Vehicle state acquisition method based on multi-source data fusion in roadside view angle

Abstract

The invention discloses a vehicle state acquisition method based on multi-source data fusion under a roadside perspective, and relates to the technical field of data analysis, and the method comprises the following steps: deploying a plurality of acquisition devices at a roadside to acquire original data of a vehicle state and an environmental condition, and preprocessing the acquired original data; and performing time synchronization operation on the preprocessed original data acquired by different sensors. According to the vehicle state acquisition method based on multi-source data fusion under the roadside perspective, the conditions of sensor data under different environmental conditions are analyzed, the interference degree on vehicle state data acquisition is analyzed, and quantitative analysis is carried out so as to implement corresponding data adjustment measures, so that the accuracy of subsequent traffic management is improved, and the vehicle state acquisition efficiency is improved. By integrating data from different sources, real-time monitoring of vehicle states and dynamic traffic management are realized, so that a scientific basis is provided for a traffic management department to formulate targeted prevention measures.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and in particular to a vehicle status acquisition method based on multi-source data fusion from a roadside perspective. Background Art

[0002] With the rapid development of intelligent transportation systems, accurate and real-time collection of vehicle status has become increasingly important. Intelligent transportation systems need to use vehicle status information to perform traffic flow management, traffic incident detection, traffic signal control, etc., in order to improve road capacity and traffic safety levels. Therefore, the development of efficient and accurate vehicle status collection methods has become the key to the development of intelligent transportation systems.

[0003] In the existing technology, multi-source data fusion involves a variety of acquisition devices and data sources. Under different traffic environments and weather conditions, the performance and data quality of sensors will change, which in turn affects the acquisition accuracy of vehicle status data. Therefore, how to analyze the fusion state during multi-source data fusion, estimate the data quality, and weaken the impact on vehicle status data acquisition is the problem we need to solve. To this end, a vehicle status acquisition method based on multi-source data fusion from a roadside perspective is proposed. Summary of the invention

[0004] To achieve the above objectives, the present invention is implemented through the following technical solutions: A vehicle status acquisition method based on multi-source data fusion from a roadside perspective, comprising the following steps: Step 1: Deploy a variety of collection devices on the roadside to collect raw data on vehicle status and environmental conditions, and pre-process the collected raw data; Step 2: Perform time synchronization on the raw data collected by different sensors after preprocessing, and use data fusion technology to fuse the raw data of different sensors; Step 3: Monitor the data anomalies in sensors and data fusion results according to the changes in different environmental conditions, analyze the severity of the anomalies, and match the corresponding recovery and adjustment measures; Step 4: Verify and analyze the quality of the adjusted and restored fusion data, control data errors, and adjust the data application strategy to meet traffic rules requirements; Step 5: Use the processed data to monitor vehicle status and implement targeted traffic management.

[0005] Preferably, in step 1, the collection and preprocessing of raw data includes: Deploy a variety of collection equipment on the roadside to ensure coverage of the required traffic area, including cameras, radars, lidars, and infrared sensors to collect raw data on vehicle status and environmental conditions; Utilize the various collection devices deployed to collect vehicle status data such as vehicle speed, location, and direction of travel, as well as environmental condition data such as light intensity, weather conditions, temperature, humidity, and wind speed; The collected raw data is transmitted wirelessly to the data processing center and preprocessed. The preprocessing includes data cleaning, format conversion and data calibration. Data cleaning removes noise, duplicate values ​​and invalid values ​​in the data to ensure the accuracy and integrity of the data. Format conversion converts the data collected by different sensors into a unified format to facilitate subsequent processing and analysis. Data calibration calibrates the data collected by the sensor to eliminate the errors and deviations of the sensor itself. Build a data warehouse, integrate the pre-processed data, and store it in the data warehouse to ensure the accessibility and security of the data.

