Equipment upgrading method, device and system

Abstract

The embodiment of the invention provides an equipment upgrading method and device and an upgrading system, and relates to the technical field of computer application, and the method comprises the steps that a main function module in terminal equipment obtains a first model and / or a second model from a version server, the first model is used for the terminal equipment to make a version upgrading decision, and the second model is used for the terminal equipment to make a version upgrading decision; the second model is used for the terminal device to make a version rollback decision, and the first model and / or the second model are / is obtained by training of a version server; the first model is used for carrying out version upgrading decision on the first function module, and / or the second model is used for carrying out version rollback decision on the first function module, and the first function module is any one of a main function module and a plurality of sub-function modules of the terminal equipment. The trained AI model is provided for the terminal equipment, and the terminal equipment autonomously carries out version upgrading or version rollback decision making, so that autonomy, differentiation and intelligentization of equipment upgrading and efficient resource allocation and iterative updating are realized.

Description

Technical Field

[0001] The embodiments of the present disclosure relate to the field of intelligent warehousing technology, and more particularly to an equipment upgrading method, device and upgrading system. Background Art

[0002] Over-the-Air Technology (OTA) is a technology that enables remote management of mobile terminal devices and SIM card data through the air interface of mobile communications.

[0003] Traditional OTA requires manual triggering of the operation control server to make upgrade decisions, which has poor timeliness and cannot repair and update in time, affecting user use; the corresponding updated functions are determined by the uploaded upgrade package, and differentiated updates cannot be performed, occupying device memory and causing the upgrade time to be too long; it can only be updated according to the upgrade package triggered by the user / supplier, and cannot find a suitable upgrade package to repair the problem according to the current system status, and cannot perform autonomous diagnosis and update. Summary of the invention

[0004] The embodiments of the present disclosure provide a method, apparatus and system for upgrading equipment. The trained AI model is deployed inside a robot. The AI ​​model determines whether an upgrade is required based on the state and performance of the robot itself, thereby upgrading the corresponding functional modules. The embodiments of the present disclosure disclose the following technical solutions:

[0005] A first aspect of an embodiment of the present disclosure provides a device upgrade method, which is applied to a main function module in a terminal device, comprising: obtaining a first model and / or a second model from a version server, the first model being used for the terminal device to make version upgrade decisions, and the second model being used for the terminal device to make version rollback decisions, the first model and / or the second model being trained by the version server; using the first model to make version upgrade decisions for a first function module, and / or using the second model to make version rollback decisions for the first function module, the first function module being the main function module of the terminal device and any one of a plurality of sub-function modules.

[0006] In some embodiments, using the first model to make a version upgrade decision for the first functional module includes: obtaining current parameters of the first functional module; based on the current parameters, using the first model to determine whether the first functional module needs to be upgraded; when it is determined that the first functional module needs to be upgraded, based on the current parameters, using the first model to determine at least one of the upgrade version, upgrade time, and upgrade location of the first functional module.

[0007] In some embodiments, based on current parameters, using the first model to determine whether the first functional module needs to be upgraded includes: based on the current state parameters, current performance parameters, current version parameters and at least one of the hardware parameters of the first functional module, when the first model is used to judge that the operating status of the first functional module meets the upgrade conditions and / or there is currently an upgradeable version, determining that the first functional module needs to be upgraded.

[0008] In some embodiments, based on current parameters, using the first model to determine at least one of the upgrade version, upgrade time, and upgrade location of the first functional module includes: determining the upgrade version of the first functional module through the first model based on the current version parameters of the first functional module and at least one of the hardware parameters; determining the upgrade time and / or upgrade location of the first functional module through the first model based on the current state parameters of the first functional module and at least one of the current performance parameters.

[0009] In some embodiments, the method further includes: sending current parameters and upgrade decision results to a version server, the upgrade decision results including at least one of indication information that the first functional module needs to be upgraded, an upgraded version of the first functional module, an upgrade time, and an upgrade location; receiving an upgrade package corresponding to the upgraded version sent by the version server; and installing the upgrade package when the upgrade time and / or upgrade location are reached.

[0010] In some embodiments, the method also includes: automatically backing up current parameters and upgrade decision results; monitoring the installation process of the upgrade package to determine whether an installation exception occurs in the upgrade package, the installation exception including at least one of an installation interruption, an installation freeze, or an installation error; in the event that an installation exception occurs in the upgrade package, determining and recording the cause of the installation exception.

[0011] In some embodiments, using the second model to make a version rollback decision for the first functional module includes: obtaining current parameters of the first functional module; based on the current parameters, using the second model to determine whether the first functional module meets the rollback conditions; when it is determined that the rollback conditions are met, based on the current parameters, using the second model to determine the target historical version before the current version, the target historical version being a historical version of the first functional module in multiple historical versions before being upgraded to the current version, whose operating status meets preset conditions.

[0012] In some embodiments, based on current parameters, using the second model to determine whether the first functional module meets the rollback conditions includes: based on the current state parameters of the first functional module, determining the fault level of the first functional module through the second model, and when the fault level is higher than a preset level threshold, or when the fault level is an installation abnormality, determining that the first functional module meets the rollback conditions; wherein, based on current parameters, using the second model to determine the target historical version before the current version includes: based on the current version parameters of the first functional module, determining the rollback historical version corresponding to the fault level through the second model, and determining from the rollback historical versions a historical version whose operating status meets the preset conditions as the target historical version.

[0013] In some embodiments, the method also includes: sending current parameters and rollback decision results to the version server, the rollback decision results including indication information that the first functional module needs to be rolled back and the target historical version; receiving the upgrade package corresponding to the target historical version sent by the version server and installing the upgrade package when the first functional module has uninstalled and deleted the upgrade package corresponding to the target historical version; or, installing the upgrade package when the first functional module has not deleted the upgrade package corresponding to the target historical version.

[0014] In some embodiments, the method further includes: obtaining historical parameters of the first functional module; sending the historical parameters to the version server, and using the historical parameters for training data so that the version server performs model training to obtain the first model and / or the second model.

[0015] In some embodiments, the current parameters and / or historical parameters include at least one of a state parameter, a performance parameter, a version parameter, and a hardware parameter, wherein the state parameter includes at least one of an abnormal state, a device state, and a task state, wherein the abnormal state is used to identify whether the first functional module is abnormal, the device state is used to identify whether the first functional module is in an idle or busy state, and the task state is used to identify the progress state, success state, or failure state of the task performed by the first functional module; the performance parameter includes at least one of a network performance, a storage performance, a processing performance, and an idle interval performance, wherein the network performance is used to identify the performance of the first communication connection between the main functional module and the plurality of sub-functional modules, and / or Or used to identify the performance of the second communication connection between the main function module and the version server, the storage performance is used to identify the remaining storage space size of the first function module, and the processing performance is used to identify the processing capacity of the first function module; the idle performance is used to identify the idle time interval between the first function module executing the current task and the next task; the version parameters include the current version number of the first function module, the operating parameters of the first function module under the current version number, at least one previous version number of the first function module before the current version number, and at least one of the operating parameters of the first function module under at least one previous version number; the hardware parameters include at least one of the model and factory of the first function module.

[0016] In some embodiments, the method also includes: sending feedback data to the version server, the feedback data including at least one of the current parameters of the first functional module, the upgrade decision result, and the rollback decision result, the feedback data is used by the version server to send an upgrade installation package and / or a rollback installation package to the terminal device, and / or the feedback data is used to update the first model and / or the second model.

[0017] A second aspect of an embodiment of the present disclosure provides a device upgrade method, which is applied to a version server, including: training a first initial model to obtain a first model, and / or training a second initial model to obtain a second model; sending the first model and / or the second model to a main function module of a terminal device, the first model is used by the main function module to make a version upgrade decision on the first function module, and the second model is used by the main function module to make a version rollback decision on the first function module, the first function module is an active energy module of the terminal device and any one of a plurality of sub-function modules.

[0018] In some embodiments, the method also includes: receiving current parameters of the first function module and an upgrade decision result sent by the main function module of the terminal device, the upgrade decision result is obtained by the main function module based on the current parameters and using the first model, and the upgrade decision result includes at least one of the indication information that the first function module needs to be upgraded, the upgraded version of the first function module, the upgrade time, and the upgrade location; and sending an upgrade package corresponding to the upgraded version to the main function module.

[0019] In some embodiments, the method also includes: receiving current parameters of the first function module and a rollback decision result sent by the main function module of the terminal device, the rollback decision result is obtained by the main function module based on the current parameters and using the second model, and the rollback decision result includes indication information that the first function module needs to be rolled back and a target historical version; sending an upgrade package corresponding to the target historical version to the main function module.

[0020] In some embodiments, the method also includes: receiving historical parameters of the first function module sent by the main function module of the terminal device, and receiving an upgrade version number and an upgrade package for upgrading the first function module sent by the compilation server; wherein, performing model training on the first initial model to obtain the first model, and / or, performing model training on the second initial model to obtain the second model includes: using at least one of the historical parameters and the upgrade version number as training data, performing model training on the first initial model to obtain the first model, and / or, performing model training on the second initial model to obtain the second model.

[0021] In some embodiments, model training is performed on the first initial model to obtain the first model, including: using at least one of the historical state parameters, historical performance parameters, historical version parameters, hardware parameters, and upgrade version number of the first functional module as training data, using a preset upgrade result label to train the first initial model, and when the number of training iterations reaches a first preset number or the training error is less than or equal to a first preset error, the training is terminated to obtain the first model.

[0022] In some embodiments, model training is performed on the second initial model to obtain the second model, including: using the historical state parameters and historical version parameters of the first functional module as training data, using a preset rollback result label, training the second initial model, and when the number of training iterations reaches a second preset number or the training error is less than or equal to the second preset error, ending the training and obtaining the second model.

