Automatic driving carrying system in intelligent traffic internet of things

Abstract

The invention belongs to the technical field of information technology, and particularly discloses an automatic driving carrying system in an intelligent traffic internet of things, which is used by the following steps of: synchronously executing biological feature recognition and digital certificate confirmation by adopting a parallel processing algorithm by collecting processing duration data of multiple verification modes to obtain a unified time reference; according to a unified time reference, acquiring a real-time data stream in a complex traffic scene, and if it is judged that the time difference is greater than a preset threshold value, activating a priority ranking algorithm, and determining a high-priority verification path; integrating the biological characteristics and the certificate information by adopting a data fusion technology through a high-priority verification path to obtain a preliminary identity confirmation result; the invention aims to solve the problem that in the prior art, the processing duration difference of biological feature recognition and digital certificate confirmation in a high dynamic environment is large, so that verification is delayed, and emergency response is not timely.

Description

Technical Field

[0001] This invention relates to the technical field of information technology, specifically to an autonomous driving vehicle system in the Internet of Things for intelligent transportation. Background Technology

[0002] In the field of intelligent transportation IoT, research on autonomous driving systems is of paramount importance. They are not only a core pillar of future transportation but also key to improving road safety and transport efficiency. With accelerating urbanization and increasing traffic demand, autonomous driving technology is highly anticipated, poised to effectively alleviate traffic congestion and reduce accidents caused by human error. However, technological development in this field still faces numerous challenges, urgently requiring breakthrough innovations to drive its practical application.

[0003] Currently, while many autonomous driving systems have made progress in identity verification and safety control, a common problem is the difficulty in balancing efficiency and security in the verification process. Existing methods often prioritize high security while neglecting the impact of verification speed on real-time operation. Especially in complex traffic scenarios, lag in system response can directly lead to safety hazards. This limitation makes the system inadequate when facing diverse needs, particularly in scenarios requiring rapid decision-making, where efficiency becomes a major bottleneck. Focusing on the technical challenges, the time difference in the identity verification process becomes a significant factor affecting the overall system performance. Different verification methods require varying amounts of time; for example, the processing speed for identity verification via biometrics or digital certificates varies depending on the technology. This time difference makes it difficult for the system to coordinate multiple verifications in parallel, thus affecting the vehicle's responsiveness at critical moments. A deeper problem is that when time differences accumulate, the system may be unable to complete identity verification within a short time, especially in emergency situations. Delays can prevent the vehicle from performing necessary operations in a timely manner, posing potential risks.

[0004] Therefore, optimizing the time differences between different verification methods while ensuring identity verification security, and ensuring that the system can respond quickly in various scenarios, has become a critical issue that autonomous driving vehicle systems urgently need to address. This problem is particularly prominent in actual business operations. For example, during peak hours or in emergencies, vehicles need to complete identity verification and make decisions within a very short time. If the system is delayed due to excessive verification time, it may miss the best response opportunity, affecting overall traffic flow and safety. Summary of the Invention

[0005] This invention provides an autonomous driving vehicle system in the Internet of Things for intelligent transportation, aiming to solve the problem in the prior art where the processing time for biometric recognition and digital certificate verification differs greatly in highly dynamic environments, leading to verification delays and untimely responses to emergencies.

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows: An autonomous driving vehicle system in the Internet of Things for intelligent transportation includes: collecting processing time data from multiple verification methods; simultaneously executing biometric recognition and digital certificate verification using a parallel processing algorithm to obtain a unified time base; acquiring real-time data streams under complex traffic scenarios based on the unified time base; activating a priority ranking algorithm to determine a high-priority verification path if the time difference exceeds a preset threshold; integrating biometric and certificate information using data fusion technology through the high-priority verification path to obtain a preliminary identity verification result; acquiring the vehicle's current speed and surrounding environment sensor data based on the preliminary identity verification result; adjusting the verification sequence to obtain an accelerated response sequence if a sudden situation signal occurs; compressing and encoding multi-source data using an information processing stage to determine an optimized verification time window based on the accelerated response sequence; acquiring historical delay records from the system log based on the optimized verification time window; triggering an adaptive adjustment mechanism to obtain real-time decision parameters if the historical delay exceeds the window limit; and combining the verification result with vehicle control commands using output fusion technology based on the real-time decision parameters to obtain the final safety response command.

[0007] In one aspect of this disclosure, the method of collecting processing time data from multiple verification methods and employing parallel processing algorithms to synchronously execute biometric identification and digital certificate verification to obtain a unified time reference includes: Processing time data is obtained through interfaces of multiple verification methods. Initial timestamps are recorded for the verification processes of biometrics and digital certificates respectively, resulting in a preliminary set of time data. Based on the initial time data set, the tasks of biometric identification and digital certificate verification are allocated using a parallel processing approach. A pre-set task queue mechanism is used to determine the execution order and time allocation of each task. If the processing time of a biometric identification task exceeds a preset threshold, the priority of the task will be adjusted to a secondary queue, resources will be reallocated, and the adjusted execution time record will be obtained. By adjusting the execution time records, the parallel tasks of biometric identification and digital certificate verification are executed simultaneously. The completion time of each task is compared to determine whether there is a time deviation. If a time deviation is detected, the deviation data is corrected by the time calibration module, and the calibrated time data is obtained by aligning with the standard timestamp. Based on the calibrated time data, the processing results of all verification methods are integrated to generate a unified time benchmark and determine the final synchronization execution standard; By using a unified time benchmark, the processing time of subsequent verification tasks is monitored, real-time time data updates are obtained, and it is determined whether the preset synchronization standard is met.

[0008] In one aspect of this disclosure, the step of acquiring real-time data streams under complex traffic scenarios based on a unified time reference, and determining that if the time difference is greater than a preset threshold, then activating a priority ranking algorithm to determine a high-priority verification path includes: By using a unified time base, real-time data streams are obtained from complex traffic scenarios. Sensor networks and data transmission modules are used to complete data collection, resulting in a preliminary traffic data set. For the initial traffic dataset, the time difference between each data point is detected. If the time difference exceeds a preset threshold, it is marked as an abnormal data point, and the abnormal data subset that needs further processing is determined. Based on the subset of abnormal data, a priority sorting mechanism is triggered, and a pre-established decision tree model is used to classify the abnormal data points and determine the set of high-priority data points. From the set of high-priority data points, extract the corresponding traffic scenario information, analyze the correlation between data points and paths, and obtain a candidate list of high-priority verification paths; Based on the candidate list of high-priority verification paths, and combined with the latest updates in the real-time data stream, the path that best matches the current traffic scenario is selected, and the final verification path scheme is determined. The final verified route plan is used to generate route execution instructions, which are then transmitted to the relevant traffic control system to complete the automated deployment of route verification.