[0006] Preferably, in step 2, the time synchronization and fusion process of the original data includes: Retrieve the pre-processed raw data collected by different sensors from the data warehouse, perform software synchronization on it, traverse each data record, record the timestamp of each sensor data, use the lidar data as the basis for time synchronization, and extract its timestamp as the reference time point; For data from other sensors (such as cameras, radars, infrared sensors, etc.), compare their timestamps with the reference time point of the LiDAR, use the interpolation algorithm to calculate the data values ​​of other sensors at the reference time point of the LiDAR, and obtain the data of all sensors at the reference time point of the LiDAR through interpolation, so as to achieve time synchronization from the software; The data acquired by different sensors are time-aligned and coordinate-aligned. The time alignment is performed based on the result of the time synchronization operation to ensure the temporal consistency of the data of each sensor, and the data of different sensors are converted into a unified coordinate system for coordinate alignment to facilitate subsequent data fusion. The data fusion algorithm using Kalman filtering method is used to define the state equation to describe the change of sensor data over time, and the observation equation to describe the relationship between sensor data and state. The state estimate and covariance matrix are initialized. For each time step, the state equation is used to predict the next state. The observation equation and observation data are used to update the state estimate and covariance matrix. The Kalman filtering process is repeated until all sensor data are fused, and then the fused data is obtained, which integrates information from different sensors to improve the accuracy and reliability of the data. The fused data is restored to the data warehouse for further analysis and processing.

[0007] Preferably, in step 3, the analysis process of the severity of the abnormal situation includes: Based on the fused data, analyze various types of environmental condition data and monitor changes in different environmental conditions; Use time series analysis methods to conduct comprehensive analysis of monitoring data and identify abnormal data, which are manifested as sudden changes in data values ​​and abnormal expansion of data fluctuation range; Integrate the abnormal data and determine the characteristics of the abnormal data, namely the number of abnormal values, the duration of the abnormality, the fluctuation range of the abnormal data, calculate the abnormal interference index, and then quantitatively analyze the severity of the abnormal data; According to the severity of abnormal data, different interference levels are divided, namely slight interference level, medium interference level and severe interference level. Combined with the calculated abnormal interference index, the corresponding interference threshold is matched for each interference level, and the corresponding recovery adjustment measures are matched for each interference level. The slight interference level corresponds to slight abnormal conditions, and simple recovery adjustment measures such as adjusting the calibration parameters of the sensor and optimizing the data acquisition frequency are taken. At the same time, monitoring is strengthened and the changing trend of abnormal data is closely monitored; the medium interference level corresponds to moderate abnormal conditions, and in-depth analysis of the abnormal causes involving sensor failure, data transmission errors or environmental changes is required. According to the analysis results, corresponding recovery adjustment measures such as replacing faulty sensors, repairing data transmission links, and adjusting monitoring strategies are taken. During the recovery process, the severe interference level corresponds to severe abnormal conditions, and the emergency plan needs to be activated immediately, and emergency recovery adjustment measures such as shutting down the faulty equipment and starting the backup system are taken. In-depth analysis is conducted to identify the causes of the abnormalities and formulate a long-term recovery plan. During the recovery process, safety monitoring is strengthened.

[0008] Preferably, the process of identifying abnormal data includes: Using the time series analysis method, the time series of monitoring data is decomposed into three parts: trend, seasonality and residual. According to the sudden change of data values ​​and abnormal expansion of data fluctuation range of abnormal data, the anomaly detection threshold T is set in combination with the three sigma principle based on normal distribution. Analyze the residual part of the time series decomposition of the monitoring data, calculate the residual at each time point, and for each time point t, calculate the median absolute deviation of the residuals of the w time points before and after it to measure the volatility of the data; The absolute value of the difference between the actual observation value at time point t and the baseline value at time point t is calculated and combined with the median absolute deviation to construct an anomaly detection index; The anomaly detection index at each time point is calculated and compared with the set anomaly detection threshold. If the anomaly detection index is greater than the anomaly detection threshold T, the data at time point t is abnormal data.

[0009] Preferably, the calculation expression of the abnormality detection index is: in, is the anomaly detection index, is the actual observed value at time point t, is the reference value at time point t, which is used to reflect the normal state of the data. is the median absolute deviation of the residuals at w time points before and after time point t, which is used to measure the volatility of the data. is the residual of w time points before and after time point t.

