of the multiple devices may be the same or different.
  • the evaluation result of the parameter combination of the plurality of devices is used to indicate the quality of the parameter combination of the plurality of devices.
  • the evaluation result of the parameter combination can also be understood as the value of the parameter combination.
  • the performance of the device may be evaluated based on one or more evaluation indicators, and an evaluation result of the parameter combination may be obtained.
  • the at least one adjusted parameter combination may be the same or different.
  • the parameter combinations of the multiple devices are the same, the working conditions of the multiple devices are different, and at least one adjusted
  • the parameter combination includes: processing the evaluation results of the parameter combinations of multiple devices to obtain a summary evaluation result; and obtaining at least one adjusted parameter combination according to the summary evaluation result.
  • Aggregate assessment results can be determined in a number of ways.
  • the weighted average of the evaluation results of the parameter combinations of the plurality of devices is used as the summary evaluation result.
  • the minimum value among the evaluation results of the parameter combinations of the plurality of devices is used as the summary evaluation result.
  • the maximum value among the evaluation results of the combination of parameters of the plurality of devices is used as the summary evaluation result.
  • the embodiment of the present application does not limit the method of determining the summary evaluation result.
  • the method further includes: sending at least one adjusted parameter combination to at least one device.
  • At least one device is determined according to at least one of the following: operating conditions of multiple devices or at least one adjusted parameter combination.
  • the evaluation result of the parameter combination of multiple devices is obtained by evaluating the test data of multiple devices, and the test data of multiple devices is obtained according to the are respectively determined based on the data collected during the calibration test in the first period of time based on the parameter combinations of multiple devices.
  • test data of the plurality of devices may be obtained by processing the data collected by the sensors during the calibration test of the plurality of devices based on the combination of parameters of the plurality of devices.
  • processing the data collected by the sensor may include processing the data collected by the sensor such as filtering, frequency reduction, or noise reduction.
  • test data of the multiple devices may be data collected by sensors during a calibration test performed by the multiple devices based on parameter combinations of the multiple devices.
  • the evaluation result of the parameter combination of the multiple devices is obtained by evaluating the test data of the multiple devices according to the working conditions of the multiple devices.
  • evaluating the test data of the device based on the working condition of the device can improve the accuracy of the evaluation result, and further improve the effect of parameter calibration.
  • the evaluation result of the parameter combination of the multiple devices may be obtained by evaluating the test data of the multiple devices according to the configuration information of the multiple vehicles.
  • the test data of the device is evaluated based on the configuration information of the device, which can improve the accuracy of the evaluation result and further improve the effect of parameter calibration.
  • evaluation results of parameter combinations of multiple devices are obtained through user feedback.
  • the evaluation result of the parameter combination can be obtained from user feedback, which can fully consider the user's feelings and is beneficial to improve user experience.
  • the method further includes: sending the sequence of actions to be performed by at least one device to the at least one device respectively, where the sequence of actions to be performed by the at least one device includes at least one device based on An action to be performed during a calibration test with an adjusted parameter combination of at least one device.
  • the sequence of actions to be performed by the at least one device may be the same or different.
  • the action sequence to be executed by the at least one device may also be determined by the at least one device itself.
  • sequence of actions to be executed by the at least one device may be preset.
  • the sequence of actions to be performed by the at least one device may be user-determined.
  • the sequence of actions to be performed by at least one device is determined according to at least one of the following: actions covered by parameter calibration requirements, working conditions or multiple The evaluation results of the parameter combination of each device.
  • the multiple devices include at least one of the following: a vehicle or a test bench.
  • the multiple devices include: sensors.
  • a method for parameter calibration including: obtaining an adjusted parameter combination, the adjusted parameter combination is obtained according to the evaluation results of the parameter combinations of multiple devices, the multiple devices include the first device, and the multiple The evaluation result of the parameter combination of a device is obtained by performing a calibration test on multiple devices based on the parameter combination of the multiple devices in the first period, and the first period is less than or equal to the first threshold; the first device is controlled based on the adjusted The parameter combination is tested for calibration.
  • the multiple devices perform the calibration test within the same period of time, which can improve the efficiency of data collection, and is conducive to real-time adjustment of the parameters of the devices based on the evaluation results of the parameter combinations of multiple devices, thereby improving the calibration efficiency. Reduce the calibration cycle and reduce the calibration cost.
  • Multiple devices in the solution of the embodiment of the present application can perform calibration tests simultaneously under different working conditions, which improves calibration efficiency and reduces time and economic costs.
  • Multiple devices in the solution of the embodiment of the present application can be in different working conditions at the same time, which can realize real-time summary of data of multiple devices and parameter optimization, which is conducive to finding compatible devices under multiple working conditions at the same time.
  • the combination of parameters can improve the stability of parameters and improve the calibration efficiency.
  • parameter calibration is performed based on the evaluation results of test data of multiple devices, which reduces the cognitive bias caused by different calibration engineers, improves the calibration quality, and helps to avoid the recall of equipment after leaving the factory .
  • the solution of the embodiment of the present application can also use an automatic optimization algorithm to complete the calibration, reduce the number of calibration engineers, reduce labor costs, and further reduce the influence of the calibration engineer's subjective preference on the calibration result.
  • the parameter combinations of multiple devices are the same, and the working conditions of the multiple devices are different, and the adjusted parameter combinations are obtained based on the summary evaluation results, and the summary evaluation results It is obtained by processing the evaluation results of a combination of parameters of a plurality of devices.
  • the first device is determined according to at least one of the following: working conditions of multiple devices or adjusted parameter combinations.
  • the evaluation result of the parameter combination of multiple devices is obtained by evaluating the test data of multiple devices, and the test data of multiple devices is obtained according to the are respectively determined based on the data collected during the calibration test in the first period of time based on the parameter combinations of multiple devices.
  • evaluation results of parameter combinations of multiple devices are obtained through user feedback.
  • the method further includes: the first device acquires a sequence of actions to be performed by the first device, and the sequence of actions to be performed by the first device includes Actions that need to be performed during the process of performing a calibration test with a combination of parameters.
  • the sequence of actions to be performed by the first device is determined according to at least one of the following: the actions covered by parameter calibration requirements, the working conditions of the first device or multiple The evaluation results of the parameter combination of each device.
  • the first device is: a vehicle or a test bench.
  • a parameter calibration device in a third aspect, includes a module or unit for performing the first aspect and the method in any one implementation manner of the first aspect.
  • a parameter calibration device in a fourth aspect, includes a module or unit configured to execute the second aspect and the method in any one of the implementation manners of the second aspect.
  • a parameter calibration device which includes: a memory for storing programs; a processor for executing the programs stored in the memory, and when the programs stored in the memory are executed, the The processor is configured to execute the method in any one implementation manner of the foregoing first aspect.
  • the device may also include a communication interface.
  • a parameter calibration device which includes: a memory for storing a program; a processor for executing the program stored in the memory, and when the program stored in the memory is executed, the The processor is configured to execute the method in any one of the implementation manners of the second aspect above.
  • the device may also include a communication interface.
  • a computer-readable medium stores instructions for execution by a device, where the instructions are used to execute the method in any one of the implementation manners of the first aspect or the second aspect.
  • a computer program product containing instructions is provided, and when the computer program product is run on a computer, the computer is made to execute the method in any one of the above-mentioned first aspect or the second aspect.
  • a chip the chip includes a processor and a communication interface, the processor reads instructions stored on the memory through the communication interface, and executes any one of the first aspect or the second aspect above method in the implementation.
  • the chip may further include a memory, the memory stores instructions, the processor is configured to execute the instructions stored in the memory, and when the instructions are executed, the processor is configured to execute the first aspect Or the method in any implementation manner in the second aspect.
  • an electronic device in a tenth aspect, includes the parameter calibration apparatus in any one of the implementation manners of the third aspect or the fourth aspect.
  • FIG. 1 is a schematic diagram of an application scenario provided by an embodiment of the present application
  • Fig. 2 is a schematic diagram of a calibration system provided by an embodiment of the present application.
  • Fig. 3 is a schematic diagram of another calibration system provided by the embodiment of the present application.
  • Fig. 4 is a schematic flow chart of a parameter calibration method provided by an embodiment of the present application.
  • Fig. 5 is a schematic flowchart of another parameter calibration method provided by the embodiment of the present application.
  • Fig. 6 is a schematic diagram of a parameter calibration device provided by the embodiment of the present application.
  • Fig. 7 is a schematic diagram of another parameter calibration device provided by the embodiment of the present application.
  • solutions of the embodiments of the present application can be applied to vehicle calibration, sensor calibration, and other fields for parameter calibration.
  • Parameter calibration refers to determining a set of parameters that are applicable under various conditions, for example, under different hardware conditions or different environmental conditions. In other words, parameter calibration refers to the process of optimizing parameters under various conditions.
  • Vehicle calibration refers to the process of optimizing the parameters in the control software in order to obtain satisfactory vehicle performance, meet the requirements of the vehicle before leaving the factory, and meet relevant national standards after the controller hardware, control software, and related sensors are determined.
  • the electronic control unit (ECU) in the vehicle also known as “driving computer”, uses the data collected by various sensors to perform calculations to obtain control signals to control the various actuators in the vehicle to perform corresponding actions.
  • the control range of the ECU may include cruise control, light control, airbag control, suspension control, exhaust control, or brake control, etc.
  • Each ECU can exchange data through the bus.
  • the parameters in the control software in the embodiment of the present application may be parameters in the ECU.
  • the existing parameter calibration process usually conducts calibration experiments under different working conditions in a serial manner. Specifically, in a calibration experiment, after the calibration engineer sets the parameters, the calibration experiment is performed, and the parameters are adjusted according to the data collected in the experiment. Repeat the process of the above calibration experiment under the same working condition. After the parameter calibration in this working condition is completed, transfer to the next working condition for calibration. The calibration efficiency of this scheme is low, and the calibration period is long.
  • the efficiency of parameter calibration in the vehicle can be improved, and the time cost of calibration can be reduced.
  • Some parameters of the sensor need to be calibrated in different environments before leaving the factory, such as the distortion calibration matrix of the camera, the sensitivity of the inertial measurement unit (IMU), and the deflection angle of the lidar on the mobile phone.
  • the distortion calibration matrix of the camera the sensitivity of the inertial measurement unit (IMU)
  • IMU inertial measurement unit
  • parameter calibration needs to be performed under different lighting conditions, different temperatures, and different humidity environments before leaving the factory, so as to obtain camera parameters that are applicable in different environments.
  • the efficiency of parameter calibration in the sensor can be improved, and the time cost of calibration can be reduced.
  • Fig. 1 shows a schematic diagram of an application scenario provided by an embodiment of the present application.
  • the working conditions of the vehicle at location 1 include: sunny, temperature 25° C., asphalt road.
  • the working conditions of the vehicle at location 2 include: sunny, temperature 6°C, tiled road.
  • the working conditions of the vehicle at location 3 include: snow, temperature of -30°C, and ice.
  • the parameters of the three vehicles can be calibrated simultaneously by using the method of the embodiment of the present application.
  • three vehicles can perform calibration tests separately, and upload relevant data, such as test data or evaluation results of parameter combinations, to the cloud, and the cloud will send the adjusted parameter combinations of each device to each vehicle, repeating Perform the above process until the adjusted parameter combination can meet the test requirements under the three working conditions, and the parameter calibration is completed.
  • relevant data such as test data or evaluation results of parameter combinations
  • the embodiment of the present application provides a parameter calibration method, which implements parameter calibration based on evaluation results of parameter combinations of multiple devices, so as to improve parameter calibration efficiency and reduce calibration time costs.
  • the calibration system 200 can be deployed on the cloud side and the device side.
  • the cloud device 210 may be implemented by one or more servers.
  • the device side includes multiple devices.
  • the plurality of devices (eg, device 220 and device 230 ) can interact with cloud device 210 .
  • the multiple devices include modules requiring parameter calibration.
  • the plurality of devices may include a smart phone, a tablet computer, a smart camera, a vehicle, a media consumption device or a wearable device and the like.
  • the plurality of devices can interact with the cloud device 210 through any communication mechanism/communication standard communication network. Ad hoc networks of communication protocols, Ethernet, WiFi, and HTTP, and various combinations of the foregoing. Such communications may be by any device capable of transmitting data to and from other computers, such as modems and wireless interfaces.
  • the cloud device 210 can also be realized by the multiple devices, for example, the device 220 executes the functions of the cloud device 210 to provide the calibration service for itself, or provides the calibration service for the device 230 . In this case, the calibration system 200 does not need to be deployed on the cloud side. In other words, the cloud device 210 is optional.
  • multiple devices may be devices of the same type, or may be devices of different types.
  • multiple devices may include multiple vehicles.
  • multiple devices may include multiple vehicles and multiple test benches.
  • FIG. 3 shows a schematic block diagram of a calibration system provided by an embodiment of the present application.
  • FIG. 3 can be regarded as a specific implementation manner of the calibration system 200 shown in FIG. 2 .
  • the calibration system 200 includes a data processing module 211 deployed on a cloud device 210, a calibration module 212, a first storage module 213 and a first communication module 214.
  • the calibration system also includes a second communication module, a second storage module, a control module and an execution module of multiple devices deployed on the device side.
  • the calibration system 200 further includes a second communication module 221, a second storage module 222, a control module 223, and an execution module 224 deployed in the device 220, and a second communication module 231, a second storage module 232, A control module 233 and an execution module 234 .
  • first and “second” in the “first communication module” and “second communication module” in the embodiment of the present application are only used to distinguish the communication module on the cloud side from the communication module on the device side, not have other limitations.
  • first and “second” in “first storage module” and “second storage module” are only used to distinguish the storage module on the cloud side from the storage module on the device side, and have no other limiting role.
  • the first communication module 214 is used to realize the communication connection between the cloud side and the device side.
  • the communication connection may be a wired connection or a wireless connection.
  • the first communication module 214 is configured to receive at least one of the following: data of the device 220 and the device 230 , or an evaluation result of a parameter combination of the device 220 and the device 230 .
  • the data of the device includes data collected by the device in a calibration test based on the current parameter combination.
  • the current parameter combination of each device may be the same or different.
  • the sequence of actions performed by the calibration test of each device may be the same or different.
  • the first communication module 214 may also be used to receive working conditions of the device 220 and the device 230 .
  • the data processing module 211 is configured to process the data received by the first communication module 214 .
  • the data received by the first communication module 214 includes data of the device 220 and the device 230 .
  • the data processing module 211 can be used to process the data of the device 220 and the device 230 to extract valid data, which is the data processed by the data processing module 211, which is the "test data" hereinafter.
  • the data processing module 211 is an optional module. For example, if the data of the device only includes valid data, that is, if the data does not need to be processed, the data processing module 211 does not need to be set. For another example, if the data received by the first communication module 214 is the evaluation result of the combination of parameters of the device 220 and the device 230 , then the data processing module 211 does not need to be set. For another example, the function of the data processing module 211 may be executed by the calibration module 212 to process the data, without additional data processing module 211 .
  • the calibration module 212 is configured to obtain at least one parameter combination according to evaluation results of parameter combinations of multiple devices (eg, device 220 and device 230 ).
  • the calibration module 212 may also be used to evaluate the test data of the multiple devices to obtain an evaluation result.
  • the marking module 212 may also be configured to determine an action sequence to be executed by at least one device among the plurality of devices.
  • the calibration module 212 may also be used to determine the at least one device.
  • step S420 and step S430 in the method 400 which will not be repeated here.
  • the first storage module 213 can be used to store at least one of the following items: test data of the equipment 220 and the equipment 230 , evaluation results of parameter combinations of the equipment 220 and the equipment 230 , working conditions or action sequences of the equipment 220 and the equipment 230 .
  • the first storage module may include multiple storage modules.
  • the first storage module may include a working condition module, which is used to store the working conditions of multiple devices, for example, the working conditions of the device 220 and the device 230 .
  • the first storage module may include a calibration data module, and the calibration data module is used to store the test data of multiple devices, for example, the test data of the device 220 and the device 230 .
  • the first communication module 214 may also be configured to respectively send the adjusted parameter combination of the at least one device to the at least one device.
  • the first communication module 214 may also be configured to respectively send the action sequence to be executed by the at least one device to the at least one device.
  • the following uses the device 220 as an example to describe each module included in the device.
  • the modules in device 230 may perform the same functions as the modules in device 220 .
  • the second communication module 221 can be used to cooperate with the first communication module 214 on the cloud side to realize the communication connection between the device 220 and the cloud device 210 .
  • the second communication module 221 may be used to receive the adjusted parameter combination of the device 220 .
  • the second communication module 221 may also be configured to send the data of the device 220 or the evaluation result of the parameter combination of the device 220 to the first communication module 214 .
  • the second communication module 221 may also be configured to send the working condition of the device 220 to the first communication module 214 .
  • the second communication module 221 may also be configured to receive an action sequence to be executed by the device 220 .
  • the parameter combination of the device 220 received by the device 220 may be written into the control module 223 .
  • the control module 223 is used to control the execution module 224 to execute the action sequence to be executed.
  • the parameters in the control module 223 are the parameters that need to be calibrated.
  • FIG. 3 is only an example, and the calibration system may not include the control module 223 .
  • device 220 may be a vehicle.
  • the control module 223 may be a vehicle ECU.
  • the executing module 224 is used for executing the sequence of actions to be executed.
  • the specific execution process can be executed in the form of automatic driving, or it can also be executed by the driver.
  • FIG. 3 is only an example, and the execution module 224 may not be included in the calibration system.
  • the second storage module 222 is used to store at least one of the following items: the working condition of the device 220 , the parameter combination of the device 220 , and the sequence of actions to be executed by the device 220 .
  • each device in FIG. 3 only uses the second storage module as one storage module as an example, and in practical applications, each device may include multiple storage modules.
  • the second storage module may include a working condition module, a parameter module and an action module.
  • the working condition module is used to send the working condition of the device 220 to the cloud through the second communication module 221 .
  • the parameter module is used to store the parameter combinations of the device 220 and write the parameters into the control module 223 .
  • the action module is used to store the action sequence to be executed in the device 220 and send the action sequence to be executed to the execution module 224 .
  • the calibration system is described by taking the terminal device including two vehicles as an example, which does not limit the solution of the embodiment of the present application.
  • terminal equipment may also include more vehicles.
  • the terminal device in FIG. 3 may also include other types of devices.
  • Fig. 4 shows a schematic flowchart of a parameter calibration method according to an embodiment of the present application.
  • the method 400 shown in FIG. 4 can also be executed by a device for parameter calibration.
  • the device for parameter calibration can be a cloud service device or a terminal device, such as a vehicle, a mobile phone, or a test bench, etc., or it can be performed by a cloud service device and a terminal device. composed system.
  • the device for parameter calibration may be deployed on cloud service equipment or terminal equipment, or may also be deployed on a system composed of cloud service equipment and terminal equipment.
  • a cloud service device may also be referred to as a cloud.
  • the method 400 may be executed by the calibration system shown in FIG. 3 .
  • the method 400 shown in FIG. 4 includes step S410 to step S440.
  • the first threshold may be 24 hours.
  • the first period may be any period, as long as the time interval of the period is less than or equal to the first threshold.
  • the multiple devices performing the calibration test within the first period of time may also be understood as the multiple devices may perform the calibration test within the same period of time, for example, the multiple devices may perform the calibration test at the same time.
  • the multiple devices are devices requiring parameter calibration.
  • the parameters in the parameter combination are the parameters to be calibrated.
  • a parameter combination includes at least one parameter.
  • the parameters in the parameter combination refer to some or all of the control parameters in the ECU.
  • the parameters in the parameter combination include the control parameters related to the working state of the transmitter in the ECU.
  • the parameter combination may include ignition advance angle and the like.
  • the multiple devices respectively execute the action sequences of the multiple devices based on the parameter combinations of the multiple devices.
  • the means for executing step S410 may be deployed in the multiple devices, or may also be deployed on other devices other than the multiple devices, for example, may be deployed in the cloud.
  • the multiple devices may be multiple vehicles.
  • the multiple devices are m devices, and the m devices may be m vehicles, where m is an integer greater than 1.
  • the method 400 can be applied in the scene of whole vehicle calibration.
  • the method 400 can be used in scenarios such as an antilock brake system (antilock brake system, ABS) scenario, a vehicle steering system scenario, or a vehicle body electronic stability control (electronic stability controller, ESC) system scenario in vehicle calibration. parameter calibration.
  • ABS antilock brake system
  • ESC vehicle body electronic stability control
  • the plurality of devices may comprise a plurality of test benches.
  • the multiple devices are m devices, and the m devices may be m test benches, where m is an integer greater than 1.
  • the plurality of devices may include vehicles and test benches.
  • the multiple devices are m devices, and the m devices may include n vehicles and m-n test benches.
  • m is an integer greater than 1
  • n is a positive integer smaller than m.
  • the method 400 may be applied in scenarios such as engine calibration or motor calibration in electric vehicles.
  • the plurality of devices may include a plurality of sensors.
  • the method 400 can be applied in the context of sensor calibration.
  • the method 400 can be used to realize the parameter calibration of the camera on the mobile phone.
  • the parameter combinations of the multiple devices may be the same or different.
  • different parameter combinations refer to different values of at least one parameter in the parameter combinations, and it is not limited to different parameter items in the parameter combinations.
  • parameter items in the parameter combinations of the multiple devices may be the same.
  • multiple devices perform parameter calibration in the ABS scene at the same time, and the parameter combinations of the multiple devices all include parameter items related to the ABS function. In this way, the calibration efficiency of the parameters in the ABS scene can be improved.
  • the parameter values in the parameter combination of each device may be randomly determined in the parameter space.
  • the parameter space can also be understood as the value range of the parameter.
  • the parameter values in the parameter combination of each device may also be determined according to the parameter values of similar vehicle models.
  • the parameter values of the parameter combinations of each device may also be set manually.
  • the initial parameter combination of each device may be determined separately according to the working conditions of each device.
  • the parameter combinations of the multiple devices may be determined in the same manner, or may be determined in different manners. The embodiment of the present application does not limit the method of determining the parameter value in the parameter combination of each device.
  • the operation sequences of the plurality of devices may be the same or different.
  • the action sequences executed during the calibration tests of the multiple devices are different, which may also be understood as different calibration tests.
  • the vehicle when performing parameter calibration in the ABS scene, the vehicle performs the following action sequence during the calibration test: the vehicle drives to the starting position, accelerates to the target speed, keeps driving at a constant speed, depresses the brake pedal until the vehicle stops, and the vehicle Drive out of the test area.
  • two action sequences including different target speeds can be regarded as different action sequences.
  • the two devices are accelerated to different target speeds during the calibration test, which can be regarded as executing different action sequences.
  • the action sequences of the multiple devices may be randomly determined in the action sequence set.
  • the action sequence of the plurality of devices may be determined according to at least one of the following: actions covered by parameter calibration requirements, working conditions of the at least one device, or evaluation results of parameter combinations of the plurality of devices.
  • the action sequences of the plurality of devices may also be artificially set.
  • the action sequences of the multiple devices may be determined in the same manner, or may be determined in different manners. This embodiment of the present application does not limit it.
  • the multiple devices meet at least one of the following: the working conditions of the multiple devices are different, the sequence of actions performed by the multiple devices during the calibration test is different, or the multiple devices have different The parameter combinations are different.
  • any two devices in the multiple devices are in different working conditions, that is, the multiple devices can be considered to be in different working conditions.
  • the method 400 is applied in a scenario of vehicle calibration, and the working conditions may include at least one of the following: road conditions, temperature, humidity, or weather.
  • Any two devices in the plurality of devices perform different action sequences during the calibration test, and it can be considered that the plurality of devices execute different action sequences.
  • devices under different working conditions execute the same action sequence based on the same parameter combination.
  • a value range of a parameter in the parameter combination is (0,1), and multiple devices under different working conditions execute the same action sequence when the parameter is 0.1.
  • the value range of a parameter in the parameter combination is (0,1), and multiple devices under the same working condition are controlled to perform the same operation when the parameter is a different value in the range (0,1). Action sequence.
  • devices under different working conditions execute different action sequences based on different parameter combinations.
  • controlling multiple devices to perform calibration tests under different working conditions can Improve the efficiency of the calibration test, obtain the evaluation results of the parameter combinations in various situations faster, or collect the evaluation results of the parameter combinations in more situations, so as to complete the parameter calibration as soon as possible, which is conducive to improving the efficiency of parameter calibration.
  • the parameter combinations of the multiple devices are the same, and the working conditions of the multiple devices are different.
  • the evaluation result of the parameter combination of the plurality of devices is used to indicate the quality of the parameter combination of the plurality of devices.
  • the evaluation result of the parameter combination can also be understood as the value of the parameter combination.
  • Evaluating the parameter combination can also be understood as evaluating the performance of the device during the calibration test based on the parameter combination.
  • the performance of the device may be evaluated based on one or more evaluation indicators, and an evaluation result of the parameter combination may be obtained.
  • the performance of the device may be scored based on each evaluation index to obtain a score corresponding to each index, and the score corresponding to each index may be used as an evaluation result of the parameter combination.
  • the scores corresponding to each indicator are processed, and the comprehensive score obtained after processing is used as the evaluation result of the parameter combination.
  • the evaluation result of the parameter combination of the multiple devices may be obtained by evaluating the test data of the multiple devices.
  • the test data of the plurality of devices is determined according to the data collected during the calibration test of the plurality of devices respectively based on the combination of parameters of the plurality of devices.
  • test data of the plurality of devices may be obtained by processing the data collected by the sensors during the calibration test of the plurality of devices based on the combination of parameters of the plurality of devices.
  • processing the data collected by the sensor may include processing the data collected by the sensor such as filtering, frequency reduction, or noise reduction.
  • the data collected by the sensor may include valid data and invalid data. In other words, the sensor collects both valid data and invalid data during the calibration test.
  • Effective data refers to data related to parameter calibration, or in other words, data that can be used to evaluate parameter combinations.
  • Invalid data refers to data that is not relevant to parameter calibration, or that cannot be used to evaluate parameter combinations.
  • filter the data collected by the sensor extract valid data, and use the valid data as test data.
  • the vehicle when calibrating the parameters of the ABS scene, the vehicle performs the following actions in sequence during the calibration test: the vehicle drives to the starting position, accelerates to the target speed, maintains a constant speed, depresses the brake pedal until the vehicle stops, The vehicle drives out of the test area. Data is collected as the vehicle performs all of the above actions. Valid data is the data collected during the period from when the brake pedal is depressed to the end until the vehicle stops. In this case, the collected data may be processed to extract valid data, and the valid data may be used as test data.
