A self-driving vehicle detection system and method

Abstract

The present invention is applicable to the technical field of autonomous driving vehicles, and provides an autonomous driving vehicle detection system and method, which obtains the historical operation logs of the target autonomous vehicle on the preset transportation route, and screens out the operation records with the operation trajectory deviation phenomenon. Among these operation records, the operation records containing the same deviation area and the number exceeding the preset value are found, and they are set as designated operation records, and the same deviation area is set as a high-frequency deviation area. Through in-depth analysis of the historical operation logs, the autonomous driving system can accurately identify the high-frequency deviation area, and through the analysis of the degree of deviation under different load conditions, the positive correlation between the deviation phenomenon and the load data is found. This method can further confirm whether the deviation is caused by the change of the curve path, so as to identify and understand the main cause of the deviation.

Description

Technical Field

[0001] The present invention belongs to the technical field of autonomous driving vehicles, and in particular, relates to an autonomous driving vehicle detection system and method. Background Art

[0002] Autonomous driving vehicles, especially autonomous driving systems for cargo transportation, have been widely used in the fields of logistics and transportation in recent years. These vehicles usually undertake tasks such as urban distribution, industrial transportation and long-distance freight, and their routes often include various complex road environments, such as sharp turns, roundabouts and winding roads with large curvatures. These road characteristics pose severe challenges to the turning ability, driving trajectory control and safety performance of autonomous driving vehicles. Under such road conditions, the turning behavior of the vehicle directly affects its driving stability and the transportation safety of the cargo. Especially under different load conditions, the steering control of the vehicle needs to be more precise and flexible to ensure safe passage through these complex sections.

[0003] However, existing autonomous driving technologies mainly rely on preset fixed turning curvatures for path control. These preset curvatures are usually based on road design and standard vehicle parameters, and do not fully consider changes in vehicle load during actual driving. As the weight of the cargo changes, the vehicle's handling characteristics, center of gravity position, and steering response will change significantly, causing the vehicle to easily deviate from the preset path on roads with large curvatures under different load conditions. Existing technologies lack the ability to intelligently analyze and identify the relationship between load and deviation, and are unable to dynamically adjust the vehicle's turning curvature on a specific curved road, resulting in the inability to optimize the vehicle in real time according to actual conditions during driving.

[0004] Therefore, current autonomous vehicles have obvious technical defects when dealing with different loads and complex road conditions: they cannot intelligently identify and adjust the curvature of turns to adapt to specific road conditions, nor can they effectively optimize driving strategies based on load changes. This technical limitation makes it difficult for autonomous vehicles to maintain accurate driving trajectories on roads with large curvatures, especially in emergency avoidance or high load conditions, increasing the risk of driving deviation and safety hazards. Summary of the invention

[0005] The purpose of the present invention is to provide an autonomous driving vehicle detection system and method, aiming to solve the problems raised in the background technology.

[0006] The present invention is implemented as follows: a method for detecting an autonomous driving vehicle, the method comprising:

[0007] Obtain the historical operation logs of the target automatic vehicle on the preset transportation route, and screen out the operation records with the operation trajectory deviation phenomenon, and among these operation records, find out the operation records containing the same deviation area and the number of which exceeds the preset value, set them as designated operation records, and set the same deviation area as the high-frequency deviation area;

[0008] determining cargo load data of the target automated vehicle for each specified run record and the degree of deviation of the target automated vehicle in the high frequency deviation area;

[0009] Analyze whether the deviation degree of the high-frequency deviation area in different specified operation records of the target automatic vehicle is positively correlated with the cargo load data. If it is determined that there is a positive correlation, analyze whether the objective influencing factors of the high-frequency deviation area are only caused by the change of the curved path;

[0010] If it is determined that the objective influencing factors of the high-frequency deviation area are only caused by the change of the curve path, a correlation curve chart between the load and the degree of deviation is generated according to the deviation degree and cargo load data in different specified operation records, and the original predetermined turning curvature of the target automatic vehicle in the high-frequency deviation area is adjusted according to the correlation curve chart.

[0011] As a further limitation of the technical solution of the embodiment of the present invention, the step of determining the cargo load data of the target automatic vehicle for each specified operation record and the degree of deviation of the target automatic vehicle in the high-frequency deviation area includes:

[0012] Determine cargo load data of the target automated vehicle in each designated operation record based on the historical operation log of the target automated vehicle on the preset transportation route;

[0013] According to the historical operation log, the GPS data of the target automatic vehicle in the high-frequency deviation area is obtained, and the maximum deviation point of the target automatic vehicle from the preset transportation route is determined according to the GPS data;

[0014] The straight-line distance from the maximum deviation point to the target transport route is set as the deviation degree of the target automatic vehicle in the high-frequency deviation area.

