Vehicle firmware upgrading method, device and equipment and computer readable medium

Abstract

The invention provides a vehicle firmware upgrading method, device and equipment and a computer readable medium, and the vehicle firmware upgrading method comprises the steps: obtaining multi-dimensional operation data and user behavior data of a vehicle, carrying out the preprocessing of the data, and carrying out the analysis through employing a multi-mode deep learning model, and obtaining a prediction result; based on the prediction result and the vehicle hardware configuration information, a differentiated firmware upgrading strategy is generated, and the firmware upgrading strategy at least comprises a to-be-upgraded module, an upgrading priority and an expected performance index; according to the firmware upgrading strategy, a firmware upgrading package is obtained in a differential transmission mode; the influence of the firmware upgrade package on the expected performance index is verified in the virtual environment, and if it is monitored that the expected performance index is abnormal, upgrade is terminated and rollback is carried out; and after the virtual environment verification is passed, selecting a vehicle idle time period or a charging time period to trigger and execute firmware upgrading, and after upgrading is completed, optimizing the multi-modal deep learning model through closed-loop feedback.

Description

Technical Field

[0001] This disclosure relates to the fields of automotive electronics and artificial intelligence technology, and in particular to a vehicle firmware upgrade method, apparatus, device, and computer-readable medium. Background Technology

[0002] As the level of automotive intelligence continues to improve, the update frequency of vehicle firmware (i.e., software embedded in the vehicle's electronic control unit) is gradually accelerating. Traditional vehicle firmware upgrade methods mainly include two modes: passive upgrades and periodic upgrades.

[0003] Passive upgrades involve manufacturers releasing firmware patches after a vehicle malfunctions or a user complaint. This approach is significantly delayed, fails to anticipate potential problems, and may lead to safety hazards or performance degradation.

[0004] Regular upgrades push out general firmware versions according to a preset cycle. They lack personalized adaptation for specific vehicles or users and do not take into account factors such as the actual operating status of the vehicle, user habits, environmental changes, and hardware aging. This can easily lead to compatibility issues or limited performance improvements after the upgrade.

[0005] In addition, traditional upgrade methods usually involve a full upgrade, which consumes a lot of network bandwidth and vehicle computing resources, resulting in low efficiency. Furthermore, the timing of upgrades is often determined by the manufacturer, making it impossible to respond promptly to the actual usage needs of the vehicle. Summary of the Invention

[0006] This disclosure aims to solve at least one of the technical problems existing in the prior art, and proposes a vehicle firmware upgrade method, apparatus, device and computer-readable medium.

[0007] In a first aspect, embodiments of this disclosure provide a vehicle firmware upgrade method, the method comprising:

[0008] The system acquires multidimensional operational data of the vehicle and user behavior data. After preprocessing the data, it uses a multimodal deep learning model to analyze the data and obtain prediction results. The prediction results include the performance degradation trend and potential failure risk of at least one key part of the vehicle.

[0009] Based on the prediction results and vehicle hardware configuration information, a differentiated firmware upgrade strategy is generated. The firmware upgrade strategy includes at least the module to be upgraded, the upgrade priority, and the expected performance indicators.

[0010] According to the firmware upgrade strategy, the firmware upgrade package is obtained through differential transmission.

[0011] Verify the impact of the firmware upgrade package on the expected performance indicators in a virtual environment. If abnormalities in the expected performance indicators are detected, terminate the upgrade and roll back.

[0012] After the virtual environment verification is passed, the firmware upgrade is triggered and executed during the vehicle's idle period or charging period. After the upgrade is completed, the multimodal deep learning model is optimized through closed-loop feedback.

[0013] In some embodiments, the acquisition of multidimensional vehicle operation data and user behavior data, followed by preprocessing of the data and analysis using a multimodal deep learning model to obtain prediction results, includes:

[0014] Multidimensional operational data and user behavior data are acquired through an in-vehicle data acquisition terminal. The data is then preprocessed, including at least one of outlier removal, missing value imputation, data standardization, and normalization.

[0015] Temporal features are extracted from the preprocessed data using the sliding window method or Fourier transform.

[0016] The temporal features are analyzed using a multimodal deep learning model to obtain the prediction results.

[0017] In some embodiments, generating a differentiated firmware upgrade strategy based on the prediction results and vehicle hardware configuration information includes:

[0018] Based on the component models, manufacturing dates, and historical upgrade records in the vehicle hardware configuration information, and combined with the performance degradation trend in the prediction results, the power output curve is dynamically adjusted to reduce energy consumption.