[0010] Preferably, the process of obtaining the abnormal interference index includes: Integrate abnormal data from different sensors within a fixed time period, including speed, location, driving direction, light intensity, weather conditions, temperature, humidity and wind speed, to form an abnormal data set. The abnormal data set contains the timestamp and abnormal value information of each abnormal data point; Analyze the abnormal data set to determine the number of abnormal values, the duration of the abnormality, the fluctuation range of the abnormal data, and evaluate the impact of abnormal data on vehicle status recognition, including the decrease in recognition accuracy and the increase in false alarm rate; Based on the evaluation results of the impact of abnormal data on vehicle state recognition, the abnormal interference index is obtained by combining the comprehensive analysis of abnormal data characteristics to quantify the severity of interference of abnormal data on vehicle state recognition.

[0011] Preferably, the calculation expression of the abnormal interference index is: in, is the abnormal interference index, is the total number of outliers in a fixed time period, is the total duration of the outlier, is the maximum deviation of the abnormal value from its baseline value, is the number of baseline outliers, representing the number of outliers expected under normal conditions, To adjust the parameters, used to control the impact of the number of outliers on the index, is the maximum value of the abnormal duration during the observation period, is the maximum value of the abnormal fluctuation range during the observation period; The three interference levels correspond to the three interference thresholds, wherein the interference thresholds include an upper threshold and a lower threshold; The three interference levels and the three interference thresholds satisfy the following relationship: Slight disturbance level ; Medium interference level ; Severe interference level ; in, is the abnormal interference index, is the lower threshold corresponding to the medium interference level and the upper threshold corresponding to the slight interference level, is the lower threshold corresponding to the severe interference level and the upper threshold corresponding to the medium interference level.

[0012] Preferably, in step 4, the process of adjusting the data application strategy to correspond to traffic rules requirements includes: Verify the restored fused data to check the completeness, accuracy, consistency and timeliness of the data. The completeness check ensures that all necessary data fields are fully filled and there are no missing values. Check whether the number of data records is consistent with expectations and ensure that no data is lost. The accuracy check verifies the accuracy of the data by comparing the data source or using known correct values, and uses statistical methods to evaluate the rationality of the data. The consistency check ensures that the data between different data sources are logically consistent and checks whether the data follows the expected format and unit. The timeliness check verifies the timeliness of the data, ensures that the data is up to date and relevant to the current traffic rules requirements, and analyzes the data quality to identify outliers in the data. Based on the results of quality analysis, identify the sources of errors and perform error correction to control data errors. This includes analyzing the links that may introduce errors during data collection, processing, transmission, and storage, identifying common error sources such as sensor failure, data loss, or data conversion errors, using data smoothing techniques to correct random errors in the data, adjusting the data collection and processing methods for systematic errors, regularly evaluating the changing trends of data errors, and taking appropriate corrective measures; Analyze traffic rules requirements, including speed limits, traffic signal control, and lane allocation, adjust the precise requirements of corresponding data according to traffic rules requirements, synchronously adjust data application strategies, implement and monitor strategy effects, and correspond to traffic rules requirements.

[0013] Preferably, in step 5, the process of implementing targeted traffic management includes: Utilize real-time data processing technology to analyze collected vehicle status data in real time, monitor basic information such as vehicle speed, location, and driving direction, dynamically adjust traffic signal control strategies based on real-time traffic flow and vehicle status data, and improve traffic flow and reduce traffic congestion through intelligent signal light systems; When a traffic accident or road congestion occurs, the vehicle status data is used to quickly locate the problem area, and through traffic induction and diversion strategies, real-time traffic diversion and guidance information is provided to drivers to guide vehicles to avoid congested sections and balance the road network load; Using camera and vehicle status data analysis results, traffic violations are monitored and recorded in real time, and the violation data is uploaded to the traffic management department for subsequent punishment and education; Regularly analyze and evaluate the effectiveness of vehicle status monitoring and traffic management, use data visualization technology to display the changing trends of key indicators of traffic flow, congestion, and violations, and adjust and optimize traffic management strategies based on the analysis results.

[0014] The present invention provides a vehicle status acquisition method based on multi-source data fusion from a roadside perspective. It has the following beneficial effects: 1. This method of vehicle status collection based on multi-source data fusion from a roadside perspective analyzes the status of sensor data under different environmental conditions, analyzes the degree of interference with vehicle status data collection, and conducts quantitative analysis to implement corresponding data adjustment measures, thereby improving the accuracy of subsequent traffic management. By integrating data from different sources, real-time monitoring of vehicle status and dynamic traffic management are achieved, thereby providing a scientific basis for traffic management departments to formulate targeted preventive measures.