[0015] As a further limitation of the technical solution of the embodiment of the present invention, the step of analyzing whether the degree of deviation of the high-frequency deviation area in different designated operation records of the target automatic vehicle is positively correlated with the cargo load data, and if it is determined that there is a positive correlation, analyzing whether the objective influencing factor of the high-frequency deviation area is only caused by the change of the curved path includes:

[0016] Analyze whether the target automatic vehicle has a greater degree of deviation in the high-frequency deviation area when the cargo load data of the target automatic vehicle is greater in different designated operation records of the target automatic vehicle;

[0017] If it is determined that the larger the cargo load data of the target automatic vehicle is, the greater the deviation degree of the target automatic vehicle in the high-frequency deviation area is, then it is determined that there is a positive correlation between the deviation degree of the high-frequency deviation area and the cargo load data;

[0018] After confirming the existence of a positive correlation, the objective influencing factors in the high-frequency deviation area were analyzed to see whether they were caused only by the change in the curve path.

[0019] As a further limitation of the technical solution of the embodiment of the present invention, after determining that there is a positive correlation, the step of analyzing whether the objective influencing factors of the high-frequency deviation area are only caused by the change of the curve path includes:

[0020] After determining that there is a positive correlation, the tilt angle data of the target vehicle in the high-frequency deviation area is obtained according to the historical operation log, and it is determined whether there is a longitudinal slope change of the target vehicle in the high-frequency deviation area according to the tilt angle data;

[0021] If it is determined that there is no longitudinal slope change in the high-frequency deviation area for the target vehicle, a driving trajectory of the target automatic vehicle in the high-frequency deviation area is plotted based on the GPS data of the target automatic vehicle in the high-frequency deviation area;

[0022] It is analyzed whether the driving trajectory of the target automatic vehicle in the high-frequency deviation area is a curved trajectory. If so, it is determined that the objective influencing factors of the high-frequency deviation area are only caused by the change of the curved path.

[0023] As a further limitation of the technical solution of the embodiment of the present invention, if it is determined that the objective influencing factor of the high-frequency deviation area is caused only by the change of the curved path, then according to the deviation degree and cargo load data in different specified operation records, a correlation curve diagram between the load and the deviation degree is generated, and the steps of adjusting the original predetermined turning curvature of the target automatic vehicle in the high-frequency deviation area according to the correlation curve diagram include:

[0024] If it is determined that the objective influencing factor of the high-frequency deviation area is only caused by the change of the curved path, the deviation degree and the corresponding cargo load data in different specified operation records are collected, and these cargo load data are sorted in ascending order;

[0025] According to the sorted cargo load data and their corresponding deviation degrees, a correlation curve between load and deviation degree is drawn;

[0026] An originally predetermined turning curvature of the target automated vehicle in the high frequency deviation region is adjusted according to the correlation curve graph.

[0027] As a further limitation of the technical solution of the embodiment of the present invention, the step of adjusting the original predetermined turning curvature of the target automatic vehicle in the high-frequency deviation area according to the correlation curve graph includes:

[0028] Calculate the overall slope of the correlation curve between load and deviation degree;

[0029] Extracting the original predetermined turning curvature of the target automated vehicle in the high-frequency deviation area from the historical operation log;

[0030] The original predetermined turning curvature is multiplied by the overall slope of the correlation curve to obtain the optimized turning curvature.

[0031] An automatic driving vehicle detection system, the system comprising: a data screening module, a data extraction module, a forward correlation analysis module, and a turning curvature adjustment module, wherein:

[0032] A data screening module is used to obtain the historical operation logs of the target automatic vehicle on the preset transportation route, and screen out the operation records with the operation trajectory deviation phenomenon, and find out the operation records containing the same deviation area and the number of which exceeds the preset value among these operation records, set them as designated operation records, and set the same deviation area as the high-frequency deviation area;

[0033] a data extraction module for determining cargo load data of the target automated vehicle for each specified run record and a degree of deviation of the target automated vehicle in the high frequency deviation area;

[0034] A positive correlation analysis module is used to analyze whether the degree of deviation in the high-frequency deviation area of ​​the target automatic vehicle in different specified operation records is positively correlated with the cargo load data. If it is determined that there is a positive correlation, it is analyzed whether the objective influencing factors of the high-frequency deviation area are only caused by the change of the curved path;

[0035] The turning curvature adjustment module is used to generate a correlation curve diagram between load and deviation degree according to the deviation degree and cargo load data in different specified operation records if it is determined that the objective influencing factor of the high-frequency deviation area is only caused by the change of the curve path, and adjust the original predetermined turning curvature of the target automatic vehicle in the high-frequency deviation area according to the correlation curve diagram.