[0019] Based on the potential failure risks and the vehicle hardware configuration information, the charging threshold of the battery management system is set;

[0020] The modules to be upgraded include battery management system firmware, advanced driver assistance system firmware, power control system firmware, and in-vehicle entertainment system firmware. The upgrade priority is sorted according to safety correlation, with the upgrade priority of braking system related firmware being higher than that of in-vehicle entertainment system firmware.

[0021] The expected performance indicators include at least the reduction in failure risk and the percentage of performance recovery of key components after the upgrade.

[0022] In some embodiments, the differential transmission method transmits only the difference data segment between the firmware upgrade package and the current firmware version of the vehicle. The integrity of the difference data segment is ensured by hash verification, and the transmission process is protected by an encryption protocol.

[0023] In some embodiments, the virtual environment is a simulation environment built based on a vehicle digital twin.

[0024] The step of verifying the impact of the firmware upgrade package on the expected performance metrics in a virtual environment, and terminating and rolling back the upgrade if abnormalities in the expected performance metrics are detected, includes:

[0025] Simulate the vehicle's operating status under preset conditions after the upgrade, and calculate the deviation rate between the actual and predicted changes in the expected performance indicators.

[0026] If the deviation rate exceeds the preset deviation threshold, the expected performance indicator is determined to be abnormal, the upgrade is automatically terminated and rolled back to the stable version before the upgrade.

[0027] In some embodiments, optimizing the multimodal deep learning model through closed-loop feedback after the upgrade includes:

[0028] After the upgrade is completed, the actual performance index data of the vehicle will be continuously collected, and the error value between the actual performance index data and the prediction result will be calculated.

[0029] If the error value exceeds a preset error threshold, the training sample set of the multimodal deep learning model is updated based on the actual performance index data, and the model is retrained to optimize the prediction accuracy.

[0030] In some embodiments, the vehicle idle time period is determined by the vehicle system, including the time period when the vehicle is turned off and not in a charging state for more than a preset time, or the time period when the vehicle is parked and the user is not in the driver's cabin.

[0031] The charging period refers to the time when the vehicle is connected to the charging equipment but the charging progress has not reached 100%.

[0032] In some embodiments, the multidimensional operating data includes battery health status, engine operating parameters, braking system wear data, environmental parameters collected by on-board sensors, and vehicle speed and acceleration time-series data.

[0033] The user behavior data includes driving time, charging frequency, frequency of rapid acceleration / deceleration operations, and preferred driving route data.

[0034] In some embodiments, the multimodal deep learning model is a hybrid architecture model of Transformer and LSTM.

[0035] The multimodal deep learning model dynamically adjusts the network layer weight parameters through a reinforcement learning mechanism. The prediction result also includes a prediction time window for when the performance of the key components degrades to a preset threshold.

[0036] Secondly, embodiments of this disclosure provide a vehicle firmware upgrade device, the device comprising:

[0037] The prediction module is configured to acquire multi-dimensional operating data of the vehicle and user behavior data, and after preprocessing the data, analyze it using a multimodal deep learning model to obtain prediction results. The prediction results include the performance degradation trend and potential failure risk of at least one key part of the vehicle.

[0038] The generation module is configured to generate a differentiated firmware upgrade strategy based on the prediction results and vehicle hardware configuration information. The firmware upgrade strategy includes at least the module to be upgraded, the upgrade priority, and the expected performance indicators.

[0039] The acquisition module is configured to acquire the firmware upgrade package via differential transmission according to the firmware upgrade strategy.

[0040] The verification module is configured to verify the impact of the firmware upgrade package on the expected performance indicators in a virtual environment, and if the expected performance indicators are detected to be abnormal, the upgrade will be terminated and rolled back.

[0041] The feedback module is configured to trigger and execute the firmware upgrade during a vehicle idle period or charging period after the virtual environment verification is passed, and optimize the multimodal deep learning model through closed-loop feedback after the upgrade is completed.

[0042] Thirdly, embodiments of this disclosure provide an electronic device, including:

[0043] One or more processors;

[0044] Memory, used to store one or more programs;

[0045] When one or more programs are executed by one or more processors, the one or more processors implement the vehicle firmware upgrade method provided in the first aspect.

[0046] Fourthly, embodiments of this disclosure provide a computer-readable medium storing a computer program, which, when executed by a processor, implements the vehicle firmware upgrade method provided in the first aspect.