[0023] In some embodiments, the current parameters and / or historical parameters include at least one of a state parameter, a performance parameter, a version parameter, and a hardware parameter, wherein the state parameter includes at least one of an abnormal state, a device state, and a task state, wherein the abnormal state is used to identify whether the first functional module is abnormal, the device state is used to identify whether the first functional module is in an idle or busy state, and the task state is used to identify the progress state, success state, or failure state of the task performed by the first functional module; the performance parameter includes at least one of a network performance, a storage performance, a processing performance, and an idle interval performance, wherein the network performance is used to identify the performance of the first communication connection between the main functional module and the plurality of sub-functional modules, and / or Or used to identify the performance of the second communication connection between the main function module and the version server, the storage performance is used to identify the remaining storage space size of the first function module, and the processing performance is used to identify the processing capacity of the first function module; the idle performance is used to identify the idle time interval between the first function module executing the current task and the next task; the version parameters include the current version number of the first function module, the operating parameters of the first function module under the current version number, at least one previous version number of the first function module before the current version number, and at least one of the operating parameters of the first function module under at least one previous version number; the hardware parameters include at least one of the model and factory of the first function module.

[0024] A third aspect of an embodiment of the present disclosure provides a device upgrade apparatus, which is applied to a main function module in a terminal device, including: an acquisition unit, used to acquire a first model and / or a second model from a version server, the first model is used for the terminal device to make version upgrade decisions, and the second model is used for the terminal device to make version rollback decisions, and the first model and / or the second model are trained by the version server; a decision unit, used to use the first model to make version upgrade decisions for the first function module, and / or use the second model to make version rollback decisions for the first function module, the first function module is the main function module of the terminal device and any one of multiple sub-function modules.

[0025] A fourth aspect of an embodiment of the present disclosure provides a device upgrade apparatus, which is applied to a version server, and includes: a training unit, which is used to perform model training on a first initial model to obtain a first model, and / or, to perform model training on a second initial model to obtain a second model; a sending unit, which is used to send the first model and / or the second model to a main function module of a terminal device, wherein the first model is used for the main function module to make a version upgrade decision on the first function module, and the second model is used for the main function module to make a version rollback decision on the first function module, and the first function module is the main function module of the terminal device and any one of a plurality of sub-function modules.

[0026] The fifth aspect of the embodiments of the present disclosure provides a device upgrade system, including a terminal device and a version server, wherein the main function module in the terminal device is used to execute the method described in any one of the embodiments in the first aspect of the present disclosure; the version server is used to execute the method described in any one of the embodiments in the second aspect of the present disclosure.

[0027] A sixth aspect of an embodiment of the present disclosure provides an electronic device, comprising: a processor and a memory, the memory being used to store computer executable instructions; the processor being used to read instructions from the memory and execute the instructions to implement the method in any one of the implementations of the first aspect or the second aspect described above.

[0028] A seventh aspect of an embodiment of the present disclosure provides a computer-readable storage medium, in which computer instructions are stored, and the computer instructions are configured to enable the computer to execute the method in any implementation of the first aspect or the second aspect.

[0029] An eighth aspect of an embodiment of the present disclosure provides a computer program product, which includes a computer program stored on a computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer executes the method in any implementation of the first or second aspect above.

[0030] The device upgrade method, apparatus and upgrade system provided by the embodiments of the present disclosure, the terminal device sends historical parameters to the server, the server performs model training, and sends the trained model to the terminal, so that the terminal can make autonomous version upgrade decisions or version rollback decisions, thereby performing version upgrades or version rollbacks on corresponding functional modules, realizing autonomous diagnosis and updates according to the current system status, as well as differentiated and preventive version upgrades. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 A schematic diagram of a warehousing system 100 provided according to an embodiment of the present disclosure;

[0032] Figure 2 A flow chart of a device upgrade method for a main function module in a terminal device proposed in an embodiment of the present disclosure;

[0033] Figure 3 A schematic diagram of a process for making a version upgrade decision for a first functional module proposed in an embodiment of the present disclosure;

[0034] Figure 4 A schematic diagram of a flow chart of a device upgrade method for a main function module in a terminal device proposed in an embodiment of the present disclosure;

[0035] Figure 5A schematic diagram of a flow chart of a device upgrade method for a main function module in a terminal device proposed in an embodiment of the present disclosure;

[0036] Figure 6 A schematic diagram of a flow chart of a device upgrade method for a main function module in a terminal device proposed in an embodiment of the present disclosure;

[0037] Figure 7 A schematic diagram of a process for making a version rollback decision for a first functional module proposed in an embodiment of the present disclosure;

[0038] Figure 8 A schematic diagram of a process for making a version rollback decision for a first functional module proposed in an embodiment of the present disclosure;

[0039] Fig. 9 A schematic diagram of a flow chart of a device upgrade method for a main function module in a terminal device proposed in an embodiment of the present disclosure;

[0040] Fig.10 A schematic diagram of a process flow of a device upgrade method applied to a version server proposed in an embodiment of the present disclosure;

[0041] Fig.11 A schematic diagram of a process flow of a device upgrade method applied to a version server proposed in an embodiment of the present disclosure;

[0042] Fig. 12A · A schematic diagram of the device upgrade method provided by the present disclosure;

[0043] Fig. 12B A flow chart of a device upgrade method provided by the present disclosure;

[0044] Fig. 12C It is the intelligent decision-making flowchart of the AI ​​model;

[0045] Fig.12D Flowchart updated for diagnostic anomalies proactively;

[0046] Fig.13 A schematic diagram of the structure of a device upgrading apparatus 1300 of a main function module in a terminal device proposed in an embodiment of the present disclosure;

[0047] Fig.14 A schematic diagram of the structure of a device upgrade apparatus 1400 applied to a version server proposed in an embodiment of the present disclosure;

[0048] Fig.15 is a structural diagram of an electronic device 1500 for implementing the above-mentioned device upgrading method according to an exemplary embodiment. DETAILED DESCRIPTION

[0049] Many specific details are described in the following description to facilitate a full understanding of the present disclosure. However, the present disclosure can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of the present disclosure, so the present disclosure is not limited by the specific implementation disclosed below.

[0050] The terms used in one or more embodiments of the present disclosure are only for the purpose of describing specific embodiments, and are not intended to limit one or more embodiments of the present disclosure. The singular forms of "a", "said" and "the" used in one or more embodiments of the present disclosure and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used in one or more embodiments of the present disclosure refers to and includes any or all possible combinations of one or more associated listed items.

[0051] It should be understood that although the terms first, second, etc. may be used to describe various information in one or more embodiments of the present disclosure, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of one or more embodiments of the present disclosure, the first may also be referred to as the second, and similarly, the second may also be referred to as the first. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0052] In the present disclosure, a device upgrade method is provided. The present disclosure also relates to a device upgrade device applied to a terminal device and a device upgrade device applied to a version server, a computing device, a computer program product, and a computer-readable storage medium, which are described in detail one by one in the following embodiments.

[0053] Figure 1 A schematic diagram of a storage system 100 provided according to an embodiment of the present disclosure is shown. Figure 1 As shown, the warehousing system includes a server 101 and at least one terminal 102 .

[0054] In some embodiments, server 101 includes a version server and a compilation server. The version server is used to train the first model and / or the second model, and the compilation server is used to automatically upload upgrade packages and related data for model training to the version server.

[0055] In some embodiments, the terminal device may be a warehouse robot, and the number of robots in the warehouse system is not limited in the embodiments of the present disclosure. The terminal device has multiple functional modules, such as a camera module, a PNC module, a clock module, etc. of the warehouse robot, and the first model and / or the second model are used by the terminal device to make version upgrade decisions and / or version rollback decisions for any module in the multiple functional modules.

[0056] In some embodiments, the terminal device 102 can be a robot for transporting cartons and / or goods, such as a RoboShuttle (RS) robot. The RS robot transports target goods and / or cartons from a target location on a fixed shelf to other locations, or transports target goods and / or cartons from other locations to a target location on a fixed shelf.

[0057] In some embodiments, the terminal device 102 can be a robot for transporting shelves and / or carriers, for example, a lurking robot, such as a P robot. The P robot is used to transport mobile shelves (also referred to as portable shelves) and / or movable carriers. For example, the P robot can transport the mobile shelf to the picking area according to the transport task sent by the server, or transport the mobile shelf to a position under the fixed shelf or other position.

[0058] In some embodiments, the terminal device 102 can be a four-way robot and / or a six-way robot, such as an X-series robot, which is a storage and retrieval robot that can move in four directions horizontally and / or six directions horizontally and vertically along a track in a three-dimensional shelf.

[0059] In some embodiments, the terminal device 102 may be a sorting robot, such as an S-series robot, for sorting cargo boxes / cargoes according to different types.

[0060] In some embodiments, the terminal device 102 may be a forklift robot, such as an F series robot; or a mobile handling robot, such as an M series robot.

[0061] In some embodiments, different functional modules can be deployed on the terminal device 102 according to different types of robots, such as a communication module for communication, a power supply module for power supply, a camera module for identifying road conditions and identifying QR codes of cargo boxes / shelves / landmarks, a PNC module for planning and controlling maps, a clock module, etc.

[0062] In some embodiments, the terminal device may also be an IoT device such as a vehicle.

[0063] The following describes the solutions of the embodiments of the present disclosure in combination with examples.

[0064] Figure 2This is a flow chart of a device upgrade method proposed in an embodiment of the present disclosure, which is applied to the main function module in the terminal device. Figure 2 As shown, the method comprises the following steps:

[0065] Step 201: Obtain the first model and / or the second model from the version server.

[0066] In some embodiments, the first model is used by the terminal device to make version upgrade decisions, and the second model is used by the terminal device to make version rollback decisions. The first model and / or the second model are trained by the version server.

[0067] In some embodiments, the version server may be a server in a server cluster in a storage system used for model training, which receives data reported by a terminal device and uses it for model training to obtain a first model and / or a second model.

[0068] In some embodiments, the main function module in the terminal device can obtain a trained first model for version upgrade decision-making from the version server, wherein the first model is used by the main function module to make version upgrade decisions for multiple function modules in the terminal device.