[0009] In one aspect of this disclosure, the method of integrating biometric features and certificate information through a high-priority verification path using data fusion technology to obtain preliminary identity verification results includes: Biometric and certificate data are acquired from a designated data source through a preset high-priority channel to complete the initial information collection and obtain the dataset to be processed. Based on the collected dataset, data fusion technology was used to process the biometric and certificate data in a unified format to generate a standardized information set and determine the preliminary integration results. If the biometric data in the standardized information set matches the certificate data to a preset threshold, it is determined that the consistency has passed and a consistency verification identifier is obtained. By using consistency verification identifiers and combining them with security verification mechanisms, a deep comparison of the standardized information set is performed to determine whether there is any abnormal data and obtain the security verification result. If the security verification result shows no abnormal data, the standardized information set will be finally confirmed through technical means to obtain the identity confirmation conclusion. Based on the identity verification results, record the processing efficiency data, generate a processing log, and determine the final output. By processing logs and archiving and storing the final output, the identity verification process is completed.

[0010] In one aspect of this disclosure, the step of acquiring the vehicle's current speed and surrounding environment sensor data based on the preliminary identity verification result, and determining whether a sudden situation signal occurs, and then adjusting the verification sequence to obtain an acceleration response sequence, includes: Acquire current vehicle status data and surrounding perception information. Collect vehicle speed and environmental data through sensors and store them in a temporary database to obtain a preliminary status dataset. For the initial state dataset, a preset threshold is used for signal detection. If a sudden situation signal is detected that exceeds the safe range, the situation judgment logic is triggered to determine whether there is a potential risk. Based on the situation assessment, if an emergency is confirmed, the identity verification information is extracted and compared with the current status data to obtain a verification consistency result. By verifying the consistency results, adjusting the verification process, generating temporary response strategies for emergencies, and determining the priority sequence for accelerating the response; For the acceleration response priority sequence, real-time updates of vehicle speed and environmental data are obtained, and specific response sequence instructions are generated in combination with temporary response strategies. According to the response sequence instructions, the data is transmitted to the vehicle control module to execute accelerated response operations and complete the immediate handling of emergencies.

[0011] In one aspect of this disclosure, the step of accelerating the response sequence and using an information processing stage to compress and encode multi-source data to determine an optimized verification time window includes: By collecting and initially organizing multi-source data, raw datasets from different sources are obtained, and the response sequences are initially classified to obtain a structured dataset. The classified dataset is used to accelerate the processing based on the characteristics of the response sequence. The data is filtered using a preset threshold to determine a high-priority subset of data. Based on the selected high-priority data subset, the information processing step is carried out to standardize the data and obtain a unified formatted dataset; By formatting the dataset and applying compression encoding techniques, the data source and encoding method are optimized to obtain compressed data units. Based on the compressed data units, combined with processing steps and sequence analysis, the dynamic range of the verification window is constructed, the rationality of the window adjustment is judged, and the preliminary time window range is determined. If the fluctuation of the response sequence within the initial time window exceeds the preset threshold, the window adjustment parameters are corrected to obtain an optimized verification time window. By optimizing the verification time window and combining it with the time optimization objective, the subsequent processing steps of multi-source data are recorded to obtain the final window configuration result.

[0012] In one aspect of this disclosure, the step of obtaining historical delay records from the system log based on the optimized verification time window, determining whether the historical delay exceeds the window limit, and then triggering an adaptive adjustment mechanism to obtain real-time decision parameters includes: Historical latency data is extracted from the system logs. The extracted data is then preliminarily cleaned to remove abnormal records, resulting in a cleaned latency dataset. Based on the organized delay dataset, and combined with the preset time window and window limit, the delay data is compared one by one. If the delay data exceeds the window limit, it is marked as an abnormal delay record, thus obtaining an abnormal delay set. For the set of abnormal delays, analyze its distribution, use statistical tools to calculate the frequency and concentrated periods of abnormal delays, and determine the key time points of abnormal delays; Based on key time points, an adaptive adjustment mechanism is triggered to obtain real-time data streams of the system's operating status. If a delay trend similar to that of the key time points exists in the real-time data stream, an adjustment signal is generated. By adjusting the signal, a pre-established decision model is invoked to obtain real-time decision parameters and output a parameter configuration scheme optimized for latency. By adopting a parameter configuration scheme, updating the system operation rules, and recording the changes in latency data after the update, the optimized latency monitoring results are obtained.

[0013] In one aspect of this disclosure, the method of combining verification results with vehicle control commands using output fusion technology based on real-time decision parameters to obtain a final safety response command includes: Real-time vehicle operation data and environmental information are collected through sensors and communication modules to obtain the raw dataset of decision parameters; Based on the collected raw dataset, preprocessing techniques are used to clean and standardize the data to obtain a structured set of decision parameters. If outliers exist in the structured set of decision parameters, they are filtered through a preset threshold range to determine the effective combination of decision parameters. For effective combinations of decision parameters, logical judgment rules are used to analyze their degree of matching with security standards and determine whether to trigger the verification process. If the verification process is triggered, the combination of decision parameters and the vehicle status are compared using a preset verification model to obtain verification result data. Based on the verification results, the output fusion method is used to integrate them with the preset vehicle control commands to generate the final safety response command.

[0014] The generated safety response command is transmitted to the vehicle execution unit through the communication interface to complete the command issuance and response.

[0015] According to one aspect of this disclosure, an autonomous driving vehicle system in an intelligent transportation Internet of Things includes: The time benchmark generation module is used to collect processing time data of various verification methods. It uses a parallel processing algorithm to execute biometric recognition and digital certificate verification synchronously to obtain a unified time benchmark. The priority path determination module is used to obtain real-time data streams in complex traffic scenarios based on the unified time base, and determine that if the time difference is greater than a preset threshold, the priority sorting algorithm is activated to determine the high-priority verification path. The identity verification module is used to integrate biometric features and certificate information through data fusion technology via the high-priority verification path to obtain a preliminary identity verification result. The response sequence adjustment module is used to obtain the vehicle's current speed and surrounding environment sensor data based on the preliminary identity confirmation result, and to determine if a sudden situation signal occurs, and then adjust the verification sequence to obtain an acceleration response sequence. The time window optimization module is used to compress and encode multi-source data using the information processing stage based on the accelerated response sequence to determine the optimized verification time window. The adaptive decision-making module is used to obtain historical delay records in the system log based on the optimized verification time window, and determine if the historical delay exceeds the upper limit of the window. If so, it triggers the adaptive adjustment mechanism to obtain real-time decision parameters. The safety response output module is used to combine the verification results with the vehicle control commands using the real-time decision parameters and output fusion technology to obtain the final safety response command.