[0047] The vehicle firmware upgrade method disclosed herein uses a modal deep learning model to predict and proactively identify vehicle problems, transforming passive fault repair into proactive performance prevention, reducing vehicle failure rates, and improving driving safety and reliability. Then, based on the prediction results and hardware configuration, a differentiated upgrade strategy is generated to avoid ineffective optimizations caused by generic upgrades, achieving personalized adaptation for the vehicle and specifically improving vehicle performance. Next, a differential transmission method is used to obtain the upgrade package, significantly reducing network traffic consumption, lowering onboard computing resource consumption, and improving upgrade efficiency. Furthermore, virtual environment verification and a dynamic rollback mechanism prevent compatibility issues or performance degradation after the upgrade, ensuring the safety of the upgrade process. Finally, the upgrade is performed during vehicle idle or charging periods, minimizing interference with user operation, improving user experience, and continuously optimizing model prediction accuracy through a closed-loop feedback mechanism to ensure the long-term adaptability of the upgrade strategy. Attached Figure Description

[0048] Figure 1 A schematic flowchart illustrating a vehicle firmware upgrade method provided in this embodiment of the disclosure;

[0049] Figure 2 This is a flowchart illustrating an optional specific implementation method of step S1 in an embodiment of this disclosure;

[0050] Figure 3 This is a flowchart illustrating an optional specific implementation method of step S2 in an embodiment of this disclosure;

[0051] Figure 4 This is a schematic diagram of the structure of a vehicle firmware upgrade device provided in an embodiment of the present disclosure;

[0052] Figure 5 This is a structural block diagram of an electronic device provided in an embodiment of the present disclosure. Detailed Implementation

[0053] To enable those skilled in the art to better understand the technical solutions of this disclosure, exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments of this disclosure to aid understanding. These should be considered merely exemplary. Therefore, those skilled in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0054] Where there is no conflict, the various embodiments of this disclosure and the features thereof in the embodiments may be combined with each other.

[0055] As used herein, the term “and / or” includes any and all combinations of one or more related enumerated entries.

[0056] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit this disclosure. As used herein, the singular forms “a” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will also be understood that when the terms “comprising” and / or “made of” are used in this specification, the presence of the stated feature, integral, step, operation, element, and / or component is specified, but the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or groups thereof is not excluded. Words such as “connected” or “linked” are not limited to physical or mechanical connections but can include electrical connections, whether direct or indirect.

[0057] Unless otherwise specified, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art. It will also be understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant art and this disclosure, and will not be interpreted as having an idealized or overly formal meaning, unless expressly so defined herein.

[0058] The collection, storage, use, processing, transmission, provision, and disclosure of user personal information in this technical solution comply with relevant laws and regulations and do not violate public order and good morals. The use of user data in this technical solution follows relevant national laws and regulations (e.g., the "Information Security Technology - Personal Information Security Specification"). For example, appropriate measures are taken for personal information access control; restrictions are imposed on the display of personal information; the purpose of using personal information does not exceed the scope of direct or reasonable association; and explicit identity targeting is eliminated when using personal information to avoid precisely identifying specific individuals.

[0059] With the continuous improvement of vehicle intelligence, vehicle firmware, as the core software embedded in the vehicle's electronic control unit, is being updated more frequently, becoming a key means to optimize vehicle performance and improve user experience. Currently, the mainstream vehicle firmware upgrade methods are mainly divided into two categories: passive upgrades and periodic upgrades.

[0060] The passive upgrade mode requires the manufacturer to release firmware patches after a vehicle malfunctions or a user complaint. This mode has significant lag, cannot predict potential problems in advance, and is prone to safety hazards or performance loss. In addition, the full upgrade process will consume a lot of network bandwidth and vehicle computing resources, resulting in resource waste.

[0061] The scheduled upgrade mode pushes out general firmware versions according to a preset cycle, lacking the ability to dynamically adapt to the actual operating status of the vehicle, user driving habits, and environmental changes. Because it does not take into account personalized factors such as hardware aging, battery degradation, and differences in driving scenarios, it often results in compatibility issues after the upgrade or limited performance improvement, making it difficult to meet users' personalized needs for vehicle performance optimization.

[0062] To address the issues of lag, lack of personalization, resource waste, and insufficient performance improvement in existing technologies, this disclosure proposes an AI-based prediction-based vehicle firmware optimization and upgrade method. This method utilizes artificial intelligence technology to achieve intelligent prediction and personalized upgrades of vehicle firmware. By combining differential transmission, virtual verification, and closed-loop feedback mechanisms, it enables vehicle firmware upgrades while ensuring accuracy, security, and resource efficiency.