  • the process of processing the data collected by the sensors may be executed by the execution device in step S410, that is, the execution device in step S410 may receive the sensor data collected by the multiple devices uploaded by the multiple devices.
  • the data collected by the sensor is processed, and the test data is obtained.
  • the process of processing the data collected by the sensors may also be performed by the multiple devices respectively.
  • the multiple devices send the processed test data to the execution device in step S410.
  • test data of the multiple devices may be data collected by sensors during a calibration test performed by the multiple devices based on parameter combinations of the multiple devices.
  • the collected data can also be used as test data.
  • the vehicle when calibrating the parameters of the ABS scene, the vehicle performs the following actions in sequence during the calibration test: the vehicle drives to the starting position, accelerates to the target speed, maintains a constant speed, depresses the brake pedal until the vehicle stops, The vehicle drives out of the test area. Data is collected during the period when the vehicle is fully depressed on the brake pedal until the vehicle stops. In this case, the collected data is valid data, and the valid data is used as test data.
  • the evaluation result of the parameter combination of the multiple devices may be obtained by evaluating the test data of the multiple devices.
  • the evaluation result of the parameter combination can also be understood as the value of the test data.
  • test data may be evaluated based on one or more evaluation indicators, and an evaluation result of the parameter combination may be obtained.
  • test data can be processed through a cost function or a value function, and the obtained cost value or value value can be used as an evaluation result of the parameter combination.
  • ABS evaluation indicators include: wheel lock time and vehicle yaw angle.
  • the calibration test is carried out to obtain the test data, from which the vehicle lock-up time and the vehicle yaw angle can be determined.
  • the braking process if the vehicle locking time in the test data is longer and the vehicle yaw angle is larger, the braking performance of the current braking system is worse, that is, the value of the test data is lower, or in other words, The current parameter combination is worse.
  • the cost function can be used to evaluate the vehicle lock time and vehicle yaw angle in the test data respectively to obtain the scores corresponding to the two indicators, which can be used as the evaluation result of the current parameter combination.
  • the cost function can also be used to evaluate the vehicle lock time and vehicle yaw angle in the test data to obtain a comprehensive score, which can be used as the evaluation result of the current parameter combination.
  • the evaluation result of the parameter combination of the multiple devices is obtained by evaluating the test data of the multiple devices according to the working conditions of the multiple devices.
  • the evaluation results of the parameter combinations corresponding to the test data under different working conditions may be different.
  • the test data of the vehicle includes the braking distance.
  • the braking distance refers to the braking distance obtained when the brake is pressed to the bottom.
  • the braking distance under the condition of the same initial speed when emergency braking is applied is evaluated according to the road surface adhesion coefficient. Specifically, in the case of the same braking distance, the evaluation result of the parameter combination under the condition of a small road surface adhesion coefficient is better.
  • the vehicle when performing parameter calibration in the ABS scene, the vehicle performs the following action sequence during the calibration test: the vehicle drives to the starting position, accelerates to the target speed, keeps driving at a constant speed, depresses the brake pedal until the vehicle Stop and drive the vehicle out of the test area.
  • the vehicle performs the same calibration test based on different parameter combinations under the two different working conditions of the gravel road and the asphalt road, that is, performs the same action sequence.
  • the braking distance collected by the sensor is a; the vehicle on the asphalt road performs based on parameter combination 2
  • the braking distance collected by the sensor is b.
  • the road adhesion coefficient of the gravel road is smaller than that of the asphalt road.
  • the evaluation score of the parameter combination 1 in the vehicle on the gravel road is higher than that of the parameter combination 2 in the vehicle on the asphalt road evaluation score.
  • evaluating the test data of the device based on the working condition of the device can improve the accuracy of the evaluation result, and further improve the effect of parameter calibration.
  • the evaluation result of the parameter combination of multiple devices may be obtained by evaluating the test data of the multiple devices according to the configuration information of the multiple vehicles.
  • the evaluation results of the parameter combinations corresponding to the test data under different configurations may also be different.
  • the test data of the vehicle includes the braking distance.
  • the configuration information of the vehicle may include the payload capacity of the vehicle. Evaluate the braking distance at the same initial speed when emergency braking is applied based on the vehicle's load capacity. Specifically, in the case of the same braking distance, the evaluation result of the parameter combination in the case of a vehicle with a larger load capacity is better.
  • the load of the vehicle is 70kg, and during the calibration test based on parameter combination 3, the collected braking distance is c; the load of the vehicle is 150kg, and the collected braking distance is c during the calibration test based on parameter combination 4
  • the braking distance is d, and when c and d are equal, the evaluation score of parameter combination 3 is smaller than that of parameter combination 4.
  • the evaluation may be performed after multiplying the braking distance by a coefficient corresponding to the braking distance.
  • the coefficient corresponding to the braking distance may be set artificially.
  • the coefficient corresponding to the braking distance is negatively correlated with the load capacity of the vehicle. For example, if the load capacity of the vehicle is 70kg, the coefficient corresponding to the braking distance is 1.
  • the collected braking distance is c, and c is multiplied by 1, and then the evaluation score of parameter combination 3 is obtained based on the processed braking distance;
  • the coefficient corresponding to the moving distance is 0.95.
  • the braking distance collected is d, and d is multiplied by 0.95, and then the evaluation result of parameter combination 4 is obtained based on the processed braking distance.
  • the test data of the device is evaluated based on the configuration information of the device, which can improve the accuracy of the evaluation result and further improve the effect of parameter calibration.
  • test data of the device may also be evaluated in other manners, which is not limited in this embodiment of the present application.
  • the evaluation result of the combination of parameters of the multiple devices may be obtained from user feedback.
  • the user's subjective feedback is required. In other words, the user needs to evaluate the user experience during the calibration test to obtain the evaluation result of the parameter combination.
  • the method 400 can be applied to the scene of vehicle calibration.
  • the user can score the driving experience, and the user's score can be used as the evaluation result of the current parameter combination .
  • the evaluation index when performing parameter calibration of the vehicle steering system, includes the steering wheel power assist effect.
  • the steering wheel assist effect can be scored by the user.
  • the evaluation result fed back by the user can be written into the device through the human-computer interaction interface.
  • the first part of the evaluation results of the parameter combinations of the multiple devices is obtained from user feedback
  • the second part of the evaluation results of the parameter combinations of the multiple devices is obtained through the test data of the multiple devices obtained from the evaluation.
  • the performance of the device can be evaluated based on multiple evaluation indicators, and the evaluation result of the parameter combination can be obtained.
  • the evaluation result may be expressed in the form of a score corresponding to the evaluation indicator.
  • the evaluation result of a parameter combination may include multiple scores.
  • Part of the scores (an example of the second part in the evaluation result) in the plurality of scores can be obtained by evaluating the test data of the plurality of devices, and the rest of the scores (an example of the first part in the evaluation result) can be Based on user feedback.
  • the evaluation manner of the above parameter combination is only an example, and other manners may also be used to evaluate the parameter combination, which is not limited in this embodiment of the present application.
  • the result of user feedback and the result of evaluating the test data of multiple devices may also be processed to obtain the evaluation result of the parameter combination.
  • the means for executing step S410 may be deployed on other devices than the multiple devices, for example, the cloud shown in FIG. 2 .
  • Step S410 can be implemented in various ways, and the specific implementation of step S410 will be illustrated below by taking the execution of step S410 on the cloud as an example.
  • the cloud may receive the evaluation results of the parameter combinations of the multiple devices.
  • the cloud may receive evaluation results of parameter combinations of multiple devices fed back by users.
  • the multiple devices may respectively send the evaluation results of the parameter combinations of the multiple devices to the cloud.
  • the multiple devices include the first device, that is, the first device can send the evaluation result of the parameter combination of the first device to the cloud.
  • the cloud may receive the test data of the multiple devices and evaluate the test data of the multiple devices to obtain an evaluation result of the parameter combination of the multiple devices.
  • the multiple devices can respectively send the test data of the multiple devices to the cloud.
  • the multiple devices include the first device, that is, the first device can send the test data of the first device to the cloud.
  • the cloud may receive evaluation results of parameter combinations of some of the multiple devices and test data of the remaining devices, and evaluate the test data of the remaining devices to obtain evaluation results of the parameter combinations of the remaining devices.
  • Some of the plurality of devices send the evaluation results of their respective parameter combinations to the cloud, and the rest of the devices send their test data to the cloud.
  • the plurality of devices includes a first device and a second device.
  • the second device sends the evaluation result of the parameter combination of the second device to the cloud.
  • the first device sends the test data of the first device to the cloud.
  • the first device sends the evaluation result of the parameter combination of the first device to the cloud.
  • the second device sends the test data of the second device to the cloud.
  • the cloud may receive the first part of the evaluation results of the parameter combinations of the multiple devices and the test data of the multiple devices, and evaluate the test data of the multiple devices to obtain the parameter combinations of the multiple devices The second part of the evaluation results.
  • the multiple devices may respectively send the first part of the evaluation results of the parameter combinations of the multiple devices and the test data of the multiple devices to the cloud.
  • the plurality of devices includes a first device.
  • the first device may send the first part of the evaluation result of the parameter combination of the first device and the test data of the first device to the cloud.
  • the cloud can receive various data in a wired or wireless manner through a communication module in the cloud, which is not limited in this embodiment of the present application.
  • the execution device of step S410 may be deployed in any device among the multiple devices.
  • the executing device in step S410 may obtain the evaluation result of the local parameter combination in various ways.
  • the executing device in step S410 may read the evaluation result of the local parameter combination.
  • the executing device in step S410 may read or receive the evaluation result of the local parameter combination from the storage module.
  • “receiving” or “reading” of local data in the embodiments of the present application is collectively referred to as "reading”.
  • the executing device in step S410 may also read the local test data and evaluate the local test data to obtain the evaluation result of the local parameter combination.
  • the executing device in step S410 may also read the first part of the evaluation result of the local parameter combination and the local test data, and evaluate the local test data to obtain the first part of the evaluation result of the local parameter combination.
  • the first part may be obtained from user feedback.
  • the specific method for the execution device of step S410 to obtain the evaluation results of the parameter combinations of the devices other than the local device among the multiple devices can refer to the above-mentioned method for obtaining on the cloud, and will not be repeated here to avoid repetition.
  • evaluation results of the parameter combinations of the plurality of devices may be obtained in a single acquisition, or may also be obtained in multiple acquisitions.
  • the method 400 includes step S411 (not shown in FIG. 4 ).
  • the working conditions of the multiple devices may be collected by sensors in the multiple devices, or may be set manually.
  • step S410 For the specific acquisition method, please refer to step S410.
  • step S410 "the evaluation result of the parameter combination of multiple devices" can be replaced with “the working conditions of multiple devices", which will not be repeated here.
  • the method 400 further includes step S412 (not shown in FIG. 4 ).
  • the configuration information includes at least one of the following: tire characteristic parameters, the load weight of the vehicle, or the distribution of the load weight of the vehicle.
  • the characteristic parameters of the tire may include at least one of the following: degree of wear, section width or aspect ratio, and the like.
  • the weight of the vehicle and the distribution of the weight of the vehicle can be expressed as [weight 70kg, distributed on the left front side of the car];
  • the load weight of the vehicle and the distribution of the load weight of the vehicle can be expressed as [weighing 150kg, distributed on the front side of the car].
  • step S410 the "assessment results of parameter combinations of multiple devices" can be replaced with “configuration information of multiple devices", which will not be repeated here.
  • the at least one adjusted parameter combination may be the same or different.
  • Step S420 can be understood as optimizing the current parameter combination according to the evaluation results of the parameter combination of multiple devices.
  • the specific optimization method is realized by using the existing optimization algorithm.
  • step S420 may be implemented by a Bayesian Optimization (Bayesian Optimization) algorithm.
  • Bayesian Optimization Bayesian Optimization
  • the Gaussian process is fitted to the evaluation results of the parameter combinations of multiple devices based on the Bayesian principle, and then the mean value and variance of the evaluation results of each value point on the parameter distribution space are obtained.
  • a value point is a parameter combination.
  • an upper confidence bound (UCB) algorithm can be used to calculate the score of each value point.
  • the mean and variance of the evaluation results of each value point are brought into the UCB formula, the scores of each value point are calculated, and at least one value point with the highest score is used as the at least one adjusted parameter combination.
  • step S420 may be realized by a genetic algorithm.
  • the multiple devices include m devices, and the parameter combinations corresponding to the best p evaluation results among the evaluation results of the parameter combinations of the m devices are crossed or mutated to obtain the at least one adjusted parameter combination.
  • p is an integer smaller than m and larger than 1.
  • step S420 can be realized through the following steps:
  • Aggregate assessment results can be determined in a number of ways.
  • the weighted average of the evaluation results of the parameter combinations of the plurality of devices is used as the summary evaluation result.
  • the minimum value among the evaluation results of the parameter combinations of the plurality of devices is used as the summary evaluation result.
  • the maximum value among the evaluation results of the combination of parameters of the plurality of devices is used as the summary evaluation result.
  • the embodiment of the present application does not limit the method of determining the summary evaluation result.
  • At least one adjusted parameter combination is obtained according to the summary evaluation results.
  • Step (2) can be understood as optimizing the current parameter combination according to the summary evaluation results.
  • specific optimization method please refer to the previous description.
  • Bayesian optimization algorithm replace the "assessment results of parameter combinations of multiple devices" in the previous optimization method with "summary evaluation results”.
  • the evaluation results of each working condition can be summarized to obtain the summary evaluation result of the parameter combination, and the current parameter combination can be optimized according to the summary evaluation result, which is conducive to obtaining a better parameter combination , thereby improving the calibration efficiency.
  • the multiple devices perform the calibration test within the same period of time, which can improve the efficiency of data collection, and is conducive to real-time adjustment of the parameters of the devices based on the evaluation results of the parameter combinations of multiple devices, thereby improving the calibration efficiency.
  • different devices can perform different action sequences, quickly cover the action sequences that need to be completed, and improve the efficiency of data collection.
  • different devices can perform calibration tests based on different parameter combinations, quickly cover the parameter space, and improve the efficiency of data collection.
  • different devices can be in different working conditions, quickly cover the working conditions required for calibration, and improve the efficiency of data collection.
  • the existing vehicle calibration process needs to be calibrated in a serial manner under different working conditions and parameters. After the calibration at one test site is completed, the vehicle can be transported to another test site for parameter calibration.
  • This application implements Multiple devices in the scheme of the example can perform calibration tests at the same time under different working conditions, which improves the calibration efficiency and reduces the time cost and economic cost.
  • parameter calibration is performed based on the evaluation results of test data of multiple devices, which reduces the cognitive bias caused by different calibration engineers, improves the calibration quality, and helps to avoid the recall of equipment after leaving the factory .
  • the solution of the embodiment of the present application can also use an automatic optimization algorithm to complete the calibration, reduce the number of calibration engineers, reduce labor costs, and further reduce the influence of the calibration engineer's subjective preference on the calibration result.
  • the method 400 also includes step S430.
  • the at least one device may be determined according to at least one of the following: working conditions of multiple devices or the at least one adjusted parameter combination.
  • the at least one adjusted parameter combination is obtained by adjusting one or more parameter values in the current parameter combination.
  • the adjusted parameter items may only be valid for low adhesion coefficient working conditions, and may be effective for high adhesion coefficient working conditions. condition is not valid, in which case the at least one device may be a device operating under a low coefficient of adhesion condition.
  • evaluation results of parameter combinations of multiple devices may be obtained multiple times.
  • the device in the parameter combination of one or more devices acquired each time may be used as the at least one device.
  • the at least one adjusted parameter combination can also be understood as the adjusted parameter combination of the at least one device.
  • Step S420 can also be understood as adjusting the parameter combination of at least one of the multiple devices according to the evaluation result of the parameter combination of the multiple devices to obtain an adjusted parameter combination of the at least one device.
  • adjusting the parameter combination of at least one of the plurality of devices here refers to adjusting the parameter combination of the at least one device stored in the execution device of step S410, and is not limited to adjusting the at least one The combination of parameters stored in the device.
  • the at least one device may also be determined in other manners, which are not limited in this embodiment of the present application.
  • the at least one device may be randomly determined.
  • the at least one adjusted parameter combination may be the same or different.
  • the at least one adjusted parameter combination may be the same, in which case the parameter combinations received by the at least one device are the same.
  • the at least one adjusted parameter combination may be different.
  • the at least one device may continue to perform a calibration test based on the adjusted combination of parameters. Method 400 may be repeatedly performed until calibration is complete.
  • the at least one device includes a first device.
  • a first device For ease of description, only the first device is used as an example for illustration, and other devices in the at least one device may perform the same actions as the first device.
  • One of the at least one adjusted parameter combination is sent to the first device.
  • the first device acquires the adjusted parameter combination, and performs a calibration test based on the adjusted parameter combination.
  • sequence of actions performed in each calibration test may be the same or different.
  • the at least one device may respectively write the adjusted parameter combination into the ECU in the at least one device, and then perform a calibration test based on the adjusted parameter combination.
  • step S430 is optional.
  • the method 400 may further include step S430.
  • the executing means of step S410 may directly update the locally stored parameter combination to the adjusted parameter combination without executing step S430.
  • the method 400 further includes step S440.
  • S440 Send the to-be-executed action sequence of at least one device to the at least one device respectively.
  • the sequence of actions to be performed by the at least one device includes actions that the at least one device needs to perform during a calibration test based on the adjusted parameter combination of the at least one device.
  • the content of the calibration test to be performed is sent to each device.
  • the sequence of actions to be performed by the at least one device may be the same or different.
  • the at least one device includes a first device.
  • a first device For ease of description, only the first device is used as an example for illustration, and other devices in the at least one device may perform the same actions as the first device.
  • the first device acquires an action sequence to be executed by the first device.
  • the sequence of actions to be performed by the first device includes actions that the first device needs to perform during a calibration test based on the adjusted parameter combination of the first device.
  • step S440 is optional.
  • the sequence of actions to be performed by the at least one device may also be determined in the following manner.
  • the action sequence to be executed by the at least one device may also be determined by the at least one device itself.
  • sequence of actions to be executed by the at least one device may be preset.
  • the sequence of actions to be performed by the at least one device may be user-determined.
  • the first device may display one or more reference motion sequences.
  • the user may select an action sequence to be executed by the first device from the one or more reference action sequences.
  • the one or more reference sequences may include unfinished action sequences in the set of action sequences of the first device.
  • the one or more reference sequences may include all action sequences in the action sequence set of the first device.
  • the to-be-executed action sequence of the at least one device is determined from an action sequence set.
  • Some or all of the multiple devices may share an action sequence set, that is, the action sequence set of some or all of the multiple devices may be the same. Alternatively, the sets of action sequences for the plurality of devices may be different. This embodiment of the present application does not limit it.
  • the to-be-executed action sequence of the at least one device is determined according to at least one of the following: the actions covered by parameter calibration requirements, the working conditions of the at least one device, or the evaluation of the combination of parameters of the multiple devices result.
  • the actions required to be covered by parameter calibration may include actions required to be covered by national standards.
  • the actions covered by the national standard need to cover low-speed, medium-speed and high-speed scenes. That is, the target vehicle speed in the action sequence needs to include three situations of low speed, medium speed and high speed. If the action sequence in which the target vehicle speed is low has been executed before, the action sequence to be executed may be an action sequence in which the target vehicle speed is a medium speed or an action sequence in which the target vehicle speed is high.
  • a road section with a low adhesion coefficient for example, on ice, requires slow acceleration, that is, the pedal depth of the accelerator pedal should not be too large.
  • Road sections with a high adhesion coefficient such as asphalt roads, require rapid acceleration, that is, the accelerator pedal requires a large stepping depth.
  • the accelerator pedal used for the acceleration action in the action sequence to be executed is less than or equal to the first pedal threshold, for example, 30%; for devices on a road with a high adhesion coefficient, to be executed
  • the stepping depth of the accelerator pedal used in the acceleration action in the action sequence is greater than or equal to the second pedal threshold, for example, 90%.
  • the action sequence performed during the calibration test can be used as the waiting list. The sequence of actions to execute.
  • the specific execution mode of the action sequence to be executed can be set as required.
  • the method 400 may be applied in a scenario of vehicle calibration, and the at least one device, that is, the at least one vehicle, may execute a sequence of actions to be executed in an automatic driving manner to obtain test data.
  • a driver of the at least one vehicle may control the at least one vehicle to perform the sequence of actions to be performed.
  • the to-be-executed action sequence may be displayed on the vehicle-mounted display of the at least one vehicle, and the driver controls the vehicle to execute the corresponding to-be-executed action sequence.
  • the execution modes adopted by the at least one device may be the same or different. For example, some vehicles can be driven automatically, and some vehicles can be driven by a driver.
  • the test data of the at least one device or the evaluation result of the parameter combination of the at least one device can be obtained.
  • the method 400 may be repeatedly executed based on the test data of the at least one device, that is, the method 400 in the embodiment of the present application may be iteratively executed. For example, execute step S410 to step S420, or repeatedly execute step S410 to step S430, or repeatedly execute step S410 to step S440, and the specific execution steps can be set as required, which is not limited in this embodiment of the present application.
  • the evaluation result of the parameter combination of at least one device may be used as the evaluation result of the parameter combination of multiple devices in step S410, and then the method 400 is repeatedly executed.
  • the evaluation result of the parameter combination of the at least one device may be added to the evaluation results of the parameter combination of multiple devices in step S410, and the method 400 may be repeatedly executed.
  • a device in step S410 may correspond to the evaluation result of the parameter combination obtained by the device in multiple calibration tests, or in other words, the parameter combination evaluation result of a device may include the parameter combination obtained from multiple calibration tests evaluation results.
  • the equipment participating in each iteration that is, "the at least one device” in method 400 may be the whole set or a subset of the equipment to be calibrated, that is, the "at least one device” in method 400. Multiple Devices" or a subset. In other words, the devices participating in each iteration may be the same or different, which is not limited in this embodiment of the present application.
  • Fig. 5 shows a schematic flowchart of a parameter calibration method provided by an embodiment of the present application.
  • the method 500 in FIG. 5 is only described by taking vehicle calibration as an example, and does not limit the solution of the embodiment of the present application.
  • the method 500 in FIG. 5 may be regarded as a specific implementation manner of the method 400 shown in FIG. 4 , and reference may be made to the method 400 for a specific description. In order to avoid unnecessary repetition, some repeated descriptions are appropriately omitted when describing the method 500 .
  • the method 500 may be executed by the calibration system shown in FIG. 3 .
  • the method 500 includes step S501 to step S511.
  • the steps in method 500 can be performed once or multiple times, that is, iteratively performed.
  • a plurality of vehicles respectively upload the operating conditions of the plurality of vehicles to the cloud.
  • the plurality of vehicles are vehicles to be calibrated.
  • the vehicles to be calibrated can be vehicles of the same model.
  • the plurality of vehicles may include vehicle 1# and vehicle 2#. Vehicle 1# and vehicle 2# upload their respective working conditions to the cloud.
  • the operating conditions of the plurality of vehicles may be collected by sensors, or may also be manually set.
  • step S501 may be executed by the first communication module and the second communication module in FIG. 3 .
  • each vehicle uploads the working condition of the vehicle to the first communication module in the cloud through its second communication module.
  • the first communication module in the cloud may store the received operating conditions of each vehicle in the first storage module.
  • the working conditions of each vehicle can be stored in the working condition module in the first storage module.
  • the first communication module and the second communication module may perform data transmission in a wireless communication manner.
  • step S501 may also include: multiple vehicles upload vehicle configuration information to the cloud.
  • the vehicle configuration information may be collected by sensors, or may also be manually set.
  • step S501 is an optional step.
  • the cloud determines initial parameter combinations of the plurality of vehicles.
  • the initial parameter combinations of the multiple vehicles may be understood as an example of the parameter combinations of the multiple devices in the method 400 .
  • the initial parameter combinations of the multiple vehicles may be the same or different.
  • the initial parameter combinations of the plurality of vehicles may be respectively determined according to the operating conditions of the plurality of vehicles.
  • the initial parameter combinations of the plurality of vehicles may be randomly selected within the parameter space.
  • the parameter space can be understood as the value range of each parameter.
  • the initial parameter combinations of the plurality of vehicles may be set according to historical parameters of similar vehicle models.
  • the initial parameter combinations of the plurality of vehicles may be determined manually.
  • step S502 may be executed by the calibration module 212 in FIG. 3 .
  • step S502 is an optional step.
  • the initial parameter combination of at least one vehicle among the plurality of vehicles may also be preset in the at least one vehicle, without being determined by the cloud.
  • step S502 the method 500 also includes step S503.
  • the cloud sends the initial parameter combinations of the multiple vehicles to the multiple vehicles respectively.
  • the cloud sends the initial parameter combination X of vehicle 1# to vehicle 1#, and sends the initial parameter combination Y of vehicle 2# to vehicle 2#.
  • the initial parameter combination X and the initial parameter combination Y may be the same or different.
  • step S503 may be performed by the first communication module and the second communication module in FIG. 3 .
  • the cloud sends the initial parameter combination of each vehicle to the second communication module of each vehicle through the first communication module.
  • the second communication module of each vehicle can store the initial parameter combination in the second storage module of each vehicle.
  • the initial parameter combination is stored in the parameter module in the second storage module.
  • the parameter module can write the initial parameter combination into the control module, for example, into the ECU.
  • the method 500 includes step S504a or step S504b.
  • the cloud determines the action sequences to be executed by the plurality of vehicles.
  • the action sequences to be executed by the multiple vehicles may be the same or different.
  • the cloud may respectively select action sequences to be executed for the plurality of vehicles from the action sequence set.
  • sequence of actions to be performed by the plurality of vehicles may be manually determined.
  • the cloud may determine the sequence of actions to be executed by the multiple vehicles according to at least one of the following: the degree of coverage of actions required by national standards, the operating conditions of the multiple vehicles, or evaluation results of parameter combinations.
  • step S440 in the method 400 which will not be repeated here.
  • step S504 may be executed by the calibration module 212 in FIG. 3 .
  • the plurality of vehicles acquire the action sequences to be executed of the plurality of vehicles fed back by the user.
  • vehicle 1# obtains the action sequence M to be executed of vehicle 1# fed back by the user.
  • Vehicle 2# obtains the action sequence N to be executed of vehicle 2# fed back by the user.
  • the vehicle may obtain the action sequence to be executed of the vehicle fed back by the user through the man-machine interface.
  • the vehicle can display one or more reference action sequences through the on-board display.
  • the one or more reference action sequences may be determined according to an action sequence set and completed action sequences.
  • step S504a or step S504b is only an example, and does not limit the solution of the embodiment of the present application.
  • the action sequence of at least one vehicle among the plurality of vehicles may also be selected by the at least one vehicle from the action sequence set.
  • the user can also directly control the vehicle to execute the action sequence to be executed. That is to say, the user does not need to input the action sequence to be executed into the action sequence to be executed, but directly controls the vehicle to execute the corresponding action.
  • the method 500 also includes step S505a.
  • the action sequence M to be executed by vehicle 1# is sent to vehicle 1#
  • the action sequence N to be executed by vehicle 2# is sent to vehicle 2#.
