A software upgrade method for an intelligent self-closed-loop vehicle light controller

Abstract

This invention discloses a software upgrade method for an intelligent self-closed-loop vehicle lighting controller, belonging to the field of vehicle lighting control technology. It includes the following steps: Step S1, the cloud-based upgrade management platform distributes the software upgrade package and upgrade strategy configuration file; Step S2, a pre-upgrade self-assessment is performed based on the current vehicle and environmental status; Step S3, after the pre-upgrade self-assessment passes, the software upgrade package is silently downloaded and installed; Step S4, after the software upgrade package is installed and upgraded, self-verification is performed using self-verification test cases. This invention provides a software upgrade method for an intelligent self-closed-loop vehicle lighting controller, aiming to solve the technical problems of insufficient intelligent upgrade decision-making, lack of immediate closed-loop verification after upgrade, and slow problem response in existing OTA upgrade methods for vehicle lighting controllers. This intelligent self-closed-loop software upgrade method can autonomously assess upgrade suitability, safely execute upgrades, and quantify and verify upgrade effects in real time, thereby significantly improving the safety, success rate, and user experience of the upgrade process.

Description

Technical Field

[0001] This invention relates to a software upgrade method for an intelligent self-closed-loop vehicle lighting controller, belonging to the field of vehicle lighting control technology. Background Technology

[0002] With the rapid development of automotive intelligence and connectivity, vehicle lighting systems have evolved from basic illumination into intelligent actuators integrating adaptive high beams (ADB), pixel-level precision anti-glare, and deep integration with advanced driver assistance systems (ADAS) environmental perception systems. Software defines the core functions and performance of such intelligent vehicle lights. To ensure continuous functional optimization and bug fixes throughout the vehicle's lifecycle, over-the-air (OTA) software updates have become a standard feature of intelligent vehicles.

[0003] Currently, mainstream OTA upgrade solutions for automotive controllers typically employ a two-tier architecture: cloud-vehicle. The typical process is as follows: Cloud push: The cloud server distributes the new version of the software firmware package to eligible vehicles based on the vehicle model, hardware and software version.

[0004] Vehicle-side download and installation: The vehicle communication module (such as T-Box) receives the upgrade package. After simple conditions are met (such as the vehicle being turned off or parked), the upgrade package is transmitted to the target electronic control unit (ECU), such as the headlight controller, which then performs the flashing and installation.

[0005] Reboot to take effect: After installation, the ECU will restart and the new software will start running.

[0006] In the existing solutions mentioned above, the upgrade decision is mainly triggered by cloud-based strategies or simple vehicle status checks (such as vehicle speed and gear). The verification after the upgrade depends on the subsequent transmission of vehicle operation data and cloud-based big data analysis, or the discovery of problems during the next round of diagnosis.

[0007] The main defects and shortcomings of the existing technology are as follows: 1. Upgrade decisions are out of sync with real-time context: Existing solutions are usually based on static rules (such as "the vehicle must be turned off"), lacking a comprehensive assessment of dynamic and complex environments (such as when the vehicle battery is about to run out, when sensors malfunction occasionally, or when the vehicle is in a complex lighting environment). This may trigger upgrades at inappropriate times, introducing safety risks or causing upgrade failures.

[0008] 2. Lack of immediate, closed-loop effect verification: After the upgrade, the system cannot immediately and quantitatively verify whether the new software works as expected. Traditional methods that rely on "no fault codes after restart" or subsequent data feedback analysis have long feedback cycles and cannot promptly detect potential functional or performance degradation issues, such as increased beam response delay caused by the new algorithm.

[0009] 3. Slow problem response and recovery: Once compatibility or performance issues arise after an upgrade, existing solutions often require waiting for user feedback or periodic diagnostics before triggering a rollback or another upgrade, affecting user experience and system reliability. Summary of the Invention

[0010] The technical problem this invention aims to solve is to overcome the shortcomings of existing technologies. This invention provides a software upgrade method for an intelligent self-closed-loop vehicle lighting controller, addressing the technical issues of insufficient intelligent upgrade decision-making, lack of real-time closed-loop verification after upgrade, and slow problem response in existing OTA upgrade methods for vehicle lighting controllers. This intelligent self-closed-loop software upgrade method can autonomously assess upgrade suitability, safely execute the upgrade, and quantify and verify the upgrade effect in real time, thereby significantly improving the safety, success rate, and user experience of the upgrade process.