[0015] 2. This vehicle status collection method based on multi-source data fusion from a roadside perspective integrates data from different sensors to obtain basic information such as vehicle speed, location, and driving direction in real time and accurately, greatly improving the efficiency and accuracy of traffic management. During peak traffic hours, the system can quickly identify and handle traffic bottlenecks, and dynamically adjust traffic light control strategies to effectively alleviate traffic congestion and improve road capacity. At the same time, accurate vehicle status monitoring can also provide more reliable data support for traffic planning and decision-making, and help continuously optimize the urban transportation system. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A method flow chart of a vehicle status acquisition method based on multi-source data fusion from a roadside perspective of the present invention; Figure 2 is a flow chart for analyzing the severity of abnormal conditions of the present invention; Figure 3 This is a flow chart for obtaining the abnormal interference index of the present invention. DETAILED DESCRIPTION

[0017] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. The embodiments of the present invention are provided for the purpose of illustration and description, and are not intended to be exhaustive or to limit the present invention to the disclosed forms. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described in order to better illustrate the principles and practical applications of the present invention, and to enable those of ordinary skill in the art to understand the present invention and thereby design various embodiments with various modifications suitable for specific uses.

[0018] The first embodiment, as Figure 1 , Figure 2 As shown, the present invention provides a technical solution: a vehicle status acquisition method based on multi-source data fusion from a roadside perspective, comprising the following steps: Step 1: Deploy collection equipment on the roadside to collect raw data on vehicle status and environmental conditions, and pre-process the collected raw data. Deploy collection equipment on the roadside to ensure that the required traffic area can be covered, including cameras, radars, lidars, and infrared sensors to collect raw data on vehicle status and environmental conditions. Use the deployed cameras, radars, lidars, and infrared sensors to collect vehicle status data such as vehicle speed, position, and driving direction, and collect environmental condition data such as light intensity, weather conditions, temperature, humidity, and wind speed. Transmit the collected raw data to the data processing center wirelessly and perform pre-processing operations. The pre-processing includes data cleaning, format conversion, and data calibration steps. Data cleaning removes noise, duplicate values, and invalid values ​​from the data to ensure the accuracy and integrity of the data. Format conversion converts data collected by different sensors into a unified format to facilitate subsequent processing and analysis. Data calibration calibrates the data collected by the sensor to eliminate the errors and deviations of the sensor itself. Build a data warehouse, integrate the pre-processed data, and store it in the data warehouse to ensure the accessibility and security of the data. Step 2: Time synchronize the raw data collected by different sensors after preprocessing, and use data fusion technology to fuse the raw data of different sensors. Retrieve the raw data collected by different sensors after preprocessing from the data warehouse, perform software synchronization time synchronization on it, traverse each data record, record the timestamp of each sensor data, use the lidar data as the basis for time synchronization, extract its timestamp as the reference time point, and compare the data of other sensors (such as cameras, radars, infrared sensors, etc.) with the reference time point of the lidar according to their timestamps. Use the interpolation algorithm to calculate the data values ​​of other sensors at the lidar reference time point. Through interpolation, obtain the data of all sensors at the lidar reference time point, realize time synchronization from the software, and align the time and coordinates of the data obtained by different sensors. Accurate, wherein, time alignment is performed based on the result of the time synchronization operation to ensure the temporal consistency of the sensor data, and the different sensor data are converted into a unified coordinate system for coordinate alignment, so as to facilitate subsequent data fusion. The data fusion algorithm of the Kalman filter method is used to define the state equation to describe the change of sensor data over time, and to define the observation equation to describe the relationship between sensor data and state, initialize the state estimate and covariance matrix, and for each time step, use the state