  • step S505a may be performed by the first communication module and the second communication module in FIG. 3 .
  • the cloud sends the action sequence to be executed of each vehicle to the second communication module of each vehicle through the first communication module.
  • the second communication module of each vehicle may store the action sequence to be executed in the second storage module of each vehicle.
  • the action sequence to be executed is stored in the action module in the second storage module.
  • the action module can send the action sequence to be executed to the execution module.
  • the plurality of vehicles are controlled to perform calibration tests respectively within the same period of time.
  • the vehicle 1# is controlled to perform the calibration test
  • the vehicle 2# is controlled to perform the calibration test.
  • the initial parameter combination X of vehicle 1# is written into the ECU of vehicle 1#; the initial parameter combination Y of vehicle 2# is written into the ECU of vehicle 2#.
  • step S506 can be understood as the multiple vehicles respectively perform calibration tests based on the parameter combinations in their respective ECUs.
  • step S506 may be to control the multiple vehicles to respectively execute the action sequences to be performed by the multiple vehicles.
  • Vehicle 1# executes the action sequence M to be executed, and vehicle 2# executes the action sequence N to be executed.
  • step S506 may be executed by the execution modules of the plurality of vehicles, that is, the vehicles are controlled in an automatic driving manner to execute the sequence of actions to be executed.
  • the to-be-executed sequence can be shown to the driver through the on-board display, and the driver controls the vehicle to execute the to-be-executed action sequence.
  • different vehicles among the plurality of vehicles may adopt different driving modes, or may adopt the same driving mode, which is not limited in this embodiment of the present application.
  • the method 500 includes step S507a or S507b.
  • the multiple vehicles respectively upload the data of the multiple vehicles to the cloud.
  • vehicle 1# uploads the data of vehicle 1# to the cloud.
  • Vehicle 2# uploads the data of vehicle 2# to the cloud.
  • the data of the vehicle may include data collected by sensors during the calibration test of the vehicle.
  • the plurality of vehicles respectively upload the action sequences executed during the calibration test to the cloud. That is to say, the sequence of actions actually performed by the vehicle during the calibration test is uploaded to the cloud.
  • the data collected by the sensor may include valid data and invalid data.
  • the data collected by the sensor is valid data, and the valid data can be used as test data.
  • step S410 in the method 400 For specific description, refer to the description in step S410 in the method 400, which will not be repeated here.
  • the data uploaded by different vehicles among the plurality of vehicles may be in different forms, for example, the data uploaded by some vehicles includes valid data and invalid data, and the data uploaded by some vehicles only includes valid data.
  • the data uploaded by the multiple vehicles may also be in the same format, which is not limited in this embodiment of the present application.
  • step S507a may be performed by the first communication module and the second communication module in FIG. 3 .
  • each vehicle uploads its own data to the first communication module in the cloud through its second communication module.
  • time when different vehicles upload data may be different because the time when each vehicle completes the action sequence to be executed may be different.
  • the multiple vehicles respectively upload the evaluation results of the parameter combinations of the multiple vehicles to the cloud.
  • vehicle 1# uploads the evaluation result of the parameter combination of vehicle 1# to the cloud.
  • Vehicle 2# uploads the evaluation result of the parameter combination of vehicle 2# to the cloud.
  • the plurality of vehicles respectively upload the action sequences executed during the calibration test to the cloud.
  • the evaluation results of the parameter combinations of the plurality of vehicles may be user feedback.
  • the plurality of vehicles may obtain evaluation results of parameter combinations of the plurality of vehicles fed back by the user.
  • the vehicle can obtain the evaluation result of the parameter combination of the vehicle fed back by the user through the man-machine interface, and upload the evaluation result of the parameter combination to the cloud.
  • step S507b may be performed by the first communication module and the second communication module in FIG. 3 .
  • each vehicle uploads the evaluation result of its parameter combination to the first communication module in the cloud through its second communication module.
  • the method 500 may include step S507a and step S507b.
  • the plurality of vehicles may combine the first part of the evaluation results of the parameters of the plurality of vehicles with the data of the plurality of devices.
  • the evaluation result of the parameter combination of the vehicle may include a first part and a second part.
  • the first part can be user feedback.
  • the vehicle may obtain the first part of the evaluation result of the parameter combination of the vehicle fed back by the user.
  • the second part can be obtained by processing and evaluating the data uploaded by the vehicle in the cloud.
  • step S507a and step S507b are only examples, and for a specific description, refer to step S410 above, which will not be repeated here.
  • the method 500 may also include step S508a.
  • the cloud processes the received vehicle data.
  • step S508a includes: performing data filtering, frequency reduction or noise reduction processing on the received data of the vehicle to obtain the received test data of the vehicle.
  • step S508a may be executed by the data processing module in FIG. 3 .
  • the data processing module can store the test data in the first storage module.
  • the data processing module may store the processed test data into the calibration data module in the first storage module.
  • the cloud may receive data uploaded by each vehicle at different times.
  • the cloud may perform step S508a on the data of all vehicles after receiving the data uploaded by all vehicles.
  • the cloud may execute step S508a after receiving the data uploaded by some of the vehicles, for example, the cloud may execute step S507a for the data of the vehicle after receiving the data uploaded by any vehicle.
  • step S508a is an optional step.
  • the data uploaded by the vehicle can be processed test data, in which case there is no need for processing by the cloud.
  • step S509 the cloud judges whether the acceptance criterion is met. If the acceptance criteria are met, the process ends and the calibration is completed. If the acceptance criteria are not met, if the method 500 includes step S507a, execute step S510a, and if the method 500 does not include step S507a, execute step S511.
  • step S509 may be executed by the calibration module in FIG. 3 .
  • step S509 may be performed in the following manner.
  • the cloud may execute step S509 on the test data of all vehicles.
  • test data of the vehicle may be stored in the first storage module.
  • the calibration module may monitor the first storage module, and execute step S509 after the first storage module receives the test data of all the vehicles in the plurality of vehicles.
  • the cloud may execute step S509 on the test data of some of the vehicles.
  • test data of the vehicle may be stored in the first storage module.
  • the calibration module may monitor the first storage module, and execute step S509 based on the test data of any vehicle after the first storage module receives the test data of any vehicle.
  • step S509 may be performed in the following manner.
  • the cloud may execute step S509 on the evaluation results of the parameter combinations of all vehicles.
  • the evaluation result of the parameter combination of the vehicle may be stored in the first storage module.
  • the calibration module may monitor the first storage module, and execute step S509 after the first storage module receives the evaluation results of the parameter combinations of all the vehicles in the plurality of vehicles.
  • the cloud may execute step S508 on the evaluation results of the parameter combinations of some of the vehicles.
  • the evaluation result of the parameter combination of the vehicle may be stored in the first storage module.
  • the calibration module can monitor the first storage module, and after the first storage module receives the evaluation result of the parameter combination of any vehicle, step S509 is executed based on the evaluation result of the parameter combination of the vehicle.
  • the cloud evaluates the test data of the vehicle, and obtains an evaluation result of the parameter combination of the vehicle.
  • the cloud evaluates the test data of the vehicle according to the working conditions of the vehicle to obtain an evaluation result of the parameter combination of the vehicle.
  • step S410 in the method 400 For a specific evaluation method, reference may be made to step S410 in the method 400, which will not be repeated here.
  • step S510a may be performed by the calibration module in FIG. 3 .
  • the cloud obtains at least one adjusted parameter combination according to the evaluation result of the parameter combination of the vehicle.
  • the cloud may execute step S511 after obtaining evaluation results of parameter combinations of some of the vehicles participating in this iteration.
  • step S511 is executed for the first time, that is, in the first iteration process, the vehicles participating in the iteration are all the vehicles to be calibrated.
  • the cloud can execute step S511 in real time after obtaining the evaluation result of any vehicle's parameter combination.
  • step S511 may be implemented by using a Bayesian optimization algorithm.
  • the cloud can use the Bayesian principle to fit the evaluation results of the current vehicle parameter combinations to obtain the mean and variance of the evaluation results of each value point in the parameter distribution space, and then determine at least one with the highest score according to the UCB formula. A value point, using the at least one value point as the at least one adjusted parameter combination.
  • the cloud can use the Bayesian principle to optimize the mean and variance of the evaluation results of each value point in the existing parameter distribution space based on the currently obtained evaluation results of the vehicle parameter combination, and then determine the parameter corresponding to the maximum value according to the UCB formula Combination, the parameter combination after the parameter combination is adjusted.
  • the evaluation results of each value point on the existing parameter distribution space can be obtained by fitting based on the evaluation results of the previously obtained parameter combinations of the vehicle.
  • step S511 may be implemented using a genetic algorithm.
  • one or more parameter combinations among the evaluation results of the at least two vehicle parameter combinations are preferably crossed or mutated,
  • the obtained parameter combination is used as the at least one adjusted parameter combination.
  • the cloud may execute step S511 after obtaining the evaluation results of the parameter combinations of all the vehicles participating in the iteration this time.
  • step S511 may be implemented by using a Bayesian optimization algorithm.
  • the cloud can use the Bayesian principle to fit the evaluation results of the parameter combinations of the vehicles participating in this iteration to obtain the mean and variance of the evaluation results of each value point in the parameter distribution space, and then determine the highest score at least according to the UCB formula.
  • the cloud can use the Bayesian principle to optimize the mean and variance of the evaluation results of each value point in the existing parameter distribution space based on the evaluation results of the parameter combinations of the vehicles participating in this iteration, and then determine the corresponding value of the maximum value according to the UCB formula
  • the parameter combination is the parameter combination after the parameter combination is adjusted.
  • the evaluation results of each value point on the existing parameter distribution space can be obtained by fitting based on the evaluation results of the previously obtained parameter combinations of the vehicle.
  • the cloud may adjust the parameter combination of the vehicle according to the evaluation result of the parameter combination obtained in the current iteration process and the evaluation result of the parameter combination obtained in the previous iteration process.
  • the cloud can adjust the parameter combination of the vehicle according to the currently obtained evaluation results of all parameter combinations.
  • step S511 may be implemented using a genetic algorithm.
  • one or more parameter combinations among the evaluation results of the at least two vehicle parameter combinations are preferably crossed or mutated,
  • the obtained parameter combination is used as the at least one adjusted parameter combination.
  • the evaluation result of the parameter combination of at least two vehicles may be obtained in one iteration process, or may be obtained in multiple iteration processes.
  • the cloud may adjust the parameter combination of the vehicle according to the evaluation result of the parameter combination obtained in the current iteration process and the evaluation result of the parameter combination obtained in the previous iteration process.
  • the cloud can adjust the parameter combination of the vehicle according to the currently obtained evaluation results of all parameter combinations.
  • step S511 may also include the following steps:
  • the weighted average of the evaluation results of the parameter combinations of the equipment participating in the iteration this time is used as the summary evaluation result.
  • the minimum value among the evaluation results of the parameter combinations of the devices participating in the iteration this time is used as the summary evaluation result.
  • the maximum value among the evaluation results of the parameter combinations of the devices participating in the iteration this time is used as the summary evaluation result.
  • the embodiment of the present application does not limit the method of determining the summary evaluation result.
  • At least one adjusted parameter combination is obtained according to the summary evaluation results.
  • Step (2) can be understood as optimizing the current parameter combination according to the summary evaluation results.
  • specific optimization method please refer to the description above.
  • Bayesian optimization algorithm replace the "assessment result of the parameter combination of the equipment" in the optimization method above with the "summary evaluation result”.
  • the at least one adjusted parameter combination is sent to at least one vehicle, that is, the vehicle participating in the next iteration, the initial parameter combination in steps S504 to S511 is replaced by the adjusted parameter combination, and step S504 to step
  • the multiple vehicles in S511 are replaced with the at least one vehicle, and step S504 to step S511 are repeatedly executed until the calibration is completed.
  • steps S504 to S511 are executed for the first time, the vehicles participating in the iteration are all the vehicles to be calibrated and the plurality of vehicles.
  • the parameter combinations of the multiple vehicles are the initial parameter combinations of the multiple vehicles.
  • step S506 to step S511 are executed later, the vehicles participating in each iteration may be some or all of the plurality of vehicles, and the parameter combination is the adjusted parameter combination obtained in step S511.
  • the at least one vehicle is determined according to at least one of the following: operating conditions of the multiple devices or the at least one adjusted parameter combination.
  • the at least one vehicle may be determined by the calibration module 212 in FIG. 3 .
  • the at least one vehicle may be a vehicle in the currently obtained evaluation result of the parameter combination of the vehicle.
  • step S511 is executed in real time after the cloud obtains the evaluation result of the parameter combination of vehicle 1#.
  • vehicle 1# can be one of the vehicles participating in the next iteration.
  • step S430 for the method of determining the at least one vehicle, reference may be made to step S430 in the foregoing method 400 , which will not be repeated here.
  • the vehicles participating in each iteration can be the same or different.
  • some or all of the multiple vehicles in the method 500 may also be replaced with test benches.
  • the test bench can be simulated and set to different working conditions. Using the test bench to replace part or all of the vehicles can be applied to scenarios such as engine calibration or motor calibration, further reducing calibration costs, improving the efficiency of data collection, and then improving calibration efficiency. .
  • the updated working conditions of the test bench can be uploaded to the cloud.
  • the working conditions of the same vehicle usually do not change during the calibration process, and multiple vehicles can upload the working conditions once in step S501.
  • the working conditions of the test benches can be set as required. After replacing some or all of the vehicles with test benches, when the working conditions of the test benches change, the updated test bench Working conditions are uploaded to the cloud in real time.
  • the data transmission between the first communication module and the second communication module may be performed by wireless communication, or data transmission may also be performed by wired communication.
  • the device for parameter calibration can be deployed in a cloud environment, which is an entity that uses basic resources to provide users with cloud services under the cloud computing model.
  • the cloud environment includes a cloud data center and a cloud service platform.
  • the cloud data center includes a large number of basic resources (including computing resources, storage resources and network resources) owned by the cloud service provider.
  • the computing resources included in the cloud data center can be a large number of computing resources. devices (such as servers).
  • the device for parameter calibration can be a server for calibrating parameters in the cloud data center; the device for parameter calibration can also be a software device deployed on a server or a virtual machine in the cloud data center, and the software device is used for parameter calibration.
  • the software device can be distributed on multiple servers, or distributed on multiple virtual machines, or distributed on virtual machines and servers.
  • the parameter calibration device is abstracted into a parameter calibration cloud service by the cloud service provider on the cloud service platform and provided to the user. After the user purchases the cloud service, the cloud environment uses the parameter calibration device to provide the user with the parameter calibration service.
  • the interface uploads the data of the device or the evaluation results of the parameter combination of the device to the cloud environment, and the parameter calibration device optimizes the parameter combination of the device, and the optimization result can be returned to the device where the user is located.
  • the device for parameter calibration is a software device
  • different modules of the device for parameter calibration can be deployed in different environments or devices.
  • part of the parameter calibration device is deployed in terminal equipment (such as: vehicles, smart phones, laptops, tablets, personal desktop computers, smart cameras), and the other part is deployed in the cloud data center (specifically deployed in the cloud data center in the server or virtual machine).
  • the functions of the method for parameter calibration are implemented in coordination among various parts of the device deployed in different environments or equipment for parameter calibration. For example, reference may be made to FIG. 3 for the deployment of the device for parameter calibration.
  • this application does not make a restrictive division of which parts of the parameter calibration device are deployed in the terminal computing device and which parts are deployed in the cloud data center. In actual application, it can be determined according to the computing capability of the terminal computing device or specific application requirements. Adaptive deployment.
  • the device for parameter calibration is a software device
  • the device for parameter calibration can also be deployed separately on a computing device in any environment (for example: separately deployed on a terminal device or separately deployed on a computing device in a cloud data center) .
  • the device of the embodiment of the present application will be described below with reference to FIG. 6 to FIG. 7 . It should be understood that the device described below can execute the method of the aforementioned embodiment of the present application. In order to avoid unnecessary repetition, repeated descriptions are appropriately omitted when introducing the device of the embodiment of the present application below.
  • Fig. 6 is a schematic block diagram of an apparatus for parameter calibration according to an embodiment of the present application.
  • the apparatus 3000 for parameter calibration shown in FIG. 6 includes an acquisition unit 3010 and a processing unit 3020 .
  • the acquisition unit 3010 and the processing unit 3020 can be used to implement the parameter calibration method of the embodiment of the present application, specifically, can be used to implement the method 400 or the method 500.
  • the obtaining unit 3010 is configured to obtain the evaluation results of the parameter combinations of multiple devices, and the evaluation results of the parameter combinations of the multiple devices are based on the parameter combinations of the multiple devices in the first period of time. obtained by performing a calibration test within , the first period of time is less than or equal to the first threshold.
  • the processing unit 3020 is configured to obtain at least one adjusted parameter combination according to evaluation results of parameter combinations of multiple devices.
  • the parameter combinations of multiple devices are the same, and the working conditions of the multiple devices are different, and the processing unit 3020 is specifically configured to: process the evaluation results of the parameter combinations of multiple devices to obtain a summary Evaluation results; at least one adjusted parameter combination is obtained based on the aggregated evaluation results.
  • the apparatus further includes: a sending unit 3030, configured to send at least one adjusted parameter combination to at least one device.
  • At least one device is determined according to at least one of the following: operating conditions of multiple devices or at least one adjusted parameter combination.
  • the evaluation result of the parameter combination of multiple devices is obtained by evaluating the test data of the multiple devices, and the test data of the multiple devices is based on the parameters of the multiple devices respectively It is determined by combining the data collected during the calibration test in the first period.
  • the evaluation results of the parameter combinations of multiple devices are obtained through user feedback.
  • the sending unit 3030 is further configured to: send the sequence of actions to be performed by at least one device to the at least one device respectively, the sequence of actions to be performed by the at least one device includes Actions that need to be performed during the calibration test with the adjusted parameter combination.
  • the to-be-executed action sequence of at least one device is determined according to at least one of the following: actions covered by parameter calibration requirements, working conditions of at least one device, or evaluation of parameter combinations of multiple devices result.
  • the multiple devices include at least one of the following: a vehicle or a test bench.
  • the obtaining unit 30103 is configured to obtain an adjusted parameter combination, the adjusted parameter combination is obtained according to the evaluation results of the parameter combinations of multiple devices, the multiple devices include the first device, and the multiple The evaluation result of the parameter combination of the device is obtained by performing a calibration test by multiple devices respectively based on the parameter combination of the multiple devices within a first period of time, and the first period of time is less than or equal to the first threshold.
  • the processing unit 3020 is configured to control the first device to perform a calibration test based on the adjusted parameter combination.
  • the parameter combinations of multiple devices are the same, and the working conditions of the multiple devices are different.
  • the adjusted parameter combination is obtained according to the summary evaluation results, and the summary evaluation results are obtained by combining the multiple devices.
  • the evaluation results of the parameter combinations are processed to obtain.
  • the first device is determined according to at least one of the following: working conditions of multiple devices or adjusted parameter combinations.
  • the evaluation result of the parameter combination of multiple devices is obtained by evaluating the test data of the multiple devices, and the test data of the multiple devices is based on the parameters of the multiple devices respectively It is determined by combining the data collected during the calibration test in the first period.
  • the evaluation results of the parameter combinations of multiple devices are obtained through user feedback.
  • the acquisition unit 3010 is also configured to: acquire the sequence of actions to be performed by the first device, the sequence of actions to be performed by the first device includes the process of the first device performing a calibration test based on the adjusted parameter combination actions to be performed in the .
  • the to-be-executed action sequence of the first device is determined according to at least one of the following: actions covered by parameter calibration requirements, working conditions of the first device, or evaluation of parameter combinations of multiple devices result.
  • the first device is: a vehicle or a test bench.
  • unit here may be implemented in the form of software and/or hardware, which is not specifically limited.
  • a "unit” may be a software program, a hardware circuit or a combination of both to realize the above functions.
  • the hardware circuitry may include application specific integrated circuits (ASICs), electronic circuits, processors (such as shared processors, dedicated processors, or group processors) for executing one or more software or firmware programs. etc.) and memory, incorporating logic, and/or other suitable components to support the described functionality.
  • ASICs application specific integrated circuits
  • processors such as shared processors, dedicated processors, or group processors for executing one or more software or firmware programs. etc.
  • memory incorporating logic, and/or other suitable components to support the described functionality.
  • the units of each example described in the embodiments of the present application can be realized by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed by hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art may use different methods to implement the described functions for each specific application, but such implementation should not be regarded as exceeding the scope of the present application.
  • Fig. 7 is a schematic diagram of the hardware structure of the parameter calibration device provided by the embodiment of the present application.
  • the parameter calibration apparatus 5000 shown in FIG. 7 includes a memory 5001 , a processor 5002 , a communication interface 5003 and a bus 5004 .
  • the memory 5001 , the processor 5002 , and the communication interface 5003 are connected to each other through a bus 5004 .
  • the memory 5001 may be a read only memory (read only memory, ROM), a static storage device, a dynamic storage device or a random access memory (random access memory, RAM).
  • the memory 5001 may store a program, and when the program stored in the memory 5001 is executed by the processor 5002, the processor 5002 is configured to execute each step of the parameter calibration method of the embodiment of the present application. Specifically, the processor 5002 may execute the method 400 shown in FIG. 4 above, or execute the method 500 shown in FIG. 5 above.
  • the processor 5002 may adopt a general-purpose central processing unit (central processing unit, CPU), a microprocessor, an application specific integrated circuit (application specific integrated circuit, ASIC), a graphics processing unit (graphics processing unit, GPU) or one or more
  • the integrated circuit is used to execute related programs to realize the parameter calibration method of the method embodiment of the present application.
  • the processor 5002 may also be an integrated circuit chip with signal processing capabilities.
  • each step of the parameter calibration method of the present application may be completed by an integrated logic circuit of hardware in the processor 5002 or instructions in the form of software.
  • the above-mentioned processor 5002 can also be a general-purpose processor, a digital signal processor (digital signal processing, DSP), an application-specific integrated circuit (ASIC), a ready-made programmable gate array (field programmable gate array, FPGA) or other programmable logic devices, Discrete gate or transistor logic devices, discrete hardware components.
  • DSP digital signal processing
  • ASIC application-specific integrated circuit
  • FPGA field programmable gate array
  • Various methods, steps, and logic block diagrams disclosed in the embodiments of the present application may be implemented or executed.
  • a general-purpose processor may be a microprocessor, or the processor may be any conventional processor, or the like.
  • the steps of the method disclosed in connection with the embodiments of the present application may be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor.
  • the software module can be located in a mature storage medium in the field such as random access memory, flash memory, read-only memory, programmable read-only memory or electrically erasable programmable memory, register.
  • the storage medium is located in the memory 5001, and the processor 5002 reads the information in the memory 5001, and combines its hardware to complete the functions required by the units included in the device shown in Figure 6, or execute the method shown in Figure 4 or Figure 5 shows the method of parameter calibration.
  • the communication interface 5003 implements communication between the apparatus 5000 and other devices or communication networks by using a transceiver device such as but not limited to a transceiver. For example, evaluation results of parameter combinations and the like can be obtained through the communication interface 5003 .
  • the bus 5004 may include a pathway for transferring information between various components of the device 5000 (eg, memory 5001, processor 5002, communication interface 5003).
  • apparatus 5000 only shows memory, processor, and communication interface, those skilled in the art should understand that the apparatus 5000 may also include other devices necessary for normal operation during specific implementation. Meanwhile, according to specific needs, those skilled in the art should understand that the apparatus 5000 may also include hardware devices for implementing other additional functions. In addition, those skilled in the art should understand that the apparatus 5000 may also only include components necessary to realize the embodiment of the present application, and does not necessarily include all the components shown in FIG. 7 .
  • the processor in the embodiment of the present application may be a central processing unit (central processing unit, CPU), and the processor may also be other general-purpose processors, digital signal processors (digital signal processor, DSP), application specific integrated circuits (application specific integrated circuit, ASIC), off-the-shelf programmable gate array (field programmable gate array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
  • a general-purpose processor may be a microprocessor, or the processor may be any conventional processor, or the like.
  • the memory in the embodiments of the present application may be a volatile memory or a nonvolatile memory, or may include both volatile and nonvolatile memories.
  • the non-volatile memory can be read-only memory (read-only memory, ROM), programmable read-only memory (programmable ROM, PROM), erasable programmable read-only memory (erasable PROM, EPROM), electrically programmable Erases programmable read-only memory (electrically EPROM, EEPROM) or flash memory.
  • Volatile memory can be random access memory (RAM), which acts as external cache memory.
  • RAM random access memory
  • static random access memory static random access memory
  • DRAM dynamic random access memory
  • DRAM synchronous dynamic random access memory Access memory
  • SDRAM synchronous dynamic random access memory
  • double data rate synchronous dynamic random access memory double data rate SDRAM, DDR SDRAM
  • enhanced synchronous dynamic random access memory enhanced SDRAM, ESDRAM
  • serial link DRAM SLDRAM
  • direct memory bus random access memory direct rambus RAM, DR RAM
  • the above-mentioned embodiments may be implemented in whole or in part by software, hardware, firmware or other arbitrary combinations.
  • the above-described embodiments may be implemented in whole or in part in the form of computer program products.
  • the computer program product comprises one or more computer instructions or computer programs.
  • the processes or functions according to the embodiments of the present application will be generated in whole or in part.
  • the computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable devices.
  • the computer instructions may be stored in or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from a website, computer, server or data center Transmission to another website site, computer, server or data center by wired (such as infrared, wireless, microwave, etc.).
  • the computer-readable storage medium may be any available medium that can be accessed by a computer, or a data storage device such as a server or a data center that includes one or more sets of available media.
  • the available media may be magnetic media (eg, floppy disk, hard disk, magnetic tape), optical media (eg, DVD), or semiconductor media.
  • the semiconductor medium may be a solid state drive.
  • the present application also provides a computer program product, the computer program product including: computer program code, when the computer program code is run on the computer, the computer is made to execute the method in any one of the foregoing method embodiments.
  • the present application also provides a computer-readable medium, the computer-readable medium stores program code, and when the program code is run on a computer, the computer is made to execute the method in any one of the foregoing method embodiments.
  • the present application also provides an electronic device, which includes the parameter calibration device in any one of the foregoing device embodiments.
  • the disclosed systems, devices and methods may be implemented in other ways.
  • the device embodiments described above are only illustrative.
  • the division of the units is only a logical function division. In actual implementation, there may be other division methods.
  • multiple units or components can be combined or May be integrated into another system, or some features may be ignored, or not implemented.
  • the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces, and the indirect coupling or communication connection of devices or units may be in electrical, mechanical or other forms.
  • the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.
  • each functional unit in each embodiment of the present application may be integrated into one processing unit, each unit may exist separately physically, or two or more units may be integrated into one unit.
  • the functions described above are realized in the form of software function units and sold or used as independent products, they can be stored in a computer-readable storage medium.
  • the technical solution of the present application is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including Several instructions are used to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods described in the various embodiments of the present application.
  • the aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (read-only memory, ROM), random access memory (random access memory, RAM), magnetic disk or optical disc and other media that can store program codes. .