[0011] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows: A software upgrade method for an intelligent self-closed-loop vehicle light controller includes the following steps: Step S1: The cloud-based upgrade management platform distributes the software upgrade package and upgrade strategy configuration file; Step S2: Perform a pre-upgrade self-assessment based on the current vehicle and environmental conditions; Step S3: After the pre-upgrade self-assessment passes, silently download and install the software upgrade package; Step S4: After the software upgrade package is installed and upgraded, self-verification is performed using self-verification test cases to calculate the confidence level of functional correctness, performance improvement, and behavior consistency. Step S5: Combine the confidence scores for functional correctness, performance improvement, and behavioral consistency to generate the final total confidence score after the upgrade; Step S6: Make an automatic decision based on the total confidence score after the upgrade and the preset threshold in the upgrade strategy configuration file.

[0012] Furthermore, in step S1, the cloud-based upgrade management platform distributes the software upgrade package and upgrade strategy configuration file, specifically including the following steps: The cloud-based upgrade management platform pushes software upgrade packages based on vehicle model, hardware version, and software baseline, and simultaneously distributes a structured upgrade strategy configuration file. The upgrade strategy configuration file includes a pre-evaluation strategy, a self-verification use case set, and a decision threshold. The pre-evaluation strategy includes the minimum battery capacity allowed for upgrades, the list of allowed vehicle states, and the ambient light stability threshold. The self-verification test case set includes a list of functional tests that need to be run automatically after the upgrade. The decision thresholds include the pre-evaluation pass threshold, the upgrade success confidence threshold, and the automatic rollback threshold.

[0013] Furthermore, in step S2, a pre-upgrade self-assessment is performed based on the current vehicle and environmental conditions, specifically including the following steps: After the vehicle gateway transmits the software upgrade package and upgrade strategy configuration file to the intelligent vehicle lighting controller, the intelligent vehicle lighting controller starts the pre-evaluation phase before performing the upgrade. During the pre-evaluation phase, the intelligent vehicle lighting controller collects multiple vehicle and environmental status data in real time, and uses the issued strategy parameters to calculate and evaluate whether the current time t is suitable for upgrading through a dynamic confidence model.

[0014] Furthermore, the expression for the dynamic confidence model is as follows: ; Among them, C pre (t) represents the pre-upgrade dynamic confidence level at time t; w i Let i be the importance of the i-th evaluation factor in the overall decision-making process; S i (t) represents the actual data collected by the intelligent vehicle lighting controller at time t via the vehicle bus and vehicle status sensors; T i The strategy parameters issued by the cloud-based upgrade management platform for the i-th evaluation factor; f i () is the normalized evaluation function for the i-th evaluation factor; The intelligent vehicle lighting controller continuously calculates C pre (t), when C pre The value of (t) is higher than the pre-evaluated pass threshold C within the preset continuous time window. pre If the threshold is reached, the upgrade conditions are deemed met.

[0015] Furthermore, in step S3, after the pre-upgrade self-assessment passes, the software upgrade package is silently downloaded and installed, specifically including the following steps: Once the current time t passes the evaluation and is suitable for upgrading, the system will download, verify, and install the software upgrade package in the background; Before installation, the system will record the key performance parameters of the current software version as baseline data for subsequent verification and upgrades.

[0016] Furthermore, in step S4, after the software upgrade package is installed and upgraded, self-verification is performed using self-verification test cases to calculate the confidence level of functional correctness, performance improvement, and behavioral consistency. Specifically, this includes the following steps: After the software upgrade package is installed, the intelligent vehicle lighting controller automatically runs the preset self-verification test cases in the new software environment, performs functional correctness verification and calculates the functional correctness confidence, performs performance index verification and calculates the performance improvement confidence. The control output results under typical scenarios are sent to the simulation prediction service in the cloud for behavior consistency verification, and the behavior consistency confidence level is calculated.