[0069] In some embodiments, the version upgrade decision can be understood as determining whether the version of the functional module deployed on the terminal device needs to be upgraded, and the time and / or location of the version upgrade can be determined based on historical parameters and current parameters of the terminal device.

[0070] In some embodiments, the version rollback decision can be understood as a judgment on whether the version of the functional module deployed on the terminal device needs to be rolled back. It can be based on the operating status of the current version of the functional module, or it can be based on the operating status of the upgraded version. The version with better operating status in the historical version is used as the target version for rollback.

[0071] In some embodiments, the main function module in the terminal device can obtain a trained second model for version rollback decision-making from the version server, wherein the second model is used by the main function module to make version rollback decisions for multiple function modules in the terminal device.

[0072] In some embodiments, the terminal device may be a warehouse robot, and the main function module may be used to manage and control multiple sub-function modules, and collect and report version information of the multiple sub-function modules for use in model training on the server.

[0073] For example, the main function module in the robot device obtains trained AI models from the network side, including a model for version upgrade decisions and a model for version rollback decisions.

[0074] Step 202: Use the first model to make a version upgrade decision for the first functional module, and / or use the second model to make a version rollback decision for the first functional module.

[0075] In some embodiments, the first functional module is a main functional module and any one of a plurality of sub-functional modules of the terminal device.

[0076] In some embodiments, the main function module of the terminal device uses the first model to make version upgrade decisions on multiple function modules deployed on the terminal device to determine whether any of the function modules needs to be upgraded, that is, to upgrade the current version of the function module to the target version, thereby realizing autonomous diagnosis, autonomous updating and differentiated updating of the device.

[0077] In some embodiments, the main function module of the terminal device uses the second model to make version rollback decisions on multiple function modules deployed on the terminal device to determine whether any of the function modules needs to be rolled back, that is, after determining that the version of the function module is upgraded, its operating effect is poor or there is a fault. When the rollback conditions are met, the upgraded version of the function module is rolled back to the target historical version, thereby achieving better operating results for the function module.

[0078] For example, the main function module in the robot device uses a model for version upgrade decisions and / or a model for version rollback decisions to make decisions on multiple function modules on the robot device to determine whether version upgrades and / or version rollbacks are required for any one or more of the multiple function modules.

[0079] For example, the transport robot receives the trained model from the server and deploys the model in the main function module of the robot. After the active energy module collects the current parameters of the camera module, PNC module, clock module, battery control module and other modules, the trained model determines whether the current version needs to be upgraded or rolled back to make the operation of the terminal device more stable. Specifically, if the current camera module fails and there is an upgradeable version, the camera module can be upgraded. If the upgraded camera module still has an abnormality or an installation abnormality occurs during the installation of the upgrade package, the camera module version can be rolled back to a historical version with good operation status.

[0080] In some embodiments, the main function module of the terminal device sends feedback data to the version server, and the feedback data includes at least one of the current parameters of the first function module, the upgrade decision result, and the rollback decision result. The feedback data is used by the version server to send an upgrade installation package and / or a rollback installation package to the terminal device, and / or the feedback data is used to update the first model and / or the second model.

[0081] In some embodiments, the terminal device sends feedback data to the version server, and the feedback data may include an upgrade decision result, so that the version server determines the version of the installation package that needs to be sent to the terminal, thereby sending the upgraded installation package to the terminal.

[0082] In some embodiments, the terminal device sends feedback data to the version server, and the feedback data may include a rollback decision result, so that the version server determines the version of the installation package that needs to be sent to the terminal, and thus sends the rollback installation package to the terminal.

[0083] In the above embodiment, through the feedback data sent by the terminal to the version server, the version server can determine the upgrade package version that needs to be provided to the terminal, so that the terminal can complete the version upgrade or rollback.

[0084] In some embodiments, the terminal device sends the upgrade decision result and current parameters obtained by using the first model to make a version upgrade decision to the version server, and the version server can use the upgrade decision result and current parameters to update the first model so that the updated first model can obtain a more accurate result when making a version upgrade decision for the first functional module. The upgrade decision result includes the upgraded installation package version and / or the upgraded time and / or the upgraded location.

[0085] In some embodiments, the terminal device sends the rollback decision result and current parameters obtained by using the second model to the version server, and the version server can use the rollback decision result and current parameters to update the second model, so that the updated second model can obtain more accurate results when making a version rollback decision on the first functional module. The rollback decision result includes a rollback indication and a rolled-back installation package version.

[0086] In the above embodiment, the feedback data sent by the terminal to the version server enables the version server to update the model, so that the updated model can obtain more reliable and accurate results during the upgrade decision and / or rollback decision process.

[0087] In the above embodiment, the terminal device obtains a trained model and uses the model to make version upgrade decisions and / or version rollback decisions on multiple functional modules in the terminal device to achieve autonomous diagnosis, autonomous updating, and differentiated updating of the terminal device, wherein the version upgrade decision and version rollback decision are parallel decision schemes, which can achieve version rollback of the functional module when the effect of the functional module is poor or there is a fault after the version upgrade, so that the functional module can be rolled back to a version with better operating effect.

[0088] Figure 3 This is a flow chart of making a version upgrade decision for the first functional module proposed in the embodiment of the present disclosure. Figure 2 The embodiment shown, Figure 3 right Figure 2 Step 202 in the embodiment is further described as follows: Figure 3 As shown, the following steps are included:

[0089] Step 301: Obtain current parameters of the first functional module.

[0090] In some embodiments, the current parameters include at least one of a state parameter, a performance parameter, a version parameter, and a hardware parameter, wherein the state parameter includes at least one of an abnormal state, a device state, and a task state, wherein the abnormal state is used to identify whether the first function module is abnormal, the device state is used to identify whether the first function module is in an idle or busy state, and the task state is used to identify the progress state, success state, or failure state of the task executed by the first function module; the performance parameters include at least one of network performance, storage performance, processing performance, and idle interval performance, wherein the network performance is used to identify the performance of the first communication connection between the main function module and the plurality of sub-function modules, and / or the performance of the second communication connection between the main function module and the version server, the storage performance is used to identify the remaining storage space size of the first function module, and the processing performance is used to identify the processing capacity of the first function module; the idle performance is used to identify the idle time interval between the first function module executing the current task and the next task; the version parameters include the current version number of the first function module, the operating parameters of the first function module under the current version number, at least one previous version number of the first function module before the current version number, and at least one of the operating parameters of the first function module under at least one previous version number; the hardware parameters include at least one of the model and the factory of the first function module.

[0091] In some embodiments, performance parameters may be performance indicators of the device, such as CPU, memory, etc.; the device status in the status parameters may be the robot status, such as TASKING, IDLE; task status may be RUN NING, SUCCESS, FAILE, for example; network status may be STRONG, WEAK, for example; idle performance may be the idle time of the robot, which is obtained based on the statistics of the robot's historical operating status; abnormal status may be an abnormal node, sub-node or fault code on the robot device; version information may be the entire package and single package information of the robot, where the version information may reflect the version number, node, upgrade time of download and upgrade, and restart time.

[0092] In some embodiments, the current parameter is a parameter of any functional module in the terminal device at the current moment.

[0093] For example, sub-function modules such as the PNC module, the clock module, and the camera module send parameters such as the current version number, the task status, and the idle time to the main function module of the robot.

[0094] Step 302: Based on current parameters, use a first model to determine whether a first functional module needs to be upgraded.

[0095] In some embodiments, the first model is a model used to make version upgrade decisions, and is provided by the version server to the main function module on the terminal device.

[0096] In some embodiments, the main function module in the terminal device can use the first model to make version upgrade decisions for the main function module and multiple sub-function modules to determine whether any one or several of the function modules need to be upgraded.

[0097] In some embodiments, based on current parameters, using the first model to determine whether the first functional module needs to be upgraded includes: based on the current state parameters, current performance parameters, current version parameters and at least one of the hardware parameters of the first functional module, when the first model is used to judge that the operating status of the first functional module meets the upgrade conditions and / or there is currently an upgradeable version, determining that the first functional module needs to be upgraded.

[0098] In some embodiments, the main function module uses at least one of the current state parameters, current performance parameters, current version parameters and hardware parameters of any function module in the terminal device as input to the first model to determine that the function module needs to be upgraded.

[0099] For example, the main function module of the robot uses a version upgrade decision model to make a version upgrade decision on the camera module, by taking the current state, current performance, current version and hardware parameters of the camera module as inputs to the model. Since the current operating condition of the camera module is poor and there is currently an upgradeable version, the model output result is that the camera module needs to be upgraded.

[0100] In some embodiments, based on current parameters, using the first model to determine whether the first functional module needs to be upgraded includes: based on the current state parameters, current performance parameters, current version parameters and at least one of the hardware parameters of the first functional module, when it is judged by the first model that the operating status of the first functional module does not meet the upgrade conditions or there is currently no upgradeable version, determining that the first functional module does not need to be upgraded.

[0101] For example, the positioning module in the robot is currently running well and has no faults. After the main function module uses the model to make a version upgrade decision for the positioning module, it is determined that the positioning module does not need to be upgraded.

[0102] For example, since the current operating status of the positioning module is poor, but there is no upgradeable version, after making a decision on the positioning module using the version upgrade decision model, it can be concluded that the positioning module does not need to be upgraded.

[0103] Step 303: When it is determined that the first functional module needs to be upgraded, at least one of an upgraded version, an upgraded time, and an upgraded location of the first functional module is determined using the first model based on current parameters.

[0104] In the above embodiment, when the first functional module needs to be upgraded, that is, the current operating status of the first functional module meets the upgrade conditions and / or there is an upgradeable version, the first model can output the upgraded version of the first functional module for upgrading based on the current parameters, as well as at least one of the upgrade time and upgrade location. That is, the first model can determine the upgrade decision based on the current parameters, that is, the first functional module is best upgraded during idle time or at location A.

[0105] For example, the camera module will be in a running state without any action for three minutes during the process of executing task A to task B. When the first model determines that the camera module needs to be upgraded, it can also determine that the camera module can be upgraded within this time period.