[0016] Compared with the prior art, the present invention has the following beneficial effects: This invention collects time data from multiple verification methods, employs a parallel processing algorithm to simultaneously execute biometric recognition and digital certificate verification, obtains a unified time benchmark, and activates a priority ranking algorithm when the time difference exceeds a threshold to determine a high-priority verification path. Then, data fusion technology is used to integrate information to obtain preliminary identity verification results. Simultaneously, the verification sequence is adjusted based on vehicle speed and environmental data to form an accelerated response sequence. The verification time window is optimized through compression coding, and an adaptive adjustment mechanism is triggered to obtain real-time decision parameters based on historical delays. Finally, output fusion technology is used to combine the verification results with vehicle control commands to generate the final safety response command. This significantly reduces verification latency, improves response speed and system safety in emergency situations, and ensures the overall reliability and efficiency of the traffic environment. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained from these drawings without creative effort.

[0018] Figure 1 This is one of the flowcharts for an autonomous driving vehicle system in the Internet of Things for intelligent transportation according to the present invention.

[0019] Figure 2 This is the second flowchart of an autonomous driving vehicle system in the Internet of Things for intelligent transportation according to the present invention.

[0020] Figure 3 This is the third flowchart of an autonomous driving vehicle system in the Internet of Things for intelligent transportation according to the present invention. Detailed Implementation

[0021] The present invention will be further described below with reference to embodiments. These embodiments are merely some, not all, of the embodiments described. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the protection scope of the present invention.

[0022] Please see Figures 1-3 As shown, this embodiment discloses an autonomous driving vehicle system in the Internet of Things for intelligent transportation, including: S1. By collecting processing time data of various verification methods, a parallel processing algorithm is used to synchronously execute biometric recognition and digital certificate confirmation to obtain a unified time benchmark; S1-1. Obtain processing time data through interfaces of multiple verification methods, record the initial timestamp for the verification process of biometrics and digital certificates respectively, and obtain a preliminary time data set; S1-2. Based on the preliminary time data set, the tasks of biometric identification and digital certificate verification are allocated using a parallel processing method. The execution order and time allocation of each task are determined using a preset task queue mechanism. S1-3. If the processing time of the biometric identification task exceeds the preset threshold, the priority of the task will be adjusted to the secondary queue, resources will be reallocated, and the adjusted execution time record will be obtained. S1-4. By adjusting the execution time record, execute the parallel tasks of biometric identification and digital certificate confirmation simultaneously, and compare the completion time of each task to determine whether there is a time deviation. S1-5. If a time deviation is detected, the deviation data is corrected by the time calibration module, and the calibrated time data is obtained by aligning with the standard timestamp. S1-6. Based on the calibrated time data, integrate the processing results of all verification methods, generate a unified time benchmark, and determine the final synchronization execution standard. S1-7. By using a unified time benchmark, monitor the processing time of subsequent verification tasks, obtain real-time time data updates, and determine whether they meet the preset synchronization standards.

[0023] For example, in the process of achieving synchronous processing of biometric recognition and digital certificate verification to obtain a unified time reference, the processing time of various verification methods is first recorded through a data acquisition system. Assuming 1,000 sets of data are collected, the average processing time of biometric recognition is 2.5 seconds with a standard deviation of 0.3 seconds, while the average processing time of digital certificate verification is 1.8 seconds with a standard deviation of 0.2 seconds. The data is stored in a distributed database for subsequent analysis.

[0024] Next, parallel processing algorithms, such as the multi-threaded MapReduce framework, are used to distribute the tasks of biometric identification and digital certificate verification to different computing nodes. Assuming there are 10 nodes, each node processes 100 sets of data. By calculating the processing time difference of each node, the maximum time difference between biometric identification and digital certificate verification is found to be 0.7 seconds. Based on this, the task allocation ratio is adjusted to ensure that the processing time of the two tends to be consistent.

[0025] Furthermore, during the synchronous execution process, the processing time of each node is calibrated using a timestamp synchronization algorithm (NTP protocol). Assuming that the time error after calibration is controlled within 0.01 seconds, the unified time base is calculated as the latest time point when all tasks are completed, i.e., 2.52 seconds, by analyzing the timestamp data. This base value is then stored in the central server for subsequent business calls.

[0026] To establish a tight logical relationship, a unified time benchmark is linked to the user authentication process of the business system. Assuming that the authentication process requires a total duration of no more than 3 seconds, analysis shows that the current benchmark value of 2.52 seconds meets the requirements. However, if a new verification method is added, the number of parallel nodes can be dynamically adjusted to 12 to reduce the duration to 2.3 seconds, ensuring system scalability.

[0027] The entire process utilizes automated scripts to collect data, allocate algorithms, and calibrate time. System logs record intermediate results of each calculation, such as time differences and baseline values, facilitating subsequent optimization and troubleshooting.

[0028] S2. Based on a unified time reference, obtain real-time data streams in complex traffic scenarios. If the time difference is greater than a preset threshold, activate the priority sorting algorithm to determine the high-priority verification path. S2-1. By using a unified time reference, real-time data streams are obtained from complex traffic scenarios. Sensor networks and data transmission modules are used to complete data collection and obtain a preliminary traffic data set. S2-2. For the initial traffic data set, detect the time difference of each data point. If the time difference exceeds the preset threshold, mark it as an abnormal data point and determine the abnormal data subset that needs further processing. S2-3. Based on the subset of abnormal data, trigger the priority sorting mechanism, use the pre-established decision tree model to classify the abnormal data points, and determine the set of high-priority data points. S2-4. Extract the corresponding traffic scenario information from the high-priority data point set, analyze the correlation between data points and paths, and obtain a candidate list of high-priority verification paths. S2-5. Based on the candidate list of high-priority verification paths and the latest updates in the real-time data stream, select the path that best matches the current traffic scenario and determine the final verification path scheme. S2-6. Based on the final verified route plan, generate route execution instructions and transmit them to the relevant traffic control system to complete the automated deployment of route verification.

[0029] For example, in complex traffic scenarios, real-time data streams are first obtained through a unified time base. Assuming that the network time protocol is used to synchronize the time of all traffic monitoring devices, the time error is ensured to be less than 1 millisecond. The data stream includes 1,000 vehicle position, speed and direction data points collected per second. The data is transmitted to the central processing system through the 5G network, and the real-time error is controlled within 0.5 seconds.