[0017] Furthermore, the process of verifying the correctness of the execution function and calculating the confidence level of the correctness of the function specifically includes the following steps: Execute all self-verification test cases defined in the upgrade strategy configuration file, and finally output the self-test result as the functional correctness confidence level C. functional The confidence level C for the correctness of the function functional The expression is as follows: C functional = ; The process of verifying the performance metrics and calculating the confidence level of the performance improvement specifically includes the following steps: Real-world testing was conducted on key performance indicators, and the results were compared with previously recorded upgrade baseline data. The final self-test output is a performance improvement confidence level C. performance The performance improvement confidence level C performance The expression is as follows: C performance = ; The process of sending the control output results under typical scenarios to the cloud-based simulation prediction service for behavioral consistency verification and calculating the behavioral consistency confidence level includes the following steps: The intelligent vehicle lighting controller simulates a typical driving scenario, generates control commands based on the new algorithm in the software upgrade package, and sends the new simulation results to the cloud simulation prediction service. Then, control commands are generated based on the algorithm of the old version of the software, and the simulation results of the old version are sent to the cloud simulation prediction service; Finally, the cloud-based simulation prediction service calculates the similarity between the simulation results of the old and new versions, and then calculates the behavioral consistency confidence score C based on the similarity. consistency .

[0018] Furthermore, in step S5, the confidence scores for functional correctness, performance improvement, and behavioral consistency are combined to generate the final overall confidence score after the upgrade, specifically including the following steps: The confidence scores for functional correctness, performance improvement, and behavioral consistency are combined to generate a final overall confidence score C after the upgrade. postThe upgraded total confidence score C post The calculation formula is as follows: ; Among them, C post This represents the total confidence level after the upgrade. C functional C represents the confidence level for functional correctness. performance To improve confidence levels for performance; C consistency Confidence level for behavioral consistency; α, β, and γ are weight coefficients, α+β+γ=1.

[0019] Furthermore, in step S6, automatic decision-making is performed based on the total confidence score after the upgrade and the preset threshold in the upgrade strategy configuration file, specifically including the following steps: If C post If the upgrade success confidence threshold is reached, the upgrade is confirmed as successful. If C post <Upgrade success confidence threshold, but C post If the value is greater than or equal to the automatic rollback threshold, it is determined that there is uncertainty in the upgrade. The system will keep the new software running but trigger an alarm, prompting manual online review. If C post If the value falls below the automatic rollback threshold, the upgrade is deemed to have failed or poses a serious risk.

[0020] By adopting the above technical solution, the present invention enables the intelligent vehicle lighting controller to autonomously and intelligently assess whether the timing of the upgrade is safe and appropriate during the software upgrade process, and to immediately conduct quantitative and multi-dimensional effect verification after the upgrade is completed, thereby forming a complete "decision-execution-verification-decision" closed loop, ultimately achieving a significant improvement in the safety, success rate and system reliability of the upgrade process. Attached Figure Description

[0021] Figure 1 This is a flowchart of a software upgrade method for an intelligent self-closed-loop vehicle light controller according to the present invention. Figure 2 This is a system principle block diagram of the software upgrade method for the intelligent self-closed-loop vehicle light controller of the present invention. Detailed Implementation

[0022] To make the content of this invention easier to understand, the invention will be further described in detail below with reference to specific embodiments and accompanying drawings.

[0023] like Figure 1 , 2As shown, this embodiment provides a software upgrade method for an intelligent self-closed-loop vehicle lighting controller. This method operates in a system consisting of a cloud server, an in-vehicle gateway, an intelligent vehicle lighting controller, and vehicle status sensors. It is an intelligent self-closed-loop software upgrade method under a cloud-vehicle collaborative architecture. It transforms a passive software flashing task into a proactive intelligent process driven by data and possessing perception, evaluation, verification, and decision-making capabilities.

[0024] Includes the following steps: Step S1: The cloud-based upgrade management platform distributes the software upgrade package and upgrade strategy configuration file; specifically, it includes the following steps: Based on vehicle model, hardware version, and software baseline, the cloud-based upgrade management platform not only pushes software upgrade packages (containing new firmware or application software) but also distributes a structured upgrade strategy configuration file. This configuration file uses JSON format and defines the pre-evaluation strategy, self-verification test cases, and decision thresholds required for the upgrade. The specific content is as follows: The pre-assessment strategy includes: the minimum battery capacity allowed for upgrades, the list of allowed vehicle states, and the ambient light stability threshold.

[0025] The self-verification test case set includes: a list of functional tests that need to be run automatically after the upgrade. The functional test table is shown in Table 1. Table 1

[0026] The decision thresholds include: pre-assessment pass threshold, upgrade success confidence threshold, and automatic rollback threshold.