[0036] As a further limitation of the technical solution of the embodiment of the present invention, the data extraction module specifically includes:

[0037] A load data acquisition unit, used to determine the cargo load data of the target automatic vehicle in each designated operation record according to the historical operation log of the target automatic vehicle on the preset transportation route;

[0038] A GPS data acquisition unit, used to acquire GPS data of the target automatic vehicle in the high-frequency deviation area according to the historical operation log, and determine the maximum deviation point of the target automatic vehicle from the preset transportation route according to the GPS data;

[0039] The deviation degree determination unit is used to set the straight-line distance from the maximum deviation point to the target transportation route as the deviation degree of the target automatic vehicle in the high-frequency deviation area.

[0040] As a further limitation of the technical solution of the embodiment of the present invention, the forward association analysis module specifically includes:

[0041] A designated operation record analysis unit is used to analyze whether the target automatic vehicle has a greater degree of deviation in a high-frequency deviation area when the cargo load data of the target automatic vehicle is greater in different designated operation records of the target automatic vehicle;

[0042] a positive correlation determination unit, for determining that the degree of deviation in the high-frequency deviation area is positively correlated with the cargo load data if it is determined that the greater the cargo load data of the target automatic vehicle, the greater the degree of deviation of the target automatic vehicle in the high-frequency deviation area;

[0043] The objective influencing factor analysis unit is used to analyze whether the objective influencing factor of the high-frequency deviation area is only caused by the change of the curve path after determining that there is a positive correlation.

[0044] As a further limitation of the technical solution of the embodiment of the present invention, the forward association analysis module also includes:

[0045] A longitudinal gradient change determination unit is used to obtain the inclination angle data of the target vehicle in the high-frequency deviation area according to the historical operation log after determining that there is a positive correlation, and determine whether there is a longitudinal gradient change of the target vehicle in the high-frequency deviation area according to the inclination angle data;

[0046] A driving trajectory drawing unit is used to draw a driving trajectory of the target automatic vehicle in the high-frequency deviation area according to the GPS data of the target automatic vehicle in the high-frequency deviation area if it is determined that there is no longitudinal slope change in the high-frequency deviation area of ​​the target vehicle;

[0047] The driving trajectory analysis unit is used to analyze whether the driving trajectory of the target automatic vehicle in the high-frequency deviation area is a curved trajectory. If so, it is determined that the objective influencing factors of the high-frequency deviation area are only caused by the change of the curved path.

[0048] Compared with the prior art, the present invention has the following beneficial effects:

[0049] Through in-depth analysis of historical operation logs, the autonomous driving system can accurately identify high-frequency deviation areas, and through analysis of the degree of deviation under different load conditions, find the positive correlation between the deviation phenomenon and the load data. This method can further confirm whether the deviation is caused by changes in the curve path, thereby identifying and understanding the main cause of the deviation. Based on this correlation, a correlation curve graph between load and deviation degree is generated, and the original predetermined turning curvature of the vehicle is adjusted through this curve graph, so that the autonomous driving vehicle can drive more stably and safely in the high-frequency deviation area.

[0050] This optimization and adjustment strategy allows autonomous vehicles to dynamically adjust turning behavior based on actual load data under different reasonable load conditions, thereby effectively reducing the occurrence of trajectory deviation and improving the overall driving accuracy, stability and safety. Especially when facing complex and winding roads, this method can ensure that the vehicle still maintains the best turning performance under different weight conditions. Through this dynamic optimization, the vehicle can not only better adapt to load changes during normal driving, but also be more calm when encountering emergencies such as emergency avoidance, and its deviation is lower. This is because the adjusted turning curvature can accurately respond to different load conditions, optimize the vehicle's trajectory control during sharp turns and avoidance, reduce the risk of deviation due to weight changes, and enhance the system's reliability and response capabilities in a changing environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 A flowchart of a method provided by an embodiment of the present invention;

[0052] Figure 2 A flow chart of determining cargo load data of a target automated vehicle and a degree of deviation of the target automated vehicle in a high-frequency deviation area in a method provided by an embodiment of the present invention;

[0053] Figure 3 A flow chart of determining whether there is a positive correlation between the degree of deviation in the high-frequency deviation area and the cargo load data in the method provided in an embodiment of the present invention;

[0054] Figure 4 A flow chart for analyzing whether the objective influencing factors of the high-frequency deviation area are only caused by the change of the curve path in the method provided in the embodiment of the present invention;

[0055] Figure 5 A flow chart of drawing a correlation curve diagram between load and deviation degree in the method provided in an embodiment of the present invention;

[0056] Figure 6 A flow chart of adjusting the original predetermined turning curvature according to the correlation curve diagram in the method provided by an embodiment of the present invention;

[0057] Figure 7An application architecture diagram of a system provided by an embodiment of the present invention;

[0058] Figure 8 A structural block diagram of a data extraction module in a system provided by an embodiment of the present invention;

[0059] Fig. 9 This is a structural block diagram of a forward correlation analysis module in a system provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0060] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0061] Figure 1 A flow chart of a method provided by an embodiment of the present invention is shown.

