Vehicle-road cooperation application scene test method and system

Abstract

The invention discloses a vehicle-road cooperation application scene test method and system, and relates to the technical field of intelligent traffic, and the method comprises the steps: in the vehicle-road cooperation application, a plurality of sensors are disposed on a vehicle, and vehicle multi-source data are collected in real time; fusing the vehicle multi-source data by using the parallel processing capability of quantum computing; constructing a virtual test environment on a high-performance computing platform based on the fused data; testing is carried out in the virtual testing environment, real-time monitoring is started, and abnormal conditions occurring in the testing process are automatically recognized through an abnormal algorithm; and according to a detected abnormal result, an intelligent intervention and error correction mechanism is automatically started. According to the invention, by introducing the parallel processing capability of quantum calculation and an advanced anomaly detection algorithm, the perception capability and decision-making level of the vehicle in a complex traffic environment are significantly improved, and strong advantages are shown in practical application.

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation technology, and particularly to a vehicle-road collaborative application scenario testing method and system. Background Art

[0002] In recent years, vehicle-road collaborative technology has made remarkable progress, especially in multi-sensor data fusion, virtual environment construction, and intelligent decision-making. Traditional vehicle-road collaborative systems usually rely on multiple sensors to collect data on the vehicle's surrounding environment in real time and generate a unified environmental perception model through data fusion algorithms. However, existing data fusion technologies have certain limitations in processing a large amount of real-time data, especially in terms of data processing speed and accuracy. In addition, although existing virtual test environments can simulate some traffic scenarios, there is still room for improvement in terms of authenticity and real-time performance in complex environments. Moreover, traditional anomaly detection algorithms are prone to false detection and missed detection in the face of complex and changing traffic scenarios, resulting in inaccurate intelligent intervention measures.

[0003] To overcome the above deficiencies, in recent years, quantum computing technology has received extensive attention due to its excellent parallel processing ability. Quantum computing can significantly improve data processing speed and accuracy, especially showing great potential in vehicle multi-source data fusion. In addition, quantum computing can also be used to optimize virtual environment construction and anomaly detection algorithms, thereby improving the robustness and real-time response ability of the system. Based on these, the present invention proposes a method for data fusion using quantum computing and constructing a high-precision virtual test environment on this basis, solving the problems of slow data processing speed, inaccurate environment simulation, and inaccurate anomaly detection existing in the prior art. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a vehicle-road collaborative application scenario testing method and system to solve the problem of efficient fusion of vehicle multi-source data in vehicle-road collaborative applications.

[0006] To solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, an embodiment of the present invention provides a method for testing a vehicle-road collaborative application scenario, which includes, in a vehicle-road collaborative application, a vehicle deploying multiple sensors to collect vehicle multi-source data in real time; using the parallel processing ability of quantum computing to fuse the vehicle multi-source data; based on the fused data, constructing a virtual test environment on a high-performance computing platform; performing tests in the virtual test environment and starting real-time monitoring to automatically identify abnormal situations occurring during the test process through an abnormal algorithm; according to the detected abnormal results, automatically starting an intelligent intervention and error correction mechanism; after completing the error correction mechanism, continuing to execute the test task and analyzing the data during the test process to make optimizations.

[0008] As a preferred solution of the vehicle-road collaborative application scenario testing method of the present invention, wherein: in the vehicle-road collaborative application, the vehicle deploys multiple sensors to collect vehicle multi-source data in real time, and the specific steps are as follows:

[0009] To achieve comprehensive perception of the vehicle's surrounding environment, multiple sensors are deployed, including millimeter-wave radar, lidar, high-definition cameras, global positioning, and inertial measurement units;

[0010] The sensors are installed at different positions on the vehicle to cover a 360-degree view, and the sampling frequency is configured to continuously collect vehicle multi-source data, including obstacle position, obstacle speed, obstacle direction, traffic signal status, vehicle's own position, vehicle's own speed, vehicle's own direction, and vehicle's own acceleration.

[0011] As a preferred solution of the vehicle-road collaborative application scenario testing method of the present invention, wherein: using the parallel processing ability of quantum computing to fuse the vehicle multi-source data, the specific steps are as follows:

[0012] Transmit the collected vehicle multi-source data to a fusion center based on quantum computing through a 5G high-speed communication network;

[0013] After the fusion center receives the vehicle multi-source data, it preprocesses the vehicle multi-source data;

[0014] Map the preprocessed vehicle multi-source data to quantum state data through quantum coding;

[0015] Adopt the superposition and entanglement characteristics of quantum algorithms to fuse the quantum state data to generate a fused quantum state.