[0027] The pre-assessment pass threshold is the core quantitative standard for the "entry threshold" before an upgrade. Only when all indicators such as vehicle hardware status (e.g., battery level, fault codes), operating mode (e.g., parking, driving), and environmental conditions (e.g., light intensity, temperature) meet or exceed this threshold will the upgrade proceed to the "upgrade execution" stage. For example, if the battery level pre-assessment pass threshold is 15%, and the actual battery level is only 14%, the system will determine "high risk of insufficient battery power" and directly terminate the upgrade to avoid equipment failure due to power outages during the upgrade process.

[0028] Upgrade success confidence threshold: This is a key criterion for "effectiveness acceptance" after the upgrade. The system compares data such as "core function test pass rate, service startup integrity, and performance indicator compliance rate" with this threshold. If the verification pass rate of core functions (such as power control and safety assistance) after the upgrade is greater than or equal to this threshold, the upgrade is considered "successful"; if it is insufficient, it indicates that there are stability risks in the upgrade, and a rollback needs to be triggered. For example, if the upgrade success confidence threshold is 95%, but the core service startup success rate after the upgrade is only 93%, the system will determine that the "upgrade effect is unsatisfactory" and trigger a rollback to ensure reliability.

[0029] Automatic rollback threshold: This is a mechanism to mitigate upgrade risks. During or after an upgrade (e.g., download interruption, installation failure), if detected risk indicators (number of failed tasks, failure frequency, abnormal log volume) exceed this threshold, the system will automatically roll back to the stable version before the upgrade, preventing localized failures from escalating into global failures. For example, if the automatic rollback threshold is set to "upgrade failure rate > 5%", when more than 5% of critical upgrade tasks are detected as incomplete, an immediate rollback will occur, preventing minor faults from causing system-wide failures.

[0030] Step S2: Perform a pre-upgrade self-assessment based on the current vehicle and environmental conditions; specifically including the following steps: After the vehicle gateway transmits the software upgrade package and upgrade policy configuration file to the intelligent headlight controller, the intelligent headlight controller does not immediately execute the upgrade, but instead initiates a pre-evaluation phase. During the pre-evaluation phase, the intelligent headlight controller collects multiple vehicle and environmental status data in real time. i (t), and utilize the issued strategy parameter T i The evaluation uses a dynamic confidence model to assess whether the current time t is suitable for upgrading. The evaluation employs a dynamic confidence model: ; Among them, C pre (t) represents the pre-upgrade dynamic confidence level at time t. The output value is between 0 and 1. The higher the value, the more the current context environment meets the conditions for safe and stable upgrade. It is a dynamic value that changes with time t. w i The importance of the i-th evaluation factor (such as battery power and vehicle speed) in the overall decision-making is not derived from a fixed value, but is obtained by training the model through machine learning (such as logistic regression analysis) on a large amount of historical upgrade success and failure case data, so that the dynamic confidence model can adapt to different vehicle models and user habits. S i (t) represents the actual data collected by the intelligent vehicle lighting controller at time t through the vehicle bus (CAN bus, LIN bus, or Ethernet bus) and vehicle status sensors, such as: the current actual battery charge percentage (e.g., 70%), current vehicle speed (e.g., 0 km / h), current gear (e.g., P gear), fault code status (e.g., none), ambient light intensity (e.g., 100 Lux), etc. T i The strategy parameters issued by the cloud-based upgrade management platform for the i-th evaluation factor, such as the minimum safety standard value that the battery charge must reach, such as 30%, are derived from vehicle engineering specifications and safety design guidelines. f i() represents the normalized evaluation function for the i-th evaluation factor, for example: for battery SOC (current remaining charge percentage), f soc = max(0, (current battery SOC-T) soc min) / (100-T soc For example: when the battery SOC is 70% and the minimum requirement is 30%, f soc ≈0.57; The intelligent vehicle lighting controller continuously calculates C pre (t), only if C pre The value of (t) is higher than the pre-evaluated pass threshold C within a preset continuous time window (e.g., 30 seconds). pre The system only determines that the upgrade conditions are met and proceeds to the next step when the threshold (e.g., 0.85) is reached; otherwise, it continues to monitor the status.

[0031] Step S3: After the pre-upgrade self-assessment passes, silently download and install the software upgrade package; specifically, it includes the following steps: Once the current time t passes the evaluation and is deemed suitable for upgrade, the system downloads, verifies, and installs the software upgrade package in the background. Before installation, the system records key performance parameters of the current software version (such as standard command response time, memory usage, etc.) as baseline data for subsequent verification.