[0062] Specifically, a method for detecting an autonomous driving vehicle comprises the following steps:

[0063] Step S100, obtain the historical operation log of the target automatic vehicle on the preset transportation route, and filter out the operation records with operation trajectory deviation. Among these operation records, find the operation records containing the same deviation area and the number exceeds the preset value, set it as the designated operation record, and set the same deviation area as the high-frequency deviation area.

[0064] In the embodiments of the present invention, the target automatic vehicle is usually a special vehicle in the field of autonomous driving, especially suitable for logistics and distribution scenarios. These vehicles are of various types, and can be automatic delivery vehicles for short-distance urban distribution, small automatic delivery robots for last-mile distribution, or unmanned vans or trucks for long-distance transportation. These vehicles are equipped with high-precision sensors, GPS systems, cameras and other monitoring equipment to collect and analyze operating data to support autonomous driving and navigation decisions.

[0065] The preset transport routes are fixed driving paths planned for the target automatic vehicles. These routes are set in advance by the system and may pass through urban roads, residential areas, commercial areas or industrial parks. They are usually designed based on optimization principles such as shortest distance, lowest cost and highest efficiency. These routes are defined before the vehicle sets off, and the navigation system guides the vehicle to follow the planned route. The historical operation log is the data recorded during the actual driving of these vehicles on the preset routes, including cargo load data, GPS data, tilt angle data, driving trajectory, location, speed, timestamp and other information of each operation of the target automatic vehicle. These log data are an important basis for analyzing vehicle performance and deviations.

[0066] To filter out operation records with trajectory deviation from historical operation logs, it is first necessary to define the deviation criteria, such as the distance threshold or speed change from the preset route. By comparing the actual trajectory of the vehicle with the preset transportation route, it is possible to identify which operation records have deviations. Among the filtered deviation records, these records are further analyzed to find operation records containing the same deviation area and count the number of occurrences. When the number of occurrences of a deviation area exceeds the preset threshold, these records are set as designated operation records, and the deviation area is defined as a high-frequency deviation area. Implementing this process usually involves technologies such as geographic information system (GIS) analysis, anomaly detection of machine learning algorithms, and data clustering to accurately identify and classify high-frequency deviation phenomena.

[0067] Furthermore, the autonomous driving vehicle detection method further includes the following steps:

[0068] Step S200, determining the cargo load data of the target automated vehicle for each designated operation record, and the degree of deviation of the target automated vehicle in the high frequency deviation area.

[0069] Specifically, Figure 2 A flow chart for determining cargo load data of a target automated vehicle and a degree of deviation of the target automated vehicle in a high frequency deviation region is shown.

[0070] Determining the cargo load data of the target automated vehicle for each designated operation record and the degree of deviation of the target automated vehicle in the high-frequency deviation area specifically includes the following steps:

[0071] Step S201, determining cargo load data of a target automatic vehicle in each designated operation record according to the historical operation log of the target automatic vehicle on a preset transportation route;

[0072] Step S202, obtaining GPS data of the target automatic vehicle in the high-frequency deviation area according to the historical operation log, and determining the maximum deviation point of the target automatic vehicle from the preset transportation route according to the GPS data;

[0073] Step S203, setting the straight-line distance from the maximum deviation point to the target transportation route as the deviation degree of the target automatic vehicle in the high-frequency deviation area.

[0074] In an embodiment of the present invention, according to the historical operation logs, GPS data of the target autonomous vehicle in the high-frequency deviation area is obtained, and these data are analyzed to determine the degree to which the vehicle deviates from the preset transportation route during driving. The specific implementation method generally includes the following steps: First, by reading the GPS coordinate data in the operation logs, the actual driving trajectory of the vehicle is drawn and compared with the preset transportation route. Using a geographic information system (GIS) or a path matching algorithm, each deviation point of the vehicle in the high-frequency deviation area is identified, and the deviation point with the largest distance between the vehicle trajectory and the preset route is found, that is, the maximum deviation point.