[0016] As a preferred solution of the vehicle-road collaborative application scenario testing method of the present invention, wherein: based on the fused data, constructing a virtual test environment on a high-performance computing platform, the specific steps are as follows:

[0017] Obtain weather data through an external meteorological service, obtain local environmental information through a roadside unit in the vehicle-road collaborative infrastructure, and obtain road characteristics through geographic information;

[0018] Integrate weather data, local environmental information, and road characteristics into an environmental parameter set through data formatting and time synchronization;

[0019] Receive the fused quantum state and the environmental parameter set through a high-performance computing platform and perform an environmental construction task;

[0020] The high-performance computing platform analyzes the fused quantum state through quantum measurement, extracts the dynamic and static characteristics of the quantum state, and combines with the environmental parameter set to adjust the vehicle's dynamic behavior;

[0021] Based on the fused quantum state and the environmental parameter set through numerical simulation, use the adjusted vehicle dynamic behavior to simulate the vehicle's dynamic behavior and generate the vehicle-environment interaction result;

[0022] The high-performance computing platform normalizes the vehicle-environment interaction result, balances the contribution ratio of the fused quantum state and the environmental parameter set, and generates a virtual test environment.

[0023] As a preferred solution of the vehicle-road collaborative application scenario test method described in the present invention, wherein: perform tests in the virtual test environment and start real-time monitoring to automatically identify abnormal situations occurring during the test through an abnormal algorithm. The specific steps are as follows:

[0024] In the constructed virtual test environment, simulate real driving scenarios and test the vehicle's behavioral responses;

[0025] Start real-time monitoring and automatically identify abnormal situations occurring during the test through an anomaly detection algorithm based on machine learning, including abnormal vehicle behavior, abnormal sensor readings, environmental perception errors, communication failures, and decision logic errors.

[0026] As a preferred solution of the vehicle-road collaborative application scenario test method described in the present invention, wherein: according to the detected abnormal results, automatically start an intelligent intervention and error correction mechanism. The specific steps are as follows:

[0027] During the virtual test environment process, when an abnormal situation is found, trigger intelligent intervention measures, including adjusting the vehicle's behavior mode, re-planning the driving route, decelerating and stopping, steering to avoid, lighting and sound warnings, emergency communication, switching the autonomous driving mode, and activating backup sensors.

[0028] As a preferred solution of the vehicle-road collaborative application scenario test method described in the present invention, wherein: after completing the error correction mechanism, continue to execute the test task and analyze the data during the test to make optimizations. The specific steps are as follows:

[0029] Continue to execute the test task and define a series of performance metrics;

[0030] Use statistical methods to evaluate the stability and reliability of test results;

[0031] Apply cluster analysis to identify patterns and trends in test data;

[0032] Explore the reasons behind test results through a causal diagram;

[0033] Based on the analysis results, generate optimization suggestions for the test plan and the vehicle control platform;

[0034] Apply the optimization suggestions to a new test cycle, execute the virtual test again, record the results, and compare the differences before and after optimization.

[0035] In a second aspect, the present invention provides a vehicle-road collaborative application scenario test system, including a data acquisition module, a data fusion module, a virtual environment construction module, a test monitoring module, an intelligent intervention and error correction module, and an analysis and optimization module; the data acquisition module is used to deploy multiple sensors on the vehicle in vehicle-road collaborative applications to collect multi-source vehicle data in real time; the data fusion module is used to fuse the multi-source vehicle data by utilizing the parallel processing ability of quantum computing; the virtual environment construction module is used to construct a virtual test environment on a high-performance computing platform based on the fused data; the test monitoring module is used to perform tests in the virtual test environment and start real-time monitoring to automatically identify abnormal situations occurring during the test through an abnormal algorithm; the intelligent intervention and error correction module is used to automatically start an intelligent intervention and error correction mechanism according to the detected abnormal results; the analysis and optimization module is used to continue to execute the test task after completing the error correction mechanism, analyze the data during the test process, and make optimizations.

[0036] In a third aspect, an embodiment of the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and: when the computer program is executed by the processor, any step of the vehicle-road collaborative application scenario test method as described in the first aspect of the present invention is implemented.

[0037] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, and: when the computer program is executed by the processor, any step of the vehicle-road collaborative application scenario test method as described in the first aspect of the present invention is implemented.

[0038] The beneficial effects of the present invention are as follows: By utilizing the parallel processing ability of quantum computing to fuse multi-source data of vehicles, the speed and accuracy of data processing are significantly improved. The quantum algorithm not only accelerates the data fusion process, but also ensures the consistency and integrity of the fused data by mapping sensor data into quantum states and performing efficient operations. This efficient quantum data fusion method reduces the latency of data processing, enhances the realism and dynamic response ability of the virtual test environment, enabling the test environment to more quickly reflect changes in the real world. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for description in the embodiments. Obviously, the following described drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0040] Figure 1 It is a flowchart of the vehicle-road cooperation application scenario test method in Embodiment 1.

[0041] Figure 2 It is a flowchart of analyzing and optimizing data during the test in Embodiment 1. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0042] In order to make the above objects, features, and advantages of the present invention more obvious and understandable, the following will make a detailed description of the specific embodiments of the present invention with reference to the drawings in the specification.