equation to predict the next state, use the observation equation and observation data to update the state estimate and covariance matrix, repeat the Kalman filter process until all sensor data are fused, and then obtain the fused data, integrate the information from different sensors, improve the accuracy and reliability of the data, and restore the fused data to the data warehouse for further analysis and processing; Step 3: In response to changes in different environmental conditions, monitor data anomalies in sensors and data fusion results, analyze the severity of anomalies, match corresponding recovery and adjustment measures, analyze various types of environmental condition data based on the fused data, monitor changes in different environmental conditions, and use time series analysis methods to conduct a comprehensive analysis of the monitoring data to identify abnormal data. Abnormal data is manifested as sudden changes in data values ​​and abnormal expansion of the data fluctuation range. The abnormal data is integrated to determine the characteristics of the abnormal data, which are the number of abnormal values, the duration of the anomaly, and the fluctuation range of the abnormal data. The interference of abnormal data on vehicle status recognition is analyzed, and the abnormal interference index is calculated to quantify the severity of the abnormal data. According to the severity of the abnormal data, different interference levels are divided, namely, slight interference level, medium interference level and severe interference level. Combined with the calculation Calculate the abnormal interference index, match the corresponding interference threshold for each interference level, and match the corresponding recovery adjustment measures for each interference level. The slight interference level corresponds to the slight abnormal situation, and take simple recovery adjustment measures such as adjusting the calibration parameters of the sensor and optimizing the data acquisition frequency. At the same time, strengthen monitoring and pay close attention to the changing trend of abnormal data; the medium interference level corresponds to the medium abnormal situation, and it is necessary to deeply analyze the abnormal causes involving sensor failure, data transmission error or environmental change. According to the analysis results, take corresponding recovery adjustment measures such as replacing the faulty sensor, repairing the data transmission link, and adjusting the monitoring strategy. During the recovery process, the severe interference level corresponds to the severe abnormal situation, and it is necessary to immediately activate the emergency plan, take emergency recovery adjustment measures such as shutting down the faulty equipment and starting the backup system, conduct in-depth analysis, find out the abnormal causes, formulate a long-term recovery plan, and strengthen safety monitoring during the recovery process; Furthermore, the process of abnormal data identification includes: Use the time series analysis method to decompose the time series of monitoring data into three parts: trend, seasonality and residual. According to the performance of sudden changes in the data value of abnormal data and abnormal expansion of the data fluctuation range, the anomaly detection threshold T is set in combination with the three sigma principle based on the normal distribution. Analyze the residual part of the time series decomposition of the monitoring data, calculate the residual at each time point, and for each time point t, calculate the median absolute deviation of the residuals of the w time points before and after it to measure the volatility of the data. Calculate the absolute value of the difference between the actual observation value at time point t and the benchmark value at time point t, and combine it with the median absolute deviation to construct an anomaly detection index. Calculate the anomaly detection index for each time point and compare it with the set anomaly detection threshold. If the anomaly detection index is greater than the anomaly detection threshold T, the data at time point t is abnormal data. Furthermore, the calculation expression of the anomaly detection index is: in, is the anomaly detection index, is the actual observed value at time point t, is the reference value at time point t, which is used to reflect the normal state of the data. is the median absolute deviation of the residuals at w time points before and after time point t, which is used to measure the volatility of the data. is the residual of w time points before and after time point t. The residual is the result of subtracting the baseline value from the original data; Step 4: Verify and analyze the quality of the adjusted and restored fusion data, control data errors, and adjust the data application strategy to meet traffic rules requirements; Step 5: Use the processed data to monitor vehicle status and implement targeted traffic management.