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Geometry (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Computer Hardware Design (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Electromagnetism (AREA)
  • Signal Processing (AREA)
  • Testing Or Calibration Of Command Recording Devices (AREA)

Abstract

本申请提供了一种参数标定的方法及装置，可以应用在智能汽车、新能源汽车、网联汽车、智能驾驶汽车等车辆上。该方法包括：获取多个设备的参数组合的评估结果，多个设备的参数组合的评估结果是通过多个设备分别基于多个设备的参数组合在第一时段内进行标定试验得到的；根据多个设备的参数组合的评估结果得到至少一个调整后的参数组合。本申请的方案能够提高参数标定的效率。

Description

参数标定的方法及装置
本申请要求于2021年09月29日提交中国专利局、申请号为202111148039.4、申请名称为“参数标定的方法及装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及数据处理领域，并且更具体地，涉及一种参数标定的方法及装置。
背景技术
参数标定指的是在多种条件下，例如，在不同的硬件条件或不同的环境条件下，寻找均能适用的一组参数的过程。以车辆标定为例，车辆标定指的是在控制器硬件、控制软件和相关传感器等器件确定后，为了得到满意的车辆性能、满足车辆出厂前的要求以及达到相关国家标准，对控制软件中的参数进行优化的过程。
车辆标定结果是影响车辆相关性能的重要因素。在车辆上市前，通常由经验丰富的标定工程师基于样车进行标定。为了使得车辆在各种工况下均满足性能要求，需要在不同的工况下进行标定实验，寻找适合的参数组合。受到工况、环境等的限制，标定过程耗时较久，标定成本较高。以车辆中的制动系统为例，通常需要经过“两夏一冬”的漫长标定过程才能完成制动系统中的参数标定，严重影响了标定效率。
因此，如何提高参数标定的效率称为一个亟待解决的问题。
发明内容
本申请提供一种参数标定的方法及装置，能够提高参数标定的效率。
第一方面，提供了一种参数标定的方法，包括：获取多个设备的参数组合的评估结果，多个设备的参数组合的评估结果是通过多个设备分别基于多个设备的参数组合在第一时段内进行标定试验得到的，第一时段小于或等于第一阈值；根据多个设备的参数组合的评估结果得到至少一个调整后的参数组合。
根据本申请实施例的方案，该多个设备在同一时段内进行标定试验，能够提高数据采集的效率，有利于基于多个设备的参数组合的评估结果实时调整设备的参数，进而提高标定效率，减少标定周期，降低标定成本。
本申请实施例的方案的多个设备可以同时处于不同的工况下同时进行标定试验，提高了标定效率，减少了时间成本与经济成本。
本申请实施例的方案的多个设备可以同时处于不同的工况下，能够实现多个设备的数据的实时汇总以及参数优化，有利于在多种工况下同时寻找兼容该多种工况的参数组合，提高参数的稳定性，进而提高标定效率。
本申请实施例的方案中，基于多个设备的测试数据的评估结果进行参数标定，减少了 不同的标定工程师带来的认知偏差，提高了标定质量，有利于避免设备出厂后再召回的情况。
本申请实施例的方案还可以采用自动优化算法完成标定，减少标定工程师的数量，降低人力成本，进一步减少标定工程师的主观偏好对标定结果的影响。
该多个设备在第一时段内进行标定试验也可以理解为该多个设备可以在同一时段内进行标定试验，例如，该多个设备可以同时进行标定试验。
参数组合中的参数即为需要标定的参数。参数组合中包括至少一个参数。
参数标定的装置可以属于该多个设备，或者，也可以不属于该多个设备。
该多个设备的参数组合可以相同，也可以不同。
该多个设备的参数组合的评估结果用于指示该多个设备的参数组合的质量。参数组合的评估结果也可以理解为参数组合的价值。
具体地，可以基于一个或多个评价指标对设备的表现进行评估，得到参数组合的评估结果。
该至少一个调整后的参数组合可以相同，也可以不同。
结合第一方面，在第一方面的某些实现方式中，多个设备的参数组合相同，多个设备所处的工况不同，以及根据多个设备的参数组合的评估结果得到至少一个调整后的参数组合，包括：将多个设备的参数组合的评估结果进行处理，得到汇总评估结果；根据汇总评估结果得到至少一个调整后的参数组合。
汇总评估结果可以采用多种方式确定。例如，将该多个设备的参数组合的评估结果的加权平均值作为汇总评估结果。再如，将该多个设备的参数组合的评估结果中的最小值作为汇总评估结果。再如，将该多个设备的参数组合的评估结果中的最大值作为汇总评估结果。本申请实施例对汇总评估结果的确定方式不做限定。
这样，有利于汇总同一个参数组合在不同工况下的评估结果，得到该参数组合的汇总评估结果，根据汇总评估结果优化当前的参数组合，有利于得到更优的参数组合，进而提高标定效率。
结合第一方面，在第一方面的某些实现方式中，方法还包括：将至少一个调整后的参数组合发送至至少一个设备。
结合第一方面，在第一方面的某些实现方式中，至少一个设备是根据以下至少一项确定的：多个设备所处的工况或至少一个调整后的参数组合。
结合第一方面，在第一方面的某些实现方式中，多个设备的参数组合的评估结果是通过对多个设备的测试数据进行评估得到的，多个设备的测试数据是根据多个设备分别基于多个设备的参数组合在第一时段内进行标定试验的过程中采集到的数据确定的。
示例性地，该多个设备的测试数据可以是通过对该多个设备分别基于该多个设备的参数组合进行标定试验的过程中传感器采集到的数据进行处理得到的。
例如，对传感器采集到的数据进行处理可以包括对传感器采集到的数据进行过滤、降频或降噪等处理。
可替换地，该多个设备的测试数据可以是为根据该多个设备分别基于该多个设备的参数组合进行标定试验的过程中传感器采集到的数据。
结合第一方面，在第一方面的某些实现方式中，多个设备的参数组合的评估结果是根 据该多个设备所处的工况对该多个设备的测试数据进行评估得到的。
在本申请实施例中，基于设备所处的工况对设备的测试数据进行评估，能够提高评估结果的准确性，进而提高参数标定的效果。
结合第一方面，在第一方面的某些实现方式中，多个设备的参数组合的评估结果可以是根据该多个车辆的配置信息对该多个设备的测试数据进行评估得到的。
在本申请实施例中，基于设备的配置信息对设备的测试数据进行评估，能够提高评估结果的准确性，进而提高参数标定的效果。
结合第一方面，在第一方面的某些实现方式中，多个设备的参数组合的评估结果是由用户反馈得到的。
在本申请实施例中，参数组合的评估结果可以由用户反馈得到，能够充分考虑用户的感受，有利于提高用户体验。
结合第一方面，在第一方面的某些实现方式中，方法还包括：将至少一个设备的待执行动作序列分别发送至至少一个设备，至少一个设备的待执行动作序列包括至少一个设备在基于至少一个设备的调整后的参数组合进行标定试验的过程中所需要执行的动作。
该至少一个设备的待执行动作序列可以相同，也可以不同。
示例性地，该至少一个设备的待执行动作序列也可以是由该至少一个设备自行确定的。
可替换地，该至少一个设备的待执行动作序列可以是预先设置的。
可替换地，该至少一个设备的待执行动作序列可以是用户确定的。
结合第一方面，在第一方面的某些实现方式中，至少一个设备的待执行动作序列是根据以下至少一项确定的：参数标定要求覆盖的动作、至少一个设备所处的工况或多个设备的参数组合的评估结果。
结合第一方面，在第一方面的某些实现方式中，多个设备包括以下至少一项：车辆或测试台架。
结合第一方面，在第一方面的某些实现方式中，多个设备包括：传感器。
第二方面，提供了一种参数标定的方法，包括：获取调整后的参数组合，调整后的参数组合是根据多个设备的参数组合的评估结果得到的，多个设备包括第一设备，多个设备的参数组合的评估结果是通过多个设备分别基于多个设备的参数组合在第一时段内进行标定试验得到的，第一时段小于或等于第一阈值；控制第一设备基于调整后的参数组合进行标定试验。
根据本申请实施例的方案，该多个设备在同一时段内进行标定试验，能够提高数据采集的效率，有利于基于多个设备的参数组合的评估结果实时调整设备的参数，进而提高标定效率，减少标定周期，降低标定成本。
本申请实施例的方案的多个设备可以同时处于不同的工况下同时进行标定试验，提高了标定效率，减少了时间成本与经济成本。
本申请实施例的方案的多个设备可以同时处于不同的工况下，能够实现多个设备的数据的实时汇总以及参数优化，有利于在多种工况下同时寻找兼容该多种工况的参数组合，提高参数的稳定性，进而提高标定效率。
本申请实施例的方案中，基于多个设备的测试数据的评估结果进行参数标定，减少了 不同的标定工程师带来的认知偏差，提高了标定质量，有利于避免设备出厂后再召回的情况。
本申请实施例的方案还可以采用自动优化算法完成标定，减少标定工程师的数量，降低人力成本，进一步减少标定工程师的主观偏好对标定结果的影响。
结合第二方面，在第二方面的某些实现方式中，多个设备的参数组合相同，多个设备所处的工况不同，调整后的参数组合是根据汇总评估结果得到的，汇总评估结果是通将多个设备的参数组合的评估结果进行处理得到的。
结合第二方面，在第二方面的某些实现方式中，第一设备是根据以下至少一项确定的：多个设备所处的工况或调整后的参数组合。
结合第二方面，在第二方面的某些实现方式中，多个设备的参数组合的评估结果是通过对多个设备的测试数据进行评估得到的，多个设备的测试数据是根据多个设备分别基于多个设备的参数组合在第一时段内进行标定试验的过程中采集到的数据确定的。
结合第二方面，在第二方面的某些实现方式中，多个设备的参数组合的评估结果是由用户反馈得到的。
结合第二方面，在第二方面的某些实现方式中，方法还包括：第一设备获取第一设备的待执行动作序列，第一设备的待执行动作序列包括第一设备在基于调整后的参数组合进行标定试验的过程中所需要执行的动作。
结合第二方面，在第二方面的某些实现方式中，第一设备的待执行动作序列是根据以下至少一项确定的：参数标定要求覆盖的动作、第一设备所处的工况或多个设备的参数组合的评估结果。
结合第二方面，在第二方面的某些实现方式中，第一设备为：车辆或测试台架。
第三方面，提供了一种参数标定的装置，所述装置包括用于执行上述第一方面以及第一方面中的任意一种实现方式中的方法的模块或单元。
第四方面，提供了一种参数标定的装置，所述装置包括用于执行上述第二方面以及第二方面中的任意一种实现方式中的方法的模块或单元。
应理解，在上述第一方面中对相关内容的扩展、限定、解释和说明也适用于第二方面、第三方面和第四方面中相同的内容。
第五方面，提供了一种参数标定的装置，该装置包括：存储器，用于存储程序；处理器，用于执行所述存储器存储的程序，当所述存储器存储的程序被执行时，所述处理器用于执行上述第一方面中的任意一种实现方式中的方法。
可选地，该装置还可以包括通信接口。
第六方面，提供了一种参数标定的装置，该装置包括：存储器，用于存储程序；处理器，用于执行所述存储器存储的程序，当所述存储器存储的程序被执行时，所述处理器用于执行上述第二方面中的任意一种实现方式中的方法。
可选地，该装置还可以包括通信接口。
第七方面，提供一种计算机可读介质，该计算机可读介质存储用于设备执行的指令，该指令用于执行第一方面或第二方面中的任意一种实现方式中的方法。
第八方面，提供一种包含指令的计算机程序产品，当该计算机程序产品在计算机上运行时，使得计算机执行上述第一方面或第二方面中的任意一种实现方式中的方法。
第九方面，提供一种芯片，所述芯片包括处理器与通信接口，所述处理器通过所述通信接口读取存储器上存储的指令，执行上述第一方面或第二方面中的任意一种实现方式中的方法。
可选地，该芯片还可以包括存储器，所述存储器中存储有指令，所述处理器用于执行所述存储器上存储的指令，当所述指令被执行时，所述处理器用于执行第一方面或第二方面中的任意一种实现方式中的方法。
第十方面，提供了一种电子设备，该电子设备包括上述第三方面或第四方面中的任意一种实现方式中的参数标定的装置。
附图说明
图1是本申请实施例提供的一种应用场景的示意图；
图2是本申请实施例提供的一种标定系统的示意图；
图3是本申请实施例提供的另一种标定系统的示意图；
图4是本申请实施例提供的一种参数标定的方法的示意性流程图；
图5是本申请实施例提供的另一种参数标定的方法的示意性流程图；
图6是本申请实施例提供的一种参数标定的装置的示意图；
图7是本申请实施例提供的另一种参数标定的装置的示意图。
具体实施方式
下面将结合附图，对本申请中的技术方案进行描述。
本申请实施例的方案能够应用于车辆标定、传感器标定以及其他进行参数标定的领域。
参数标定指的是确定在多种条件下，例如，在不同的硬件条件或不同的环境条件下，均能适用的一组参数。或者说，参数标定指的是在多种条件下对参数进行优化的过程。
下面对车辆标定和传感器标定这两种应用场景进行说明。
(1)车辆标定
车辆标定指的是在控制器硬件、控制软件和相关传感器等器件确定后，为了得到满意的车辆性能、满足车辆出厂前的要求以及达到相关国家标准，对控制软件中的参数进行优化的过程。
车辆中的电子控制单元(electronic control unit，ECU)，又称为“行车电脑”，利用各种传感器采集的数据进行运算得到控制信号，以控制车辆中的各个执行机构执行相应的动作。示例性地，ECU的控制范围可以包括巡航控制、灯光控制、安全气囊控制、悬架控制、排气控制或制动控制等。各个ECU之间可以通过总线进行数据交互。本申请实施例中的控制软件中的参数可以为ECU中的参数。
现有的参数标定过程通常以串行方式在不同的工况下进行标定实验。具体地，在一次标定实验中，标定工程师设置参数后，进行标定实验，根据实验中收集到的数据调整参数。在同一工况下循环上述标定实验的过程，待该工况下的参数标定完成后，转移至下一个工况进行标定。该方案的标定效率较低，标定周期较长。
利用本申请实施例的方案，能够提高车辆中的参数标定的效率，减少标定的时间成本。
(2)传感器标定
传感器的部分参数在出厂前需要完成不同环境下的标定，例如，摄像头的畸变校准矩阵、惯性测量单元(inertial measurement unit，IMU)的敏感度、手机载激光雷达的偏角等。
以手机上的摄像头为例，在出厂前需要在不同光照条件、不同温度、不同湿度等环境下进行参数标定，以得到在不同环境下均适用的摄像头参数。
利用本申请实施例的方案，能够提高传感器中的参数标定的效率，减少标定的时间成本。
图1示出了本申请实施例提供的一种应用场景的示意图。
如图1所示，同款车型的三辆车分别处于地点1、地点2和地点3三个测试场中进行参数标定。如图1所示，位于地点1的车辆所处的工况包括：晴，温度为25℃，柏油路面。位于地点2的车辆所处的工况包括：晴，温度为6℃，瓷砖路面。位于地点3的车辆所处的工况包括：雪，温度为-30℃，冰面。
该三辆车可以采用本申请实施例的方法同时进行参数标定。例如，三辆车可以分别执行标定试验，并将相关数据，例如测试数据或参数组合的评估结果等，上传至云端，由云端将各个设备的调整后的参数组合分别发送至各辆车，重复执行上述过程，直至调整后的参数组合在该三种工况下均能满足测试要求，完成参数标定。具体描述可以参见后文中的方法400或方法500。
本申请实施例提供了一种参数标定的方法，基于多个设备的参数组合的评估结果实现参数标定，以提高参数标定效率，减少标定的时间成本。
为了更好地描述本申请实施例的方案，下面结合图2对本申请实施例提供的标定系统进行说明。
如图2所示，标定系统200可以部署于云端侧和设备侧。示例性地，云端设备210可以由一个或多个服务器实现。设备侧包括多个设备。该多个设备(例如，设备220和设备230)可以与云端设备210进行交互。该多个设备中包括需要进行参数标定的模块。示例性地，该多个设备可以包括智能手机、平板电脑、智能摄像头、车辆、媒体消费设备或可穿戴设备等。该多个设备可以通过任何通信机制/通信标准的通信网络与云端设备210进行交互，通信网络可以是因特网、万维网、内联网、虚拟专用网络、广域网、局域网、使用一个或多个公司的专有通信协议的专用网络、以太网、WiFi和HTTP、以及前述的各种组合。这种通信可以由能够传送数据到其它计算机和从其它计算机传送数据的任何设备，诸如调制解调器和无线接口。
需要注意的是，云端设备210的所有功能也可以由该多个设备实现，例如，设备220执行云端设备210的功能为自身提供标定服务，或者为设备230提供标定服务。在该情况下，标定系统200无需部署于云端侧。换言之，云端设备210为可选项。
需要说明的是，多个设备(例如，设备220和设备230)可以为同一类型的设备，或者，也可以为不同类型的设备。例如，多个设备可以包括多个车辆。再如，多个设备可以包括多个车辆和多个测试台架。
图3示出了本申请实施例提供的一种标定系统的示意性框图。图3可以视为图2所示的标定系统200的一种具体实现方式。
如图3所示，标定系统200包括部署于云端设备210的数据处理模块211、标定模块 212、第一存储模块213和第一通信模块214。标定系统还包括部署于设备侧的多个设备的第二通信模块、第二存储模块、控制模块和执行模块。例如，标定系统200还包括部署于设备220中的第二通信模块221、第二存储模块222、控制模块223和执行模块224，以及设备230中的第二通信模块231、第二存储模块232、控制模块233和执行模块234。
应理解，本申请实施例中的“第一通信模块”和“第二通信模块”中的“第一”和“第二”仅用于区分云端侧的通信模块和设备侧的通信模块，不具有其他限定作用。相应地，“第一存储模块”和“第二存储模块”中的“第一”和“第二”仅用于区分云端侧的存储模块和设备侧的存储模块，不具有其他限定作用。
第一通信模块214用于实现云端侧和设备侧的通信连接。该通信连接可以为有线连接，也可以为无线连接。
具体地，第一通信模块214用于接收以下至少一项：设备220和设备230的数据，或者，设备220和设备230的参数组合的评估结果。
示例性地，设备的数据包括设备基于当前的参数组合进行标定试验采集到的数据。
各个设备当前的参数组合可以相同，也可以不同。各个设备的标定试验所执行的动作序列可以相同，也可以不同。
可选地，第一通信模块214还可以用于接收设备220和设备230所处的工况。
数据处理模块211用于对第一通信模块214接收的数据进行处理。
示例性地，第一通信模块214接收的数据包括设备220和设备230的数据。数据处理模块211可以用于对设备220和设备230的数据进行处理，提取有效数据，该有效数据即为数据处理模块211处理后的数据，也就是后文中的“测试数据”。
需要说明的是，数据处理模块211为可选模块。例如，若设备的数据中仅包括有效数据，即若无需对数据进行处理，无需设置数据处理模块211。再如，若第一通信模块214接收的数据为设备220和设备230的参数组合的评估结果，则无需设置数据处理模块211。再如，可以由标定模块212执行数据处理模块211的功能，对数据进行处理，无需另外设置数据处理模块211。
标定模块212用于根据多个设备(例如，设备220和设备230)的参数组合的评估结果得到至少一个参数组合。
可选地，标定模块212还可以用于对该多个设备的测试数据进行评估，得到评估结果。
可选地，标定模块212还可以用于确定该多个设备中的至少一个设备的待执行动作序列。
可选地，标定模块212还可以用于确定该至少一个设备。
标定模块212的具体描述可以参见方法400中的步骤S420和步骤S430，此处不再赘述。
第一存储模块213可以用于存储以下至少一项：设备220和设备230的测试数据、设备220和设备230的参数组合的评估结果、设备220和设备230所处的工况或者动作序列集合。
应理解，图3中仅以第一存储模块为一个存储模块作为示例，在实际应用中，第一存储模块可以包括多个存储模块。例如，第一存储模块可以包括工况模块，工况模块用于存储多个设备所处的工况，例如，设备220和设备230所处的工况。再如，第一存储模块可 以包括标定数据模块，标定数据模块用于存储多个设备的测试数据，比如，设备220和设备230的测试数据。
可选地，第一通信模块214还可以用于将该至少一个设备的调整后的参数组合分别发送至该至少一个设备。
可选地，第一通信模块214还可以用于将该至少一个设备的待执行动作序列分别发送至该至少一个设备。
下面以设备220为例对设备所包括的各个模块进行说明。设备230中的模块可以执行与设备220中的模块相同的功能。
第二通信模块221可以用于与云端侧的第一通信模块214配合，实现设备220和云端设备210之间的通信连接。
具体地，第二通信模块221可以用于接收设备220的调整后的参数组合。第二通信模块221还可以用于向第一通信模块214发送设备220的数据或设备220的参数组合的评估结果。
可选地，第二通信模块221还可以用于向第一通信模块214发送设备220所处的工况。
可选地，第二通信模块221还可以用于接收设备220的待执行动作序列。
设备220接收到的设备220的参数组合可以写入控制模块223中。控制模块223用于控制执行模块224执行待执行动作序列。控制模块223中的参数即为需要进行标定的参数。图3中仅为示例，标定系统中也可以不包括控制模块223。
示例性地，设备220可以为车辆。在该情况下，控制模块223可以为车载ECU。
执行模块224用于执行待执行动作序列。具体的执行过程可以以自动驾驶的形式执行，或者，也可以由驾驶员执行。
图3中仅为示例，标定系统中也可以不包括执行模块224。