[0043] In the following description, many specific details are set forth to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar generalizations without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0044] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that can be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment that excludes other embodiments.

[0045] Embodiment 1, referring to Figure 1 and Figure 2 , is the first embodiment of the present invention. This embodiment provides a vehicle-road cooperation application scenario test method, including the following steps:

[0046] S1. In vehicle-road cooperation applications, a vehicle deploys multiple sensors to collect multi-source data of the vehicle in real time.

[0047] Furthermore, to achieve a comprehensive perception of the vehicle's surrounding environment, multiple sensors are deployed, including millimeter-wave radar, lidar, high-definition cameras, global positioning, and inertial measurement units;

[0048] The sensors are installed at different positions on the vehicle to cover a 360-degree view, and the sampling frequency is configured to continuously collect multi-source data of the vehicle, including obstacle position, obstacle speed, obstacle direction, traffic signal status, vehicle's own position, vehicle's own speed, vehicle's own direction, and vehicle's own acceleration.

[0049] It should be noted that in vehicle-road collaborative applications, by deploying a variety of sensors including millimeter-wave radar, lidar, high-definition cameras, global positioning system, and inertial measurement units, a comprehensive perception of the vehicle's surrounding environment is achieved; these sensors are strategically installed at different positions on the vehicle to ensure 360-degree coverage without dead angles, and continuously collect data such as the position, speed, and direction of obstacles, as well as the vehicle's own state information at a high sampling frequency; the goal of this deployment is to provide real-time, accurate, and comprehensive environmental perception data to support the vehicle in making timely and precise decisions, enhancing the safety and reliability of autonomous driving.

[0050] S2. Utilize the parallel processing ability of quantum computing to fuse the vehicle's multi-source data.

[0051] Furthermore, the collected multi-source data of the vehicle is transmitted to a quantum computing-based fusion center through a 5G high-speed communication network;

[0052] Specifically, through the 5G high-speed communication network, the high bandwidth (up to 10 Gbps) and low latency (within 1 ms) characteristics of 5G meet the real-time requirements of vehicle-road collaboration;

[0053] To ensure data integrity and security, the vehicle's multi-source data is protected using timestamp marking (recording the acquisition time) and encryption protocols (such as TLS or AES) to prevent tampering or loss during transmission;

[0054] Lossless compression is used to reduce bandwidth occupancy while retaining key information for transmission to the quantum computing-based fusion center;

[0055] After the fusion center receives the vehicle's multi-source data, it preprocesses the vehicle's multi-source data (including denoising, time synchronization, and format unification to eliminate heterogeneity and noise interference between sensors);

[0056] Specifically, filtering (such as Kalman filtering or median filtering) is used to smooth the data;

[0057] Due to the different sampling frequencies of different sensors (e.g., LiDAR at 10 frames per second and cameras at 30 frames per second), the data is unified to the same time base through interpolation algorithms (such as linear interpolation) or timestamp alignment;

[0058] Convert heterogeneous data (such as the range-angle data of radar and camera pixel data) into a unified vector representation (such as three-dimensional coordinates or velocity vectors) for subsequent quantum encoding;

[0059] Detect and remove obvious outliers (such as invalid data caused by overexposure of the camera due to strong light) through statistical methods (such as the Z-score);

[0060] Map the preprocessed multi-source vehicle data to quantum state data through quantum encoding;

[0061] Utilize the superposition and entanglement properties of quantum algorithms to fuse the quantum state data and generate a fused quantum state. The expression is:

[0062]

[0063] where, |ψ f > is the fused quantum state, which is the final quantum state after the fusion process and is used to represent the fused multi-source vehicle data, is the normalization factor that ensures the final quantum state, is the summation symbol, indicating the sum over all i from 1 to n, w i is the weight factor of the i-th data point, i is the index of the data point, n is the total number of all sensor data participating in the fusion, which indicates how many data points are used to construct the final fused quantum state, |D i > is the quantum state corresponding to the i-th data point, and each sensor data D i is mapped to a quantum state |D i >, and this quantum state contains the relevant information of this data point.