[0030] Next, the system compares the received data timestamps with the baseline time. The preset threshold is 2 seconds. If a data stream time difference of 2.3 seconds is detected, it is judged as abnormal and the priority sorting algorithm is automatically triggered. This algorithm calculates the priority based on vehicle density and accident risk weight. For example, the area with a density of more than 50 vehicles per square kilometer and a speed of less than 10 km / h has a weight of 0.8. Combined with the historical accident data weight of 0.2, road sections with a comprehensive score of more than 0.7 are marked as high priority.

[0031] Subsequently, the system activates the high-priority verification path and calls a deep learning model, such as a convolutional neural network (CNN), to analyze the video stream of the high-priority road segment and identify congestion or accident characteristics. Assuming the model's prediction accuracy reaches 85%, if the identified congestion length exceeds 500 meters, the system automatically adjusts the traffic light cycle, shortening the red light time to 20 seconds and increasing the green light time to 40 seconds. At the same time, it pushes detour suggestions to surrounding vehicles. The detour path is calculated based on the shortest path using the Dijkstra algorithm, reducing the overall traffic time by 15%.

[0032] To form a logical chain, if the traffic flow does not improve after the signal adjustment, the system will further call the backup route planning module to analyze the carrying capacity of secondary roads and ensure that the traffic share is increased by more than 10%.

[0033] Through the above steps, a closed-loop process is formed from data collection to route verification, ensuring the efficiency of traffic management.

[0034] S3. Through a high-priority verification path, data fusion technology is used to integrate biometric features and certificate information to obtain preliminary identity confirmation results; S3-1. Obtain biometric and certificate data from a specified data source through a preset high-priority channel to complete the initial information collection and obtain the dataset to be processed. S3-2. Based on the collected dataset, data fusion technology is used to process the biometric and certificate data in a unified format to generate a standardized information set and determine the preliminary integration results. S3-3. If the matching degree between the biometric data and the certificate data in the standardized information set reaches a preset threshold, it is determined that the consistency has passed and a consistency verification mark is obtained. S3-4. By using the consistency verification identifier and combining it with the security verification mechanism, perform a deep comparison of the standardized information set to determine whether there is any abnormal data and obtain the security verification result. S3-5. If the security verification result shows no abnormal data, the standardized information set will be finally confirmed through the technical path to obtain the identity confirmation conclusion. S3-6. Based on the identity verification conclusion, record the processing efficiency data, generate the processing log, and determine the final output result; S3-7. By processing the logs, the final output results are archived and stored to complete the identity verification process.

[0035] For example, the specific implementation method of integrating biometric features and certificate information through data fusion technology to obtain preliminary identity verification results via a high-priority verification path can be as follows: First, the system will prioritize the processing of identity verification requests in high-risk or high-importance scenarios, such as user identity verification involving financial transactions. The priority threshold is set to 95%, meaning that only verification requests with a confidence level exceeding this value will enter the fast channel, increasing the data processing speed to 1,000 requests per second and reducing queuing time.

[0036] Next, in the data fusion stage, the system collects users' biometric data, such as fingerprints and facial recognition, as well as certificate information, such as ID card numbers and electronic signatures. Fingerprint recognition uses the Minutiae algorithm to extract feature points, and the matching rate needs to reach more than 98%. Facial recognition uses the deep learning model ResNet-50 with a feature vector dimension of 512 and a similarity threshold of 0.85. At the same time, certificate information uses OCR technology to extract text and compare it with the database, and the accuracy rate needs to reach 99.5%.

[0037] Subsequently, the system weights and fuses the verification results of biometrics and certificate information, with biometrics accounting for 60% and certificate information accounting for 40%, and calculates the comprehensive confidence score. For example, if a user's fingerprint matching score is 0.92, face recognition score is 0.88, and certificate information score is 0.96, then the comprehensive score is 0.92×0.3+0.88×0.3+0.96×0.4=0.925. If this score exceeds the preset threshold of 0.9, the identity is initially confirmed.

[0038] The analysis process will further incorporate historical verification data. If a user has more than 3 successful verification records in the past 30 days, the system will automatically increase the confidence level by 0.02, ultimately resulting in a confirmed result.

[0039] To ensure logical rigor, the system also links the verification log with the risk assessment module. If an anomaly is detected, such as the certificate information and biometric features not matching more than twice, a secondary verification mechanism is triggered, calling a backup algorithm such as LBP feature extraction for supplementary comparison to ensure the reliability of the results.

[0040] S4. Based on the preliminary identity verification result, obtain the vehicle's current speed and surrounding environment sensor data, determine if a sudden situation signal occurs, adjust the verification sequence, and obtain the acceleration response sequence. S4-1. Acquire vehicle current status data and surrounding perception information. Collect vehicle speed and environmental data through sensors and store them in a temporary database to obtain a preliminary status dataset. S4-2. For the initial state dataset, a preset threshold is used to detect signals. If a sudden situation signal is detected that exceeds the safe range, the situation judgment logic is triggered to determine whether there is a potential risk. S4-3. Based on the situation assessment, if an emergency is confirmed, extract the identity verification information and compare it with the current status data to obtain a verification consistency result. S4-4. By verifying the consistency results, adjust the verification process, generate temporary response strategies for emergencies, and determine the priority sequence for accelerating the response. S4-5. For the acceleration response priority sequence, obtain real-time updates of vehicle speed and environmental data, and generate specific response sequence instructions in combination with temporary response strategies; S4-6. Based on the response sequence instructions, transmit them to the vehicle control module to execute the acceleration response operation and complete the immediate handling of the emergency.

[0041] For example, regarding the acquisition and analysis of vehicle current speed and surrounding environment data, as well as the adjustment of verification sequence and generation of acceleration response sequence when sudden situation signals occur, assuming that we collect data in real time through the vehicle system, the vehicle's current speed is obtained as 60 km / h through the speed sensor, and the surrounding environment sensors include LiDAR and camera. The LiDAR detects an obstacle 100 meters ahead, and the camera identifies the obstacle as a stationary object. The environmental data also includes the road slipperiness of 0.7 (range 0 to 1, 1 being the most slippery).

[0042] The system first calculates the relative safe distance between the vehicle and the obstacle using an algorithm. The formula is: Safe distance = speed × reaction time + braking distance, where the reaction time is set to 1.5 seconds, and the braking distance is calculated based on the current speed and the degree of slipperiness as speed squared / (2 × friction coefficient). The friction coefficient is taken as 0.4 (for slippery surfaces). The resulting braking distance is 60 × 60 / (2 × 0.4) = 4500 meters, and the safe distance is 60 × 1.5 + 4500 = 4590 meters, which is much larger than the actual distance of 100 meters, and is therefore judged as high risk.