[0063] Figure 1 This is a flowchart illustrating a vehicle firmware upgrade method provided in an embodiment of this disclosure, as shown below. Figure 1 As shown, the above method includes steps S1-S5, as detailed below:

[0064] Step S1: Obtain multi-dimensional operating data of the vehicle and user behavior data. After data preprocessing, use a multi-modal deep learning model for analysis to obtain prediction results. The prediction results include the performance degradation trend and potential failure risk of at least one key part of the vehicle.

[0065] Step S2: Based on the prediction results and vehicle hardware configuration information, generate a differentiated firmware upgrade strategy. The firmware upgrade strategy shall include at least the module to be upgraded, the upgrade priority, and the expected performance indicators.

[0066] Step S3: Obtain the firmware upgrade package via differential transmission according to the firmware upgrade strategy.

[0067] Step S4: Verify the impact of the firmware upgrade package on the expected performance indicators in a virtual environment. If abnormal expected performance indicators are detected, terminate the upgrade and roll back.

[0068] Step S5: After the virtual environment verification is passed, select the vehicle's idle time period or charging time period to trigger and execute the firmware upgrade. After the upgrade is completed, optimize the multimodal deep learning model through closed-loop feedback.

[0069] The vehicle firmware upgrade method provided in this disclosure uses a modal deep learning model to predict and proactively identify vehicle problems, transforming passive fault repair into proactive performance prevention, reducing vehicle failure rates, and improving driving safety and reliability. Then, based on the prediction results and hardware configuration, a differentiated upgrade strategy is generated to avoid ineffective optimizations caused by generic upgrades, achieving personalized adaptation for the vehicle and specifically improving vehicle performance. Next, a differential transmission method is used to obtain the upgrade package, significantly reducing network traffic consumption, lowering onboard computing resource consumption, and improving upgrade efficiency. Furthermore, virtual environment verification and a dynamic rollback mechanism prevent compatibility issues or performance degradation after the upgrade, ensuring the safety of the upgrade process. Finally, the upgrade is performed during vehicle idle or charging periods, minimizing interference with user operation, improving user experience, and continuously optimizing model prediction accuracy through a closed-loop feedback mechanism to ensure the long-term adaptability of the upgrade strategy.

[0070] Figure 2 This is a flowchart illustrating an optional specific implementation method of step S1 in an embodiment of this disclosure. In some embodiments, step S1 may specifically include steps S101-S103:

[0071] Step S101: Obtain multi-dimensional operation data and user behavior data through the vehicle-mounted data acquisition terminal, and preprocess the data. The preprocessing includes at least one of outlier removal, missing value imputation, data standardization and normalization.

[0072] Step S102: Extract time-series features from the preprocessed data using the sliding window method or Fourier transform.

[0073] Step S103: Use a multimodal deep learning model to analyze the temporal features and obtain the prediction results.

[0074] In some embodiments, multidimensional operating data includes battery health status, engine operating parameters, braking system wear data, environmental parameters collected by onboard sensors, and vehicle speed and acceleration time-series data; user behavior data includes driving duration, charging frequency, frequency of rapid acceleration / deceleration operations, and preferred driving route data.

[0075] The aforementioned multi-dimensional operational data can be obtained through vehicle-mounted data acquisition terminals such as vehicle sensors and OBD interfaces, while user behavior data can be collected through user interaction systems.

[0076] In some embodiments, the multimodal deep learning model is a hybrid architecture model of Transformer and LSTM. The multimodal deep learning model dynamically adjusts the network layer weight parameters through a reinforcement learning mechanism. The prediction results also include a prediction time window when the performance of key parts degrades to a preset threshold.

[0077] The reason for adopting a combination of multi-source data acquisition and multimodal deep learning models in step S1 is that a single data type cannot fully reflect the vehicle's operating status and user habits. The hybrid architecture model can fully explore the deep correlation features in time series data, while the reinforcement learning mechanism can enable the model to dynamically adapt to new vehicle models and new operating conditions. This solves the problem that existing technologies cannot predict vehicle performance problems in advance, and realizes the transformation from fault repair to performance prevention. This allows for the early identification of potential fault risks and performance degradation trends, providing accurate data support for subsequent personalized upgrades and reducing the failure rate.

[0078] Figure 3 This is a flowchart illustrating an optional implementation of step S2 in an embodiment of this disclosure. In some embodiments, step S2 may specifically include steps S201-S204:

[0079] Step S201: Based on the component model, manufacturing time and historical upgrade records in the vehicle hardware configuration information, and combined with the performance degradation trend in the prediction results, dynamically adjust the power output curve to reduce energy consumption.