[0064] Specifically, assume that three quantum states have been obtained from the multi-source vehicle data (such as an obstacle at a distance of 5 meters, a speed of 2 meters per second, and a direction of 30 degrees), just like three small boxes filled with information, each representing a data point. These small boxes have been prepared by the quantum computer, each carrying its own information and ready to be used at any time. Now, what needs to be done is to use the quantum computer to mix the information in these three small boxes into a big box. This big box can represent all the data simultaneously and is more powerful than looking at each box separately. This is the fusion process;

[0065] Quantum computers rely on superposition and entanglement to make them "speak" simultaneously. Superposition is like having three radio channels, each broadcasting a different voice. An ordinary computer has to listen to them one by one, but a quantum computer can turn on all three channels at the same time, allowing all the voices to overlap and form a new voice. Quantum state fusion works the same way. The states in three small boxes are put into the quantum computer and made to "speak" simultaneously. However, they can't just be randomly superimposed; there must be a sense of importance. For example, the distance of 5 meters is measured by lidar and is very accurate, so we listen to it more. The speed of 2 meters per second is measured by an ordinary radar and has some errors, so we listen to it less. The angle of 30 degrees is seen by the camera and is very important, but it depends on the situation. This degree of importance is like attaching a label to each box, telling the quantum computer which box should speak loudly and which should speak softly. These labels are called "weights" and are determined based on the credibility of the multi-source data of the vehicle.

[0066] Then, the quantum computer gets to work and uses some special tools (like switches or knobs) to mix the information from the three boxes according to their importance. Imagine there are three water cups, the first one has more water, the second one has less, and the third one has a moderate amount. Pour them into a big cup, and the water mixes together, but we can still see how much each cup contributed. The quantum computer does the same thing, superimposing the states in the three small boxes proportionally to form a new state.

[0067] During this process, entanglement also helps. Entanglement is like tying an invisible string to these three boxes, making them no longer separate individuals but friends that influence each other. For example, distance and speed are actually related. The quantum computer uses entanglement to let them communicate, ensuring that the resulting big box not only contains the three pieces of information but also reflects the connection between them. In this way, the fused result is smarter and can help better judge the situation of obstacles.

[0068] Finally, we get a big box, which is the fused quantum state. On the surface, it doesn't look like an ordinary number; we can't say if it's 5 or 2. But it hides the shadows of all the data, like a super information packet. The quantum computer knows how to use it. To make it useful, the quantum computer will automatically adjust the size of this big box to ensure it's neither too big nor too small, just like adjusting the volume to make the sound just right. This big box can be passed on to the next step, such as simulating how the vehicle will react or deciding how to drive, because it can see the whole picture at once and is much stronger than the individual boxes.

[0069] It should be noted that by utilizing the parallel processing ability of quantum computing, the multi-source vehicle data on the vehicle is transmitted to the quantum computing fusion center, and quantum algorithms are used for data fusion. The purpose of this step is to efficiently integrate information from different sensors through the advantages of quantum computing, form a unified and high-precision quantum state, thereby improving the accuracy and real-time performance of vehicle decision-making.

[0070] S3. Based on the fused data, construct a virtual test environment on a high-performance computing platform.

[0071] Furthermore, obtain weather data (such as rainfall intensity, wind speed) through external meteorological services (such as the weather forecast service of the meteorological bureau), obtain local environmental information (such as traffic flow, road conditions) through the roadside unit in vehicle-road collaborative infrastructure (highway V2X pilot), and obtain road characteristics (such as road friction coefficient, slope) through geographical information (such as maps);

[0072] Integrate the weather data, local environmental information, and road characteristics into an environmental parameter set through data formatting and time synchronization to ensure compatibility with the fused quantum state in terms of time and format;

[0073] Receive the fused quantum state and the environmental parameter set through the high-performance computing platform and execute the environmental construction task;

[0074] Specifically, receive the fused quantum state through the high-performance computing platform. The fused quantum state integrates the multi-source vehicle data collected by vehicle sensors and contains a comprehensive representation of the perception information. At the same time, receive the environmental parameter set. The high-performance computing platform couples the fused quantum state with the environmental parameter set through a calculation function and executes the environmental construction task, laying the foundation for the generation of the virtual test environment;

[0075] The high-performance computing platform analyzes the fused quantum state through quantum measurement, extracts the dynamic characteristics (such as obstacle speed, vehicle motion state) and static characteristics (such as obstacle position, traffic signal) of the quantum state, and combines with the environmental parameter set (such as road friction coefficient, rainfall intensity, traffic flow) to adjust the vehicle's dynamic behavior, such as braking response under wet conditions or stability adjustment in strong winds, providing accurate input for numerical simulation;

[0076] The high-performance computing platform simulates the dynamic behavior of the vehicle based on the fused quantum state and the environmental parameter set through numerical simulation, and generates the interaction results between the vehicle and the environment;

[0077] Specifically, the simulation process comprehensively considers the perception information provided by the fused quantum state and the external constraints of the environmental parameter set, generates dynamic behavior data such as the vehicle's motion trajectory, acceleration change, and steering response, as well as the digital mapping of the environmental conditions, as the interaction results between the vehicle and the environment;

[0078] The high-performance computing platform normalizes the interaction results between the vehicle and the environment, balances the contribution ratio of the fused quantum state and the environmental parameter set, and generates a virtual test environment;

[0079] Specifically, the processed interaction results are integrated into a virtual test environment, which includes the dynamic behavior of the vehicle and the digital mapping of environmental parameters, supports real-time updates, and can be used for the test tasks of vehicle-road collaborative applications;