[0043] Next, if the system receives a signal of an emergency (such as a sudden movement of an obstacle ahead), it triggers an adjustment to the verification sequence. The original verification sequence was a three-step verification of speed, distance, and environment. Now, it is adjusted to prioritize the verification of distance and speed, shortening the verification time to within 0.5 seconds. The algorithm uses weight allocation, with distance weight set at 0.6 and speed weight at 0.4, to calculate the risk index as 100×0.6+60×0.4=84 (out of 100). This exceeds the threshold of 70, confirming that an emergency response is required.

[0044] Finally, an acceleration response sequence is generated. The system calculates the acceleration demand based on the current power output and risk index. Assuming the current power output is 50% and the target power output is 80%, the system uses a linear interpolation algorithm to set the power increase rate to increase by 5% every 0.1 seconds, that is, to complete the adjustment from 50% to 80% within 0.6 seconds, forming an acceleration sequence [50, 55, 60, 65, 70, 75, 80], which is then synchronously transmitted to the power control unit for execution, and logs are recorded for subsequent analysis.

[0045] Through the above process, the system achieves fully automated processing from data acquisition to response adjustment, ensuring logical rigor and real-time performance.

[0046] S5. By accelerating the response sequence, the information processing stage is used to compress and encode multi-source data to determine the optimized verification time window; S5-1. By collecting and initially organizing multi-source data, obtain raw datasets from different sources, perform preliminary classification on the response sequences, and obtain a structured dataset. S5-2. Using the classified dataset, accelerate the processing based on the characteristics of the response sequence, and use a preset threshold to filter the data and determine a high-priority subset of data. S5-3. Based on the selected high-priority data subset, implement the information processing step, perform standardization operations on the data, and obtain a unified formatted dataset; S5-4. By formatting the dataset and applying compression encoding technology, the data source and encoding method are optimized to obtain compressed data units; S5-5. Based on the compressed data units, combined with the processing steps and sequence analysis, construct the dynamic range of the verification window, judge the rationality of the window adjustment, and determine the preliminary time window range. S5-6. If the fluctuation of the response sequence within the initial time window exceeds the preset threshold, the window adjustment parameters are corrected to obtain the optimized verification time window. S5-7. By using the optimized verification time window and combining it with the time optimization target, the subsequent processing steps of multi-source data are recorded to obtain the final window configuration result.

[0047] For example, the following specific methods can be used to implement accelerated response sequences and compress and encode multi-source data to optimize the verification time window.

[0048] First, regarding the accelerated response sequence, assuming the system needs to process data streams from multiple sensors, generating 1000 data points per second, time series analysis algorithms, such as the sliding window averaging method, can be used to aggregate data points every 10 seconds into a representative value, reducing the amount of data processing. Calculations show that the average response time per window is reduced from 0.5 seconds to 0.2 seconds. Analysis indicates that this method effectively improves the system's real-time performance.

[0049] Secondly, when compressing and encoding multi-source data in the information processing stage, assuming there is data from three sensors: temperature, humidity, and pressure, with each data sampled at 500Hz, the Huffman coding algorithm can be used to compress the data without loss. The calculation results show that the compression rate reaches 60%, that is, the original data volume is reduced from 1.5MB per minute to 0.6MB. The integrity of the data before and after compression is analyzed to ensure that no information is lost, while reducing the burden of storage and transmission.

[0050] Finally, when determining the optimized verification time window, based on the aforementioned compressed data and combined with business requirements, the verification cycle was set. Assuming the original time window was 30 seconds, by analyzing response time and data processing efficiency, a dynamic adjustment algorithm was adopted to optimize the time window to 20 seconds. The verification accuracy was increased from 95% to 98%. The analysis showed that the system's ability to capture abnormal data was enhanced after shortening the window. At the same time, combined with business scenarios, such as the need for rapid response to equipment failures in industrial monitoring, the time window adjustment was ensured to match the actual application.

[0051] By employing the methods described above, a complete logical chain is formed, from data acquisition to processing and time optimization, ensuring system efficiency and accuracy.

[0052] S6. Based on the optimized verification time window, obtain historical delay records in the system log, and determine if the historical delay exceeds the window limit. Then, trigger the adaptive adjustment mechanism to obtain real-time decision parameters. S6-1. Extract historical delay data from the system log, perform preliminary cleaning on the extracted data, remove abnormal records, and obtain the sorted delay dataset. S6-2. Based on the sorted delay dataset, and combined with the preset time window and window limit, compare the delay data one by one. If the delay data exceeds the window limit, mark it as an abnormal delay record and obtain the abnormal delay set. S6-3. For the set of abnormal delays, analyze its distribution, use statistical tools to calculate the frequency and concentrated period of abnormal delays, and determine the key time points of abnormal delays. S6-4. Based on key time points, trigger the adaptive adjustment mechanism, obtain the real-time data stream of the system operation status, and determine if there is a delay trend in the real-time data stream similar to the key time points, then generate an adjustment signal. S6-5. By adjusting the signal, the pre-established decision model is invoked to obtain real-time decision parameters and output a parameter configuration scheme optimized for latency. S6-6. Adopt the parameter configuration scheme, update the system operation rules, record the changes in latency data after the update, and obtain the optimized latency monitoring results.

[0053] For example, under the optimized verification time window mechanism, the system will first automatically extract historical delay records from the system log database within the past 24 hours according to the preset upper limit of the window, such as 10 seconds. Assuming that the query results show 1,000 delay data, with a maximum delay of 12 seconds, an average delay of 6.5 seconds, and a standard deviation of 2.3 seconds, the system will periodically execute data extraction through the built-in SQL query script and store the results in a temporary cache table for subsequent analysis.

[0054] Next, the system will automatically compare each delay record with the window limit of 10 seconds. Using a simple comparison algorithm, if the proportion of records exceeding 10 seconds reaches 5%, that is, 50 records, then the historical delay is judged to be abnormal, and the adaptive adjustment mechanism is triggered. The specific calculation process is: the number of records exceeding the limit divided by the total number of records, 50 / 1000=0.05, that is, 5%, which reaches the trigger threshold.

[0055] Subsequently, the system enters the adaptive adjustment phase, calling the preset decision parameter adjustment algorithm. For example, based on the linear regression model, it inputs historical latency data and the current system load rate (assumed to be 70%), predicts that the new upper limit of the time window should be adjusted to 12 seconds. The calculation formula is: new upper limit = current upper limit + (average latency - current upper limit) x load rate x 0.5, that is, 10 + (6.5-10) x 0.7 x 0.5 = 8.775 seconds, which is rounded to 9 seconds. However, considering the minimum adjustment step size of 2 seconds based on business requirements, it is finally adjusted to 12 seconds.