[0080] Step S202: Based on potential fault risks and vehicle hardware configuration information, set the charging threshold of the battery management system.

[0081] Step S203: The modules to be upgraded include the battery management system firmware, advanced driver assistance system firmware, power control system firmware, and in-vehicle entertainment system firmware. The upgrade priority is sorted according to the degree of safety correlation, with the upgrade priority of the braking system-related firmware being higher than that of the in-vehicle entertainment system firmware.

[0082] Step S204, the expected performance indicators include at least the reduction in failure risk of key components and the percentage of performance recovery after the upgrade.

[0083] The aforementioned differentiated firmware upgrade strategy reduces vehicle energy consumption by dynamically adjusting the power output curve and extends battery life by appropriately setting the charging threshold of the battery management system. The upgrade strategy clearly identifies the modules to be upgraded, prioritizing them according to their safety relevance. Firmware upgrades related to the braking system have a higher priority than those for the in-vehicle entertainment system, ensuring that safety-related functions are optimized first. Furthermore, the upgrade strategy includes expected performance indicators, at least covering the reduction in failure risk and the percentage of performance recovery for key components after the upgrade.

[0084] The reason for this design in step S2 is that existing general upgrade strategies ignore the individualized characteristics of vehicle hardware differences and performance degradation, resulting in poor upgrade outcomes. By combining hardware configuration with AI prediction results to generate differentiated strategies, customized upgrades can be achieved for each vehicle, thereby avoiding ineffective upgrades, specifically optimizing vehicle performance, ensuring driving safety through priority ranking, and meeting users' personalized performance improvement needs.

[0085] In some embodiments, the differential transmission method in step S3 is to transmit only the difference data segment between the firmware upgrade package and the current firmware version of the vehicle. The difference data segment is ensured to be intact through hash verification, and the transmission process is protected by an encryption protocol.

[0086] Differential transmission addresses the issue of excessive network bandwidth and onboard computing resources consumed by current full-scale upgrades. Encryption protocols and hash verification ensure the security and integrity of data transmission. Furthermore, differential transmission saves network bandwidth, reduces network resource consumption, and prevents data leakage or corruption during transmission, improving the efficiency and security of upgrade package acquisition.

[0087] In some embodiments, the virtual environment is a simulation environment built based on a vehicle digital twin, then step S4 may specifically include:

[0088] The system simulates the vehicle's operation under preset conditions after the upgrade, calculates the deviation rate between the actual and predicted changes in the expected performance indicators, and determines that the expected performance indicators are abnormal if the deviation rate exceeds the preset deviation threshold. The upgrade is then automatically terminated and the system is rolled back to the stable version before the upgrade.

[0089] Because existing technologies lack effective verification mechanisms before upgrades, compatibility issues or performance degradation are likely to occur after upgrades. Therefore, a simulation environment built using a digital twin is used to realistically simulate vehicle operating conditions. By building a virtual simulation environment to simulate vehicle operation, upgrade risks can be identified in advance, avoiding impacts on normal vehicle use after the upgrade. Combined with a dynamic rollback mechanism, this provides dual protection for the safety of the upgrade process and the stability of vehicle performance.

[0090] Furthermore, in some embodiments, after the virtual environment verification is passed, a firmware upgrade is triggered and executed during either a vehicle idle period or a charging period. The vehicle idle period is determined by the onboard system and includes periods when the vehicle is off and not charging for more than a preset time, or when the vehicle is parked and the user is not in the driver's cabin. The charging period is when the vehicle is connected to the charging device and the charging progress has not reached 100%.

[0091] Choosing the above-mentioned time period for the upgrade can avoid performing the upgrade while the user is driving, which could affect the driving experience or cause safety hazards, thereby reducing interference with the user's vehicle use and ensuring that the upgrade process does not affect the normal use of the vehicle.

[0092] In some embodiments, step S5, after the upgrade is completed, optimizes the multimodal deep learning model through closed-loop feedback, including:

[0093] After the upgrade is completed, the system continuously collects actual vehicle performance data and calculates the error between the actual performance data and the prediction results. If the error exceeds the preset error threshold, the training sample set of the multimodal deep learning model is updated based on the actual performance data, and the model is retrained to optimize the prediction accuracy.