第二存储模块222用于存储以下至少一项设备220所处的工况、设备220的参数组合、设备220的待执行动作序列。
应理解，图3中的各个设备仅以第二存储模块为一个存储模块作为示例，在实际应用中，各个设备中可以包括多个存储模块。例如，第二存储模块可以包括工况模块、参数模块和动作模块。工况模块用于将设备220所处的工况通过第二通信模块221发送至云端。参数模块用于存储设备220的参数组合，并将该参数写入控制模块223中。动作模块用于存储设备220的待执行动作序列，并将待执行动作序列发送至执行模块224。
应理解，图3中以终端设备包括两个车辆为例对标定系统进行说明，不对本申请实施例的方案构成限定。例如，终端设备还可以包括更多的车辆。再如，图3中的终端设备也可以包括其他类型的设备。
图4示出了本申请实施例的一种参数标定的方法的示意性流程图。图4所示的方法400也可以由参数标定的装置执行，该参数标定的装置可以为云服务设备或终端设备，例如，车辆、手机或测试台架等，也可以由云服务设备和终端设备构成的系统。或者，该参数标定的装置可以部署于云服务设备或终端设备上，或者，也可以部署于云服务设备和终端设备构成的系统上。云服务设备也可以称为云端。
示例性地，方法400可以由图3所示的标定系统执行。
图4所示的方法400包括步骤S410至步骤S440。
S410，获取多个设备的参数组合的评估结果，多个设备的参数组合的评估结果是通过该多个设备分别基于该多个设备的参数组合在第一时段内进行标定试验得到的。第一时段小于或等于第一阈值。
例如，第一阈值可以为24小时。
第一时段可以为任一时段，只要该时段的时间间隔小于或等于第一阈值即可。该多个设备在第一时段内进行标定试验也可以理解为该多个设备可以在同一时段内进行标定试验，例如，该多个设备可以同时进行标定试验。
该多个设备为需要进行参数标定的设备。参数组合中的参数即为需要标定的参数。参数组合中包括至少一个参数。
具体地，在车辆标定中，参数组合中的参数指的是ECU中的控制参数的部分或全部。
例如，在发送机标定中，参数组合中的参数包括ECU中与发送机的工作状态相关的控制参数。例如，参数组合可以包括点火提前角等。
在该多个设备分别基于该多个设备的参数组合进行标定试验的过程中，该多个设备分别基于该多个设备的参数组合执行该多个设备的动作序列。
步骤S410的执行装置可以部署于该多个设备中，或者，也可以部署于该多个设备以外的其他设备上，例如，可以部署于云端。
可选地，该多个设备可以为多个车辆。
例如，该多个设备为m个设备，该m个设备可以为m个车辆，m为大于1的整数。
在该情况下，方法400可以应用于整车标定的场景中。
例如，方法400可以用于实现整车标定中的制动防抱死系统(antilock brake system，ABS)场景、车辆转向系统场景或车身电子稳定性控制(electronic stability controller，ESC)系统场景等场景中的参数标定。
可选地，该多个设备可以包括多个测试台架。
例如，该多个设备为m个设备，该m个设备可以为m个测试台架，m为大于1的整数。
可选地，该多个设备可以包括车辆和测试台架。
例如，该多个设备为m个设备，该m个设备可以包括n个车辆和m-n个测试台架。m为大于1的整数，n为小于m的正整数。
在该多个设备包括测试台架的情况下，方法400可以应用于发动机标定或者电动汽车中的电动机标定等场景中。
可选地，该多个设备可以包括多个传感器。在该情况下，方法400可以应用于传感器标定的场景中。
以手机上的摄像头为例，在出厂前需要在不同的环境下，例如，在不同光照条件、不同温度或不同湿度下进行参数标定。方法400即可用于实现手机上的摄像头的参数标定。
该多个设备的参数组合可以相同，也可以不同。
在本申请实施例中，参数组合不同，指的是，参数组合中的至少一个参数的值不同，而并非限定参数组合中的参数项不同。
为了提高参数标定效率，该多个设备的参数组合中的参数项可以是相同的。例如，多个设备同时进行ABS场景中的参数标定，该多个设备的参数组合中均包括与ABS功能相 关的参数项，这样，可以提高与ABS场景中的参数的标定效率。
示例性地，各个设备的参数组合中的参数值可以是在参数空间内随机确定的。参数空间也可以理解为参数的取值范围。或者，各个设备的参数组合中的参数值也可以是根据类似车型的参数值确定的。或者，各个设备的参数组合的参数值也可以是人为设定的。或者，各个设备的初始参数组合可以是根据各个设备所处的工况分别确定的。该多个设备的参数组合可以是采用相同的方式确定的，也可以是采用不同的方式确定的。本申请实施例对各个设备的参数组合中的参数值的确定方式不做限定。
该多个设备的动作序列可以相同，也可以不同。
该多个设备的标定试验的过程中所执行的动作序列不同，也可以理解为标定试验不同。
下面以整车标定的场景为例对动作序列进行说明。
例如，在ABS场景中进行参数标定时，车辆在进行标定试验的过程中执行如下动作序列：车辆驾驶到起始位置，加速至目标速度，保持匀速行驶，制动踏板踩到底直至车辆停止，车辆行驶出测试区域。
其中，两个包括不同目标速度的动作序列可以视为不同的动作序列。换言之，两个设备在标定试验过程中分别加速至不同的目标速度，即可视为执行不同的动作序列。
示例性地，该多个设备的动作序列可以是在动作序列集合中随机确定的。或者，该多个设备的动作序列可以是根据以下至少一项确定的：参数标定要求覆盖的动作、所述至少一个设备所处的工况或所述多个设备的参数组合的评估结果。或者，该多个设备的动作序列也可以是人为设定的。该多个设备的动作序列可以是采用相同的方式确定的，也可以是采用不同的方式确定的。本申请实施例对此不作限定。
可选地，该多个设备满足以下至少一项：该多个设备所处的工况不同，该多个设备在进行标定试验的过程中所执行的动作序列不同，或者，该多个设备的参数组合不同。
该多个设备中的任意两个设备处于不同的工况下，即可以认为该多个设备处于不同的工况下。
示例性地，方法400应用于整车标定的场景中，工况可以包括以下至少一项：路况、温度、湿度或天气等。
该多个设备中的任意两个设备的参数组合不同，即可认为该多个设备的参数组合不同。
该多个设备中的任意两个设备在进行标定试验的过程中所执行的动作序列不同，即可认为该多个设备所执行的动作序列不同。
该多个设备的工况、参数组合以及动作序列是否相同可以根据需要组合。
示例性地，处于不同工况下的设备基于相同的参数组合执行相同的动作序列。
例如，参数组合中的一个参数的取值范围为(0,1)，处于不同工况下的多个设备在该参数为0.1的情况下执行相同的动作序列。
示例性地，处于相同工况下的多个设备基于不同的参数组合执行相同的动作序列。
例如，参数组合中的一个参数的取值范围为(0,1)，控制处于相同工况下的多个设备分别在该参数为(0,1)范围内不同的值的情况下执行相同的动作序列。
示例性地，处于不同工况下的设备基于不同的参数组合执行不同的动作序列。
在本申请实施例中，控制多个设备在不同的工况下进行标定试验，或者，控制该多个设备基于不同的参数组合进行标定试验，或者，控制多个设备执行不同的动作序列，能够提高标定试验的效率，更快获取各种情况下的参数组合的评估结果，或者说，采集更多的情况下的参数组合的评估结果，以便尽快完成参数标定，有利于提高参数标定的效率。
可选地，该多个设备的参数组合相同，且该多个设备所处的工况不同。
这样，有利于汇总同一个参数组合在不同工况下的评估结果。
该多个设备的参数组合的评估结果用于指示该多个设备的参数组合的质量。参数组合的评估结果也可以理解为参数组合的价值。
对参数组合进行评估也可以理解为对基于该参数组合进行标定试验的过程中的设备的表现进行评估。
具体地，可以基于一个或多个评价指标对设备的表现进行评估，得到参数组合的评估结果。
示例性地，可以基于各个评价指标对设备的表现进行打分，得到各个指标对应的分数，各个指标对应的分数可以作为参数组合的评估结果。或者，将各个指标对应的分数进行处理，将处理后得到的综合分数作为参数组合的评估结果。
可选地，该多个设备的参数组合的评估结果可以是通过对该多个设备的测试数据进行评估得到的。该多个设备的测试数据是根据该多个设备分别基于该多个设备的参数组合进行标定试验的过程中采集到的数据确定的。
示例性地，该多个设备的测试数据可以是通过对该多个设备分别基于该多个设备的参数组合进行标定试验的过程中传感器采集到的数据进行处理得到的。
例如，对传感器采集到的数据进行处理可以包括对传感器采集到的数据进行过滤、降频或降噪等处理。
下面以数据过滤为例进行说明。
传感器采集到的数据中可能包括有效数据和无效数据。换言之，在标定试验的过程中传感器既采集了有效数据，也采集了无效数据。有效数据指的是与参数标定相关的数据，或者说，能够用于评估参数组合的数据。无效数据指的是与参数标定无关的数据，或者说，无法用于评估参数组合的数据。
在该情况下，对传感器采集到的数据进行过滤，提取有效数据，将该有效数据作为测试数据。
例如，在对ABS场景的参数进行标定时，车辆在标定试验的过程中依次执行如下动作：车辆驾驶到起始位置，加速到目标速度，保持一段匀速行驶，制动踏板踩到底直至车辆停止，车辆行驶出测试区域。在车辆执行上述所有动作的过程中采集数据。有效数据为制动踏板踩到底直至车辆停止这一段过程中采集的数据。在该情况下，可以对采集到的数据进行处理，提取出有效数据，将该有效数据作为测试数据。
在一种实现方式中，对传感器采集到的数据进行处理的过程可以由步骤S410的执行装置执行，也就是说，步骤S410的执行装置可以接收该多个设备上传的该多个设备的传感器采集到的数据，并对传感器采集到的数据进行处理，得到测试数据。
在另一种实现方式中，对传感器采集到的数据进行处理的过程也可以由该多个设备分别执行。该多个设备将处理后得到的测试数据发送至步骤S410的执行装置。
可替换地，该多个设备的测试数据可以是为根据该多个设备分别基于该多个设备的参数组合进行标定试验的过程中传感器采集到的数据。
若传感器采集到的数据中仅包括有效数据，换言之，在标定试验的过程中仅采集有效数据，也可以将采集到的数据作为测试数据。
例如，在对ABS场景的参数进行标定时，车辆在标定试验的过程中依次执行如下动作：车辆驾驶到起始位置，加速到目标速度，保持一段匀速行驶，制动踏板踩到底直至车辆停止，车辆行驶出测试区域。在车辆执行制动踏板踩到底直至车辆停止这一段过程中采集数据。在该情况下，采集到的数据即为有效数据，将该有效数据作为测试数据。
在标定试验的过程中仅采集有效数据，这样可以减少对采集到的数据进行处理的操作，提高了数据处理效率，即提高了参数标定的效率。
如前所述，该多个设备的参数组合的评估结果可以是通过对该多个设备的测试数据进行评估得到的。
在该情况下，参数组合的评估结果也可以理解为测试数据的价值。
具体地，可以基于一个或多个评价指标对测试数据进行评估，得到参数组合的评估结果。
例如，可以通过代价(cost)函数或者价值(value)函数对测试数据进行处理，得到的cost值或value值可以作为参数组合的评估结果。
下面以ABS场景中的参数标定为例进行说明。
ABS的评价指标包括：车轮抱死时间和车辆横摆角的大小。
基于当前的ABS的参数组合进行标定试验得到测试数据，从测试数据中可以确定车辆抱死时间和车辆横摆角。在制动过程中，若测试数据中的车辆抱死时间越长，车辆横摆角越大，则当前的制动系统的制动性能越差，即该测试数据的价值越低，或者说，当前的参数组合越差。
例如，可以利用cost函数分别对测试数据中的车辆抱死时间和车辆横摆角进行评估，得到两个指标对应的分数，这两个指标对应的分数可以作为当前的参数组合的评估结果。或者，也可以利用cost函数对测试数据中的车辆抱死时间和车辆横摆角进行评估，得到综合分数，该综合分数可以作为当前的参数组合的评估结果。
进一步地，多个设备的参数组合的评估结果是根据该多个设备所处的工况对该多个设备的测试数据进行评估得到的。
对于同一个评价指标，即使不同工况下的测试数据相同，不同工况下的测试数据对应的参数组合的评估结果也可以是不同的。
下面以ABS场景中的参数标定为例进行说明。
ABS场景中，车辆的测试数据包括制动距离。制动距离指的是刹车踩到底的状态下得到的制动距离。根据路面附着系数对采取紧急制动时的初速度相同的情况下的制动距离进行评估。具体地，在制动距离相同的情况下，路面附着系数较小的工况下的参数组合的评估结果更好。
如前所述，在ABS场景中进行参数标定时，车辆在进行标定试验的过程中执行如下动作序列：车辆驾驶到起始位置，加速至目标速度，保持匀速行驶，制动踏板踩到底直至车辆停止，车辆行驶出测试区域。
制动踏板踩到底直至车辆停止即为采取紧急制动，采取紧急制动时的初速度即为上述动作序列中的目标速度。
例如，车辆在处于砂石路面和处于柏油路面这两种不同工况下分别基于不同的参数组合进行相同的标定试验，即执行相同的动作序列。具体地，处于砂石路面这种工况下的车辆基于参数组合1执行上述动作序列后，通过传感器采集到的制动距离为a；处于柏油路面这种工况下的车辆基于参数组合2执行上述动作序列后，通过传感器采集到的制动距离为b。砂石路面的路面附着系数小于柏油路面的路面附着系数，在a与b相等的情况下，处于砂石路面的车辆中的参数组合1的评估分数高于处于柏油路面的车辆中的参数组合2的评估分数。需要说明的是，此处在不同的工况下得到相同的测试数据是由不同的参数组合导致的。也就是说，在不同工况下基于不同的参数组合进行相同的标定实验，即使测试数据相同，不同的参数组合的评估分数仍有可能不同。
在本申请实施例中，基于设备所处的工况对设备的测试数据进行评估，能够提高评估结果的准确性，进而提高参数标定的效果。
进一步地，多个设备的参数组合的评估结果可以是根据该多个车辆的配置信息对该多个设备的测试数据进行评估得到的。
对于同一个评价指标，即使不同配置下得到的测试数据相同，不同配置下的测试数据对应的参数组合的评估结果也可以是不同的。
下面以ABS场景中的参数标定为例进行说明。
ABS场景中，车辆的测试数据包括制动距离。车辆的配置信息可以包括车辆的载重重量。根据车辆的载重重量数对采取紧急制动时的初速度相同的情况下的制动距离进行评估。具体地，在制动距离相同的情况下，车辆的载重重量较大的情况下的参数组合的评估结果更好。
例如，车辆的载重重量为70kg，基于参数组合3进行标定试验的过程中，采集到的制动距离为c；车辆的载重重量为150kg，基于参数组合4进行标定试验的过程中，采集到的制动距离为d，在c和d相等的情况下，参数组合3的评估分数小于参数组合4的评估分数。具体地，可以将制动距离乘以制动距离对应的系数之后，再进行评估。制动距离对应的系数可以是人为设定的。制动距离对应的系数与车辆的载重重量呈负相关关系。例如，车辆的载重重量为70kg，制动距离对应的系数为1。基于参数组合3进行标定试验的过程中，采集到的制动距离为c，将c乘以1，然后基于处理后的制动距离得到参数组合3的评估分数；车辆的载重重量为150kg，制动距离对应的系数为0.95，基于参数组合4进行标定试验的过程中，采集到的制动距离为d，将d乘以0.95，然后基于处理后的制动距离得到参数组合4的评估结果。
在本申请实施例中，基于设备的配置信息对设备的测试数据进行评估，能够提高评估结果的准确性，进而提高参数标定的效果。
应理解，上述参数组合的评估方式仅为示例，还可以根据其他方式对设备的测试数据进行评估，本申请实施例对此不作限定。
可选地，该多个设备的参数组合的评估结果可以是由用户反馈得到的。
在一些应用场景中进行参数标定时，需要用户的主观感受反馈。换言之，需要用户评估标定试验过程中的使用体验，以得到参数组合的评估结果。
示例性地，方法400可以应用于整车标定的场景中，在车辆基于当前的参数组合进行标定试验的过程中，用户可以对驾驶体验进行打分，用户的打分可以作为当前的参数组合的评估结果。
例如，在进行车辆转向系统的参数标定时，评价指标包括方向盘助力效果。在标定试验过程中，方向盘助力效果可以由用户打分。
示例性地，用户反馈的评估结果可以通过人机交互接界面写入设备中。
这样，能够充分考虑用户的感受，有利于提高用户体验。
可选地，该多个设备的参数组合的评估结果中的第一部分是由用户反馈得到的，该多个设备的参数组合的评估结果中的第二部分是通过对该多个设备的测试数据进行评估得到的。
如前所述，可以基于多个评价指标对设备的表现进行评估，得到参数组合的评估结果。
示例性地，评估结果可以以评估指标对应的分数的形式表示，换言之，一个参数组合的评估结果可以包括多个分数。该多个分数中的部分分数(评估结果中的第二部分的一例)可以是通过对该多个设备的测试数据进行评估得到的，其余部分分数(评估结果中的第一部分的一例)可以是由用户反馈得到的。
应理解，以上参数组合的评估方式仅为示例，还可以采用其他方式对参数组合进行评估，本申请实施例对此不做限定。例如，还可以对用户反馈的结果和对多个设备的测试数据进行评估得到的结果进行处理，以得到参数组合的评估结果。
在一种实现方式中，步骤S410的执行装置可以部署于该多个设备以外的其他设备上，例如，图2所示的云端。
步骤S410可以通过多种方式实现，下面以云端执行步骤S410为例，对步骤S410的具体实现方式进行举例说明。
示例性地，云端可以接收该多个设备的参数组合的评估结果。
例如，云端可以接收用户反馈的多个设备的参数组合的评估结果。
该多个设备可以分别将该多个设备的参数组合的评估结果发送至云端。该多个设备包括第一设备，即第一设备可以将第一设备的参数组合的评估结果发送至云端。
可替换地，云端可以接收该多个设备的测试数据，并对该多个设备的测试数据进行评估，得到该多个设备的参数组合的评估结果。
该多个设备可以分别将该多个设备的测试数据发送到云端。该多个设备包括第一设备，即第一设备可以将第一设备的测试数据发送至云端。
可替换地，云端可以接收该多个设备中的部分设备的参数组合的评估结果以及其余设备的测试数据，并对该其余设备的测试数据进行评估，得到该其余设备的参数组合的评估结果。
该多个设备中的部分设备将该各自的参数组合的评估结果发送至云端，其余设备将各自的测试数据发送至云端。
例如，该多个设备包括第一设备和第二设备。第二设备将第二设备的参数组合的评估结果发送至云端。第一设备将第一设备的测试数据发送至云端。或者，第一设备将第一设备的参数组合的评估结果发送至云端。第二设备将第二设备的测试数据发送至云端。
可替换地，云端可以接收该多个设备的参数组合的评估结果中的第一部分和该多个设 备的测试数据，并对该多个设备的测试数据进行评估，得到该多个设备的参数组合的评估结果中的第二部分。
该多个设备可以分别将该多个设备的参数组合的评估结果中的第一部分和该多个设备的测试数据发送至云端。
例如，该多个设备包括第一设备。第一设备可以将第一设备的参数组合的评估结果中的第一部分和第一设备的测试数据发送至云端。
云端可以通过云端中的通信模块以有线或无线的方式接收各种数据，本申请实施例对此不做限定。
在另一种实现方式中，步骤S410的执行装置可以部署于该多个设备中的任一设备。
步骤S410的执行装置可以通过多种方式获取本地的参数组合的评估结果。
示例性地，步骤S410的执行装置可以读取本地的参数组合的评估结果。例如，步骤S410的执行装置可以从存储模块中读取或者说接收本地的参数组合的评估结果。为了便于描述，本申请实施例中对于本地的数据的“接收”或者“读取”统称为“读取”。
可替换地，步骤S410的执行装置也可以读取本地的测试数据，并对本地的测试数据进行评估，得到本地的参数组合的评估结果。
可替换地，步骤S410的执行装置也可以读取本地的参数组合的评估结果中的第一部分以及本地的测试数据，并对本地的测试数据进行评估，得到本地的参数组合的评估结果中的第二部分。第一部分可以是由用户反馈得到的。
步骤S410的执行装置获取该多个设备中的除本地设备以外的其他设备的参数组合的评估结果的具体方式可以参考前述云端的获取方式，为了避免重复，此处不再赘述。
需要说明的是，该多个设备的参数组合的评估结果可以是单次获取得到的，或者，也可以是多次获取得到的。
可选地，方法400包括步骤S411(图4中未示出)。
S411，获取多个设备所处的工况。
示例性地，该多个设备所处的工况可以是由该多个设备中的传感器采集，或者，也可以人为设定。
具体的获取方式可以参考步骤S410，将步骤S410中的“多个设备的参数组合的评估结果”替换为“多个设备所处的工况”即可，此处不再赘述。
可选地，方法400还包括步骤S412(图4中未示出)。
S412，获取该多个设备的配置信息。
示例性地，在ABS场景中，配置信息包括以下至少一项：轮胎特性参数、车辆的载重重量或车辆的载重重量的分布情况。
例如，轮胎特性参数可以包括以下至少一项：磨损程度、断面宽度或扁平比等。
例如，在车上仅有驾驶员的情况下，车辆的载重重量以及车辆的载重重量分布情况可以表示为[体重70kg，分布在轿厢左前侧]；在车上载有驾驶员以及副驾驶员的情况下，车辆的载重重量以及车辆的载重重量分布情况可以表示为[体重150kg，分布在轿厢前侧]。
具体的获取方式可以参考步骤S410，将步骤S410中的“多个设备的参数组合的评估结果”替换为“多个设备的配置信息”即可，此处不再赘述。
S420，根据多个设备的参数组合的评估结果得到至少一个调整后的参数组合。
该至少一个调整后的参数组合可以相同，也可以不同。
步骤S420可以理解为根据多个设备的参数组合的评估结果对当前的参数组合进行优化。具体的优化方式采用现有的优化算法实现。
示例性地，步骤S420可以通过贝叶斯优化(Bayesian Optimization)算法实现。
具体地，基于贝叶斯原理根据多个设备的参数组合的评估结果拟合高斯过程，进而得到参数分布空间上各个取值点的评估结果的均值和方差。一个取值点即为一个参数组合。计算各个取值点的得分，将得分最高的取值点作为调整后的参数组合。例如，可以利用上置信界(upper confidence bound，UCB)算法计算各个取值点的得分。具体地，将各个取值点的评估结果的均值和方差带入UCB公式中，计算各个取值点的得分，将得分最高的至少一个取值点作为该至少一个调整后的参数组合。