[0056] Meanwhile, the system will update the adjusted parameters to the configuration center in real time and link them to the business monitoring module. If the latency still exceeds the standard after adjustment, the system will further link the resource scheduling module to increase the server resource allocation ratio, such as increasing the CPU quota by 10%, to ensure system stability.

[0057] Through the above series of automated processes, from data extraction to parameter adjustment and resource linkage, a complete technical closed loop is formed, ensuring that the system can still maintain efficient operation in high-latency scenarios.

[0058] S7. By using real-time decision parameters and output fusion technology, the verification results are combined with vehicle control commands to obtain the final safety response command; S7-1. Real-time acquisition of vehicle operation data and environmental information through sensors and communication modules to obtain the raw dataset of decision parameters; S7-2. Based on the collected raw dataset, preprocessing techniques are used to clean and standardize the data to obtain a structured set of decision parameters. S7-3. If there are outliers in the structured set of decision parameters, filter them through a preset threshold range to determine the effective combination of decision parameters. S7-4. For effective combinations of decision parameters, use logical judgment rules to analyze their degree of matching with security standards and determine whether to trigger the verification process. S7-5. If the verification process is triggered, the combination of decision parameters and the vehicle status are compared using a preset verification model to obtain verification result data. S7-6. Based on the verification result data, the output fusion method is used to integrate it with the preset vehicle control commands to generate the final safety response command.

[0059] S7-7. The generated safety response command is transmitted to the vehicle execution unit through the communication interface to complete the command issuance and response.

[0060] For example, in the process of combining the verification results with vehicle control commands through output fusion technology using real-time decision parameters to obtain the final safe response command, the vehicle operation data is first collected in real time by sensors, such as the vehicle speed of 60 kilometers per hour and the road slope of 5 degrees. Combined with weather data such as rainfall of 10 millimeters per hour, a preset decision algorithm is used to conduct risk assessment and calculate the risk index. The formula is: Risk Index = Vehicle Speed ​​X 0.5 + Slope X 2 + Rainfall X 1.5. The result is: Risk Index = 60 X 0.5 + 5 X 2 + 10 X 1.5 = 55.

[0061] If the risk index exceeds the threshold of 50, the verification process is triggered. The system automatically calls historical data models to compare with the current scenario and concludes that "slowing down is required".

[0062] Next, the verification results are fused with the vehicle control command. Assuming the current control command is "maintain speed", the priority is calculated by the fusion algorithm. The priority formula is safety weight × verification result + control weight × command value, where the safety weight is 0.7, the control weight is 0.3, the quantized value of the verification result is -10 (decelerate), and the command value is 0 (maintain speed). The calculated fusion value is 0.7 × (-10) + 0.3 × 0 = -7.

[0063] Ultimately, based on the condition that the fusion value is less than 0, the system automatically generates a safety response command of "decelerate to 50 kilometers per hour" and sends it to the execution unit through the vehicle controller to ensure the safe operation of the vehicle.

[0064] The above process forms a complete logical chain, from data collection to risk assessment, then to verification and integration, and finally to instruction generation. Each step is closely linked, and historical data models are introduced as business connections to ensure the accuracy of decision-making. The entire process is automated and requires no manual intervention.

[0065] In some optional embodiments, the above solution is described below through a specific scenario. The scenario is set as follows: when an autonomous vehicle is driving on a highway, it suddenly encounters an obstacle in front of it. The system achieves rapid identity verification and secure response through each step. An autonomous vehicle is traveling at 100 kilometers per hour on a highway, with onboard sensors (such as lidar and cameras) monitoring the surrounding environment in real time. Suddenly, the system detects a stationary obstacle (such as a malfunctioning vehicle) 150 meters ahead. At the same time, the vehicle needs to authenticate itself to ensure the legitimacy of its operation. The purpose is to quickly complete authentication (biometric recognition and digital certificate confirmation) and generate safety response instructions (such as slowing down or changing lanes) in case of emergencies to avoid a collision.

[0066] Step 1: Parallel Processing and Unified Time Base Generation Specific implementation: The system collects processing time data for biometric recognition (such as driver facial recognition) and digital certificate verification (such as vehicle digital certificates) through a verification interface. It is assumed that biometric recognition takes an average of 2.5 seconds (standard deviation 0.3 seconds) and digital certificate verification takes an average of 1.8 seconds (standard deviation 0.2 seconds).

[0067] Parallel processing algorithms (such as the multi-threaded MapReduce framework) are employed, distributing tasks across 10 computing nodes, with each node processing 100 sets of data. A task queue mechanism dynamically adjusts priorities: if a biometric recognition task times out (threshold 2.2 seconds), its priority is reduced to a secondary queue, and resources are reallocated.

[0068] The time calibration module (based on the NTP protocol) is used to align the timestamps. After calibration, the time error is controlled within 0.01 seconds, and a unified time base of 2.52 seconds is generated (based on the slowest task completion time).

[0069] Parallel processing algorithm: Simultaneously executes multiple verification methods to solve the problem of processing time differences.

[0070] Unified time base: Provides a consistent time reference for subsequent steps, ensuring system synchronization.

[0071] Step 2: Real-time data stream analysis and high-priority verification path determination Specific implementation: Based on a unified time reference, the system acquires real-time data streams (1000 data points per second, including vehicle position, speed, and obstacle distance) from the sensor network. Data points with time differences exceeding a preset threshold (2 seconds) are flagged as abnormal.

[0072] Activate the priority ranking algorithm: Use a decision tree model to classify outlier data points and calculate priority weights (e.g., vehicle density weight 0.8, accident risk weight 0.2). Road segments with a comprehensive score exceeding 0.7 are marked as high priority.

[0073] Traffic scenario information is extracted from high-priority data points to generate a candidate list of verification paths. Combined with real-time updates, the optimal path (such as a direct obstacle avoidance path) is selected, and instructions are generated and transmitted to the traffic control system.

[0074] Step 3: Data Fusion and Preliminary Identity Verification Specific implementation: Biometric data (fingerprints, faces) and certificate data (digital certificates) are acquired through high-priority channels. Data fusion techniques are used to unify the format (such as fingerprint feature point extraction and facial feature vector standardization).

[0075] Matching score: fingerprint score 0.92, face score 0.88, certificate score 0.96. After weighted fusion (60% biometric weight and 40% certificate weight), the overall score is 0.925, which exceeds the threshold of 0.9, and the consistency is passed.

[0076] The security verification mechanism performs a deep comparison of the data. If no anomalies are found, a preliminary identity confirmation result is generated and a processing log is recorded.