[0094] Because vehicle operating conditions and user behavior are dynamically changing, fixed models struggle to maintain predictive accuracy over the long term. Therefore, iterative model iteration using real-world data feedback allows the model to continuously adapt to new usage scenarios. This enables dynamic model optimization, improves the accuracy of subsequent predictions, ensures continuous optimization of upgrade strategies, and creates a virtuous cycle of prediction, upgrade, feedback, and optimization, guaranteeing the long-term adaptability of upgrade strategies.

[0095] The vehicle firmware upgrade method provided in this disclosure employs a multimodal deep learning model combined with multi-source data acquisition and temporal feature extraction. This enables accurate prediction of vehicle performance degradation trends and potential fault risks. A differentiated upgrade strategy ensures that each upgrade is targeted, avoiding ineffective upgrades. Verification in a virtual environment based on a digital twin proactively identifies upgrade risks. An automatic rollback mechanism in abnormal situations ensures the vehicle remains in a stable operating state, while the upgrade priority based on safety correlation further enhances driving safety. Firmware upgrades are performed within a set time period, creating a seamless upgrade mode that reduces vehicle interference. Personalized upgrade strategies optimize performance based on user driving habits and actual vehicle conditions, improving fuel efficiency, power performance, and driving comfort to meet individual user needs. Furthermore, differential transmission technology saves significant network traffic, and edge computing reduces cloud dependence, minimizing network bandwidth and onboard computing resource waste, thus improving the economics of upgrades. A reinforcement learning mechanism enables the model to continuously adapt to new vehicle models and operating conditions, while a closed-loop feedback mechanism achieves iterative model optimization. Applicable to gasoline vehicles, electric vehicles, and autonomous vehicles, this provides automakers with differentiated technical service advantages, helping them enhance their competitiveness in the intelligent vehicle market.

[0096] Figure 4 This is a schematic diagram of the structure of a vehicle firmware upgrade device provided in an embodiment of the present disclosure, as shown below. Figure 4 As shown, the device includes:

[0097] The prediction module 10 is configured to acquire multi-dimensional operating data of the vehicle and user behavior data, preprocess the data, and then analyze it using a multimodal deep learning model to obtain prediction results. The prediction results include the performance degradation trend and potential failure risk of at least one key part of the vehicle.

[0098] The generation module 20 is configured to generate a differentiated firmware upgrade strategy based on the prediction results and vehicle hardware configuration information. The firmware upgrade strategy includes at least the module to be upgraded, the upgrade priority, and the expected performance indicators.

[0099] The acquisition module 30 is configured to acquire the firmware upgrade package via differential transmission according to the firmware upgrade strategy.

[0100] The verification module 40 is configured to verify the impact of the firmware upgrade package on the expected performance indicators in a virtual environment. If the expected performance indicators are detected to be abnormal, the upgrade will be terminated and rolled back.

[0101] The feedback module 50 is configured to trigger and execute the firmware upgrade during a vehicle idle period or charging period after the virtual environment verification is passed, and optimize the multimodal deep learning model through closed-loop feedback after the upgrade is completed.

[0102] The vehicle firmware upgrade device provided in this disclosure uses a modal deep learning model to predict and proactively identify vehicle problems, transforming passive fault repair into proactive performance prevention, reducing vehicle failure rates, and improving driving safety and reliability. Then, based on the prediction results and hardware configuration, a differentiated upgrade strategy is generated to avoid ineffective optimizations caused by generic upgrades, achieving personalized adaptation for the vehicle and specifically improving vehicle performance. Next, a differential transmission method is used to obtain the upgrade package, significantly reducing network traffic consumption, lowering onboard computing resource consumption, and improving upgrade efficiency. Furthermore, virtual environment verification and a dynamic rollback mechanism prevent compatibility issues or performance degradation after the upgrade, ensuring the safety of the upgrade process. Finally, the upgrade is performed during vehicle idle or charging periods, minimizing interference with user operation, improving user experience, and continuously optimizing model prediction accuracy through a closed-loop feedback mechanism to ensure the long-term adaptability of the upgrade strategy.

[0103] Based on the same inventive concept, this disclosure also provides an electronic device. Figure 5 This is a structural block diagram of an electronic device provided in an embodiment of this disclosure. Figure 5As shown, this disclosure provides an electronic device including: one or more processors 101, a memory 102, and one or more I / O interfaces 103. The memory 102 stores one or more programs, which, when executed by the one or more processors, cause the one or more processors to implement any of the vehicle firmware upgrade methods described in the above embodiments; the one or more I / O interfaces 103 are connected between the processor and the memory, configured to enable information interaction between the processor and the memory.