[0106] In some embodiments, the positioning module is in an idle state when the robot is in a charging state. When the first model determines that the positioning module needs to be upgraded, it can also determine that the module can be upgraded when it is within the location range of the charging pile.

[0107] In the above embodiment, the trained model can make version upgrade decisions for multiple functional modules in the terminal device, and determine whether an upgrade is required, as well as at least one of the version number, upgrade time, and upgrade location required, based on the current status, performance, and version and hardware conditions of the functional modules. This achieves the purpose of autonomously performing differentiated diagnosis and differentiated updates on functional modules by the terminal device.

[0108] Figure 4 The following is a flow chart of the device upgrade method proposed in the embodiment of the present disclosure. Figure 2 , Figure 3 The embodiment shown, Figure 4 right Figure 3 Step 303 in is further described, such as Figure 4 As shown, the following steps are included:

[0109] Step 401: Based on the current version parameter of the first functional module and at least one of the hardware parameters, determine the upgraded version of the first functional module through a first model.

[0110] In some embodiments, the main function module uses the current version number and device model of the first function module as input of the first model, and the model output result is the upgraded version of the first function module, that is, the version number of the upgrade package.

[0111] For example, the main function module on the robot collects the current version number of the camera module, the operating parameters under the current version number, the model and at least one of the factories, and uses it as the input of the version upgrade decision model to obtain the model output result, namely the upgraded version number of the camera module.

[0112] Step 402: Based on at least one of the current state parameter and the current performance parameter of the first functional module, determine the upgrade time and / or upgrade location of the first functional module through the first model.

[0113] In some embodiments, the main function module uses at least one of the current state parameters and current performance parameters of the first function module as input to the first model, and the model output result is an upgrade decision, that is, at least one of the time and location of the upgrade of the first function module.

[0114] For example, the main function module on the robot collects at least one of the current state parameters and current performance parameters of the camera module, and uses it as the input of the version upgrade decision model. The model output result can be obtained, that is, at least one of the time and location of the version upgrade. For example, the upgrade time is the interval time between executing task A and task B, which is 3 minutes, and the upgrade location is the intersection between the aisle and the shelf.

[0115] In the above embodiment, the trained model can not only judge whether the functional module currently needs to be upgraded, but also determine the upgrade decision based on the parameter information of the current functional module, including the upgrade time and / or upgrade location, for version upgrade of the first functional module, to achieve intelligent and efficient optimization and update, and avoid the version upgrade affecting the normal operation of the functional module.

[0116] Figure 5 A schematic diagram of a device upgrade method proposed in an embodiment of the present disclosure. Figure 2-Figure 4 The embodiment shown, as Figure 5 As shown, the method also includes the following steps:

[0117] Step 501: Send current parameters and upgrade decision results to the version server.

[0118] In some embodiments, the upgrade decision result includes at least one of the indication information that the first functional module needs to be upgraded, the upgraded version of the first functional module, the upgrade time, and the upgrade location.

[0119] For example, the main function module of the robot makes a version upgrade decision on the camera module through the first model, determines the upgrade package of the camera module, and the upgrade decision, including the operation that can perform the version upgrade within the 3-minute interval between the execution of task A and task B, and sends the version number of the corresponding upgrade package and the upgrade time information to the version server.

[0120] Step 502: Receive an upgrade package corresponding to the upgraded version sent by the version server.

[0121] In some embodiments, the version server sends an upgrade package corresponding to the upgraded version to the first functional module based on the received upgraded version of the first functional module and the upgrade decision result, so that the first functional module can be used for version upgrade.

[0122] For example, the version server sends the upgrade package of the camera module to the main function module in the robot, and the main function module is responsible for managing and controlling the version upgrade of the camera module.

[0123] Step 503: When the upgrade time and / or upgrade location are reached, install the upgrade package.

[0124] In some embodiments, the main function module executes a version upgrade of the first function module based on the decision result of the first model when the upgrade time and / or upgrade location are met.

[0125] For example, the camera module performs a version upgrade 3 minutes after executing Task A and Task B; the positioning module performs a version upgrade when it reaches the location of the charging pile.

[0126] In the above embodiment, the terminal device makes version upgrade decisions for multiple functional modules deployed thereon based on the model provided by the server. Based on the decision results, the version server provides an upgrade package for the corresponding upgraded version, and performs version upgrades on the functional modules according to the upgrade time and / or upgrade location in the decision results. This can achieve intelligent and efficient version upgrades and avoid the upgrade process affecting the normal operation of the functional modules.

[0127] Figure 6 A schematic diagram of a device upgrade method proposed in an embodiment of the present disclosure, based on Figure 2-Figure 5 The embodiment shown, as Figure 6 As shown, the method also includes the following steps:

[0128] Step 601, automatically back up the current parameters and upgrade decision results.

[0129] In some embodiments, automatic backup of current parameters and upgrade decision results can be achieved by the main function module of the terminal device saving the current parameters collected during the version upgrade decision process and the upgrade decision results obtained after the version upgrade decision, so as to report backup data to the server, continuously optimize and update the first model and / or the second model, so that the trained model is more suitable for the current operating environment and terminal device.

[0130] For example, the terminal device automatically backs up important data, such as version parameters, task parameters, status parameters, etc., during the version upgrade of the camera module or the positioning module.

[0131] Step 602: monitor the installation process of the upgrade package to determine whether an installation anomaly occurs in the upgrade package.

[0132] In some embodiments, the installation anomaly includes at least one of an installation interruption, an installation freeze, or an installation error.

[0133] In some embodiments, the main function module may monitor the installation process of the upgrade package, monitor whether the installation is successful, and monitor whether an abnormality occurs during the installation process to determine whether the version of the first function module is successfully upgraded.

[0134] In some embodiments, the main function module monitors the installation process of the upgrade package to collect potential errors in the first function module when executing the version upgrade, such as installation interruption caused by power outage, or installation jam or installation error caused by poor communication conditions.

[0135] For example, during the installation of the camera module upgrade package, the terminal device is low on battery, resulting in a power outage and the camera module installation process is interrupted, causing the camera module version upgrade to fail. The main function module determines that an installation abnormality has occurred in the camera module upgrade package by monitoring the installation process.

[0136] For example, when multiple functional modules are performing version upgrades simultaneously, the main functional module may monitor the process of installing the upgrade packages of the multiple modules simultaneously, and determine whether an installation abnormality occurs in each functional module.

[0137] Step 603: When an installation exception occurs in the upgrade package, determine and record the cause of the installation exception.

[0138] In some embodiments, determining the cause of the installation exception may be determining a potential error in a functional module in the terminal device during a version upgrade process, so as to record the cause of the installation exception.

[0139] In some embodiments, the cause of the installation anomaly may be a power outage or poor device communication conditions.

[0140] For example, when an installation exception occurs in the upgrade package of the camera module, the main function module determines that the terminal device is low on power, which causes the version upgrade to fail, while monitoring the installation process. The main function module stores the reason for the camera module upgrade failure.

[0141] For example, when multiple functional modules are upgraded simultaneously and there is an installation anomaly among them, the main functional module may record the functional module in which the installation anomaly occurs, including the cause of the anomaly and the corresponding functional module.

[0142] In the above embodiment, by automatically backing up the current parameters and upgrade decision results and monitoring the upgrade process, the causes of abnormal situations occurring during the upgrade process are recorded for subsequent reporting of backup data and records to the server, which is used for the server to train the first model and / or the second model to continuously optimize the model.

[0143] Figure 7 A schematic diagram of a process for making a version rollback decision for a first functional module proposed in an embodiment of the present disclosure. Figure 7 based on Figure 2-Figure 6 The embodiment shown, Figure 2 Step 202 in the embodiment is further described as follows: Figure 7 As shown, the following steps are included:

[0144] Step 701, obtaining current parameters of the first functional module.

[0145] In some embodiments, the current parameter may be a current status parameter, wherein the status parameter includes at least one of an abnormal status, a device status, and a task status, wherein the abnormal status is used to identify whether the first functional module is abnormal, the device status is used to identify whether the first functional module is in an idle or busy state, and the task status is used to identify the progress status, success status, or failure status of the task executed by the first functional module.

[0146] In some embodiments, the current parameters may also include version parameters and performance parameters, wherein the version parameters include at least one of the current version number of the first functional module, the operating parameters of the first functional module under the current version number, at least one previous version number of the first functional module before the current version number, and the operating parameters of the first functional module under at least one previous version number.

[0147] For example, the robot main function module obtains the current parameters of the camera module, including the current version number, current performance parameters, whether there is an abnormality, whether it is in an idle state or a busy state, and the status of the current task execution. For example, the camera module currently has an installation abnormality of the upgrade package version and is currently in an idle state, and the version number is version 6.

[0148] Step 702: Based on the current parameters, use the second model to determine whether the first functional module meets the rollback condition.

[0149] In some embodiments, based on current parameters, using a second model to determine whether the first functional module reaches a rollback condition includes: based on current state parameters of the first functional module, determining a fault level of the first functional module through a second model, and determining that the first functional module reaches a rollback condition when the fault level is higher than a preset level threshold or when the fault level is an installation abnormality.

[0150] In some embodiments, the preset level threshold may be a critical value of a preset fault level, that is, when the threshold is reached, it indicates that the current version of the functional module is in poor operating condition and needs to be rolled back. Specifically, the server may preset multiple fault types corresponding to the fault levels. When a fault occurs during monitoring or a fault occurs during operation of the terminal device, the second model may determine the fault level based on the fault type and compare the fault level with the preset level threshold. That is, when the fault level reaches the preset level threshold, it may be determined that the current version of the corresponding functional module is in poor operating condition and needs to be rolled back.

[0151] In some embodiments, the second model can determine whether the first functional module needs to be rolled back to a historical version based on the status parameters of the first functional module. For example, the current status of the first functional module is abnormal and the fault level is higher than a preset level threshold, thereby determining that the first functional module needs to roll back the version.

[0152] In some embodiments, an installation exception occurs in the first functional module during the version upgrade process, and the second model determines that the first functional module needs to perform a version rollback based on this.