[0080] It should be noted that by integrating multi-source environmental data and quantum state perception information, a virtual test environment is constructed, effectively improving the test efficiency and reliability of vehicle-road collaborative applications. The formatting and time synchronization of the environmental parameter set ensure data consistency. The analysis of quantum measurements and dynamic behavior adjustment optimize the vehicle response accuracy. The numerical simulation of the high-performance computing platform makes full use of the quantum state and environmental constraints to generate realistic vehicle motion trajectories and environmental mappings. The normalization process balances the contributions of various factors to ensure environmental stability. The finally generated virtual test environment supports real-time updates, covers dynamic behavior and environmental parameters, provides reliable support for vehicle decision-making and performance evaluation in complex scenarios, significantly reduces the cost and risk of actual road tests, and promotes the development of intelligent transportation.

[0081] S4. Conduct tests in the virtual test environment and start real-time monitoring to automatically identify abnormal situations occurring during the test process through an abnormal algorithm.

[0082] Furthermore, in the constructed virtual test environment, simulate real driving scenarios and test the vehicle's behavioral responses;

[0083] Start real-time monitoring and automatically identify abnormal situations occurring during the test process through an anomaly detection algorithm based on machine learning, including abnormal vehicle behavior, abnormal sensor readings, environmental perception errors, communication failures, and decision logic errors.

[0084] The anomaly detection algorithm determines whether an anomaly occurs by calculating the distance between each test instance and the training data and judging through a preset threshold;

[0085] To determine the threshold τ in the abnormal algorithm, it is necessary to define a range of normal behaviors based on historical data. In this scenario, a statistical-based method can be used to determine a specific threshold; the specific steps are as follows:

[0086] Collect vehicle sensor data under normal operating conditions from the historical dataset, including but not limited to radar data, lidar data, camera data, GPS data, etc.; preprocess the data, including data cleaning, missing value processing, etc.;

[0087] Extract meaningful features from the raw data, such as the rate of change of distance, the rate of change of speed, the rate of change of direction, etc. These features can help better describe vehicle behavior;

[0088] Use unsupervised learning methods (such as Isolation Forest or One-Class SVM) to train a model that can learn the distribution of normal behavior; for example, using the Isolation Forest algorithm, a model that can distinguish normal data from abnormal data can be constructed;

[0089] Use the trained model to score new data points to obtain an anomaly score S; this score represents the degree of deviation between the new data point and the normal behavior in the training set;

[0090] Based on the anomaly score distribution of the training data set, select a threshold τ such that all data points with scores lower than τ are considered normal, while data points higher than τ are considered abnormal; specifically, the 99th percentile of the anomaly scores in the training set can be selected as the threshold τ, which can ensure that only a very small number of extreme cases are labeled as abnormal.

[0091] The specific threshold is as follows:

[0092] Assume that the training data set contains a large amount of normal driving data, and calculate the anomaly score S for each data point through the Isolation Forest model; if the maximum value of the anomaly scores in the training data set is found to be 0.95;

[0093] This means that any data point with a score S > 0.5 will be considered an abnormal situation. Specifically, if the behavior of the vehicle in the virtual test environment causes the anomaly score of its sensor data to exceed 0.5, then the system will trigger the intelligent intervention mechanism and take corresponding corrective measures. If the score at the 99th percentile is 0.5, then τ can be set to 0.5;

[0094] When S < τ (S < 0.5), the data point is considered part of normal behavior and does not require special handling;

[0095] When S ≥ τ (S ≥ 0.5), the data point is considered abnormal behavior, triggering an alarm mechanism or intelligent intervention measures;

[0096] It should be noted that the specific implementation steps of the anomaly detection algorithm are as follows:

[0097] Collect vehicle sensor data under normal operating conditions;

[0098] Clean the data and handle missing values;

[0099] Extract meaningful features from the original data, such as the rate of change of distance, the rate of change of speed, the rate of change of direction, etc.;

[0100] Use an unsupervised learning method to train a model to learn the distribution of normal behaviors;

[0101] Use the trained model to score new data points to obtain an anomaly score S;

[0102] Based on the anomaly score distribution of the training dataset, select a threshold τ such that all data points with scores lower than τ are considered normal, while those higher than τ are considered anomalous

[0103] If the anomaly score S of the new data point is greater than or equal to the threshold τ, then the data point is considered anomalous;

[0104] For example, in the case of using the Isolation Forest model, the model will construct one or more decision trees and try to isolate each data point as quickly as possible; normal data points usually require more splits to be isolated, so they have a longer average path length; on the contrary, anomalous data points will be isolated faster and have a shorter average path length; by comparing the average path length of the new data point with that of the normal data points, its degree of anomaly can be evaluated and the threshold τ can be set accordingly.