[0104] The processor 101 is a device with data processing capabilities, including but not limited to a central processing unit (CPU); the memory 102 is a device with data storage capabilities, including but not limited to random access memory (RAM, more specifically SDRAM, DDR, etc.), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), and flash memory (FLASH); the I / O interface (read / write interface) 103 is connected between the processor 101 and the memory 102, and can realize information interaction between the processor 101 and the memory 102, including but not limited to a data bus (Bus).

[0105] In some embodiments, the processor 101, memory 102, and I / O interface 103 are interconnected via bus 104, and thus connected to other components of the computing device.

[0106] In some embodiments, the one or more processors 101 include a field-programmable gate array.

[0107] This disclosure also provides a computer-readable medium. The computer-readable medium stores a computer program, which, when executed by a processor, implements the steps of any of the vehicle firmware upgrade methods described above. The computer-readable storage medium may be volatile or non-volatile.

[0108] This disclosure also provides a computer program product, including computer-readable code, or a non-volatile computer-readable storage medium carrying computer-readable code, wherein when the computer-readable code is run in the processor of an electronic device, the processor in the electronic device executes the above-described vehicle firmware upgrade method.

[0109] Those skilled in the art will understand that all or some of the steps, systems, and apparatuses disclosed above, and their functional modules / units, can be implemented as software, firmware, hardware, or suitable combinations thereof. In hardware implementations, the division between functional modules / units mentioned above does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed collaboratively by several physical components. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit (ASIC). Such software can be distributed on a computer-readable storage medium, which may include computer storage media (or non-transitory media) and communication media (or transient media).

[0110] As is known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information, such as computer-readable program instructions, data structures, program modules, or other data. Computer storage media includes, but is not limited to, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), static random access memory (SRAM), flash memory or other memory technologies, portable compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer. Furthermore, it is known to those skilled in the art that communication media typically contain computer-readable program instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0111] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0112] Computer program instructions used to perform the operations of this disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, status setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The computer-readable program instructions may execute entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing the status information of the computer-readable program instructions to implement various aspects of this disclosure.

[0113] The computer program product described herein can be implemented specifically through hardware, software, or a combination thereof. In one alternative embodiment, the computer program product is specifically embodied in a computer storage medium; in another alternative embodiment, the computer program product is specifically embodied in a software product, such as a software development kit (SDK), etc.

[0114] Various aspects of this disclosure are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0115] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processor of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner; thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0116] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0117] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0118] Example embodiments have been disclosed herein, and while specific terminology has been used, it is for illustrative purposes only and should be construed as such, and is not intended to be limiting. In some instances, it will be apparent to those skilled in the art that features, characteristics, and / or elements described in connection with particular embodiments may be used alone, or in combination with features, characteristics, and / or elements described in connection with other embodiments, unless otherwise expressly indicated. Therefore, those skilled in the art will understand that various changes in form and detail may be made without departing from the scope of this disclosure as set forth by the appended claims.

Claims

1. A vehicle firmware upgrade method, characterized by, The method comprises: acquiring multi-dimensional running data and user behavior data of a vehicle, pre-processing the data, and analyzing the pre-processed data using a multi-modal deep learning model to obtain a prediction result, the prediction result including a performance degradation trend and a potential failure risk of at least one key part of the vehicle; generating a differentiated firmware upgrade strategy based on the prediction result and vehicle hardware configuration information, the firmware upgrade strategy including at least a module to be upgraded, an upgrade priority, and an expected performance indicator; obtaining a firmware upgrade package through differential transmission according to the firmware upgrade strategy; verifying the influence of the firmware upgrade package on the expected performance indicator in a virtual environment, and terminating the upgrade and rolling back if the expected performance indicator is found to be abnormal; after the verification in the virtual environment is passed, triggering and executing the firmware upgrade during an idle period or a charging period of the vehicle, and optimizing the multi-modal deep learning model through closed-loop feedback after the upgrade is completed.

2. The method of claim 1, wherein, The acquisition of multi-dimensional running data and user behavior data of a vehicle, the pre-processing of the data, and the analysis of the pre-processed data using a multi-modal deep learning model to obtain a prediction result include: acquiring multi-dimensional running data and user behavior data through a vehicle-mounted data acquisition terminal, pre-processing the data, the pre-processing including at least one of outlier rejection, missing value filling, data standardization, and normalization processing; extracting time series features from the pre-processed data using a sliding window method or Fourier transform; analyzing the time series features using a multi-modal deep learning model to obtain the prediction result.