[0075] The straight-line distance from the maximum deviation point to the preset transportation route is defined as the deviation degree of the target autonomous vehicle in the high-frequency deviation area. The basis for this setting method is that the maximum deviation point represents the most serious situation in which the vehicle deviates from the preset path during driving, reflecting the extreme degree of trajectory deviation. By calculating this straight-line distance, the deviation phenomenon can be quantified, providing a clear measurement standard for evaluating the running stability and navigation accuracy of the vehicle in the high-frequency deviation area. This method is based on trajectory analysis and distance measurement, and can intuitively reflect the deviation characteristics of the vehicle, which is an effective means for evaluating path deviation.

[0076] Furthermore, the autonomous vehicle detection method further includes the following steps:

[0077] Step S300, analyze whether there is a positive correlation between the deviation degree of the target autonomous vehicle in the high-frequency deviation area and the cargo load data in different specified operation records. If it is determined that there is a positive correlation, analyze whether the objective influencing factors of this high-frequency deviation area are only caused by the change of the curved path.

[0078] Specifically, Figure 3 The flowchart shows the determination of whether there is a positive correlation between the deviation degree of the high-frequency deviation area and the cargo load data.

[0079] Among them, analyzing whether there is a positive correlation between the deviation degree of the target autonomous vehicle in the high-frequency deviation area and the cargo load data in different specified operation records. If it is determined that there is a positive correlation, analyzing whether the objective influencing factors of this high-frequency deviation area are only caused by the change of the curved path specifically includes the following steps:

[0080] Step S301, analyze whether the deviation degree of the target autonomous vehicle in the high-frequency deviation area is greater when the cargo load data of the target autonomous vehicle is larger in different specified operation records;

[0081] Step S302, if it is determined that the deviation degree of the target autonomous vehicle in the high-frequency deviation area is also greater when the cargo load data of the target autonomous vehicle is larger, it is determined that there is a positive correlation between the deviation degree of the high-frequency deviation area and the cargo load data;

[0082] Step S303: after determining that there is a positive correlation, analyzing whether the objective influencing factors of the high-frequency deviation area are only caused by the curve path change.

[0083] In an embodiment of the present invention, when analyzing different designated operation records of the target automated vehicle, by comparing the cargo load data and the degree of deviation of the vehicle in the high-frequency deviation area, the relationship between the two can be evaluated. If it is observed that as the cargo load increases, the degree of deviation of the vehicle in the high-frequency deviation area also increases, it can be determined that the degree of deviation is positively correlated with the cargo load data.

[0084] The basis for judging this positive correlation is the correlation analysis in statistics, that is, when one variable (cargo load) changes, another variable (deviation degree) also shows a trend of regular changes. In this scenario, a larger cargo load may affect the stability, braking distance and turning radius of the vehicle, resulting in a more significant deviation. The judgment process usually uses regression analysis, correlation coefficient calculation and other methods to verify the numerical relationship between load and deviation data to determine whether there is a significant positive trend.

[0085] This judgment is also based on the principles of physics and vehicle dynamics: as the load increases, the vehicle's inertia increases, the responsiveness to cornering and trajectory control decreases, and therefore the degree of deviation increases. This physical effect is reflected in actual operation as a positive correlation between load and degree of deviation. Therefore, based on the combination of statistical analysis results and actual dynamics principles, it is reasonable to judge the existence of this positive correlation.

[0086] Specifically, Figure 4 A flow chart is shown for analyzing whether the objective influencing factors of the high-frequency deviation area are caused only by the curve path change.

[0087] After determining that there is a positive correlation, analyzing whether the objective influencing factors of the high-frequency deviation area are only caused by the change of the curve path specifically includes the following steps:

[0088] Step S3031, after determining that there is a positive correlation, obtaining the inclination angle data of the target vehicle in the high-frequency deviation area according to the historical operation log, and judging whether there is a longitudinal slope change of the target vehicle in the high-frequency deviation area according to the inclination angle data;

[0089] Step S3032, if it is determined that there is no longitudinal slope change of the target vehicle in the high-frequency deviation area, a driving trajectory of the target automatic vehicle in the high-frequency deviation area is drawn according to the GPS data of the target automatic vehicle in the high-frequency deviation area;

[0090] Step S3033, analyzing whether the driving trajectory of the target automatic vehicle in the high-frequency deviation area is a curved trajectory. If so, it is determined that the objective influencing factor of the high-frequency deviation area is only caused by the change of the curved path.