[0019] The second embodiment is based on the first embodiment. Figure 3 As shown, the process of obtaining the abnormal interference index includes: Integrate abnormal data from different sensors within a fixed time period, including speed, location, driving direction, light intensity, weather conditions, temperature, humidity and wind speed, to form an abnormal data set. The abnormal data set contains the timestamp and abnormal value information of each abnormal data point. Analyze the abnormal data set to determine the number of abnormal values, the duration of the abnormality, and the abnormal data fluctuation range. Evaluate the impact of abnormal data on vehicle status recognition, including the decrease in recognition accuracy and the increase in false alarm rate. Based on the evaluation results of the impact of abnormal data on vehicle status recognition, combine the abnormal data characteristics for comprehensive analysis to obtain the abnormal interference index, and quantify the severity of the interference of abnormal data on vehicle status recognition. Furthermore, the calculation expression of the abnormal interference index is: in, is the abnormal interference index, is the total number of outliers in a fixed time period, is the total duration of the outlier, is the maximum deviation of the abnormal value from its baseline value, is the number of baseline outliers, representing the number of outliers expected under normal conditions, To adjust the parameters, used to control the impact of the number of outliers on the index, is the maximum value of the abnormal duration during the observation period, is the maximum value of the abnormal fluctuation range during the observation period, The S-shaped sigmoid function converts the number of outliers to a value between 0 and 1. When the number of outliers is much higher than the baseline, the value is close to 1. The exponent is adjusted by the root formula and the maximum duration. When the abnormality lasts longer, the value of this item will decrease. It directly reflects the size of the abnormal fluctuation range and standardizes it to between 0 and 1. The value range is between 0 and 1, where 0 indicates no abnormal interference and 1 indicates extreme abnormal interference. As the severity of the abnormal data increases, The value of will also increase, so that the severity of abnormal data can be quantitatively analyzed to evaluate the interference of abnormal data on vehicle status recognition and take corresponding measures accordingly; The three interference levels correspond to three interference thresholds, wherein the interference threshold includes an upper threshold and a lower threshold; The three interference levels and the three interference thresholds satisfy the following relationship: Slight disturbance level ; Medium interference level ; Severe interference level ; in, is the abnormal interference index, is the lower threshold corresponding to the medium interference level and the upper threshold corresponding to the slight interference level, The lower threshold corresponding to the severe interference level and the upper threshold corresponding to the medium interference level are analyzed by analyzing the abnormal conditions of sensor data under different environmental conditions and the degree of interference of abnormal data on vehicle status data collection, and quantitative analysis is performed to implement corresponding data adjustment measures, thereby improving the accuracy of subsequent traffic management. By integrating data from different sources, real-time monitoring of vehicle status and dynamic traffic management are achieved, thus providing a scientific basis for traffic management departments to formulate targeted preventive measures; In step 4, the process of adjusting the data application strategy to correspond to traffic rules requirements includes: Verify the restored fused data to check the completeness, accuracy, consistency and timeliness of the data. The completeness check ensures that all necessary data fields are fully filled and there are no missing values. Check whether the number of data records is consistent with expectations and ensure that no data is lost. The accuracy check verifies the accuracy of the data by comparing the data source or using known correct values, and uses statistical methods to evaluate the rationality of the data. The consistency check ensures that the data between different data sources is logically consistent and checks whether the data follows the expected format and unit. The timeliness check verifies the timeliness of the data, ensures that the data is up to date and relevant to the current traffic regulations, and analyzes the data quality to identify anomalies in the data. value, identify the source of error based on the results of quality analysis, and perform error correction to control data errors, including analyzing the links that may introduce errors during data collection, processing, transmission and storage, identifying common error sources such as sensor failure, data loss or data conversion errors, using data smoothing technology to correct random errors in the data, and adjusting the data collection and processing methods for systematic errors. Regularly evaluate the changing trend of data errors and take corresponding corrective measures. Analyze traffic rules requirements, including speed limits, traffic signal control, and lane allocation, adjust the corresponding data accuracy requirements according to traffic rules requirements, and simultaneously adjust data application strategies, implement and monitor strategy effects, corresponding to traffic rules requirements; In step five, the process of implementing targeted traffic management includes: Utilize real-time data processing technology to analyze the collected vehicle status data in real time, monitor the basic information of vehicle speed, location, and driving direction, dynamically adjust the control strategy of traffic lights according to real-time traffic flow and vehicle status data, improve traffic fluidity and reduce traffic congestion through intelligent traffic light system, use vehicle status data to quickly locate problem areas when traffic accidents or road congestion occur, and provide drivers with real-time traffic diversion and guidance information through traffic induction and diversion strategies to guide vehicles to avoid congested sections and balance the road network load, use cameras and vehicle status data analysis results to monitor and record traffic violations in real time, upload violation data to traffic management departments for subsequent punishment and education, regularly analyze and evaluate vehicle status monitoring and traffic management effects, use data visualization technology to display the changing trends of key indicators of traffic flow, congestion, and violations, and adjust and optimize traffic management strategies based on the analysis results.

[0020] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field and related fields without creative work should fall within the scope of protection of the present invention. Structures, devices and operating methods not specifically described and explained in the present invention shall be implemented according to conventional means in the field unless otherwise specified and limited.