3. The method of claim 1, wherein, The generation of a differentiated firmware upgrade strategy based on the prediction result and vehicle hardware configuration information includes: dynamically adjusting a power output curve to reduce energy consumption according to the component model, the time of manufacture, and the historical upgrade record in the vehicle hardware configuration information, in combination with the performance degradation trend in the prediction result; setting a charging threshold for a battery management system according to the potential failure risk and the vehicle hardware configuration information. The module to be upgraded includes a battery management system firmware, an advanced driver assistance system firmware, a power control system firmware, and a vehicle infotainment system firmware, and the upgrade priority is sorted according to safety correlation, wherein the upgrade priority of the brake system related firmware is higher than that of the vehicle infotainment system firmware. The expected performance indicator includes at least a failure risk reduction amplitude and a performance recovery percentage of the key part after the upgrade.

4. The method of claim 1, wherein, The differential transmission method is to only transmit the difference data segment between the firmware upgrade package and the current firmware version of the vehicle, the difference data segment is verified for integrity through a hash check, and an encryption protocol is used for data protection during transmission.

5. The method of claim 1, wherein, The virtual environment is a simulation environment constructed based on a vehicle digital twin, The verification of the influence of the firmware upgrade package on the expected performance indicator in the virtual environment, and the termination of the upgrade and the rollback if the expected performance indicator is found to be abnormal include: simulating the running state of the vehicle after the upgrade under a preset working condition, and calculating the deviation rate of the actual change value and the predicted change value of the expected performance indicator; if the deviation rate exceeds a preset deviation threshold, it is determined that the expected performance indicator is abnormal, and the upgrade is automatically terminated and rolled back to the stable version before the upgrade.

6. The method of claim 1, wherein, The multi-modal deep learning model is optimized through closed-loop feedback after the upgrade is completed, and the method comprises the following steps: After the upgrade is completed, the actual performance index data of the vehicle is continuously collected, and the error value of the actual performance index data and the prediction result is calculated; If the error value exceeds the preset error threshold, the training sample set of the multi-modal deep learning model is updated based on the actual performance index data, and the model is retrained to optimize the prediction accuracy.

7. The method of claim 1, wherein, The idle period of the vehicle is determined by the vehicle-mounted system, including the period when the vehicle is turned off and not in the charging state for more than a preset time, or the period when the vehicle is in the parking state and the user is not in the cabin; The charging period is the period when the vehicle is connected to the charging device and the charging progress does not reach 100%.

8. The method of claim 1, wherein, The multi-dimensional operation data includes battery health status, engine operation parameters, brake system wear data, environmental parameters collected by vehicle-mounted sensors, and vehicle speed and acceleration time series data; The user behavior data includes driving time, charging frequency, frequent operation of sudden acceleration / sudden deceleration, and preferred driving route data.

9. The method of claim 1, wherein, The multi-modal deep learning model is a hybrid architecture model of Transformer and LSTM, The multi-modal deep learning model dynamically adjusts the network layer weight parameters through the reinforcement learning mechanism, and the prediction result further includes the prediction time window when the performance of the key part degrades to the preset threshold.

10. A vehicle firmware upgrade apparatus, characterized by comprising: The device comprises: A prediction module configured to obtain multi-dimensional operation data and user behavior data of a vehicle, pre-process the data, and analyze the data using a multi-modal deep learning model to obtain a prediction result, the prediction result including the performance degradation trend and potential fault risk of at least one key part of the vehicle; A generation module configured to generate a differentiated firmware upgrade strategy based on the prediction result and vehicle hardware configuration information, the firmware upgrade strategy including at least the module to be upgraded, the upgrade priority, and the expected performance index; An acquisition module configured to acquire a firmware upgrade package through differential transmission based on the firmware upgrade strategy; A verification module configured to verify the impact of the firmware upgrade package on the expected performance index in a virtual environment, and terminate the upgrade and roll back if the expected performance index is abnormal; A feedback module configured to trigger and execute the firmware upgrade during the idle period or the charging period of the vehicle after the virtual environment verification is passed, and optimize the multi-modal deep learning model through closed-loop feedback after the upgrade is completed.

11. An electronic device, comprising: Comprise: One or more processors; Memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the vehicle firmware upgrade method of any one of claims 1 to 9.

12. A computer readable medium having stored thereon a computer program, characterized in that The computer program is executed by the processor to implement the steps in the vehicle firmware upgrade method of any one of claims 1 to 9.

Images