[0077] Step 4: Emergency Detection and Accelerated Response Sequence Generation Specific implementation: The system obtains the vehicle's current speed (100 km / h) and the distance to the obstacle (150 m), and calculates the safe distance: Safe distance = speed × reaction time (1.5 seconds) + braking distance (calculated as 4500 meters based on a friction coefficient of 0.4). If the actual distance is much smaller than the safe distance, an emergency signal is triggered.

[0078] The verification sequence has been adjusted: the original verification sequence (speed-distance-environment) has been simplified to prioritize the verification of distance and speed, and the verification time has been shortened to 0.5 seconds.

[0079] Generate acceleration response sequence: Based on the risk index (distance weight 0.6, speed weight 0.4, calculated to be 84, which exceeds the threshold of 70), generate a power boost sequence (increasing by 5% every 0.1 seconds from 50% to 80%), and transmit it to the vehicle control module.

[0080] Step 5: Multi-source data compression coding and verification time window optimization Specific implementation: Multi-source data (radar, camera data) is collected and classified, and the data is aggregated using the sliding window averaging method (data points are aggregated into representative values ​​every 10 seconds), reducing the processing volume by 80%.

[0081] By applying compression coding techniques (such as Huffman coding), the compression rate reaches 60%, and the data volume is reduced from 1.5MB per minute to 0.6MB.

[0082] The verification time window was optimized from 30 seconds to 20 seconds through dynamic adjustment of the algorithm, improving the verification accuracy from 95% to 98%.

[0083] Step 6: Historical Delay Analysis and Adaptive Adjustment Mechanism Specific implementation: Historical latency data (1000 records from the past 24 hours) was extracted from the system logs. After cleaning, it was found that 5% of the records had a latency exceeding the window limit (10 seconds).

[0084] Trigger the adaptive adjustment mechanism: Using a linear regression model, input historical latency and current load rate (70%), predict a new window upper limit of 12 seconds.

[0085] After updating system parameters and monitoring latency changes, the latency exceedance rate decreased to 2% after optimization.

[0086] Step 7: Output Fusion and Security Response Command Generation Specific implementation: The real-time decision parameters (risk index 55, calculated based on vehicle speed, slope, and rainfall) trigger the verification process, and the verification result is "need to slow down".

[0087] Using output fusion technology: safety weight 0.7 × verification result (-10) + control weight 0.3 × instruction value (0) = -7, a safety response instruction "decelerate to 80 km / h" is generated.

[0088] The command is sent to the vehicle's execution unit via the communication interface, and the vehicle immediately decelerates.

[0089] Through the above steps, in scenarios involving sudden obstacles, the system completes identity verification in just 2.52 seconds (up from 4.3 seconds), reducing response time by 40%; the accelerated response sequence allows the vehicle to adjust its power within 0.6 seconds to avoid a collision. Adaptive adjustment ensures continuous system optimization, resolving historical latency issues.