示例性地，步骤S420可以通过遗传算法实现。
例如，该多个设备包括m个设备，将该m个设备的参数组合的评估结果中的最好的p个评估结果对应的参数组合进行交叉或变异等操作，得到该至少一个调整后的参数组合。p为小于m，且大于1的整数。
应理解，还可以采用其他优化方式得到调整后的参数组合，本申请实施例对优化过程的具体实现方式不做限定。
可选地，多个设备的参数组合相同，所述多个设备所处的工况不同。在该情况下，步骤S420可以通过以下步骤实现：
(1)将多个设备的参数组合的评估结果进行处理，得到汇总评估结果。
汇总评估结果可以采用多种方式确定。例如，将该多个设备的参数组合的评估结果的加权平均值作为汇总评估结果。再如，将该多个设备的参数组合的评估结果中的最小值作为汇总评估结果。再如，将该多个设备的参数组合的评估结果中的最大值作为汇总评估结果。本申请实施例对汇总评估结果的确定方式不做限定。
(2)根据汇总评估结果得到至少一个调整后的参数组合。
步骤(2)可以理解为根据汇总评估结果对当前的参数组合进行优化。具体的优化方式可以参考前文中的描述，例如，采用贝叶斯优化算法时，将前文的优化方法中的“多个设备的参数组合的评估结果”替换为“汇总评估结果”即可。
由于各个设备的参数组合是相同的，可以对处在各个工况的评估结果进行汇总，得到该参数组合的汇总评估结果，根据汇总评估结果优化当前的参数组合，有利于得到更优的参数组合，进而提高标定效率。
根据本申请实施例的方案，该多个设备在同一时段内进行标定试验，能够提高数据采集的效率，有利于基于多个设备的参数组合的评估结果实时调整设备的参数，进而提高标定效率，减少标定周期，降低标定成本。例如，不同的设备可以执行不同的动作序列，快速覆盖需要完成的动作序列，提高数据采集的效率。再如，不同的设备可以基于不同的参数组合进行标定试验，快速覆盖参数空间，提高数据采集的效率。再如，不同的设备可以处于不同的工况中，快速覆盖标定所需的工况，提高数据采集的效率。
现有的车辆标定过程需要以串行的方式在不同的工况、参数等情况下完成标定，在一处测试场标定完成后，才能将车辆运到另一测试场进行参数标定，本申请实施例的方案的多个设备可以同时处于不同的工况下同时进行标定试验，提高了标定效率，减少了时间成 本与经济成本。
现有的标定方案中通常要在一个工况下完成标定后再到另一个工况中进行标定，标定完成后需要返回之前的工况中进行验证调整。本申请实施例的方案的多个设备可以同时处于不同的工况下，能够实现多个设备的数据的实时汇总以及参数优化，有利于在多种工况下同时寻找兼容该多种工况的参数组合，提高参数的稳定性，进而提高标定效率。
此外，标定工程师进行标定时的主观偏好会影响标定结果的准确性。本申请实施例的方案中，基于多个设备的测试数据的评估结果进行参数标定，减少了不同的标定工程师带来的认知偏差，提高了标定质量，有利于避免设备出厂后再召回的情况。
此外，本申请实施例的方案还可以采用自动优化算法完成标定，减少标定工程师的数量，降低人力成本，进一步减少标定工程师的主观偏好对标定结果的影响。
方法400还包括步骤S430。
S430，将该至少一个调整后的参数组合发送至至少一个设备。
可选地，该至少一个设备可以是根据以下至少一项确定的：多个设备所处的工况或该至少一个调整后的参数组合。
例如，该至少一个调整后的参数组合是通过调整当前的参数组合中的一个或多个参数值得到的，被调整的参数项可能仅对低附着系数的工况有效，对高附着系数的工况无效，在该情况下，该至少一个设备可以为处于低附着系数的工况下的设备。
再如，多个设备的参数组合的评估结果可以多次获取到的。在该情况下，可以将每次获取到的一个或多个设备的参数组合中的设备作为该至少一个设备。该至少一个调整后的参数组合也可以理解为该至少一个设备的调整后的参数组合。步骤S420也可以理解为，根据该多个设备的参数组合的评估结果调整该多个设备中的至少一个设备的参数组合，得到该至少一个设备的调整后的参数组合。需要说明的是，此处的“调整该多个设备中的至少一个设备的参数组合”指的是调整步骤S410的执行装置中存储的该至少一个设备的参数组合，而并非限定调整该至少一个设备中存储的参数组合。
应理解，以上仅为示例，该至少一个设备还可以通过其他方式确定，本申请实施例不做限定。例如，该至少一个设备可以是随机确定的。
如前所述，该至少一个调整后的参数组合可以相同，也可以不同。
例如，该至少一个调整后的参数组合可以是相同的，在该情况下，该至少一个设备接收到的参数组合是相同的。
再如，该至少一个调整后的参数组合与该至少一个设备可以是一一对应的，在该情况下，该至少一个设备接收到的参数组合可以是不同的。
该至少一个设备可以基于调整后的参数组合继续进行标定试验。方法400可以重复执行，直至标定完成为止。
该至少一个设备包括第一设备，为了便于描述，此处仅以第一设备为例进行说明，该至少一个设备中的其他设备可以执行与第一设备相同的动作。
将至少一个调整后的参数组合中的一个调整后的参数组合发送至第一设备。第一设备获取该调整后的参数组合，基于调整后的参数组合进行标定试验。
对于同一设备而言，每次标定试验所执行的动作序列可以相同，也可以不同。
以方法400应用于车辆标定为例，该至少一个设备可以将调整后的参数组合分别写入 该至少一个设备中的ECU中，进而可以基于调整后的参数组合执行标定试验。
应理解，步骤S430为可选项。步骤S410的执行装置未部署于第一设备为不同设备时，方法400还可以包括步骤S430。步骤S410的执行装置部署于第一设备时，步骤S410的执行装置可以直接将本地存储的参数组合更新为调整后的参数组合，无需执行步骤S430。
可选地，方法400还包括步骤S440。
S440，将至少一个设备的待执行动作序列分别发送至该至少一个设备。该至少一个设备的待执行动作序列包括该至少一个设备在基于该至少一个设备的调整后的参数组合进行标定试验的过程中所需要执行的动作。
换言之，将待执行的标定试验的内容发送至各个设备。
该至少一个设备的待执行动作序列可以相同，也可以不同。
该至少一个设备包括第一设备，为了便于描述，此处仅以第一设备为例进行说明，该至少一个设备中的其他设备可以执行与第一设备相同的动作。
第一设备获取第一设备的待执行动作序列。第一设备的待执行动作序列包括第一设备在基于第一设备的调整后的参数组合进行标定试验的过程中所需要执行的动作。
应理解，步骤S440为可选项。该至少一个设备的待执行动作序列还可以由以下方式确定。
示例性地，该至少一个设备的待执行动作序列也可以是由该至少一个设备自行确定的。
可替换地，该至少一个设备的待执行动作序列可以是预先设置的。
可替换地，该至少一个设备的待执行动作序列可以是用户确定的。
例如，第一设备可以显示一个或多个参考动作序列。用户可以从该一个或多个参考动作序列中选择第一设备的待执行动作序列。该一个或多个参考序列可以包括第一设备的动作序列集合中未完成的动作序列。或者，该一个或多个参考序列可以包括第一设备的动作序列集合中的全部动作序列。
这样可以为用户在选择待执行动作序列时提供参考，提高用户体验，避免执行重复的动作序列，提高标定效率。
可选地，该至少一个设备的待执行动作序列是从动作序列集合中确定的。
该多个设备中的部分或全部设备可以共享动作序列集合，即该多个设备中的部分或全部设备的动作序列集合可以是相同的。或者，该多个设备的动作序列集合可以是不同的。本申请实施例对此不做限定。
可选地，该至少一个设备的待执行动作序列是根据以下至少一项确定的：参数标定要求覆盖的动作、所述至少一个设备所处的工况或所述多个设备的参数组合的评估结果。
示例性地，参数标定要求覆盖的动作可以包括国家标准要求覆盖的动作。
以ABS场景中的参数标定为例，国家标准要求覆盖的动作需要覆盖低速、中速和高速场景。即在动作序列中的目标车速需要包括低速、中速和高速三种情况。若之前已执行过目标车速为低速的动作序列，则待执行动作序列可以为目标车速为中速的动作序列或者目标车速为高速的动作序列。
不同的工况对待执行动作序列的要求不同。
以ABS场景中的参数标定为例，低附着系数的路段，例如，冰面上，要求缓慢加速， 即油门踏板的踩踏深度不宜过大。高附着系数的路段，例如，柏油路，要求快速加速，即油门踏板需要较大的踩踏深度。对于处于低附着系数的路段的设备，待执行动作序列中的加速动作所使用的油门踏板的踩踏深度小于或等于第一踏板阈值，比如，30％；对于高附着系数的路段的设备，待执行动作序列中的加速动作所使用的油门踏板的踩踏深度大于或等于第二踏板阈值，比如，90％。
若基于一个参数组合执行进行标定试验后得到的参数组合的评估结果较差，例如，评估分数小于或等于评估阈值，则在调整参数组合后，可以将该标定试验过程中执行的动作序列作为待执行的动作序列。
待执行动作序列的具体执行方式可以根据需要设置。
示例性地，方法400可以应用于整车标定的场景中，该至少一个设备，即该至少一个车辆，可以以自动驾驶的方式执行待执行动作序列，得到测试数据。
或者，该至少一个车辆上的驾驶员可以控制该至少一个车辆执行该待执行动作序列。
例如，该至少一个车辆的车载显示器上可以显示该待执行动作序列，由驾驶员控制车辆执行相应的待执行动作序列。
该至少一个设备若包括两个及两个以上的设备，则该至少一个设备中的各个设备所采取的执行方式可以相同，也可以不同。例如，部分车辆可以采用自动驾驶的方式，部分车辆可以采用驾驶员驾驶的方式。
在该至少一个设备执行完标定试验后，可以得到该至少一个设备的测试数据或者该至少一个设备的参数组合的评估结果。进一步地，可以基于该至少一个设备的测试数据重复执行方法400，也就是说，本申请实施例的方法400可以迭代执行。例如，执行步骤S410至步骤S420，或者，重复执行步骤S410至步骤S430，或者，重复执行步骤S410至步骤S440，具体的执行步骤可以根据需要设置，本申请实施例对此不做限定。例如，可以将该至少一个设备的参数组合的评估结果作为该步骤S410中的多个设备的参数组合的评估结果，进而重复执行方法400。再如，可以将该至少一个设备的参数组合的评估结果添加至步骤S410中的多个设备的参数组合的评估结果，进行重复执行方法400。在该情况下，步骤S410中的一个设备可能对应该设备在多次标定试验中得到的参数组合的评估结果，或者说，一个设备的参数组合的评估结果可能包括多次标定试验得到的参数组合的评估结果。
需要说明的是，在重复执行方法400过程中，每次参与迭代的设备，即方法400中的“该至少一个设备”可以为待标定的设备的全集或子集，即该方法400中的“多个设备”的全集或子集。换言之，每次参与迭代的设备可以相同，也可以不同，本申请实施例对此不做限定。
图5示出了本申请实施例提供的一种参数标定的方法的示意性流程图。图5中的方法500仅以车辆标定为例进行说明，不对本申请实施例的方案构成限定。图5中的方法500可以视为图4所示的方法400的一种具体实现方式，具体描述可以参考方法400。为了避免不必要的重复，在描述方法500时适当省略部分重复的描述。
示例性地，方法500可以由图3所示的标定系统执行。
方法500包括步骤S501至步骤S511。方法500中的步骤可以执行一次，也可以执行多次，即迭代执行。
S501，多个车辆分别将该多个车辆所处的工况上传至云端。
该多个车辆为待标定的车辆。待标定的车辆可以为同款车型的车辆。如图5所示，该多个车辆可以包括车辆1#和车辆2#。车辆1#和车辆2#分别将各自所处的工况上传至云端。
应理解，图5中仅以两辆车作为示例，不对本申请实施例中的待标定的车辆的数量构成限定。
示例性地，该多个车辆所处的工况可以是由传感器采集，或者，也可以由人工设定。
示例性地，步骤S501可以由图3中的第一通信模块和第二通信模块执行。
换言之，各个车辆通过各自的第二通信模块将车辆所处的工况上传至云端的第一通信模块。云端的第一通信模块可以将接收到各个车辆所处的工况存储至第一存储模块。具体地，可以将各个车辆所处的工况存储至第一存储模块中的工况模块中。
示例性地，车辆在进行标定试验的过程中处于移动状态，第一通信模块和第二通信模块之间可以采用无线通信的方式进行数据传输。
进一步地，步骤S501还可以包括：多个车辆将车辆配置信息上传至云端。
示例性地，车辆配置信息可以是由传感器采集，或者，也可以由人工设定。
需要说明的是，步骤S501为可选步骤。
S502，云端确定该多个车辆的初始参数组合。
该多个车辆的初始参数组合可以理解为方法400中的该多个设备的参数组合的示例。
该多个车辆的初始参数组合可以相同，也可以不同。
示例性地，该多个车辆的初始参数组合可以是根据该多个车辆所处的工况分别确定的。
可替换地，该多个车辆的初始参数组合可以是在参数空间内随机选择的。参数空间可以理解为各个参数的取值范围。
可替换地，该多个车辆的初始参数组合可以是根据类似车型的历史参数设定的。
可替换地，该多个车辆的初始参数组合可以是人为确定的。
应理解，以上初始参数组合的确定方式仅为示例，在实际应用中，还可以采用其他方式初始化参数组合，本申请实施例对此不做限定。
示例性地，步骤S502可以由图3中的标定模块212执行。
需要说明的是，步骤S502为可选步骤。例如，该多个车辆中的至少一个车辆的初始参数组合也可以预先设置在该至少一个车辆中，无需由云端确定。
在方法500包括步骤S502的情况下，方法500还包括步骤S503。
S503，云端将该多个车辆的初始参数组合分别发送至该多个车辆。
如图5所示，云端将车辆1#的初始参数组合X发送至车辆1#，将车辆2#的初始参数组合Y发送至车辆2#。初始参数组合X和初始参数组合Y可以相同，也可以不同。
示例性地，步骤S503可以由图3中的第一通信模块和第二通信模块执行。
换言之，云端通过第一通信模块将各个车辆的初始参数组合分别发送至各个车辆的第二通信模块。各个车辆的第二通信模块可以将初始参数组合存储至各个车辆的第二存储模块中。具体地，将初始参数组合存储至第二存储模块中的参数模块。
进一步地，参数模块可以将初始参数组合写入控制模块中，例如，写入ECU中。
可选地，方法500包括步骤S504a或步骤S504b。
S504a，云端确定该多个车辆的待执行动作序列。
该多个车辆的待执行动作序列可以相同，也可以不同。
示例性地，云端可以从动作序列集合中分别为该多个车辆选择待执行动作序列。
可替换地，该多个车辆的待执行动作序列可以是人为确定的。
可选地，云端可以根据以下至少一项确定该多个车辆的待执行动作序列：国家标准要求的动作覆盖程度、该多个车辆所处的工况或者参数组合的评估结果。
具体描述可以参见方法400中的步骤S440，此处不再赘述。
示例性地，步骤S504可以由图3中的标定模块212执行。
S504b，该多个车辆获取用户反馈的该多个车辆的待执行动作序列。
如图5所示，车辆1#获取用户反馈的车辆1#的待执行动作序列M。车辆2#获取用户反馈的车辆2#的待执行动作序列N。
示例性地，车辆可以通过人机界面获取用户反馈的车辆的待执行动作序列。
进一步地，车辆可以通过车载显示器显示一个或多个参考动作序列。该一个或多个参考动作序列可以是根据动作序列集合和已完成的动作序列确定的。
应理解，步骤S504a或步骤S504b仅为示例，不对本申请实施例的方案构成限定。例如，该多个车辆中的至少一个车辆的动作序列也可以由该至少一个车辆在动作序列集合中自行选择。再如，用户也可以直接控制车辆执行待执行动作序列。也就是说，用户无需将待执行动作序列输入待执行动作序列，而是直接控制车辆执行相应的动作。
在方法500包括步骤S504a的情况下，方法500还包括步骤S505a。
S505a，将该多个车辆的待执行动作序列分别发送至该多个车辆。
如图5所示，将车辆1#的待执行动作序列M发送至车辆1#，将车辆2#的待执行动作序列N发送至车辆2#。
示例性地，步骤S505a可以由图3中的第一通信模块和第二通信模块执行。
换言之，云端通过第一通信模块将各个车辆的待执行动作序列分别发送至各个车辆的第二通信模块。各个车辆的第二通信模块可以将待执行动作序列存储至各个车辆的第二存储模块中。具体地，将待执行动作序列存储至第二存储模块中的动作模块。
进一步地，动作模块可以将待执行动作序列发送至执行模块。
S506，控制该多个车辆分别进行标定试验。
具体地，控制该多个车辆在同一时段内分别进行标定试验。
如图5所示，控制车辆1#进行标定试验，控制车辆2#进行标定试验。
车辆1#的初始参数组合X写入车辆1#的ECU中；车辆2#的初始参数组合Y写入车辆2#的ECU中。
由于该多个车辆的初始参数组合已经写入ECU中，步骤S506可以理解为该多个车辆分别基于各自的ECU中的参数组合分别进行标定试验。
例如，步骤S506可以为控制该多个车辆分别执行该多个车辆的待执行动作序列。车辆1#执行待执行动作序列M，车辆2#执行待执行动作序列N。
示例性地，步骤S506可以由该多个车辆的执行模块执行，即以自动驾驶的方式控制车辆执行待执行动作序列。
可替换地，可以将该待执行序列通过车载显示器展示给驾驶员，由驾驶员控制车辆执 行待执行动作序列。
需要说明的是，该多个车辆中的不同车辆可以采用不同的驾驶方式，也可以采用相同的驾驶方式，本申请实施例对此不做限定。
方法500包括步骤S507a或S507b。
S507a，该多个车辆分别将该多个车辆的数据上传至云端。
如图5所示，车辆1#将车辆1#的数据上传至云端。车辆2#将车辆2#的数据上传至云端。
示例性地，车辆的数据可以包括车辆在进行标定试验的过程中传感器采集的数据。
进一步地，该多个车辆分别将标定试验过程中所执行的动作序列上传至云端。也就是说，将车辆在标定试验过程中实际执行的动作序列上传至云端。
在一种实现方式中，传感器采集的数据可以包括有效数据和无效数据。
在另一种实现方式中，传感器采集到的数据为有效数据，该有效数据可以作为测试数据。
具体描述可以参见方法400中的步骤S410中的描述，此处不再赘述。
需要说明的是，该多个车辆中的不同车辆上传的数据可以是不同形式的，例如，部分车辆上传的数据包括有效数据和无效数据，部分车辆上传的数据仅包括有效数据。或者，该多个车辆上传的数据也可以是相同形式的，本申请实施例对此不作限定。
示例性地，步骤S507a可以由图3中的第一通信模块和第二通信模块执行。
换言之，各个车辆通过各自的第二通信模块将各自的数据上传至云端的第一通信模块。
应理解，由于各个车辆执行完待执行动作序列的时刻可能不同等原因，不同车辆上传数据的时刻可能是不同的。
S507b，该多个车辆分别将该多个车辆的参数组合的评估结果上传至云端。
如图5所示，车辆1#将车辆1#的参数组合的评估结果上传至云端。车辆2#将车辆2#的参数组合的评估结果上传至云端。
进一步地，该多个车辆分别将标定试验过程中所执行的动作序列上传至云端。
该多个车辆的参数组合的评估结果可以是用户反馈的。在该情况下，该多个车辆可以获取用户反馈的该多个车辆的参数组合的评估结果。
例如，车辆可以通过人机界面获取用户反馈的该车辆的参数组合的评估结果，并将该参数组合的评估结果上传至云端。
示例性地，步骤S507b可以由图3中的第一通信模块和第二通信模块执行。
换言之，各个车辆通过各自的第二通信模块将各自的参数组合的评估结果上传至云端的第一通信模块。
应理解，由于各个车辆执行完待执行动作序列的时刻可能不同等原因，不同车辆上传评估结果的时刻可能是不同的。
在另一种实现方式中，方法500可以包括步骤S507a和步骤S507b。
具体地，该多个车辆可以将该多个车辆的参数组合的评估结果中的第一部分和该多个设备的数据。
车辆的参数组合的评估结果可以包括第一部分和第二部分。第一部分可以是用户反馈 的。在该情况下，车辆可以获取用户反馈的该车辆的参数组合的评估结果的第一部分。第二部分可以由云端对该车辆上传的数据进行处理以及评估得到的。
应理解，以上步骤S507a和步骤S507b仅为示例，具体描述可以参见前文中的步骤S410，此处不再赘述。
在方法500包括步骤S507a的情况下，方法500还可以包括步骤S508a。
S508a，云端对接收到的车辆的数据进行处理。
示例性地，步骤S508a包括：对接收到的车辆的数据进行数据过滤、降频或降噪等处理，得到接收到的车辆的测试数据。
示例性地，步骤S508a可以由图3中的数据处理模块执行。进一步地，数据处理模块可以将测试数据存储至第一存储模块中。具体地，数据处理模块可以将处理后得到的测试数据存储至第一存储模块中的标定数据模块中。
需要说明的是，由于不同车辆上传数据的时刻可能是不同的，云端接收到各个车辆上传的数据的时刻可能是不同的。
在一种实现方式中，云端可以在接收到所有车辆上传的数据之后对所有车辆的数据执行步骤S508a。
在另一种实现方式中，云端可以在接收到其中部分车辆上传的数据之后执行步骤S508a，例如，云端在接收到任一车辆上传的数据之后对该车辆的数据执行步骤S507a。
需要说明的是，步骤S508a为可选步骤。例如，车辆上传的数据可以为处理后得到的测试数据，在该情况下无需由云端进行处理。
S509，云端判断是否满足验收标准。若满足验收标准，则结束流程，即完成标定。若不满足验收标准，在方法500包括步骤S507a的情况下，执行步骤S510a，在方法500不包括步骤S507a的情况下，执行步骤S511。
示例性地，步骤S509可以由图3中的标定模块执行。
若方法500包括步骤S507a，则步骤S509可以通过以下方式执行。
在一种实现方式中，云端可以对所有车辆的测试数据执行步骤S509。
例如，车辆的测试数据可以存储于第一存储模块中。标定模块可以监控第一存储模块，在第一存储模块接收到该多个车辆中的所有车辆的测试数据之后执行步骤S509。
在另一种实现方式中，云端可以对其中部分车辆的测试数据执行步骤S509。