[0153] For example, an installation exception occurs in the process of executing version upgrade of the camera module, and the version rollback decision model determines that the camera module needs to be rolled back based on this.

[0154] In some embodiments, based on current parameters, using a second model to determine whether the first functional module reaches the rollback condition also includes: based on the current state parameters of the first functional module, determining the fault level of the first functional module through the second model, and when the fault level is lower than a preset level threshold or when no fault occurs, determining that the first functional module does not reach the rollback condition.

[0155] In some embodiments, the abnormal state in the current state parameter of the first functional module is non-abnormal, that is, the first functional module does not fail during the version upgrade process, then the second model can determine that the first functional module does not need to perform version rollback based on this.

[0156] Step 703: When it is determined that the rollback condition is met, a target historical version before the current version is determined using the second model based on the current parameters.

[0157] In some embodiments, the target historical version is a historical version of the first functional module whose operating status meets preset conditions among multiple historical versions before the first functional module is upgraded to the current version.

[0158] In some embodiments, based on current parameters, using a second model to determine a target historical version before the current version includes: based on the current version parameters of the first functional module, determining a rollback historical version corresponding to the fault level through the second model, and determining from the rollback historical versions a historical version whose operating status meets preset conditions as the target historical version.

[0159] In some embodiments, the second model is based on the current version number of the first functional module, at least one previous version number before the current version, and the corresponding operating parameters. Among the at least one previous version number, the previous version number whose operating parameters show that the operating status meets the preset conditions is determined as the target historical version, wherein the operating parameters showing that the operating status meets the preset conditions can indicate that under this version, the operating status of the first functional module is better than the operating status under the current version, and its version can be rolled back to this version to achieve stable operation of the device.

[0160] For example, the version rollback decision model deployed on the main function module of the robot can determine the priority of the version based on historical version information and parameters of the historical version running status, and rank the version numbers with good running status as the first to be rolled back. When the sub-function module meets the rollback conditions, the top-ranked historical version will be determined as the rolled back version.

[0161] For example, the version rollback decision model can also use the historical version whose operating status meets the preset conditions as the rollback version. The preset conditions can be the conditions that need to be met when performing different tasks based on historical parameters. For example, when the camera module is performing a handling task in the robot, it needs to first meet the requirement of accurate obstacle identification. Therefore, in the historical version, the version that can accurately identify obstacles will be determined as the target historical version that meets the preset conditions; or when the camera module is performing a sorting task in the robot, it needs to first meet the requirement of accurate QR code positioning. Therefore, in the historical version, the version that can accurately locate the QR code on the shelf, cargo box or landmark will be determined as the target historical version that meets the preset conditions.

[0162] In the above embodiment, the second model is used to make version rollback decisions for the first functional module. By collecting the current parameters, it is determined whether there is an abnormal condition, and the version number in the historical version whose operating status meets the preset conditions is determined as the historical version to be rolled back to achieve stable operation of the device.

[0163] Figure 8 A schematic diagram of a process for making a version rollback decision for a first functional module proposed in an embodiment of the present disclosure. Figure 8 based on Figure 7 The embodiment shown, as Figure 8 As shown, the method also includes the following steps:

[0164] Step 801: Send current parameters and rollback decision results to the version server.

[0165] In some embodiments, the rollback decision result includes indication information that the first functional module needs to be rolled back and a target historical version.

[0166] In some embodiments, the main function module of the terminal device will send the decision result obtained by using the second model to make a version rollback decision on the first function module to the version server, so as to send a rollback instruction and version information of the target historical version, such as the version number, to the version server.

[0167] Step 802, when the first function module has uninstalled and deleted the upgrade package corresponding to the target historical version, receive the upgrade package corresponding to the target historical version sent by the version server and install the upgrade package; or, when the first function module has not deleted the upgrade package corresponding to the target historical version, install the upgrade package.

[0168] In some embodiments, the first functional module will periodically clean up the installed upgrade packages, uninstall and delete the corresponding upgrade packages to avoid occupying memory. When the first functional module has deleted the upgrade package corresponding to the target historical version, the first functional module needs to roll back the version. It is necessary to receive the upgrade package corresponding to the target historical version sent to it by the version server based on the rollback decision result sent by it, and perform version rollback after receiving the upgrade package and install the upgrade package.

[0169] In some embodiments, when the first functional module has not reached the cleanup cycle and has not deleted the upgrade package corresponding to the target historical version, the first functional module can install the corresponding upgrade package based on the rollback decision result of the second model to achieve version rollback.

[0170] In the above embodiment, the first functional module can complete the version rollback based on the rollback decision result of the second model by sending the decision result to the version server to request the version server to provide the upgrade package corresponding to the target historical version and install the upgrade package, or it can directly install the corresponding upgrade package in the uncleaned memory to achieve version rollback and make the device run stably.

[0171] Fig. 9 A schematic diagram of a device upgrade method according to an embodiment of the present disclosure. Fig. 9 based on Figure 2-Figure 8 The embodiment shown, as Fig. 9 As shown, the method also includes the following steps:

[0172] Step 901, obtaining historical parameters of the first functional module.

[0173] In some embodiments, historical parameters may be parameter information collected by the first functional module during the version upgrade and / or version rollback process based on the version upgrade decision results and / or version rollback decision results, including data automatically backed up during the version upgrade process and abnormal data in the installation process.

[0174] In some embodiments, the main function module in the terminal device can collect historical parameters of multiple function modules on the terminal device side, including but not limited to its own historical parameters and historical parameters of function modules that have executed or not executed version upgrades and / or version rollbacks.

[0175] In some embodiments, the historical parameters may include Figure 2-Figure 8 In the illustrated embodiment, the main function module collects the current parameters of the first function module.

[0176] In some embodiments, the historical parameters include at least one of a state parameter, a performance parameter, a version parameter, and a hardware parameter, wherein the state parameter includes at least one of an abnormal state, a device state, and a task state, wherein the abnormal state is used to identify whether the first function module is abnormal, the device state is used to identify whether the first function module is in an idle or busy state, and the task state is used to identify the progress state, success state, or failure state of the task executed by the first function module; the performance parameters include at least one of network performance, storage performance, processing performance, and idle interval performance, wherein the network performance is used to identify the performance of the first communication connection between the main function module and the plurality of sub-function modules, and / or the performance of the second communication connection between the main function module and the version server, the storage performance is used to identify the remaining storage space size of the first function module, and the processing performance is used to identify the processing capacity of the first function module; the idle performance is used to identify the idle time interval between the first function module executing the current task and the next task; the version parameters include at least one of the current version number of the first function module, the operating parameters of the first function module under the current version number, at least one previous version number of the first function module before the current version number, and the operating parameters of the first function module under at least one previous version number; the hardware parameters include at least one of the model and the factory of the first function module.

[0177] Step 902: Send the historical parameters to the version server.

[0178] In some embodiments, historical parameters are used for training data so that the version server performs model training to obtain the first model and / or the second model.

[0179] In some embodiments, the version server trains the initial model based on the historical parameters sent by the terminal device to obtain a first model for version upgrade decisions and / or a second model for version rollback decisions.

[0180] In the above embodiment, the terminal device can send the collected historical parameters on the terminal device side to the version server for model training to obtain a model for version upgrade decisions and / or for version rollback decisions. The version server provides the model to the terminal device side for the terminal device to perform autonomous diagnosis, autonomous updating, and autonomous optimization.

[0181] Fig.10 A schematic diagram of the process flow of the device upgrade method proposed in the embodiment of the present disclosure, which is applied to the version server, such as Fig.10 As shown, the following steps are included:

[0182] Step 1001: Perform model training on the first initial model to obtain a first model, and / or perform model training on the second initial model to obtain a second model.

[0183] In some embodiments, the architecture diagram of the first initial model is as follows: Fig. 12C As shown, including data preprocessing and intelligent upgrade decision network, the first initial model is trained based on historical parameter information to obtain a first model for equipment upgrade decision.

[0184] In some embodiments, the architecture diagram of the second initial model is as follows: Fig.12D As shown, it includes a robot state relationship diagram and an intelligent diagnosis rollback network, wherein the intelligent diagnosis rollback network includes a gated convolutional network, a graph convolutional network, batch normalization, a pooling layer, and a fully connected layer. The version server trains the second initial model based on historical parameter information to obtain a second model for device rollback decision-making.

[0185] Step 1002: Send the first model and / or the second model to the main function module of the terminal device.

[0186] In some embodiments, the first model is used by the main function module to make version upgrade decisions on the first function module, and the second model is used by the main function module to make version rollback decisions on the first function module. The first function module is the main function module of the terminal device and any one of multiple sub-function modules.

[0187] In some embodiments, the method in which the main function module of the terminal device uses the first model to make a version upgrade decision can be seen in Figures 2 to 5 Implementation in the illustrated embodiment.

[0188] In some embodiments, the version server receives the current parameters of the first function module and the upgrade decision result sent by the main function module of the terminal device; and sends an upgrade package corresponding to the upgraded version to the main function module.

[0189] Among them, the upgrade decision result is obtained by the main functional module based on current parameters and using the first model, and the upgrade decision result includes at least one of the indication information that the first functional module needs to be upgraded, the upgraded version of the first functional module, the upgrade time, and the upgrade location.

[0190] In some embodiments, the method in which the main function module of the terminal device uses the second model to make a version rollback decision can be seen in Figures 6 to 9 Implementation in the illustrated embodiment.

[0191] In some embodiments, the version server receives the current parameters of the first function module and the rollback decision result sent by the main function module of the terminal device; and sends an upgrade package corresponding to the target historical version to the main function module.

[0192] The rollback decision result is obtained by the main function module based on current parameters and using the second model, and the rollback decision result includes indication information that the first function module needs to be rolled back and the target historical version.

[0193] In the above embodiment, the version server provides the trained first model and / or second model to the main function module of the terminal device, so that the main function module can make version upgrade decisions and / or version rollback decisions for multiple function modules on the terminal device side, thereby achieving the purpose of autonomous diagnosis, autonomous updating, and autonomous optimization on the terminal side.