[0091] In an embodiment of the present invention, to determine whether the target vehicle has a longitudinal slope change in the high-frequency deviation area, it can be achieved by analyzing the vehicle's tilt angle data (such as the pitch angle). These data are usually collected by the vehicle's inertial measurement unit (IMU) or other sensors, reflecting the longitudinal posture changes of the vehicle during driving. The specific implementation steps include: first, extracting the vehicle's tilt angle data in the high-frequency deviation area, performing time series analysis on these data, and calculating key statistical indicators such as the average value, maximum value, and minimum value of the tilt angle. Then, by setting a slope change threshold (such as the pitch angle exceeds a certain degree), it is determined whether the vehicle has experienced a significant longitudinal slope change. If the tilt angle remains in a stable state or the change is not significant, it can be considered that there is no longitudinal slope change in the area.

[0092] To analyze whether the vehicle's driving trajectory is a curved trajectory, the path curvature calculation and trajectory morphology analysis of the vehicle's GPS data can be performed. First, the GPS coordinate data in the high-frequency deviation area is extracted from the vehicle's historical operation log, and the actual driving trajectory is drawn using this data. Next, the curvature of the trajectory is calculated to determine the curvature of the trajectory. The higher the curvature, the greater the curvature of the trajectory. The trajectory can be analyzed using a curvature formula or a trajectory fitting algorithm (such as polynomial fitting) to evaluate whether the trajectory's morphology exhibits curved features. If the curvature value reaches a certain standard, the trajectory can be confirmed to be a curved trajectory.

[0093] It is of great significance to determine that the deviation in the high-frequency deviation area is only caused by the change of the curved path. This conclusion helps to rule out other potential influencing factors (such as changes in longitudinal slope, road obstacles, environmental conditions, etc.) and attribute the deviation phenomenon to the curvature characteristics of the road design itself. This means that the deviation is not caused by vehicle failure, operating error or external interference, but is a normal reaction of the vehicle in response to changes in the curved path of the road. This understanding helps to optimize the control strategy of the vehicle in the curved area in a targeted manner, such as improving driving performance by adjusting the turning curvature, optimizing speed control, etc., thereby improving the vehicle's operating stability and safety. This conclusion provides a clear guiding direction for the optimization of vehicle path planning and autonomous driving control systems, and helps to reduce unnecessary deviations.

[0094] Furthermore, the autonomous driving vehicle detection method further includes the following steps:

[0095] Step S400, if it is determined that the objective influencing factor of the high-frequency deviation area is only caused by the change of the curve path, a correlation curve diagram between the load and the degree of deviation is generated according to the deviation degree and cargo load data in different specified operation records, and the original predetermined turning curvature of the target automatic vehicle in the high-frequency deviation area is adjusted according to the correlation curve diagram.

[0096] Specifically, Figure 5 A flow chart for plotting a correlation graph of load and deflection degree is shown.

[0097] If it is determined that the objective influencing factor of the high-frequency deviation area is only caused by the change of the curved path, a correlation curve diagram between the load and the deviation degree is generated according to the deviation degree and the cargo load data in different designated operation records, and the original predetermined turning curvature of the target automatic vehicle in the high-frequency deviation area is adjusted according to the correlation curve diagram. Specifically, the following steps are included:

[0098] Step S401, if it is determined that the objective influencing factor of the high-frequency deviation area is only caused by the curve path change, the deviation degree and the corresponding cargo load data in different designated operation records are collected, and these cargo load data are sorted in ascending order;

[0099] Step S402, drawing a correlation curve diagram between the load and the degree of deviation according to the sorted cargo load data and the corresponding degree of deviation;

[0100] Step S403: adjusting the original predetermined turning curvature of the target automatic vehicle in the high-frequency deviation area according to the correlation curve diagram.

[0101] In the embodiment of the present invention, after determining that the objective influencing factor of the high-frequency deviation area is only caused by the change of the curved path, it is necessary to collect the deviation degree and the corresponding cargo load data in different specified operation records, and sort these data in ascending order. The sorted data is used to draw a correlation curve chart between load and deviation degree to analyze the influence of load on deviation degree.

[0102] These data should be coherent and consistent when analyzed, especially the gap between load data should not be too large. Adjacent load data should be kept within a reasonable range to avoid extreme values ​​or abnormal values ​​that deviate significantly from other data. If the cargo load data differ too much or there are extremely large data points, the curve chart may lose coherence and fail to accurately reflect the true relationship between load and deviation. Such abnormal data may be caused by specific operating errors or rare situations and are not universal. Therefore, they should be handled or excluded with caution when drawing and analyzing to ensure the accuracy and practical significance of the correlation analysis. Maintaining a reasonable interval and continuity of the data can more truly reflect the deviation characteristics of the vehicle under different load conditions and provide a reliable basis for optimization.