[0032] Baseline data refers to key performance indicators measured against the vehicle's existing (older version) software prior to the software upgrade. For example, the average or typical value of the "response time when switching from high beam to ADB mode" measured and recorded multiple times before the upgrade.

[0033] Step S4: After the software upgrade package is installed and upgraded, self-verification is performed using self-verification test cases to calculate the confidence scores for functional correctness, performance improvement, and behavioral consistency. Specifically, this includes the following steps: After the software upgrade package is installed, the intelligent vehicle lighting controller automatically runs a series of preset self-verification test cases (such as lighting up all LED pixels and performing standard beam mode switching) in the new software environment, instead of waiting for user interaction. This process involves quantitative testing from three dimensions: functional correctness verification and calculation of functional correctness confidence; performance index verification and calculation of performance improvement confidence; and sending the control output results under typical scenarios to the cloud-based simulation prediction service for behavioral consistency verification and calculation of behavioral consistency confidence.

[0034] 1. Perform functional correctness verification and calculate the functional correctness confidence score, which includes the following steps: Automatically and sequentially execute all self-verification test cases defined in the upgrade strategy configuration file (e.g., sequentially illuminating each lighting zone, cyclically switching all predefined beam modes, testing communication interfaces, etc.), and finally output the self-test result as a functional correctness confidence level C. functional C functional = .

[0035] 2. Perform performance metric verification and calculate the confidence level of performance improvement, specifically including the following steps: Actual tests were conducted on key performance indicators (such as response time when switching from high beam to ADB mode), and the results were compared with previously recorded upgrade baseline data (for indicators like response time where "lower is better," a weighted average was taken if there were multiple indicators). The final self-test result output is a performance improvement confidence level C. performance C performance = .

[0036] 3. Send the control output results under typical scenarios to the cloud-based simulation prediction service for behavioral consistency verification, and calculate the behavioral consistency confidence score. The specific steps include the following: The intelligent vehicle lighting controller simulates a typical driving scenario (such as an oncoming vehicle), generates control commands (such as an anti-glare shading matrix) based on the new algorithm in the software upgrade package, and sends the new simulation results to the cloud simulation prediction service. Then, it generates control commands based on the algorithm of the old software version and sends the simulation results of the old version to the cloud simulation prediction service. Finally, the cloud simulation prediction service calculates the similarity between the simulation results of the two versions, and then calculates the behavioral consistency confidence score C based on the similarity. consistency The specific similarity evaluation scheme is as follows: First, the output of the intelligent headlight controller needs to be converted into a computable mathematical vector. For example, for an ADB (Adaptive High Beam) system, its core output may be a shading matrix generated for the scene in front at a certain moment. This shading matrix can be represented as a two-dimensional array, where each element represents the brightness value of a pixel or region (e.g., 0 represents shading, and 255 represents full brightness).

[0037] Next, vectorization is performed, flattening this two-dimensional array into a one-dimensional vector. For example, a 10x10 shading matrix is ​​transformed into a vector containing 100 values.

[0038] Next, the cosine similarity between the two shading matrices under the old and new versions is calculated. This is done by calculating the difference in direction (cosine of the included angle) between the two one-dimensional vectors under the old and new versions to evaluate the cosine similarity. The formula for calculating cosine similarity is as follows: ; Where A is the vector of the actual output after the upgrade; B is the vector output by the simulation prediction in the old version; A·B is the dot product of vectors A and B (the sum of corresponding elements after multiplying them). ||A|| is the magnitude (Euclidean length) of vector A, and ||B|| is the magnitude of vector B; Furthermore, regarding the calculation results: If the similarity = 1, it means that the two one-dimensional vectors in the old and new versions have the same direction, which means that the behavior of the old and new software packages is completely consistent. If similarity≈0, it means that the two one-dimensional vectors under the old and new versions are nearly orthogonal, which means that the behavioral patterns are very different. If the similarity is -1, it means that the two one-dimensional vectors in the old and new versions are in completely opposite directions.

[0039] Specifically, in vehicle lighting control scenarios, taking anti-glare shading as an example, non-negative similarity is usually the focus. Therefore, the calculation result can be obtained through a simple linear mapping, i.e., C consistency = .