[0194] Fig.11 A schematic diagram of a device upgrade method according to an embodiment of the present disclosure, based on Fig.10 The embodiment shown, as Fig.11 As shown, the method also includes the following steps:

[0195] Step 1101: receiving historical parameters of a first function module sent by a main function module of a terminal device, and receiving an upgrade version number and an upgrade package for upgrading the first function module sent by a compilation server.

[0196] In some embodiments, the historical parameters include at least one of the following: state parameters, performance parameters, version parameters, and hardware parameters, wherein the state parameters include at least one of the following: abnormal state, device state, and task state, wherein the abnormal state is used to identify whether the first function module is abnormal, the device state is used to identify whether the first function module is in an idle or busy state, and the task state is used to identify the progress state, success state, or failure state of the task executed by the first function module; the performance parameters include at least one of the following: network performance, storage performance, processing performance, and idle interval performance, wherein the network performance is used to identify the performance of the first communication connection between the main function module and the plurality of sub-function modules, and / or the performance of the second communication connection between the main function module and the version server, the storage performance is used to identify the remaining storage space size of the first function module, and the processing performance is used to identify the processing capacity of the first function module; the idle performance is used to identify the idle time interval between the first function module executing the current task and the next task; the version parameters include at least one of the current version number of the first function module, the operating parameters of the first function module under the current version number, at least one previous version number of the first function module before the current version number, and the operating parameters of the first function module under at least one previous version number; the hardware parameters include at least one of the model and the factory of the first function module.

[0197] In some embodiments, the historical parameters sent by the main function module of the terminal device may be those of the first function module, and the first function module may be the main function module and at least one of multiple sub-function modules, so that the version server can use the historical parameters to train the initial model.

[0198] In some embodiments, the upgrade version number and upgrade package for upgrading the first functional module sent by the compilation server can be used by the version server to train the initial model.

[0199] Step 1102, using at least one of the historical parameters and the upgrade version number as training data, performing model training on the first initial model to obtain a first model, and / or performing model training on the second initial model to obtain a second model.

[0200] In some embodiments, model training is performed on the first initial model to obtain the first model, including: using at least one of the historical state parameters, historical performance parameters, historical version parameters, hardware parameters, and upgrade version number of the first functional module as training data, using a preset upgrade result label to train the first initial model, and when the number of training iterations reaches a first preset number or the training error is less than or equal to a first preset error, the training is terminated to obtain the first model.

[0201] In some embodiments, the first model obtained after training is used for Figure 2-Figure 8In the illustrated embodiment, a version upgrade decision is made for the first functional module. Based on current parameters of the first functional module, it is determined whether an upgrade is required and an upgrade decision is made. The upgrade decision includes the upgrade time and / or upgrade location.

[0202] In some embodiments, model training is performed on the second initial model to obtain the second model, including: using the historical state parameters and historical version parameters of the first functional module as training data, using a preset rollback result label, training the second initial model, and when the number of training iterations reaches a second preset number or the training error is less than or equal to the second preset error, ending the training and obtaining the second model.

[0203] In some embodiments, the trained second model is used to Figure 2-Figure 8 In the illustrated embodiment, a version rollback decision is made for the first functional module, and based on the current parameters of the first functional module, it is determined whether a version rollback is required and the target historical version to be rolled back.

[0204] In summary, the device upgrade method provided by the present invention is that the version server trains the initial model based on the historical parameters uploaded by the terminal, obtains the first model for version upgrade decision and / or the second model for version rollback decision, and sends the first model and / or the second model to the main function module of the terminal device, and the terminal side makes autonomous version upgrade decisions and version rollback decisions, thereby realizing autonomous and differentiated device upgrades, as well as efficient resource allocation and intelligent optimization.

[0205] The following is a specific implementation of a device upgrade method provided by the present disclosure:

[0206] Fig. 12A The schematic diagram of the device upgrade method provided by the embodiment of the present disclosure is as follows: Fig. 12A As shown, the main device is communicatively connected with the base station. The main device and multiple sub-devices are functional modules. Through data collection and self-diagnosis, the modules or functions that need to be upgraded are determined, and the corresponding upgrade packages are downloaded to upgrade the modules. The base station performs autonomous learning based on the data uploaded by the main device, diagnostic information, and data automatically uploaded by the compilation server.

[0207] Fig. 12B A flow chart of a device upgrade method provided in an embodiment of the present disclosure, such as Fig. 12B As shown, the method comprises the following steps:

[0208] Step 1: Monitoring and decision making:

[0209] The AI ​​model decides whether to start an OTA upgrade based on data such as the device's current status, historical upgrade records, and device performance indicators; it monitors device status and performance indicators in real time.

[0210] Fig. 12C This is the intelligent decision-making flowchart of the AI ​​model, such as Fig. 12C As shown in the figure, the original data collected includes remote basic data, robot status data, and user habit data. The original data is preprocessed and spliced ​​before being sent to the OTA intelligent upgrade decision network. Feature extraction and dimension reduction are performed through the linear layer, and after normalization through the Softw ax layer, it is determined whether the OTA is upgraded and the upgraded version package, and is automatically pushed to the corresponding device.

[0211] Step 2: Upgrade preparation:

[0212] Automatically select upgrade packages: The decision system automatically selects the appropriate upgrade package based on the device model, version, and other parameters;

[0213] Upgrade strategy: AI models can predict the best time and location for upgrades based on device usage patterns.

[0214] Step 3: Upgrade execution:

[0215] Download upgrade package: automatically download the upgrade package from the server;

[0216] Data backup: automatically back up important data;

[0217] Install Upgrades: Automatically install upgrade packages and monitor the progress and any potential errors during the installation.

[0218] Step 4: Verification and rollback:

[0219] Automatic rollback: If a problem is detected, the system can automatically roll back to the previously available version to ensure stable operation of the device.

[0220] like Fig.12D Flowchart updated for diagnostic anomalies, such as Fig.12D As shown, the robot state data is spliced ​​into a robot state relationship diagram, and the fault classification is obtained through the intelligent diagnosis rollback network, and the corresponding version rollback is determined.

[0221] The intelligent diagnosis rollback network consists of a gated convolutional network, a graph convolutional network, batch normalization, a pooling layer, and a fully connected layer. The gated convolutional network dynamically adjusts neighborhood information, the graph convolutional network extracts and fuses features, the pooling layer can average features and fix vector length, and the fully connected layer performs classification.

[0222] Step 5: Learning and Optimization:

[0223] Continuous optimization: Continuously optimize AI decision-making models and upgrade processes based on feedback data.

[0224] The device upgrade method provided by the above-mentioned embodiment of the present disclosure can realize personalized upgrade experience; accurate fault diagnosis and repair; intelligent optimization and update; efficient resource allocation; and realize completely unmanned and rapid iterative update.

[0225] Fig.13 This is a schematic diagram of the structure of a device upgrade apparatus 1300 proposed in an embodiment of the present disclosure, which is applied to a main functional module in a terminal device, such as Fig.13 As shown, the device comprises:

[0226] The acquisition unit 1301 is used to obtain the first model and / or the second model from the version server. The first model is used by the terminal device to make version upgrade decisions, and the second model is used by the terminal device to make version rollback decisions. The first model and / or the second model are trained by the version server.

[0227] The decision unit 1302 is used to make a version upgrade decision for the first functional module using the first model, and / or to make a version rollback decision for the first functional module using the second model. The first functional module is the main functional module of the terminal device and any one of multiple sub-functional modules.

[0228] In some embodiments, the decision unit is also used to obtain current parameters of the first functional module; based on the current parameters, use the first model to determine whether the first functional module needs to be upgraded; when it is determined that the first functional module needs to be upgraded, based on the current parameters, use the first model to determine at least one of the upgrade version, upgrade time, and upgrade location of the first functional module.

[0229] In some embodiments, the decision unit is also used to determine that the first functional module needs to be upgraded based on at least one of the current state parameters, current performance parameters, current version parameters and hardware parameters of the first functional module, when it is judged through the first model that the operating status of the first functional module meets the upgrade conditions and / or there is currently an upgradeable version.

[0230] In some embodiments, the decision unit is also used to determine the upgraded version of the first functional module through the first model based on the current version parameters of the first functional module and at least one of the hardware parameters; and to determine the upgrade time and / or upgrade location of the first functional module through the first model based on the current state parameters of the first functional module and at least one of the current performance parameters.

[0231] In some embodiments, the acquisition unit is also used to send current parameters and upgrade decision results to the version server, the upgrade decision results including indication information that the first functional module needs to be upgraded, the upgraded version of the first functional module, at least one of the upgrade time and upgrade location; receive the upgrade package corresponding to the upgraded version sent by the version server; and install the upgrade package when the upgrade time and upgrade location are reached.

[0232] In some embodiments, the decision unit is also used to automatically back up current parameters and upgrade decision results; monitor the installation process of the upgrade package to determine whether an installation exception occurs in the upgrade package, and the installation exception includes at least one of installation interruption, installation freeze or installation error; in the event that an installation exception occurs in the upgrade package, determine and record the cause of the installation exception.

[0233] In some embodiments, the decision unit is also used to obtain current parameters of the first functional module; based on the current parameters, use the second model to determine whether the first functional module meets the rollback conditions; when it is determined that the rollback conditions are met, based on the current parameters, use the second model to determine the target historical version before the current version, the target historical version being a historical version of the first functional module among multiple historical versions before being upgraded to the current version, whose operating status meets the preset conditions.

[0234] In some embodiments, the decision unit is also used to determine the fault level of the first functional module through the second model based on the current state parameters of the first functional module, and determine that the first functional module meets the rollback condition when the fault level is higher than a preset level threshold or when the fault level is an installation abnormality; wherein, based on the current parameters, using the second model to determine the target historical version before the current version includes: based on the current version parameters of the first functional module, determining the rollback historical version corresponding to the fault level through the second model, and determining from the rollback historical versions a historical version whose operating status meets the preset conditions as the target historical version.