[0103] Specifically, Figure 6 A flow chart for adjusting the original predetermined turning curvature according to the correlation curve diagram is shown.

[0104] Wherein, adjusting the original predetermined turning curvature of the target automatic vehicle in the high-frequency deviation area according to the correlation curve diagram specifically includes the following steps:

[0105] Step S4031, calculating the overall slope of the correlation curve between load and deviation degree;

[0106] Step S4032, extracting the original predetermined turning curvature of the target automatic vehicle in the high-frequency deviation area from the historical operation log;

[0107] Step S4033: multiply the original predetermined turning curvature by the overall slope of the correlation curve to obtain an optimized turning curvature.

[0108] In the embodiment of the present invention, the original predetermined turning curvature of the high-frequency deviation area refers to the ideal or planned turning curvature for a specific high-frequency deviation area set in the vehicle planning and navigation system. This is usually preset based on factors such as vehicle design parameters, road design, and traffic rules, with the purpose of guiding the vehicle to drive safely along the predetermined path in this area without affecting safety and efficiency. This curvature takes into account the curve of the road, as well as the operability and stability of the vehicle, and is the basic guidance in the vehicle control system on how to turn in this area.

[0109] The benefit of multiplying the original predetermined turning curvature with the overall slope of the correlation curve to obtain the optimized turning curvature is mainly reflected in the dynamic adjustment of vehicle behavior to adapt to the actual driving needs under different load conditions. The original predetermined turning curvature may not fully consider the changes in vehicle behavior under different loads, especially when the load is large, the vehicle's handling performance and dynamic response will be different. By analyzing the correlation between load and deviation degree, a slope indicator that describes this change trend can be obtained.

[0110] This slope reflects the rate of change of the deviation degree as the load increases, that is, the proportion of the deviation degree increasing for each unit increase in load. By multiplying this slope with the original curvature, the original curvature can be adjusted to make it closer to the actual driving situation. This adjustment helps to improve the driving accuracy and safety of the vehicle under high load conditions. The adjusted curvature can better reflect the actual vehicle dynamic characteristics, optimize the driving trajectory, reduce deviation accidents caused by improper driving on curves, and enhance the adaptability and stability of autonomous driving vehicles in complex road conditions.

[0111] The basis of this optimization measure is that the correlation results obtained through analysis of actual driving data are based on actual driving conditions and are more valuable as references than a single theoretical model or preset parameters, and can significantly improve navigation accuracy and driving safety.

[0112] In summary, by carefully analyzing the historical operation data of autonomous vehicles, especially the correlation between load and deviation, we can significantly improve the driving accuracy and safety of vehicles under various load conditions. This method enables autonomous vehicles to automatically adjust the curvature according to different loads and pass through winding roads in the best way, thereby reducing the risk of deviation and improving driving stability.

[0113] More specifically, the optimized turning curvature is not only suitable for regular driving conditions, but also significantly improves the vehicle's responsiveness and control accuracy in emergency avoidance situations. This optimization measure effectively enhances the vehicle's ability to respond to emergencies by keeping the vehicle's deviation level low when quick avoidance is required, thereby greatly improving the safety of passengers and pedestrians, while enhancing the public's trust and acceptance of autonomous vehicle technology.

[0114] In short, by optimizing turning curvature with this data-based method, autonomous vehicles can better adapt to complex road environments and demonstrate higher safety and reliability in both daily transportation and in the face of emergencies. This has a positive effect on the advancement of autonomous driving technology and its popularization and acceptance in practical applications.

[0115] Furthermore, Figure 7 The application architecture diagram of the system provided by the embodiment of the present invention is shown.

[0116] In another preferred embodiment of the present invention, an automatic driving vehicle detection system includes:

[0117] The data screening module 100 is used to obtain the historical operation logs of the target automatic vehicle on the preset transportation route, and screen out the operation records with the operation trajectory deviation phenomenon. Among these operation records, the operation records containing the same deviation area and the number exceeding the preset value are found, and they are set as designated operation records, and the same deviation area is set as a high-frequency deviation area.

[0118] In the embodiments of the present invention, the target automatic vehicle is usually a special vehicle in the field of autonomous driving, especially suitable for logistics and distribution scenarios. These vehicles are of various types, and can be automatic delivery vehicles for short-distance urban distribution, small automatic delivery robots for last-mile distribution, or unmanned vans or trucks for long-distance transportation. These vehicles are equipped with high-precision sensors, GPS systems, cameras and other monitoring equipment to collect and analyze operating data to support autonomous driving and navigation decisions.