[0105] Isolation Forest is an unsupervised learning method mainly used for anomaly detection; its basic idea is to construct decision trees by randomly selecting features and randomly splitting feature values until all data points are isolated (i.e., each node contains only one data point);

[0106] The working principle is as follows: Isolation Forest constructs decision trees by recursively splitting the data space; each split randomly selects a feature and a random value on that feature, which can effectively distinguish anomalous points from normal points because anomalous points often do not require many splits to be isolated.

[0107] Measuring the anomaly score: The anomaly score of each data point is determined by its path length in the tree; normal data points usually require more splits to be isolated, so they have a longer average path length; on the contrary, anomalous data points will be isolated faster and have a shorter average path length. The lower the score, the more likely it is to be an anomalous point.

[0108] It should be noted that in the virtual test environment, by simulating real driving scenarios and starting real-time monitoring, combined with anomaly detection algorithms based on machine learning, the aim is to automatically identify anomalies in aspects such as vehicle behavior, sensor readings, environmental perception, communication, and decision-making logic during the test process; this process first extracts normal operation characteristics from historical data, uses unsupervised learning techniques such as Isolation Forest to train a model to identify anomaly patterns, and sets a threshold τ to distinguish normal and abnormal behaviors; once the detected anomaly score S exceeds the threshold, the system will trigger an alarm or intelligent intervention measures, thereby verifying the stability and safety of the vehicle system and preventing potential problems in advance.

[0109] S5. Automatically activate the intelligent intervention and error correction mechanism according to the detected abnormal results.

[0110] Furthermore, during the virtual test environment process, when an abnormal situation is detected, intelligent intervention measures will be triggered, including adjusting the vehicle behavior pattern, re-planning the driving route, decelerating and stopping, steering to avoid, lighting and sound warnings, emergency communication, switching the autopilot mode, and activating backup sensors.

[0111] It should be noted that adjusting the vehicle behavior pattern: if it is detected that the vehicle deviates from the expected trajectory, it can be corrected by adjusting the vehicle's speed or direction;

[0112] Re-planning the driving route: when encountering obstacles or other dangerous situations, the system can recalculate a safer path;

[0113] Decelerating and stopping: when it is detected that there is an emergency ahead, the vehicle should immediately decelerate or come to a complete stop;

[0114] Steering to avoid: when an obstacle appears on the driving path, the vehicle can automatically steer to avoid the obstacle;

[0115] Lighting and sound warnings: send warning signals to other drivers to prompt them to pay attention to the emergency operations being carried out by the vehicle;

[0116] Emergency communication: send an alarm to nearby vehicles or infrastructure to notify them of potential risks;

[0117] Autopilot mode switching: if it is detected that the system cannot handle a situation, switch to the manual driving mode or hand over control to a higher-level decision-making system;

[0118] Activating backup sensors: if the main sensor fails or is interfered with, activate the backup sensor to maintain the sensing function.

[0119] It should be noted that the actual description of the intelligent intervention measures:

[0120] Adjust the vehicle behavior mode: Suppose the vehicle detects a pedestrian suddenly crossing the street ahead, it will automatically decelerate and change direction to avoid collision;

[0121] Re-plan the driving route: If the road ahead is blocked due to an accident, the vehicle will recalculate a detour route based on the latest map information;

[0122] Decelerate and stop: When the sensor detects an obstacle ahead, the vehicle will automatically decelerate until it comes to a complete stop;

[0123] Steer to avoid: On the highway, if a vehicle in the adjacent lane suddenly changes lanes and approaches, the autonomous vehicle will slightly adjust the steering wheel to maintain a safe distance;

[0124] Lights and sound warnings: If the vehicle is about to perform an emergency brake, it will turn on the brake lights and send a warning signal to other vehicles;

[0125] Emergency communication: When the vehicle encounters an emergency, it will send an emergency communication signal to other nearby vehicles to remind them to pay attention to safety;

[0126] Autopilot mode switch: If a system failure is detected or external conditions do not allow autopilot, the vehicle will prompt the driver to take over control;

[0127] Activate backup sensors: When the mainly used sensors have problems, the system will switch to backup sensors to ensure continuous acquisition of necessary environmental information.

[0128] It should be noted that in the virtual test environment, once an abnormal situation is detected, the system will automatically start the intelligent intervention and error correction mechanism; these measures include adjusting the vehicle behavior mode, re-planning the driving route, decelerating or stopping, steering to avoid, emitting lights and sound warnings, initiating emergency communication, switching the autopilot mode, and activating backup sensors; these intervention measures are aimed at quickly responding to potential risks, ensuring that the vehicle can take reasonable actions when facing emergencies, protecting the safety of occupants and other road users, and maintaining the stability and reliability of the system.

[0129] S6. After completing the error correction mechanism, continue to execute the test task, and analyze the data during the test to make optimizations.