Claims

1. A vehicle status acquisition method based on multi-source data fusion from a roadside perspective, characterized in that: The following steps are involved: Step 1: Deploy a variety of collection devices on the roadside to collect raw data on vehicle status and environmental conditions, and pre-process the collected raw data; Step 2: Perform time synchronization on the raw data collected by different sensors after preprocessing, and use data fusion technology to fuse the raw data of different sensors; Step 3: Monitor the data anomalies in sensors and data fusion results according to the changes in different environmental conditions, analyze the severity of the anomalies, and match the corresponding recovery and adjustment measures; Step 4: Verify and analyze the quality of the adjusted and restored fusion data, control data errors, and adjust the data application strategy to meet traffic rules requirements; Step 5: Use the processed data to monitor vehicle status and implement targeted traffic management.

2. The vehicle status acquisition method based on multi-source data fusion from a roadside perspective according to claim 1 is characterized by: In step 1, the collection and preprocessing of raw data includes: Deploy a variety of data collection equipment on the roadside, including cameras, radars, lidars, and infrared sensors, to collect raw data on vehicle status and environmental conditions; Utilize the various collection devices deployed to collect vehicle status data such as vehicle speed, location, and direction of travel, as well as environmental condition data such as light intensity, weather conditions, temperature, humidity, and wind speed; The collected raw data is transmitted to the data processing center by wireless means and preprocessed, including data cleaning, format conversion and data calibration steps; Build a data warehouse, integrate the preprocessed data, and store it in the data warehouse.

3. The vehicle status acquisition method based on multi-source data fusion from a roadside perspective according to claim 2 is characterized in that: In step 2, the time synchronization and fusion process of the original data includes: Retrieve the pre-processed raw data collected by different sensors from the data warehouse, perform software synchronization on it, traverse each data record, record the timestamp of each sensor data, use the lidar data as the basis for time synchronization, and extract its timestamp as the reference time point; For the data of other sensors, compare their timestamps with the reference time point of the LiDAR, and use the interpolation algorithm to calculate the data values ​​of other sensors at the reference time point of the LiDAR. Through interpolation, the data of all sensors at the reference time point of the LiDAR are obtained, and time synchronization is achieved from the software. Performing time alignment and coordinate alignment on data acquired by different sensors, wherein the time alignment is performed based on the result of the time synchronization operation, and the different sensor data are converted into a unified coordinate system for coordinate alignment; Using the data fusion algorithm of Kalman filtering, define the state equation to describe the change of sensor data over time, and define the observation equation to describe the relationship between sensor data and state, initialize the state estimate and covariance matrix, for each time step, use the state equation to predict the next state, use the observation equation and observation data to update the state estimate and covariance matrix, repeat the Kalman filtering process until all sensor data are fused, and then obtain the fused data; The fused data is restored to the data warehouse for further analysis and processing.

4. The vehicle status acquisition method based on multi-source data fusion from a roadside perspective according to claim 3 is characterized by: In step 3, the analysis process of the severity of the abnormal situation includes: Based on the fused data, analyze various types of environmental condition data and monitor changes in different environmental conditions; Use time series analysis methods to conduct comprehensive analysis of monitoring data and identify abnormal data, which are manifested as sudden changes in data values ​​and abnormal expansion of data fluctuation range; Integrate the abnormal data and determine the characteristics of the abnormal data, namely the number of abnormal values, the duration of the abnormality, and the fluctuation range of the abnormal data. Analyze the interference of abnormal data on vehicle status recognition, calculate the abnormal interference index, and then quantify and analyze the severity of the abnormal data; According to the severity of abnormal data, different interference levels are divided, namely slight interference level, medium interference level and severe interference level. Combined with the calculated abnormal interference index, the corresponding interference threshold is matched for each interference level, and the corresponding recovery adjustment measures are matched for each interference level.

5. The vehicle status acquisition method based on multi-source data fusion from a roadside perspective according to claim 4 is characterized in that: The process of abnormal data identification includes: Using the time series analysis method, the time series of monitoring data is decomposed into three parts: trend, seasonality and residual. According to the sudden change of data values ​​and abnormal expansion of data fluctuation range of abnormal data, the anomaly detection threshold T is set in combination with the three sigma principle based on normal distribution. Analyze the residual part of the time series decomposition of the monitoring data, calculate the residual at each time point, and for each time point t, calculate the median absolute deviation of the residuals of the w time points before and after it to measure the volatility of the data; The absolute value of the difference between the actual observation value at time point t and the baseline value at time point t is calculated and combined with the median absolute deviation to construct an anomaly detection index; The anomaly detection index at each time point is calculated and compared with the set anomaly detection threshold. If the anomaly detection index is greater than the anomaly detection threshold T, the data at time point t is abnormal data.