[0235] In some embodiments, the acquisition unit is also used to send current parameters and rollback decision results to the version server, where the rollback decision results include indication information that the first functional module needs to be rolled back and the target historical version; when the first functional module has uninstalled and deleted the upgrade package corresponding to the target historical version, receiving the upgrade package corresponding to the target historical version sent by the version server and installing the upgrade package; or, when the first functional module has not deleted the upgrade package corresponding to the target historical version, installing the upgrade package.

[0236] In some embodiments, the acquisition unit is further used to acquire historical parameters of the first functional module; send the historical parameters to the version server, and use the historical parameters as training data so that the version server performs model training to obtain the first model and / or the second model.

[0237] In some embodiments, the current parameters and / or historical parameters include at least one of a state parameter, a performance parameter, a version parameter, and a hardware parameter, wherein the state parameter includes at least one of an abnormal state, a device state, and a task state, wherein the abnormal state is used to identify whether the first functional module is abnormal, the device state is used to identify whether the first functional module is in an idle or busy state, and the task state is used to identify the progress state, success state, or failure state of the task performed by the first functional module; the performance parameter includes at least one of a network performance, a storage performance, a processing performance, and an idle interval performance, wherein the network performance is used to identify the performance of the first communication connection between the main functional module and the plurality of sub-functional modules, and / or Or used to identify the performance of the second communication connection between the main function module and the version server, the storage performance is used to identify the remaining storage space size of the first function module, and the processing performance is used to identify the processing capacity of the first function module; the idle performance is used to identify the idle time interval between the first function module executing the current task and the next task; the version parameters include the current version number of the first function module, the operating parameters of the first function module under the current version number, at least one previous version number of the first function module before the current version number, and at least one of the operating parameters of the first function module under at least one previous version number; the hardware parameters include at least one of the model and factory of the first function module.

[0238] In some embodiments, the acquisition unit is also used to send feedback data to the version server, the feedback data includes at least one of the current parameters of the first functional module, the upgrade decision result, and the rollback decision result. The feedback data is used by the version server to send an upgrade installation package and / or a rollback installation package to the terminal device, and / or the feedback data is used to update the first model and / or the second model.

[0239] Fig.14 A schematic diagram of the structure of a device upgrade apparatus 1400 proposed in an embodiment of the present disclosure, which is applied to a version server, such as Fig.14 As shown, the device comprises:

[0240] The training unit 1401 is used to perform model training on the first initial model to obtain a first model, and / or to perform model training on the second initial model to obtain a second model.

[0241] Sending unit 1402 is used to send the first model and / or the second model to the main function module of the terminal device, the first model is used by the main function module to make version upgrade decisions on the first function module, and the second model is used by the main function module to make version rollback decisions on the first function module. The first function module is the main function module of the terminal device and any one of multiple sub-function modules.

[0242] In some embodiments, the sending unit is also used to receive current parameters and an upgrade decision result of the first function module sent by the main function module of the terminal device. The upgrade decision result is obtained by the main function module based on the current parameters and using the first model. The upgrade decision result includes at least one of the indication information that the first function module needs to be upgraded, the upgraded version of the first function module, the upgrade time, and the upgrade location; and send an upgrade package corresponding to the upgraded version to the main function module.

[0243] In some embodiments, the sending unit is also used to receive the current parameters of the first function module and the rollback decision result sent by the main function module of the terminal device. The rollback decision result is obtained by the main function module based on the current parameters and using the second model. The rollback decision result includes the indication information that the first function module needs to be rolled back and the target historical version; and the upgrade package corresponding to the target historical version is sent to the main function module.

[0244] In some embodiments, the sending unit is further used to receive the historical parameters of the first function module sent by the main function module of the terminal device, and receive the upgrade version number and upgrade package for the first function module to be upgraded sent by the compilation server.

[0245] In some embodiments, the training unit is also used to use at least one of the historical parameters and the upgrade version number as training data to perform model training on the first initial model to obtain the first model, and / or to perform model training on the second initial model to obtain the second model.

[0246] In some embodiments, the training unit is also used to use at least one of the historical state parameters, historical performance parameters, historical version parameters, hardware parameters, and upgrade version number of the first functional module as training data, and use a preset upgrade result label to train the first initial model. When the number of training iterations reaches a first preset number or the training error is less than or equal to a first preset error, the training is terminated and the first model is obtained.

[0247] In some embodiments, the training unit is also used to use the historical state parameters and historical version parameters of the first functional module as training data, and use a preset rollback result label to train the second initial model. When the number of training iterations reaches a second preset number or the training error is less than or equal to the second preset error, the training is terminated and the second model is obtained.

[0248] In some embodiments, the current parameters and / or historical parameters include at least one of a state parameter, a performance parameter, a version parameter, and a hardware parameter, wherein the state parameter includes at least one of an abnormal state, a device state, and a task state, wherein the abnormal state is used to identify whether the first functional module is abnormal, the device state is used to identify whether the first functional module is in an idle or busy state, and the task state is used to identify the progress state, success state, or failure state of the task performed by the first functional module; the performance parameter includes at least one of a network performance, a storage performance, a processing performance, and an idle interval performance, wherein the network performance is used to identify the performance of the first communication connection between the main functional module and the plurality of sub-functional modules, and / or Or used to identify the performance of the second communication connection between the main function module and the version server, the storage performance is used to identify the remaining storage space size of the first function module, and the processing performance is used to identify the processing capacity of the first function module; the idle performance is used to identify the idle time interval between the first function module executing the current task and the next task; the version parameters include the current version number of the first function module, the operating parameters of the first function module under the current version number, at least one previous version number of the first function module before the current version number, and at least one of the operating parameters of the first function module under at least one previous version number; the hardware parameters include at least one of the model and factory of the first function module.

[0249] Fig.15 1 is a schematic diagram of a structure of an electronic device 1500 for implementing the above-mentioned device upgrade method according to an exemplary embodiment. The components of the electronic device 1500 include but are not limited to a memory 1510 and a processor 1520. The processor 1520 is connected to the memory 1510 via a bus 1530, and the database 1550 is used to store data.

[0250] The electronic device 1500 also includes an access device 1540, which enables the electronic device 1500 to communicate via one or more networks 1560. Examples of these networks include a public switched telephone network (PSTN), a local area network (LAN), a wide area network (WAN), a personal area network (PAN), or a combination of communication networks such as the Internet. The access device 1540 may include one or more of any type of network interface (e.g., a network interface card (NIC)) of wired or wireless, such as an IEEE 802.11 wireless local area network (WLAN) wireless interface, a world-wide interoperability for microwave access (Wi-MAX) interface, an Ethernet interface, a universal serial bus (USB) interface, a cellular network interface, a Bluetooth interface, a near field communication (NFC) interface, and the like.

[0251] In one embodiment of the present disclosure, the above components of the electronic device 1500 and Fig.15 Other components not shown in the figure may also be connected to each other, for example, via a bus. It should be understood that Fig.15 The electronic device structure block diagram shown is only for the purpose of illustration, and is not intended to limit the scope of the present disclosure. Those skilled in the art may add or replace other components as needed.

[0252] Electronic device 1500 may be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., a tablet computer, a personal digital assistant, a laptop computer, a notebook computer, a netbook, etc.), a mobile phone (e.g., a smart phone), a wearable computing device (e.g., a smart watch, smart glasses, etc.), or other types of mobile devices, or a stationary computing device such as a desktop computer or a personal computer (PC). Electronic device 1500 may also be a mobile or stationary server.

[0253] When the processor 1520 executes the computer instructions, the steps of the workstation resetting method are implemented.

[0254] The above is a schematic scheme of an electronic device of this embodiment. It should be noted that the technical scheme of the electronic device and the technical scheme of the above-mentioned device upgrade method belong to the same concept, and the details of the technical scheme of the computing electronic device that are not described in detail can be referred to the description of the technical scheme of the above-mentioned device upgrade method.

[0255] The embodiments of the present disclosure further provide a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to enable a computer to execute the device upgrade method described in the above embodiments of the present disclosure.

[0256] The embodiments of the present disclosure further provide a computer program product, including a computer program, which executes the device upgrade method described in the above embodiments of the present disclosure when a processor is used to execute the computer program.

[0257] The above describes specific embodiments of the present disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0258] The computer instructions include computer program codes, which may be in source code form, object code form, executable files or some intermediate forms, etc. The computer readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0259] It should be noted that, for the above-mentioned method embodiments, for the sake of simplicity of description, they are all expressed as a series of action combinations, but those skilled in the art should know that the present disclosure is not limited by the described action sequence, because according to the present disclosure, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present disclosure.

[0260] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0261] The preferred embodiments of the present disclosure disclosed above are only used to help explain the present disclosure. The optional embodiments do not describe all the details in detail, nor do they limit the invention to only the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of the present disclosure. The present disclosure selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present disclosure, so that those skilled in the art can well understand and use the present disclosure. The present disclosure is limited only by the claims and their full scope and equivalents.

Claims

1. A device upgrading method, characterized in that: The method is applied to a main function module in a terminal device, and comprises: Acquire a first model and / or a second model from a version server, wherein the first model is used by the terminal device to make a version upgrade decision, and the second model is used by the terminal device to make a version rollback decision, and the first model and / or the second model are trained by the version server; The first model is used to make a version upgrade decision for a first functional module, and / or the second model is used to make a version rollback decision for the first functional module, where the first functional module is a main functional module of the terminal device and any one of multiple sub-functional modules.

2. The method according to claim 1, characterized in that The using the first model to make a version upgrade decision for the first functional module includes: Obtaining current parameters of the first functional module; Based on the current parameters, using the first model to determine whether the first functional module needs to be upgraded; When it is determined that the first functional module needs to be upgraded, at least one of an upgraded version, an upgraded time, and an upgraded location of the first functional module is determined using the first model based on the current parameters.

3. The method according to claim 2, characterized in that The determining, based on the current parameters, using the first model whether the first functional module needs to be upgraded comprises: Based on at least one of the current state parameters, current performance parameters, current version parameters and hardware parameters of the first functional module, when it is judged through the first model that the operating status of the first functional module meets the upgrade conditions and / or there is currently an upgradeable version, it is determined that the first functional module needs to be upgraded.