[0040] Step S5: Combine the confidence scores for functional correctness, performance improvement, and behavioral consistency to generate the final overall confidence score after the upgrade; specifically, this includes the following steps: The confidence scores for functional correctness, performance improvement, and behavioral consistency are combined to generate a final overall confidence score C after the upgrade. post The overall confidence score after the upgrade is C. post The calculation formula is as follows: ; Among them, C post The total confidence level after the upgrade is in the range [0,1]. C functional C represents the confidence level for functional correctness. performance To improve confidence levels for performance; C consistency Confidence level for behavioral consistency; α, β, and γ are the weighting coefficients for the three dimensions of function, performance, and consistency, respectively. α + β + γ = 1. Their values ​​are determined by domain experts based on principles such as the safety primacy of the vehicle lighting system and user experience (e.g., α = 0.5, β = 0.3, γ = 0.2).

[0041] Step S6: Make an automatic decision based on the total confidence score after the upgrade and the preset threshold in the upgrade strategy configuration file to form a closed loop; specifically including the following steps: If C post If the upgrade success confidence threshold (e.g., 0.90) is reached, the upgrade is considered successful. If C post <Upgrade success confidence threshold, but C post If the value is ≥ the automatic rollback threshold (e.g., 0.75), the upgrade is deemed to have uncertainty. The system will keep the new software running but trigger an alarm, prompting the user or server to conduct a manual online review via the vehicle's infotainment system.

[0042] If C post If the value falls below the automatic rollback threshold, the upgrade is deemed a failure or a serious risk exists. The system automatically triggers a fast rollback mechanism, restoring to the stable version before the upgrade, and reports detailed failure logs and verification data.

[0043] The technical solution described in this invention includes all logs, status data, and C of the entire process described above. post All scores are sent back to the cloud for iterative optimization of evaluation factors. i And policy thresholds enable the system to continuously learn and evolve.

[0044] The specific embodiments described above further illustrate the technical problems, technical solutions, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A software upgrade method for an intelligent self-closed-loop vehicle light controller, characterized in that, Includes the following steps: Step S1: The cloud-based upgrade management platform distributes the software upgrade package and upgrade strategy configuration file; Step S2: Perform a pre-upgrade self-assessment based on the current vehicle and environmental conditions; Step S3: After the pre-upgrade self-assessment passes, silently download and install the software upgrade package; Step S4: After the software upgrade package is installed and upgraded, self-verification is performed using self-verification test cases to calculate the confidence level of functional correctness, performance improvement, and behavior consistency. Step S5: Combine the confidence scores for functional correctness, performance improvement, and behavioral consistency to generate the final total confidence score after the upgrade; Step S6: Make an automatic decision based on the total confidence score after the upgrade and the preset threshold in the upgrade strategy configuration file.

2. The software upgrade method for the intelligent self-closed-loop vehicle light controller according to claim 1, characterized in that, In step S1, the cloud-based upgrade management platform distributes the software upgrade package and upgrade strategy configuration file, specifically including the following steps: The cloud-based upgrade management platform pushes software upgrade packages based on vehicle model, hardware version, and software baseline, and simultaneously distributes a structured upgrade strategy configuration file. The upgrade strategy configuration file includes a pre-evaluation strategy, a self-verification use case set, and a decision threshold. The pre-evaluation strategy includes the minimum battery capacity allowed for upgrades, the list of allowed vehicle states, and the ambient light stability threshold. The self-verification test case set includes a list of functional tests that need to be run automatically after the upgrade. The decision thresholds include the pre-evaluation pass threshold, the upgrade success confidence threshold, and the automatic rollback threshold.

3. The software upgrade method for the intelligent self-closed-loop vehicle light controller according to claim 1, characterized in that, In step S2, a pre-upgrade self-assessment is performed based on the current vehicle and environmental conditions, specifically including the following steps: After the vehicle gateway transmits the software upgrade package and upgrade strategy configuration file to the intelligent vehicle lighting controller, the intelligent vehicle lighting controller starts the pre-evaluation phase before performing the upgrade. During the pre-evaluation phase, the intelligent vehicle lighting controller collects multiple vehicle and environmental status data in real time and uses the issued strategy parameters to calculate and evaluate whether the current time t is suitable for upgrading through a dynamic confidence model.