6. The vehicle status acquisition method based on multi-source data fusion from a roadside perspective according to claim 5 is characterized by: The calculation expression of the anomaly detection index is: in, is the anomaly detection index, is the actual observed value at time point t, is the reference value at time point t, is the median absolute deviation of the residuals at w time points before and after time point t, is the residual of w time points before and after time point t.

7. The vehicle status acquisition method based on multi-source data fusion from a roadside perspective according to claim 6 is characterized by: The process of obtaining the abnormal interference index includes: Integrate abnormal data from different sensors within a fixed time period, including speed, location, driving direction, light intensity, weather conditions, temperature, humidity and wind speed, to form an abnormal data set. The abnormal data set contains the timestamp and abnormal value information of each abnormal data point; Analyze the abnormal data set to determine the number of abnormal values, the duration of the abnormality, the fluctuation range of the abnormal data, and evaluate the impact of abnormal data on vehicle status recognition, including the decrease in recognition accuracy and the increase in false alarm rate; Based on the evaluation results of the impact of abnormal data on vehicle state recognition, the abnormal interference index is obtained by combining the comprehensive analysis of abnormal data characteristics to quantify the severity of interference of abnormal data on vehicle state recognition.

8. The vehicle status acquisition method based on multi-source data fusion from a roadside perspective according to claim 7 is characterized by: The calculation expression of the abnormal interference index is: in, is the abnormal interference index, is the total number of outliers in a fixed time period, is the total duration of the outlier, is the maximum deviation of the abnormal value from its baseline value, is the number of baseline outliers, representing the number of outliers expected under normal conditions, To adjust the parameters, used to control the impact of the number of outliers on the index, is the maximum value of the abnormal duration during the observation period, is the maximum value of the abnormal fluctuation range during the observation period; The three interference levels correspond to the three interference thresholds, wherein the interference thresholds include an upper threshold and a lower threshold; The three interference levels and the three interference thresholds satisfy the following relationship: Slight disturbance level ; Medium interference level ; Severe interference level ; in, is the abnormal interference index, is the lower threshold corresponding to the medium interference level and the upper threshold corresponding to the slight interference level, is the lower threshold corresponding to the severe interference level and the upper threshold corresponding to the medium interference level.

9. The vehicle status acquisition method based on multi-source data fusion from a roadside perspective according to claim 8 is characterized by: In step 4, the process of adjusting the data application strategy to correspond to traffic rules requirements includes: Verify the restored fusion data and check the integrity, accuracy, consistency and timeliness of the data; Based on the results of quality analysis, identify the source of error and perform error correction to control data errors; Analyze traffic rules requirements, including speed limits, traffic signal control, and lane allocation, adjust the precise requirements of corresponding data according to traffic rules requirements, synchronously adjust data application strategies, implement and monitor strategy effects, and correspond to traffic rules requirements.

10. The vehicle status acquisition method based on multi-source data fusion from a roadside perspective according to claim 9, characterized in that: In step 5, the process of implementing targeted traffic management includes: Use real-time data processing technology to analyze the collected vehicle status data in real time, monitor the basic information of vehicle speed, location, and driving direction, and dynamically adjust the control strategy of traffic lights based on real-time traffic flow and vehicle status data; When a traffic accident or road congestion occurs, the vehicle status data is used to quickly locate the problem area, and through traffic induction and diversion strategies, real-time traffic diversion and guidance information is provided to drivers to guide vehicles to avoid congested sections and balance the road network load; Using camera and vehicle status data analysis results, traffic violations are monitored and recorded in real time, and the violation data is uploaded to the traffic management department; Regularly analyze and evaluate the effectiveness of vehicle status monitoring and traffic management, use data visualization technology to display the changing trends of key indicators of traffic flow, congestion, and violations, and adjust and optimize traffic management strategies based on the analysis results.

Images