4. The method according to claim 2, characterized in that: The determining, based on the current parameters, at least one of an upgrade version, an upgrade time, and an upgrade location of the first functional module using the first model comprises: Based on at least one of the current version parameter and the hardware parameter of the first functional module, determining an upgraded version of the first functional module through the first model; Based on at least one of the current state parameter and the current performance parameter of the first functional module, the upgrade time and / or upgrade location of the first functional module is determined by the first model.

5. The method according to any one of claims 2 to 4, characterized in that The method further comprises: Sending the current parameters and an upgrade decision result to the version server, where the upgrade decision result includes at least one of indication information that the first functional module needs to be upgraded, an upgraded version of the first functional module, an upgrade time, and an upgrade location; Receiving an upgrade package corresponding to the upgraded version sent by the version server; When the upgrade time and / or the upgrade location are reached, the upgrade package is installed.

6. The method according to claim 5, characterized in that The method further comprises: Automatically backing up the current parameters and the upgrade decision results; Monitoring the installation process of the upgrade package to determine whether an installation anomaly occurs to the upgrade package, wherein the installation anomaly includes at least one of an installation interruption, an installation freeze, or an installation error; In the event that an installation exception occurs to the upgrade package, the cause of the installation exception is determined and recorded.

7. The method according to claim 1, characterized in that The using the second model to make a version rollback decision on the first functional module includes: Obtaining current parameters of the first functional module; Based on the current parameters, using the second model to determine whether the first functional module reaches a rollback condition; When it is determined that the rollback condition is met, based on the current parameters, the second model is used to determine the target historical version before the current version, and the target historical version is a historical version of the first functional module whose operating status meets the preset conditions among multiple historical versions before the upgrade to the current version.

8. The method according to claim 7, characterized in that: The determining, based on the current parameters, using the second model to determine whether the first functional module reaches a rollback condition comprises: Based on the current state parameters of the first functional module, determining the fault level of the first functional module through the second model, and determining that the first functional module reaches the rollback condition when the fault level is higher than a preset level threshold or when the fault level is an installation abnormality; Wherein, based on the current parameter, using the second model to determine the target historical version before the current version includes: Based on the current version parameters of the first functional module, the rollback historical version corresponding to the fault level is determined through the second model, and the historical version whose operating status meets the preset conditions is determined from the rollback historical versions as the target historical version.

9. The method according to claim 7 or 8, characterized in that: The method further comprises: Sending the current parameters and the rollback decision result to the version server, wherein the rollback decision result includes the indication information that the first functional module needs to be rolled back and the target historical version; In the case where the first functional module has uninstalled and deleted the upgrade package corresponding to the target historical version, receive the upgrade package corresponding to the target historical version sent by the version server and install the upgrade package; or, in the case where the first functional module has not deleted the upgrade package corresponding to the target historical version, install the upgrade package.

10. The method according to claim 1, characterized in that The method further comprises: Obtaining historical parameters of the first functional module; The historical parameters are sent to the version server, and the historical parameters are used for training data so that the version server performs model training to obtain the first model and / or the second model.

11. The method according to any one of claims 2, 7 and 10, characterized in that: The current parameter and / or the historical parameter include at least one of a state parameter, a performance parameter, a version parameter, and a hardware parameter, wherein: The state parameter includes at least one of an abnormal state, a device state, and a task state, wherein the abnormal state is used to identify whether the first functional module is abnormal, the device state is used to identify whether the first functional module is in an idle or busy state, and the task state is used to identify a progress state, a success state, or a failure state of a task executed by the first functional module; The performance parameter includes at least one of network performance, storage performance, processing performance, and idle interval performance. The network performance is used to identify the performance of the first communication connection between the main function module and the multiple sub-function modules, and / or is used to identify the performance of the second communication connection between the main function module and the version server. The storage performance is used to identify the remaining storage space size of the first function module. The processing performance is used to identify the processing capacity of the first function module; the idle performance is used to identify the idle time interval between the first function module executing the current task and the next task; The version parameters include at least one of the current version number of the first functional module, the operating parameters of the first functional module under the current version number, at least one previous version number of the first functional module before the current version number, and the operating parameters of the first functional module under the at least one previous version number; The hardware parameters include at least one of a model and a factory of the first functional module.

12. The method according to claim 1, characterized in that The method further comprises: Feedback data is sent to the version server, wherein the feedback data includes at least one of the current parameters of the first functional module, an upgrade decision result, and a rollback decision result, and the feedback data is used by the version server to send an upgrade installation package and / or a rollback installation package to the terminal device, and / or the feedback data is used to update the first model and / or the second model.

13. A device upgrade method, characterized in that: Applied to a version server, the method includes: Performing model training on the first initial model to obtain a first model, and / or, performing model training on the second initial model to obtain a second model; The first model and / or the second model are sent to the main function module of the terminal device, the first model is used by the main function module to make version upgrade decisions on the first function module, and the second model is used by the main function module to make version rollback decisions on the first function module, the first function module is the main function module of the terminal device and any one of multiple sub-function modules.

14. The method according to claim 13, characterized in that The method further comprises: receiving current parameters and an upgrade decision result of the first function module sent by the main function module of the terminal device, wherein the upgrade decision result is obtained by the main function module based on the current parameters and using the first model, and the upgrade decision result includes at least one of indication information that the first function module needs to be upgraded, an upgraded version of the first function module, an upgrade time, and an upgrade location; Send an upgrade package corresponding to the upgraded version to the main function module.

15. The method according to claim 13, characterized in that The method further comprises: Receive current parameters and a rollback decision result of the first function module sent by the main function module of the terminal device, where the rollback decision result is obtained by the main function module based on the current parameters and using the second model, and the rollback decision result includes indication information that the first function module needs to be rolled back and a target historical version; Send the upgrade package corresponding to the target historical version to the main function module.

16. The method according to claim 13, characterized in that The method further comprises: Receiving historical parameters of the first function module sent by the main function module of the terminal device, and receiving an upgrade version number and an upgrade package for upgrading the first function module sent by a compile server; The performing model training on the first initial model to obtain the first model, and / or the performing model training on the second initial model to obtain the second model comprises: Using at least one of the historical parameters and the upgraded version number as training data, performing model training on the first initial model to obtain the first model, and / or performing model training on the second initial model to obtain the second model.

17. The method according to claim 16, characterized in that The performing model training on the first initial model to obtain the first model comprises: The historical state parameters, historical performance parameters, historical version parameters and hardware parameters of the first functional module and at least one of the upgraded version numbers are used as training data, and a preset upgrade result label is used to train the first initial model. When the number of training iterations reaches a first preset number or the training error is less than or equal to a first preset error, the training is terminated and the first model is obtained.

18. The method according to claim 16, characterized in that The performing model training on the second initial model to obtain the second model comprises: The historical state parameters and historical version parameters of the first functional module are used as training data, and the preset rollback result label is used to train the second initial model. When the number of training iterations reaches a second preset number or the training error is less than or equal to the second preset error, the training is terminated and the second model is obtained.

19. The method according to any one of claims 14 to 18, characterized in that The current parameter and / or the historical parameter include at least one of a state parameter, a performance parameter, a version parameter, and a hardware parameter, wherein: The state parameter includes at least one of an abnormal state, a device state, and a task state, wherein the abnormal state is used to identify whether the first functional module is abnormal, the device state is used to identify whether the first functional module is in an idle or busy state, and the task state is used to identify a progress state, a success state, or a failure state of a task executed by the first functional module; The performance parameter includes at least one of network performance, storage performance, processing performance, and idle interval performance. The network performance is used to identify the performance of the first communication connection between the main function module and the multiple sub-function modules, and / or is used to identify the performance of the second communication connection between the main function module and the version server. The storage performance is used to identify the remaining storage space size of the first function module. The processing performance is used to identify the processing capacity of the first function module; the idle performance is used to identify the idle time interval between the first function module executing the current task and the next task; The version parameters include at least one of the current version number of the first functional module, the operating parameters of the first functional module under the current version number, at least one previous version number of the first functional module before the current version number, and the operating parameters of the first functional module under the at least one previous version number; The hardware parameters include at least one of a model and a factory of the first functional module.

20. A device upgrading apparatus, characterized in that: The main functional modules used in terminal equipment include: an acquisition unit, configured to acquire a first model and / or a second model from a version server, wherein the first model is used by the terminal device to make a version upgrade decision, and the second model is used by the terminal device to make a version rollback decision, and the first model and / or the second model are obtained by training the version server; A decision unit is used to use the first model to make a version upgrade decision for a first functional module, and / or use the second model to make a version rollback decision for the first functional module, wherein the first functional module is a main functional module of the terminal device and any one of multiple sub-functional modules.

21. A device upgrading apparatus, characterized in that: Applicable to version servers, including: A training unit, configured to perform model training on the first initial model to obtain a first model, and / or perform model training on the second initial model to obtain a second model; A sending unit is used to send the first model and / or the second model to the main function module of the terminal device, the first model is used by the main function module to make version upgrade decisions on the first function module, and the second model is used by the main function module to make version rollback decisions on the first function module, and the first function module is the main function module of the terminal device and any one of multiple sub-function modules.

22. A device upgrade system, characterized in that: It includes terminal equipment and version server, among which, The main function module in the terminal device is used to perform the method according to any one of claims 1 to 12; The version server is used to execute the method according to any one of claims 13 to 19.

23. An electronic device, comprising: processor and memory, wherein The memory is used to store computer executable instructions; The processor is configured to read the instruction from the memory and execute the instruction to implement the method according to any one of claims 1 to 19.

24. A computer-readable storage medium, wherein: The storage medium stores computer program instructions, and when a computer reads the instructions, the method according to any one of claims 1 to 19 is executed.

25. A computer program product comprising a computer program, wherein: When the computer program is executed by a processor, the method according to any one of claims 1 to 19 is implemented.

Images