例如，车辆的测试数据可以存储于第一存储模块中。标定模块可以监控第一存储模块，在第一存储模块接收到任一车辆的测试数据之后基于该车辆的测试数据执行步骤S509。
若方法500包括步骤S5076b，则步骤S509可以通过以下方式执行。
在一种实现方式中，云端可以对所有车辆的参数组合的评估结果执行步骤S509。
例如，车辆的参数组合的评估结果可以存储于第一存储模块中。标定模块可以监控第一存储模块，在第一存储模块接收到该多个车辆中的所有车辆的参数组合的评估结果之后执行步骤S509。
在另一种实现方式中，云端可以对其中部分车辆的参数组合的评估结果执行步骤S508。
例如，车辆的参数组合的评估结果可以存储于第一存储模块中。标定模块可以监控第一存储模块，在第一存储模块接收到任一车辆的参数组合的评估结果之后基于该车辆的参 数组合的评估结果执行步骤S509。
S510a，云端对车辆的测试数据进行评估，得到车辆的参数组合的评估结果。
可选地，云端根据该车辆所处的工况对该车辆的测试数据进行评估，得到该车辆的参数组合的评估结果。
具体评估方式可以参见方法400中的步骤S410，此处不再赘述。
示例性地，步骤S510a可以由图3中的标定模块执行。
S511，云端根据车辆的参数组合的评估结果得到至少一个调整后的参数组合。
在一种实现方式中，云端可以在得到本次参与迭代的车辆中的部分车辆的参数组合的评估结果后执行步骤S511。在首次执行步骤S511时，即在第一次迭代过程中，参与迭代的车辆为待标定的所有车辆。例如，云端可以在得到任一车辆的参数组合的评估结果后实时执行步骤S511。
示例性地，步骤S511可以采用贝叶斯优化算法实现。
例如，云端可以基于当前得到的车辆的参数组合的评估结果利用贝叶斯原理拟合得到参数分布空间上各取值点的评估结果的均值和方差，进而根据UCB公式确定得分最高的至少一个取值点，将该至少一个取值点作为该至少一个调整后的参数组合。
或者，云端可以基于当前得到的车辆的参数组合的评估结果利用贝叶斯原理优化已有的参数分布空间上各取值点的评估结果的均值和方差，进而根据UCB公式确定最大值对应的参数组合，将该参数组合调整后的参数组合。
已有的参数分布空间上各取值点的评估结果可以是基于之前得到的车辆的参数组合的评估结果拟合得到的。
可替换地，步骤S511可以采用遗传算法实现。
例如，在云端中存储有至少两个车辆的参数组合的评估结果的情况下，将该至少两个车辆的参数组合的评估结果中较优的一个或多个参数组合进行交叉或变异等操作，将得到的参数组合作为该至少一个调整后的参数组合。
在另一种实现方式中，云端可以在得到本次参与迭代的车辆中的全部车辆的参数组合的评估结果后执行步骤S511。
示例性地，步骤S511可以采用贝叶斯优化算法实现。
例如，云端可以基于本次参与迭代的车辆的参数组合的评估结果利用贝叶斯原理拟合得到参数分布空间上各取值点的评估结果的均值和方差，进而根据UCB公式确定得分最高的至少一个取值点，将该至少一个取值点作为该至少一个调整后的参数组合。
或者，云端可以基于本次参与迭代的车辆的参数组合的评估结果利用贝叶斯原理优化已有的参数分布空间上各取值点的评估结果的均值和方差，进而根据UCB公式确定最大值对应的参数组合，将该参数组合调整后的参数组合。
已有的参数分布空间上各取值点的评估结果可以是基于之前得到的车辆的参数组合的评估结果拟合得到的。
也就是说，在重复执行步骤S511的过程中，云端可以根据本次迭代过程中得到的参数组合的评估结果和之前迭代过程中得到的参数组合的评估结果调整车辆的参数组合。换言之，云端可以根据当前已得到的所有参数组合的评估结果调整车辆的参数组合。
可替换地，步骤S511可以采用遗传算法实现。
例如，在云端中存储有至少两个车辆的参数组合的评估结果的情况下，将该至少两个车辆的参数组合的评估结果中较优的一个或多个参数组合进行交叉或变异等操作，将得到的参数组合作为该至少一个调整后的参数组合。
该至少两个车辆的参数组合的评估结果可以是一次迭代过程中得到的，也可以是多次迭代过程中得到的。
也就是说，在重复执行步骤S511的过程中，云端可以根据本次迭代过程中得到的参数组合的评估结果和之前迭代过程中得到的参数组合的评估结果调整车辆的参数组合。换言之，云端可以根据当前已得到的所有参数组合的评估结果调整车辆的参数组合。
进一步地，在该本次参与迭代的车辆的参数组合相同，且本次参与迭代的车辆所处的工况不同的情况下，步骤S511还可以包括以下步骤：
(1)将本次参与迭代的车辆的参数组合的评估结果进行处理，得到汇总评估结果。
例如，将本次参与迭代的设备的参数组合的评估结果的加权平均值作为汇总评估结果。再如，将该本次参与迭代的设备的参数组合的评估结果中的最小值作为汇总评估结果。再如，将该本次参与迭代的设备的参数组合的评估结果中的最大值作为汇总评估结果。本申请实施例对汇总评估结果的确定方式不做限定。
(2)根据汇总评估结果得到至少一个调整后的参数组合。
步骤(2)可以理解为根据汇总评估结果对当前的参数组合进行优化。具体的优化方式可以参考前文中的描述，例如，采用贝叶斯优化算法时，将前文的优化方法中的“设备的参数组合的评估结果”替换为“汇总评估结果”即可。
将该至少一个调整后的参数组合发送至至少一个车辆，即下一次参与迭代的车辆，将步骤S504至步骤与S511中的初始参数组合替换为调整后的参数组合，以及，将骤S504至步骤与S511中的多个车辆替换为该至少一个车辆，重复执行步骤S504至步骤S511，直至标定完成。换言之，在首次执行步骤S504至步骤S511时，参与迭代的车辆为待标定的所有车辆，及该多个车辆。该多个车辆的参数组合即为该多个车辆的初始参数组合。在之后执行步骤S506至步骤S511时，每次参与迭代的车辆可以为该多个车辆中的部分或全部，参数组合为步骤S511得到的调整后的参数组合。
示例性地，该至少一个车辆是根据以下至少一项确定的：多个设备所处的工况或该至少一个调整后的参数组合。
该至少一个车辆可以由图3中的标定模块212确定。
可替换地，该至少一个车辆可以为当前得到的车辆的参数组合的评估结果中的车辆。例如，云端得到车辆1#的参数组合的评估结果后实时执行步骤S511，在该情况下，车辆1#可以作为下一次参与迭代的车辆中的一个。
该至少一个车辆的确定方法可以参考前述方法400中的步骤S430，此处不再赘述。
每次参与迭代的车辆可以是相同的，也可以是不同的。
需要说明的是，方法500中仅以车辆作为示例对本申请实施例的方法进行说明，不对本申请实施例的方案构成限定。
在一种实现方式中，还可以将方法500中的多个车辆中的部分或全部替换为测试台架。
测试台架可以模拟设定为不同的工况，利用测试台架替换部分或全部车辆，可以应用 于发动机标定或电动机标定等场景中，进一步降低标定成本，提高数据采集的效率，进而提升标定效率。
由于测试台架可以模拟设定为不同的工况，在测试台架的工况发生变化时，可以将更新后的测试台架的工况上传至云端。同一车辆所处的工况在标定过程中通常不会发生变化，多个车辆可以在步骤S501中上传一次工况。而测试台架的工况可以根据需要设定，将该多个车辆中的部分或全部替换为测试台架后，当测试台架的工况发生变化后，可以将更新后的测试台架的工况实时上传至云端。
利用测试台架进行标定试验时，第一通信模块和第二通信模块之间可以采用无线通信的方式进行数据传输，或者，也可以采用有线通信的方式进行数据传输。
参数标定的装置可部署在云环境中，云环境是云计算模式下利用基础资源向用户提供云服务的实体。云环境包括云数据中心和云服务平台，所述云数据中心包括云服务提供商拥有的大量基础资源(包括计算资源、存储资源和网络资源)，云数据中心包括的计算资源可以是大量的计算设备(例如服务器)。
参数标定的装置可以是云数据中心中用于对参数进行标定的服务器；参数标定的装置还可以是部署在云数据中心中的服务器或者虚拟机上的软件装置，该软件装置用于对参数进行标定，该软件装置可以分布式地部署在多个服务器上、或者分布式地部署在多个虚拟机上、或者分布式地部署在虚拟机和服务器上。
参数标定的装置由云服务提供商在云服务平台抽象成一种参数标定的云服务提供给用户，用户购买该云服务后，云环境利用参数标定的装置向用户提供参数标定的服务，可以通过通信接口将设备的数据或设备的参数组合的评估结果等上传至云环境，由参数标定的装置对设备的参数组合进行优化，优化结果可以返回至用户所在的设备。
当参数标定的装置为软件装置时，参数标定的装置的不同模块可以部署在不同的环境或设备中。例如：参数标定的装置中的一部分部署在终端设备(如：车辆、智能手机、笔记本电脑、平板电脑、个人台式电脑、智能摄相机)，另一部分部署在云数据中心(具体部署在云数据中心中的服务器或虚拟机上)。
部署在不同环境或设备的参数标定的装置的各个部分之间协同实现参数标定的方法的功能。例如，参数标定的装置的部署情况可以参考图3。
可以理解的是，本申请不对参数标定的装置的哪些部分部署在终端计算设备和哪些部分部署在云数据中心进行限制性的划分，实际应用时可根据终端计算设备的计算能力或具体应用需求进行适应性的部署。
当参数标定的装置为软件装置时，参数标定的装置也可以单独部署在任意环境的一个计算设备上(例如：单独部署在一个终端设备上或者单独部署在云数据中心中的一个计算设备上)。
下面结合图6至图7对本申请实施例的装置进行说明。应理解，下面描述的装置能够执行前述本申请实施例的方法，为了避免不必要的重复，下面在介绍本申请实施例的装置时适当省略重复的描述。
图6是本申请实施例的参数标定的装置的示意性框图。图6所示的参数标定的装置3000包括获取单元3010和处理单元3020。
获取单元3010和处理单元3020可以用于执行本申请实施例的参数标定的方法，具体 地，可以用于执行方法400或方法500。
在一种实现方式中，获取单元3010，用于获取多个设备的参数组合的评估结果，多个设备的参数组合的评估结果是通过多个设备分别基于多个设备的参数组合在第一时段内进行标定试验得到的，第一时段小于或等于第一阈值。处理单元3020，用于根据多个设备的参数组合的评估结果得到至少一个调整后的参数组合。
可选地，作为一个实施例，多个设备的参数组合相同，多个设备所处的工况不同，以及处理单元3020具体用于：将多个设备的参数组合的评估结果进行处理，得到汇总评估结果；根据汇总评估结果得到至少一个调整后的参数组合。
可选地，作为一个实施例，装置还包括：发送单元3030，用于将至少一个调整后的参数组合发送至至少一个设备。
可选地，作为一个实施例，至少一个设备是根据以下至少一项确定的：多个设备所处的工况或至少一个调整后的参数组合。
可选地，作为一个实施例，多个设备的参数组合的评估结果是通过对多个设备的测试数据进行评估得到的，多个设备的测试数据是根据多个设备分别基于多个设备的参数组合在第一时段内进行标定试验的过程中采集到的数据确定的。
可选地，作为一个实施例，多个设备的参数组合的评估结果是由用户反馈得到的。
可选地，作为一个实施例，发送单元3030还用于：将至少一个设备的待执行动作序列分别发送至至少一个设备，至少一个设备的待执行动作序列包括至少一个设备在基于至少一个设备的调整后的参数组合进行标定试验的过程中所需要执行的动作。
可选地，作为一个实施例，至少一个设备的待执行动作序列是根据以下至少一项确定的：参数标定要求覆盖的动作、至少一个设备所处的工况或多个设备的参数组合的评估结果。
可选地，作为一个实施例，多个设备包括以下至少一项：车辆或测试台架。
在另一种实现方式中，获取单元30103，用于获取调整后的参数组合，调整后的参数组合是根据多个设备的参数组合的评估结果得到的，多个设备包括第一设备，多个设备的参数组合的评估结果是通过多个设备分别基于多个设备的参数组合在第一时段内进行标定试验得到的，第一时段小于或等于第一阈值。
处理单元3020，用于控制第一设备基于调整后的参数组合进行标定试验。
可选地，作为一个实施例，多个设备的参数组合相同，多个设备所处的工况不同，调整后的参数组合是根据汇总评估结果得到的，汇总评估结果是通将多个设备的参数组合的评估结果进行处理得到的。
可选地，作为一个实施例，第一设备是根据以下至少一项确定的：多个设备所处的工况或调整后的参数组合。
可选地，作为一个实施例，多个设备的参数组合的评估结果是通过对多个设备的测试数据进行评估得到的，多个设备的测试数据是根据多个设备分别基于多个设备的参数组合在第一时段内进行标定试验的过程中采集到的数据确定的。
可选地，作为一个实施例，多个设备的参数组合的评估结果是由用户反馈得到的。
可选地，作为一个实施例，获取单元3010还用于：获取第一设备的待执行动作序列，第一设备的待执行动作序列包括第一设备在基于调整后的参数组合进行标定试验的过程 中所需要执行的动作。
可选地，作为一个实施例，第一设备的待执行动作序列是根据以下至少一项确定的：参数标定要求覆盖的动作、第一设备所处的工况或多个设备的参数组合的评估结果。
可选地，作为一个实施例，第一设备为：车辆或测试台架。
需要说明的是，上述装置3000以功能单元的形式体现。这里的术语“单元”可以通过软件和/或硬件形式实现，对此不作具体限定。
例如，“单元”可以是实现上述功能的软件程序、硬件电路或二者结合。所述硬件电路可能包括应用特有集成电路(application specific integrated circuit，ASIC)、电子电路、用于执行一个或多个软件或固件程序的处理器(例如共享处理器、专有处理器或组处理器等)和存储器、合并逻辑电路和/或其它支持所描述的功能的合适组件。
因此，在本申请的实施例中描述的各示例的单元，能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。
图7是本申请实施例提供的参数标定的装置的硬件结构示意图。图7所示的参数标定的装置5000(该装置5000具体可以是一种计算机设备)包括存储器5001、处理器5002、通信接口5003以及总线5004。其中，存储器5001、处理器5002、通信接口5003通过总线5004实现彼此之间的通信连接。
存储器5001可以是只读存储器(read only memory，ROM)，静态存储设备，动态存储设备或者随机存取存储器(random access memory，RAM)。存储器5001可以存储程序，当存储器5001中存储的程序被处理器5002执行时，处理器5002用于执行本申请实施例的参数标定的方法的各个步骤。具体地，处理器5002可以执行上文中图4所示的方法400，或者执行上文中图5所示的方法500。
处理器5002可以采用通用的中央处理器(central processing unit，CPU)，微处理器，应用专用集成电路(application specific integrated circuit，ASIC)，图形处理器(graphics processing unit，GPU)或者一个或多个集成电路，用于执行相关程序，以实现本申请方法实施例的参数标定的方法。
处理器5002还可以是一种集成电路芯片，具有信号的处理能力。在实现过程中，本申请的参数标定的方法的各个步骤可以通过处理器5002中的硬件的集成逻辑电路或者软件形式的指令完成。
上述处理器5002还可以是通用处理器、数字信号处理器(digital signal processing，DSP)、专用集成电路(ASIC)、现成可编程门阵列(field programmable gate array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件。可以实现或者执行本申请实施例中的公开的各方法、步骤及逻辑框图。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。结合本申请实施例所公开的方法的步骤可以直接体现为硬件译码处理器执行完成，或者用译码处理器中的硬件及软件模块组合执行完成。软件模块可以位于随机存储器，闪存、只读存储器，可编程只读存储器或者电可擦写可编程存储器、寄存器等本领域成熟的存储介质中。该存储介质位于存储器5001，处理器5002读取存储器5001中的信息，结合其硬件完成图6所示的装置中包括的单元所需执行的功 能，或者，执行本申请方法实施例的图4或图5所示的参数标定的方法。
通信接口5003使用例如但不限于收发器一类的收发装置，来实现装置5000与其他设备或通信网络之间的通信。例如，可以通过通信接口5003获取参数组合的评估结果等。
总线5004可包括在装置5000各个部件(例如，存储器5001、处理器5002、通信接口5003)之间传送信息的通路。
应注意，尽管上述装置5000仅仅示出了存储器、处理器、通信接口，但是在具体实现过程中，本领域的技术人员应当理解，装置5000还可以包括实现正常运行所必须的其他器件。同时，根据具体需要，本领域的技术人员应当理解，装置5000还可包括实现其他附加功能的硬件器件。此外，本领域的技术人员应当理解，装置5000也可仅仅包括实现本申请实施例所必须的器件，而不必包括图7中所示的全部器件。
应理解，本申请实施例中的处理器可以为中央处理单元(central processing unit，CPU)，该处理器还可以是其他通用处理器、数字信号处理器(digital signal processor，DSP)、专用集成电路(application specific integrated circuit，ASIC)、现成可编程门阵列(field programmable gate array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。
还应理解，本申请实施例中的存储器可以是易失性存储器或非易失性存储器，或可包括易失性和非易失性存储器两者。其中，非易失性存储器可以是只读存储器(read-only memory，ROM)、可编程只读存储器(programmable ROM，PROM)、可擦除可编程只读存储器(erasable PROM，EPROM)、电可擦除可编程只读存储器(electrically EPROM，EEPROM)或闪存。易失性存储器可以是随机存取存储器(random access memory，RAM)，其用作外部高速缓存。通过示例性但不是限制性说明，许多形式的随机存取存储器(random access memory，RAM)可用，例如静态随机存取存储器(static RAM，SRAM)、动态随机存取存储器(DRAM)、同步动态随机存取存储器(synchronous DRAM，SDRAM)、双倍数据速率同步动态随机存取存储器(double data rate SDRAM，DDR SDRAM)、增强型同步动态随机存取存储器(enhanced SDRAM，ESDRAM)、同步连接动态随机存取存储器(synchlink DRAM，SLDRAM)和直接内存总线随机存取存储器(direct rambus RAM，DR RAM)。
上述实施例，可以全部或部分地通过软件、硬件、固件或其他任意组合来实现。当使用软件实现时，上述实施例可以全部或部分地以计算机程序产品的形式实现。所述计算机程序产品包括一个或多个计算机指令或计算机程序。在计算机上加载或执行所述计算机指令或计算机程序时，全部或部分地产生按照本申请实施例所述的流程或功能。所述计算机可以为通用计算机、专用计算机、计算机网络、或者其他可编程装置。所述计算机指令可以存储在计算机可读存储介质中，或者从一个计算机可读存储介质向另一个计算机可读存储介质传输，例如，所述计算机指令可以从一个网站站点、计算机、服务器或数据中心通过有线(例如红外、无线、微波等)方式向另一个网站站点、计算机、服务器或数据中心进行传输。所述计算机可读存储介质可以是计算机能够存取的任何可用介质或者是包含一个或多个可用介质集合的服务器、数据中心等数据存储设备。所述可用介质可以是磁性介质(例如，软盘、硬盘、磁带)、光介质(例如，DVD)、或者半导体介质。半导体介质 可以是固态硬盘。
本申请还提供一种计算机程序产品，该计算机程序产品包括：计算机程序代码，当该计算机程序代码在计算机上运行时，使得该计算机执行前述任意一个方法实施例中的方法。
本申请还提供一种计算机可读介质，该计算机可读介质存储有程序代码，当该程序代码在计算机上运行时，使得该计算机执行前述任意一个方法实施例中的方法。
本申请还提供一种电子设备，该电子设备包括前述任意一个装置实施例中的参数标定的装置。
除非另有定义，本文所使用的所有的技术和科学术语与属于本申请的技术领域的技术人员通常理解的含义相同。本文中在本申请的说明书中所使用的术语只是为了描述具体的实施例的目的，不是旨在于限制本申请。
本领域普通技术人员可以意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的系统、装置和单元的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
在本申请所提供的几个实施例中，应该理解到，所揭露的系统、装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。
所述功能如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(read-only memory，ROM)、随机存取存储器(random access memory，RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述，仅为本申请的具体实施方式，但本申请的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本申请揭露的技术范围内，可轻易想到变化或替换，都应涵盖 在本申请的保护范围之内。因此，本申请的保护范围应以所述权利要求的保护范围为准。