4. The software upgrade method for the intelligent self-closed-loop vehicle light controller according to claim 3, characterized in that, The expression for the dynamic confidence model is as follows: ； Among them, C pre (t) represents the pre-upgrade dynamic confidence level at time t; w i Let i be the importance of the i-th evaluation factor in the overall decision-making process; S i (t) represents the actual data collected by the intelligent vehicle lighting controller at time t via the vehicle bus and vehicle status sensors; T i The strategy parameters issued by the cloud-based upgrade management platform for the i-th evaluation factor; f i () is the normalized evaluation function for the i-th evaluation factor; The intelligent vehicle lighting controller continuously calculates C pre (t), when C pre The value of (t) is higher than the pre-evaluated pass threshold C within the preset continuous time window. pre If the threshold is reached, the upgrade conditions are deemed met.

5. The software upgrade method for the intelligent self-closed-loop vehicle light controller according to claim 1, characterized in that, In step S3, after the pre-upgrade self-assessment passes, the software upgrade package is silently downloaded and installed, specifically including the following steps: Once the current time t passes the evaluation and is suitable for upgrading, the system will download, verify, and install the software upgrade package in the background; Before installation, the system will record the key performance parameters of the current software version as baseline data for subsequent verification and upgrades.

6. The software upgrade method for the intelligent self-closed-loop vehicle light controller according to claim 1, characterized in that, In step S4, after the software upgrade package is installed and upgraded, self-verification is performed using self-verification test cases to calculate the confidence levels of functional correctness, performance improvement, and behavioral consistency. Specifically, this includes the following steps: After the software upgrade package is installed, the intelligent vehicle lighting controller automatically runs preset self-verification test cases in the new software environment, performs functional correctness verification and calculates the functional correctness confidence, performs performance index verification and calculates the performance improvement confidence. The control output results under typical scenarios are sent to the simulation prediction service in the cloud for behavior consistency verification, and the behavior consistency confidence level is calculated.

7. The software upgrade method for the intelligent self-closed-loop vehicle light controller according to claim 6, characterized in that, The process of verifying the correctness of the execution function and calculating the confidence level of the correctness of the function specifically includes the following steps: Execute all self-verification test cases defined in the upgrade strategy configuration file, and finally output the self-test result as the functional correctness confidence level C. functional The confidence level C for the correctness of the function functional The expression is as follows: C functional = ； The process of verifying the performance metrics and calculating the confidence level of the performance improvement specifically includes the following steps: Real-world testing was conducted on key performance indicators, and the results were compared with previously recorded upgrade baseline data. The final self-test output is a performance improvement confidence level C. performance The performance improvement confidence level C performance The expression is as follows: C performance = ； The process of sending the control output results under typical scenarios to the cloud-based simulation prediction service for behavioral consistency verification and calculating the behavioral consistency confidence level includes the following steps: The intelligent vehicle lighting controller simulates a typical driving scenario, generates control commands based on the new algorithm in the software upgrade package, and sends the new simulation results to the cloud simulation prediction service. Then, control commands are generated based on the algorithm of the old version of the software, and the simulation results of the old version are sent to the cloud simulation prediction service; Finally, the cloud-based simulation prediction service calculates the similarity between the simulation results of the old and new versions, and then calculates the behavioral consistency confidence score C based on the similarity. consistency .

8. The software upgrade method for the intelligent self-closed-loop vehicle light controller according to claim 1, characterized in that, In step S5, the confidence scores for functional correctness, performance improvement, and behavioral consistency are combined to generate the final overall confidence score after the upgrade. This process includes the following steps: The confidence scores for functional correctness, performance improvement, and behavioral consistency are combined to generate a final overall confidence score C after the upgrade. post The upgraded total confidence score C post The calculation formula is as follows: ； Among them, C post This represents the total confidence level after the upgrade. C functional C represents the confidence level for functional correctness. performance To improve confidence levels for performance; C consistency Confidence level for behavioral consistency; α, β, and γ are weight coefficients, α+β+γ=1.

9. The software upgrade method for the intelligent self-closed-loop vehicle light controller according to claim 1, characterized in that, In step S6, an automatic decision is made based on the total confidence score after the upgrade and the preset threshold in the upgrade strategy configuration file, specifically including the following steps: If C post If the upgrade success confidence threshold is reached, the upgrade is confirmed as successful. If C post <Upgrade success confidence threshold, but C post If the value is greater than or equal to the automatic rollback threshold, it is determined that there is uncertainty in the upgrade. The system will keep the new software running but trigger an alarm, prompting manual online review. If C post If the value falls below the automatic rollback threshold, the upgrade is deemed to have failed or poses a serious risk.

Images