[0090] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An autonomous driving transportation system in an intelligent transportation internet of things, characterized in that, The application relates to a method for realizing real-time identity verification in complex traffic scenes. The method comprises the following steps: Collecting processing time data of multiple verification methods, using a parallel processing algorithm to synchronously execute biometric feature recognition and digital certificate confirmation, and obtaining a unified time reference; According to the unified time reference, real-time data flow in a complex traffic scene is obtained, and if the time difference is greater than a preset threshold, a priority sorting algorithm is activated to determine a high-priority verification path; Through the high-priority verification path, data fusion technology is used to integrate biometric features and certificate information to obtain a preliminary identity confirmation result; According to the preliminary identity confirmation result, the current speed of the vehicle and the surrounding environment sensor data are obtained, and if a sudden situation signal appears, the verification sequence is adjusted to obtain an accelerated response sequence; Through the accelerated response sequence, information processing links are used to compress and encode multi-source data to determine an optimized verification time window; According to the optimized verification time window, historical delay records in the system log are obtained, and if the historical delay exceeds the upper limit of the window, an adaptive adjustment mechanism is triggered to obtain real-time decision parameters; 2.The automatic driving system in an intelligent transportation Internet of Things of claim 1, wherein: Through the real-time decision parameters, output fusion technology is used to combine the verification result and the vehicle control instruction to obtain a final safety response instruction. The method comprises the following steps: Through the interface of multiple verification methods, processing time data is obtained, initial time stamps are recorded for the verification processes of biometric features and digital certificates, and a preliminary time data set is obtained; According to the preliminary time data set, parallel processing is used to distribute the tasks of biometric feature recognition and digital certificate confirmation, a preset task queue mechanism is used to determine the execution order and time distribution of each task; If the processing time of the biometric feature recognition task exceeds the preset threshold, the priority of the task is adjusted to a secondary queue, resources are redistributed, and adjusted execution time records are obtained; Through the adjusted execution time records, the parallel tasks of biometric feature recognition and digital certificate confirmation are synchronously executed, the completion time points of each task are compared, and whether there is a time deviation is judged; If a time deviation is detected, the deviation data is corrected through a time calibration module, a standard time stamp alignment method is used to obtain calibrated time data; According to the calibrated time data, the processing results of all verification methods are integrated to generate a unified time reference and determine the final synchronous execution standard; 3.The automatic driving system in an intelligent transportation Internet of Things of claim 1, wherein: Through the unified time reference, the processing time of subsequent verification tasks is monitored, real-time time data updates are obtained, and whether the preset synchronous standard is met is judged. The method comprises the following steps: Through the unified time reference, real-time data flow in a complex traffic scene is obtained, a sensor network and a data transmission module are used to complete data collection, and a preliminary traffic data set is obtained; For the preliminary traffic data set, the time difference of each data point is detected, and if the time difference exceeds the preset threshold, it is marked as an abnormal data point, and an abnormal data subset that needs further processing is determined; According to the abnormal data subset, a priority sorting mechanism is triggered, and a pre-established decision tree model is used to classify the abnormal data points to determine a high-priority data point set; From the high-priority data point set, corresponding traffic scene information is extracted, the correlation between data points and paths is analyzed, and a candidate list of high-priority verification paths is obtained; For the candidate list of high-priority verification paths, the latest updates in the real-time data stream are combined to filter out the path that best matches the current traffic scene, and the final verification path scheme is determined; Through the final verification path scheme, path execution instructions are generated and transmitted to related traffic control systems to complete the automated deployment of path verification. 4.The automatic driving system in an intelligent transportation Internet of Things of claim 1, wherein: The high-priority verification path is used to integrate biological features and certificate information using data fusion technology to obtain a preliminary identity confirmation result, including: Through a pre-set high-priority channel, biological feature and certificate data are obtained from specified data sources to complete initial information collection and obtain a data set for processing; According to the collected data set, data fusion technology is used to process biological features and certificate data in a unified format to generate a standardized information set and determine a preliminary integration result; If the matching degree of biological feature data and certificate data in the standardized information set reaches a preset threshold, it is determined to be consistent and a consistency verification identifier is obtained; Through the consistency verification identifier, the standardized information set is compared in depth in combination with a security verification mechanism to determine whether there is abnormal data and obtain a security verification result; If the security verification result shows no abnormal data, the standardized information set is finally confirmed through a technical path to obtain an identity confirmation conclusion; According to the identity confirmation conclusion, processing efficiency data is recorded to generate a processing log and determine a final output result; Through the processing log, the final output result is archived and stored to complete the identity confirmation process. 5.The automatic driving system in an intelligent transportation Internet of Things of claim 1, wherein: According to the preliminary identity confirmation result, the current speed of the vehicle and the surrounding environment sensor data are obtained, and if a sudden condition signal appears, the verification sequence is adjusted to obtain an acceleration response sequence, including: Obtain vehicle state data and surrounding perception information, collect vehicle speed and environmental data through sensors, and store them in a temporary database to obtain a preliminary state data set; For the preliminary state data set, a preset threshold is used for signal detection, and if a sudden condition signal is detected that exceeds the safety range, a condition judgment logic is triggered to determine whether there is a potential risk; According to the condition judgment result, if it is confirmed that there is a sudden condition, the identity verification information is compared with the current state data to obtain a verification consistency result; Through the verification consistency result, the verification process is adjusted to generate a temporary response strategy for the sudden condition and determine the priority sequence of the acceleration response; For the acceleration response priority sequence, real-time updates of vehicle speed and environmental data are obtained, and in combination with the temporary response strategy, specific response sequence instructions are generated; According to the response sequence instruction, the vehicle control module is transmitted to perform an acceleration response operation to complete the instant processing of the emergency situation. 6.The automatic driving system in an intelligent transportation Internet of Things of claim 1, wherein: The response sequence is compressed and encoded by using the information processing link to determine the optimized verification time window, including: Through the collection and preliminary arrangement of multi-source data, the original data set from different sources is obtained, the response sequence is preliminarily classified, and the structured data set is obtained; Using the classified data set, the characteristics of the response sequence are processed, the data is filtered using a preset threshold, and a high-priority data subset is determined; According to the filtered high-priority data subset, the information processing link is implemented to standardize the data and obtain a unified formatted data set; Through the formatted data set, the compression encoding technology is applied to optimize the data source and encoding method, and the compressed data unit is obtained; According to the compressed data unit, the dynamic range of the verification window is constructed by combining the processing link and sequence analysis, the rationality of the window adjustment is judged, and the preliminary time window range is determined; If the response sequence fluctuation in the preliminary time window range exceeds the preset threshold, the window adjustment parameter is corrected to obtain the optimized verification time window; Through the optimized verification time window, the subsequent processing link of multi-source data is recorded in combination with the time optimization target to obtain the final window configuration result. 7.The automatic driving system in an intelligent transportation Internet of Things of claim 1, wherein: According to the optimized verification time window, the historical delay record in the system log is obtained, and if the historical delay exceeds the upper limit of the window, the adaptive adjustment mechanism is triggered to obtain real-time decision parameters, including: Extract the historical delay data from the system log, preliminarily clean the extracted data, and remove abnormal records to obtain the cleaned delay data set; According to the cleaned delay data set, the preset time window and the upper limit of the window are combined to compare the delay data one by one, and if the delay data exceeds the upper limit of the window, it is marked as an abnormal delay record to obtain an abnormal delay set; For the abnormal delay set, analyze its distribution, calculate the frequency and concentrated period of abnormal delay using statistical tools, and determine the key time point of abnormal delay; According to the key time point, the adaptive adjustment mechanism is triggered to obtain the real-time data stream in the system running state, and if there is a delay trend similar to the key time point in the real-time data stream, an adjustment signal is generated; Through the adjustment signal, the pre-established decision model is called to obtain real-time decision parameters, and a parameter configuration scheme for delay optimization is output; Using the parameter configuration scheme, the system running rules are updated, the updated delay data changes are recorded, and the optimized delay monitoring result is obtained. 8.The automatic driving system in an intelligent transportation Internet of Things of claim 1, wherein: Through the real-time decision parameters, the verification result is combined with the vehicle control instruction by using the output fusion technology to obtain the final safety response instruction, including: Real-time collection of vehicle running data and environmental information through sensors and communication modules to obtain the original data set of decision parameters; According to the collected original data set, the data is cleaned and standardized by using preprocessing technology to obtain a structured decision parameter set; If there is an abnormal value in the structured decision parameter set, the effective decision parameter combination is determined by filtering through a preset threshold range; For the effective decision parameter combination, the matching degree with the safety standard is analyzed by using a logical judgment rule to determine whether to trigger a verification process; If the verification process is triggered, the decision parameter combination and the vehicle state are compared through a preset verification model to obtain verification result data; According to the verification result data, an output fusion method is used to integrate it with the preset vehicle control instruction to generate the final safety response instruction. The generated safety response instruction is transmitted to the vehicle execution unit through the communication interface to complete the issuance and response of the instruction. 9.The automatic driving system in an intelligent transportation Internet of Things of claim 1, wherein: The system comprises: A time reference generation module for collecting processing time data of multiple verification methods, performing biological feature recognition and digital certificate confirmation synchronously by using a parallel processing algorithm to obtain a unified time reference; A priority path determination module for obtaining real-time data flow under complex traffic scenes according to the unified time reference, and determining to activate a priority sorting algorithm to determine a high-priority verification path if the time difference is greater than a preset threshold; An identity confirmation module for integrating biological features and certificate information by using data fusion technology through the high-priority verification path to obtain a preliminary identity confirmation result; A response sequence adjustment module for obtaining the current speed of the vehicle and the surrounding environment sensor data according to the preliminary identity confirmation result, and determining to adjust the verification sequence if a sudden situation signal appears to obtain an accelerated response sequence; A time window optimization module for compressing and encoding multi-source data by using an information processing link through the accelerated response sequence to determine an optimized verification time window; An adaptive decision module for obtaining historical delay records in the system log according to the optimized verification time window, and determining to trigger an adaptive adjustment mechanism if the historical delay exceeds the upper limit of the window to obtain real-time decision parameters; A safety response output module for combining the verification result and the vehicle control instruction by using output fusion technology through the real-time decision parameters to obtain the final safety response instruction.

Images