Claims (36)

  1. 一种参数标定的方法，其特征在于，包括：
    获取多个设备的参数组合的评估结果，所述多个设备的参数组合的评估结果是通过所述多个设备分别基于所述多个设备的参数组合在第一时段内进行标定试验得到的，所述第一时段小于或等于第一阈值；
    根据所述多个设备的参数组合的评估结果得到至少一个调整后的参数组合。
  2. 根据权利要求1所述的方法，其特征在于，所述多个设备的参数组合相同，所述多个设备所处的工况不同，以及
    所述根据所述多个设备的参数组合的评估结果得到至少一个调整后的参数组合，包括：
    将所述多个设备的参数组合的评估结果进行处理，得到汇总评估结果；
    根据所述汇总评估结果得到所述至少一个调整后的参数组合。
  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：
    将所述至少一个调整后的参数组合发送至至少一个设备。
  4. 根据权利要求3所述的方法，其特征在于，所述至少一个设备是根据以下至少一项确定的：
    所述多个设备所处的工况或所述至少一个调整后的参数组合。
  5. 根据权利要求1至4中任一项所述的方法，其特征在于，所述多个设备的参数组合的评估结果是通过对所述多个设备的测试数据进行评估得到的，所述多个设备的测试数据是根据所述多个设备分别基于所述多个设备的参数组合在第一时段内进行标定试验的过程中采集到的数据确定的。
  6. 根据权利要求1至4中任一项所述的方法，其特征在于，所述多个设备的参数组合的评估结果是由用户反馈得到的。
  7. 根据权利要求3至6中任一项所述的方法，其特征在于，所述方法还包括：
    将所述至少一个设备的待执行动作序列分别发送至所述至少一个设备，所述至少一个设备的待执行动作序列包括所述至少一个设备在基于所述至少一个设备的调整后的参数组合进行标定试验的过程中所需要执行的动作。
  8. 根据权利要求7所述的方法，其特征在于，所述至少一个设备的待执行动作序列是根据以下至少一项确定的：
    参数标定要求覆盖的动作、所述至少一个设备所处的工况或所述多个设备的参数组合的评估结果。
  9. 根据权利要求1至8中任一项所述的方法，其特征在于，所述多个设备包括以下至少一项：车辆或测试台架。
  10. 一种参数标定的方法，其特征在于，包括：
    获取调整后的参数组合，所述调整后的参数组合是根据多个设备的参数组合的评估结果得到的，所述多个设备包括第一设备，所述多个设备的参数组合的评估结果是通过所述多个设备分别基于所述多个设备的参数组合在第一时段内进行标定试验得到的，所述第一 时段小于或等于第一阈值；
    控制所述第一设备基于所述调整后的参数组合进行标定试验。
  11. 根据权利要求10所述的方法，其特征在于，所述多个设备的参数组合相同，所述多个设备所处的工况不同，所述调整后的参数组合是根据汇总评估结果得到的，所述汇总评估结果是通将所述多个设备的参数组合的评估结果进行处理得到的。
  12. 根据权利要求10或11所述的方法，其特征在于，所述第一设备是根据以下至少一项确定的：
    所述多个设备所处的工况或所述调整后的参数组合。
  13. 根据权利要求10至12中任一项所述的方法，其特征在于，所述多个设备的参数组合的评估结果是通过对所述多个设备的测试数据进行评估得到的，所述多个设备的测试数据是根据所述多个设备分别基于所述多个设备的参数组合在第一时段内进行标定试验的过程中采集到的数据确定的。
  14. 根据权利要求10至12中任一项所述的方法，其特征在于，所述多个设备的参数组合的评估结果是由用户反馈得到的。
  15. 根据权利要求10至14中任一项所述的方法，其特征在于，所述方法还包括：
    获取所述第一设备的待执行动作序列，所述第一设备的待执行动作序列包括所述第一设备在基于所述调整后的参数组合进行标定试验的过程中所需要执行的动作。
  16. 根据权利要求15所述的方法，其特征在于，所述第一设备的待执行动作序列是根据以下至少一项确定的：
    参数标定要求覆盖的动作、所述第一设备所处的工况或所述多个设备的参数组合的评估结果。
  17. 根据权利要求10至16中任一项所述的方法，其特征在于，所述第一设备为：车辆或测试台架。
  18. 一种参数标定的装置，其特征在于，包括：
    获取单元，用于获取多个设备的参数组合的评估结果，所述多个设备的参数组合的评估结果是通过所述多个设备分别基于所述多个设备的参数组合在第一时段内进行标定试验得到的，所述第一时段小于或等于第一阈值；
    处理单元，用于根据所述多个设备的参数组合的评估结果得到至少一个调整后的参数组合。
  19. 根据权利要求18所述的装置，其特征在于，所述多个设备的参数组合相同，所述多个设备所处的工况不同，以及
    所述处理单元具体用于：
    将所述多个设备的参数组合的评估结果进行处理，得到汇总评估结果；
    根据所述汇总评估结果得到所述至少一个调整后的参数组合。
  20. 根据权利要求18或19所述的装置，其特征在于，所述装置还包括：
    发送单元，用于将所述至少一个调整后的参数组合发送至至少一个设备。
  21. 根据权利要求20所述的装置，其特征在于，所述至少一个设备是根据以下至少一项确定的：
    所述多个设备所处的工况或所述至少一个调整后的参数组合。
  22. 根据权利要求18至21中任一项所述的装置，其特征在于，所述多个设备的参数组合的评估结果是通过对所述多个设备的测试数据进行评估得到的，所述多个设备的测试数据是根据所述多个设备分别基于所述多个设备的参数组合在第一时段内进行标定试验的过程中采集到的数据确定的。
  23. 根据权利要求18至21中任一项所述的装置，其特征在于，所述多个设备的参数组合的评估结果是由用户反馈得到的。
  24. 根据权利要求20至23中任一项所述的装置，其特征在于，所述发送单元还用于：
    将所述至少一个设备的待执行动作序列分别发送至所述至少一个设备，所述至少一个设备的待执行动作序列包括所述至少一个设备在基于所述至少一个设备的调整后的参数组合进行标定试验的过程中所需要执行的动作。
  25. 根据权利要求24所述的装置，其特征在于，所述至少一个设备的待执行动作序列是根据以下至少一项确定的：
    参数标定要求覆盖的动作、所述至少一个设备所处的工况或所述多个设备的参数组合的评估结果。
  26. 根据权利要求18至25中任一项所述的装置，其特征在于，所述多个设备包括以下至少一项：车辆或测试台架。
  27. 一种参数标定的装置，其特征在于，包括：
    获取单元，用于获取调整后的参数组合，所述调整后的参数组合是根据多个设备的参数组合的评估结果得到的，所述多个设备包括第一设备，所述多个设备的参数组合的评估结果是通过所述多个设备分别基于所述多个设备的参数组合在第一时段内进行标定试验得到的，所述第一时段小于或等于第一阈值；
    处理单元，用于控制所述第一设备基于所述调整后的参数组合进行标定试验。
  28. 根据权利要求27所述的装置，其特征在于，所述多个设备的参数组合相同，所述多个设备所处的工况不同，所述调整后的参数组合是根据汇总评估结果得到的，所述汇总评估结果是通将所述多个设备的参数组合的评估结果进行处理得到的。
  29. 根据权利要求27或28所述的装置，其特征在于，所述第一设备是根据以下至少一项确定的：
    所述多个设备所处的工况或所述调整后的参数组合。
  30. 根据权利要求27至29中任一项所述的装置，其特征在于，所述多个设备的参数组合的评估结果是通过对所述多个设备的测试数据进行评估得到的，所述多个设备的测试数据是根据所述多个设备分别基于所述多个设备的参数组合在第一时段内进行标定试验的过程中采集到的数据确定的。
  31. 根据权利要求27至29中任一项所述的装置，其特征在于，所述多个设备的参数组合的评估结果是由用户反馈得到的。
  32. 根据权利要求27至31中任一项所述的装置，其特征在于，所述获取单元还用于：
    获取所述第一设备的待执行动作序列，所述第一设备的待执行动作序列包括所述第一设备在基于所述调整后的参数组合进行标定试验的过程中所需要执行的动作。
  33. 根据权利要求32所述的装置，其特征在于，所述第一设备的待执行动作序列是根据以下至少一项确定的：
    参数标定要求覆盖的动作、所述第一设备所处的工况或所述多个设备的参数组合的评估结果。
  34. 根据权利要求27至33中任一项所述的装置，其特征在于，所述第一设备为：车辆或测试台架。
  35. 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中存储用于计算设备执行的指令，所述指令用于实现如权利要求1至9或权利要求10至17中任一项所述的参数标定的方法。
  36. 一种包含指令的计算机程序产品，其特征在于，当所述计算机程序产品在计算机上运行时，使得计算机执行如权利要求1至9或权利要求10至17中任一项所述的参数标定的方法。
PCT/CN2022/112810 2021-09-29 2022-08-16 参数标定的方法及装置 WO2023051065A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
EP22874460.3A EP4386612A1 (en) 2021-09-29 2022-08-16 Parameter calibration method and apparatus

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202111148039.4 2021-09-29
CN202111148039.4A CN113987753A (zh) 2021-09-29 2021-09-29 参数标定的方法及装置

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US18/610,200 Continuation US20240230383A1 (en) 2021-09-29 2024-03-19 Parameter calibration method and apparatus

Publications (1)

Publication Number Publication Date
WO2023051065A1 true WO2023051065A1 (zh) 2023-04-06

Family

ID=79737105

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/112810 WO2023051065A1 (zh) 2021-09-29 2022-08-16 参数标定的方法及装置

Country Status (3)

Country Link
EP (1) EP4386612A1 (zh)
CN (1) CN113987753A (zh)
WO (1) WO2023051065A1 (zh)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113987753A (zh) * 2021-09-29 2022-01-28 华为技术有限公司 参数标定的方法及装置
CN114919343B (zh) * 2022-03-29 2023-07-07 东风柳州汽车有限公司 一种轮胎温度监控方法、装置及存储介质
CN114780326B (zh) * 2022-04-27 2023-03-24 镁佳(北京)科技有限公司 一种跨平台标定测试方法、装置及设备

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130135474A1 (en) * 2011-11-28 2013-05-30 Clarion Co., Ltd. Automotive Camera System and Its Calibration Method and Calibration Program
CN110322587A (zh) * 2018-03-28 2019-10-11 广州汽车集团股份有限公司 驾驶过程中的评价记录方法、装置、设备及存储介质
CN111951430A (zh) * 2019-04-30 2020-11-17 广州汽车集团股份有限公司 一种车辆驾驶性评价方法及其系统
CN113987753A (zh) * 2021-09-29 2022-01-28 华为技术有限公司 参数标定的方法及装置

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130135474A1 (en) * 2011-11-28 2013-05-30 Clarion Co., Ltd. Automotive Camera System and Its Calibration Method and Calibration Program
CN110322587A (zh) * 2018-03-28 2019-10-11 广州汽车集团股份有限公司 驾驶过程中的评价记录方法、装置、设备及存储介质
CN111951430A (zh) * 2019-04-30 2020-11-17 广州汽车集团股份有限公司 一种车辆驾驶性评价方法及其系统
CN113987753A (zh) * 2021-09-29 2022-01-28 华为技术有限公司 参数标定的方法及装置

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
AUTOBIT: "The three cities have been synchronized in two years, and the road test of Yuanrong Qixing has exceeded 1 million kilometers. How can the dark horse of L4 autonomous driving come first?", TENCENT CORPORATION, 27 January 2021 (2021-01-27), XP093053138, Retrieved from the Internet <URL:https://view.inews.qq.com/a/20210127A06XB100> [retrieved on 20230609] *

Also Published As

Publication number Publication date
CN113987753A (zh) 2022-01-28
EP4386612A1 (en) 2024-06-19

Similar Documents

Publication Publication Date Title
WO2023051065A1 (zh) 参数标定的方法及装置
CN110895147B (zh) 自动驾驶车辆的用相机捕获图像数据的图像数据获取逻辑
US20210171050A1 (en) System for evaluating vehicle performance
CN113219962A (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
Fox et al. Crowdsourcing undersampled vehicular sensor data for pothole detection
CN116476843A (zh) 目标滑移估计
US10097671B2 (en) Compression algorithms for vehicle-bus-messaging of preview data
Wang et al. Fair: Towards impartial resource allocation for intelligent vehicles with automotive edge computing
CN113968231A (zh) 一种符合驾驶员习惯的智能驾驶员模型参数确定方法
CN109552326B (zh) 一种无线通信信道衰减下的车辆变速巡航控制方法
US20240230383A1 (en) Parameter calibration method and apparatus
CN112347558B (zh) 一种基于概率波的闭环优化方法、系统、电子设备和介质
CN115489340A (zh) 多电机扭矩分配方法、装置、计算机设备、存储介质
US20220073073A1 (en) Device and Method for Controlling Platooning
CN114841514A (zh) 模型训练和车辆舒适性评价方法、装置、设备及存储介质
US10834550B2 (en) Vehicle feature control
CN113074950A (zh) 试车场汽车制动试验的模拟方法
US20230392940A1 (en) Systems and methods for generating fuel efficiency score based on cell phone sensor data
KR20210052770A (ko) Ecu 데이터 학습 시스템 및 ecu 데이터 학습 방법
CN112364561B (zh) 控车动作修正方法、装置、电子设备和存储介质
CN114228722B (zh) 驾驶风格划分方法、装置、设备、存储介质和程序产品
WO2023024618A1 (zh) 一种数据处理方法及相关装置
JP6470221B2 (ja) 通信装置、通信方法、及びプログラム
WO2023106021A1 (ja) 車載装置、路側装置、制御方法およびコンピュータプログラム
JP7074166B2 (ja) サーバ、車両の制御装置、および車両の機械学習システム

Legal Events

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

Ref document number: 22874460

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 2022874460

Country of ref document: EP

ENP Entry into the national phase

Ref document number: 2022874460

Country of ref document: EP

Effective date: 20240311

NENP Non-entry into the national phase

Ref country code: DE