[0130] Continue to execute the test task, define a series of performance indicators, such as vehicle response time, path planning efficiency, collision avoidance success rate, etc. Each indicator should have a clear calculation method and standard unit;

[0131] Use statistical methods to evaluate the stability and reliability of the test results;

[0132] Apply cluster analysis to identify patterns and trends in the test data;

[0133] Through the cause-and-effect diagram, explore the reasons behind the test results. For example, determine whether a certain sensor failure directly leads to a specific type of abnormal behavior;

[0134] Based on the analysis results, generate test plans and optimization suggestions for the vehicle control platform;

[0135] Apply the optimization suggestions to the new test cycle, execute the virtual test again, record the results, and compare the differences before and after optimization.

[0136] It should be noted that the performance indicators are defined as follows: for example, "vehicle response time" can be defined as the time required from the system detecting an event to taking corresponding actions, "path planning efficiency" can be defined as the ability to plan the shortest or fastest path, and "collision avoidance success rate" refers to the proportion of successfully avoiding collisions.

[0137] The application of statistical methods is as follows: in order to evaluate the stability and reliability of test results, hypothesis testing (such as t-test) can be used to determine whether there are significant differences in the results between different test groups; analysis of variance (ANOVA) can be used to compare the mean differences between multiple sample groups; regression analysis can also be used to explore the strength of the relationship between different variables.

[0138] It should be noted that after completing the intelligent error correction mechanism, the test tasks will continue to be executed and the test data will be analyzed in depth; by defining performance indicators such as vehicle response time and path planning efficiency, and using statistical methods to evaluate the stability and reliability of the results, the effectiveness of the test can be ensured; using cluster analysis to identify data patterns and combining with the cause-and-effect diagram to explore the root causes helps to understand the connection between sensor failures and abnormal behaviors; based on the analysis results, optimization suggestions are put forward and implemented in the new test cycle, and the differences before and after optimization are recorded and compared, so as to continuously improve the performance and safety of the autonomous driving system.

[0139] This embodiment also provides a vehicle-road collaborative application scenario test system, including: a data acquisition module, a data fusion module, a virtual environment construction module, a test monitoring module, an intelligent intervention and error correction module, and an analysis and optimization module; the data acquisition module, in vehicle-road collaborative applications, deploys multiple sensors on the vehicle to collect multi-source vehicle data in real time; the data fusion module uses the parallel processing ability of quantum computing to fuse the multi-source vehicle data; the virtual environment construction module constructs a virtual test environment on a high-performance computing platform based on the fused data; the test monitoring module conducts tests in the virtual test environment and starts real-time monitoring to automatically identify abnormal situations that occur during the test through abnormal algorithms; the intelligent intervention and error correction module automatically activates the intelligent intervention and error correction mechanism according to the detected abnormal results; the analysis and optimization module, after completing the error correction mechanism, continues to execute the test task and analyzes the data during the test to make optimizations.

[0140] This embodiment also provides a computer device applicable to the vehicle-road collaborative application scenario test method, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the vehicle-road collaborative application scenario test method proposed in the above embodiment.

[0141] This computer device can be a terminal, and this computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected through a system bus. Among them, the processor of this computer device is used to provide computing and control capabilities. The memory of this computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of this computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, a carrier network, NFC (Near Field Communication), or other technologies. The display screen of this computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of this computer device can be a touch layer covering the display screen, or a button, a trackball, or a touchpad provided on the housing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0142] This embodiment also provides a storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the vehicle-road collaborative application scenario testing method proposed in the above embodiment. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disc.

[0143] In summary, by utilizing the parallel processing ability of quantum computing to fuse vehicle multi-source data, the present invention significantly improves the speed and accuracy of data processing. The quantum algorithm not only accelerates the data fusion process, but also ensures the consistency and integrity of the fused data by mapping sensor data into quantum states and performing efficient operations. This efficient quantum data fusion method reduces the data processing delay, enhances the realism and dynamic response ability of the virtual test environment, and enables the test environment to more quickly reflect the changes in the real world.

[0144] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. A test method for vehicle-road collaborative application scenarios, characterized in that: including, In vehicle-road collaborative applications, vehicles deploy multiple sensors to collect multi-source vehicle data in real time; Utilize the parallel processing ability of quantum computing to fuse multi-source vehicle data; Based on the fused data, construct a virtual test environment on a high-performance computing platform; Conduct tests in the virtual test environment and start real-time monitoring to automatically identify abnormal situations that occur during the test through abnormal algorithms; According to the detected abnormal results, automatically start the intelligent intervention and error correction mechanism; After completing the error correction mechanism, continue to execute the test task, analyze the data during the test, and make optimizations.

2. The vehicle-road collaborative application scenario test method according to claim 1, wherein: In the vehicle-road collaborative application, where vehicles deploy multiple sensors to collect multi-source vehicle data in real time, the specific steps are as follows: To achieve comprehensive perception of the vehicle's surrounding environment, deploy multiple sensors, including millimeter-wave radar, lidar, high-definition cameras, global positioning, and inertial measurement units; Install the sensors at different positions on the vehicle to cover a 360-degree view, and configure the sampling frequency to continuously collect multi-source vehicle data, including obstacle position, obstacle speed, obstacle direction, traffic signal status, vehicle's own position, vehicle's own speed, vehicle's own direction, and vehicle's own acceleration.

3. The vehicle-road collaborative application scenario testing method according to claim 2, wherein: The specific steps for utilizing the parallel processing ability of quantum computing to fuse multi-source vehicle data are as follows: Transmit the collected multi-source vehicle data to the fusion center based on quantum computing through a 5G high-speed communication network; After the fusion center receives the multi-source vehicle data, preprocess the multi-source vehicle data; Map the preprocessed multi-source vehicle data to quantum state data through quantum encoding; Adopt the superposition and entanglement characteristics of quantum algorithms to fuse the quantum state data and generate a fused quantum state.

4. The vehicle-road collaborative application scenario testing method according to claim 3, wherein: The specific steps for constructing a virtual test environment on a high-performance computing platform based on the fused data are as follows: Obtain weather data through external weather services, obtain local environment information through roadside units in vehicle-road collaborative infrastructure, and obtain road characteristics through geographic information; Integrate weather data, local environment information, and road characteristics into an environmental parameter set through data formatting and time synchronization; Receive the fused quantum state and environmental parameter set through the high-performance computing platform and execute the environment construction task; The high-performance computing platform analyzes the fused quantum state through quantum measurement, extracts the dynamic and static characteristics of the quantum state, combines with the environmental parameter set, and adjusts the vehicle's dynamic behavior; Based on the fused quantum state and environmental parameter set through numerical simulation, use the adjusted vehicle's dynamic behavior to simulate the vehicle's dynamic behavior and generate the vehicle-environment interaction result; The high-performance computing platform normalizes the vehicle-environment interaction result, balances the contribution ratio of the fused quantum state and environmental parameter set, and generates a virtual test environment.

5. The vehicle-road collaborative application scenario test method according to claim 4, wherein: The specific steps for conducting tests in the virtual test environment and starting real-time monitoring to automatically identify abnormal situations that occur during the test are as follows: In the constructed virtual test environment, simulate real driving scenarios and test the vehicle's behavior response; Start real-time monitoring and automatically identify abnormal situations during the test through a machine learning-based anomaly detection algorithm, including abnormal vehicle behavior, abnormal sensor readings, environmental perception errors, communication failures, and decision logic errors.

6. The vehicle-road collaborative application scenario testing method according to claim 5, wherein: According to the detected abnormal results, automatically start the intelligent intervention and error correction mechanism. The specific steps are as follows: During the virtual test environment process, when an abnormal situation is detected, trigger intelligent intervention measures, including adjusting the vehicle behavior mode, re-planning the driving route, decelerating and stopping, steering to avoid, lighting and sound warnings, emergency communication, automatic driving mode switching, and activating backup sensors.

7. The vehicle-road collaborative application scenario test method according to claim 6, characterized in that: After completing the error correction mechanism, continue to execute the test task and analyze the data during the test process to make optimizations. The specific steps are as follows: Continue to execute the test task and define a series of performance indicators; Use statistical methods to evaluate the stability and reliability of the test results; Apply clustering analysis to identify patterns and trends in the test data; Through a causal diagram, explore the reasons behind the test results; Based on the analysis results, generate optimization suggestions for the test plan and the vehicle control platform; Apply the optimization suggestions to a new test cycle, execute the virtual test again, and record the results to compare the differences before and after optimization.

8. A vehicle-road collaborative application scenario testing system, based on the vehicle-road collaborative application scenario testing method according to any one of claims 1 to 7, characterized in that: Including a data acquisition module, a data fusion module, a virtual environment construction module, a test monitoring module, an intelligent intervention and error correction module, and an analysis and optimization module; The data acquisition module is used to deploy multiple sensors on the vehicle in a vehicle-road collaborative application to collect multi-source vehicle data in real time; The data fusion module is used to fuse the multi-source vehicle data by utilizing the parallel processing ability of quantum computing; The virtual environment construction module is used to construct a virtual test environment on a high-performance computing platform based on the fused data; The test monitoring module is used to conduct tests in the virtual test environment and start real-time monitoring to automatically identify abnormal situations that occur during the test through an anomaly algorithm; The intelligent intervention and error correction module is used to automatically start the intelligent intervention and error correction mechanism according to the detected abnormal results; The analysis and optimization module is used to continue to execute the test task after completing the error correction mechanism and analyze the data during the test process to make optimizations.

9. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the vehicle-road collaborative application scenario test method described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the vehicle-road collaborative application scenario test method described in any one of claims 1 to 7.

Images