[0119] The preset transport routes are fixed driving paths planned for the target automatic vehicles. These routes are set in advance by the system and may pass through urban roads, residential areas, commercial areas or industrial parks. They are usually designed based on optimization principles such as shortest distance, lowest cost and highest efficiency. These routes are defined before the vehicle sets off, and the navigation system guides the vehicle to follow the planned route. The historical operation log is the data recorded during the actual driving of these vehicles on the preset routes, including cargo load data, GPS data, tilt angle data, driving trajectory, location, speed, timestamp and other information of each operation of the target automatic vehicle. These log data are an important basis for analyzing vehicle performance and deviations.

[0120] To filter out operation records with trajectory deviation from historical operation logs, it is first necessary to define the deviation criteria, such as the distance threshold or speed change from the preset route. By comparing the actual trajectory of the vehicle with the preset transportation route, it is possible to identify which operation records have deviations. Among the filtered deviation records, these records are further analyzed to find operation records containing the same deviation area and count the number of occurrences. When the number of occurrences of a deviation area exceeds the preset threshold, these records are set as designated operation records, and the deviation area is defined as a high-frequency deviation area. Implementing this process usually involves technologies such as geographic information system (GIS) analysis, anomaly detection of machine learning algorithms, and data clustering to accurately identify and classify high-frequency deviation phenomena.

[0121] Furthermore, the autonomous driving vehicle detection system further includes:

[0122] The data extraction module 200 is used to determine the cargo load data of the target automated vehicle for each specified operation record, and the degree of deviation of the target automated vehicle in the high-frequency deviation area.

[0123] Specifically, Figure 8 It shows a structural block diagram of the data extraction module 200 in the system provided by the embodiment of the present invention.

[0124] Among them, in the preferred embodiment provided by the present invention, the data extraction module 200 specifically includes:

[0125] The load data acquisition unit 201 is used to determine the cargo load data of the target automatic vehicle in each designated operation record according to the historical operation log of the target automatic vehicle on the preset transportation route;

[0126] The GPS data acquisition unit 202 is used to acquire GPS data of the target automatic vehicle in the high-frequency deviation area according to the historical operation log, and determine the maximum deviation point of the target automatic vehicle from the preset transportation route according to the GPS data;

[0127] The deviation degree determination unit 203 is used to set the straight-line distance from the maximum deviation point to the target transportation route as the deviation degree of the target automatic vehicle in the high-frequency deviation area.

[0128] In an embodiment of the present invention, GPS data of the target automatic vehicle in the high-frequency deviation area is obtained based on the historical operation log, and the data is analyzed to determine the degree to which the vehicle deviates from the preset transportation route during driving. The specific implementation method generally includes the following steps: First, by reading the GPS coordinate data in the operation log, the actual driving trajectory of the vehicle is drawn and compared with the preset transportation route. Using a geographic information system (GIS) or a path matching algorithm, each deviation point of the vehicle in the high-frequency deviation area is identified, and the deviation point with the largest distance between the vehicle trajectory and the preset route, that is, the maximum deviation point, is found.

[0129] The straight-line distance from the maximum deviation point to the preset transportation route is defined as the degree of deviation of the target automatic vehicle in the high-frequency deviation area. The basis for this setting method is that the maximum deviation point represents the most serious situation of the vehicle deviating from the preset path during driving, reflecting the extreme degree of trajectory deviation. By calculating this straight-line distance, the deviation phenomenon can be quantified, providing a clear metric for evaluating the vehicle's operating stability and navigation accuracy in the high-frequency deviation area. This method is based on trajectory analysis and distance measurement, can intuitively reflect the deviation characteristics of the vehicle, and is an effective means of evaluating path deviation.

[0130] Furthermore, the autonomous driving vehicle detection system further includes:

[0131] The positive correlation analysis module 300 is used to analyze whether the degree of deviation in the high-frequency deviation area of ​​the target automatic vehicle in different specified operation records is positively correlated with the cargo load data. If it is determined that there is a positive correlation, it is analyzed whether the objective influencing factors of the high-frequency deviation area are only caused by the change of the curve path.

[0132] Specifically, Fig. 9 It shows a structural block diagram of the forward correlation analysis module 300 in the system provided by the embodiment of the present invention.

[0133] Among them, in the preferred embodiment provided by the present invention, the forward correlation analysis module 300 specifically includes:

[0134] The designated operation record analysis unit 301 is used to analyze whether the deviation degree of the target automatic vehicle in the high-frequency deviation area is greater when the cargo load data of the target automatic vehicle is greater in different designated operation records of the target automatic vehicle;

[0135] A positive correlation determination unit 302 is used to determine that if it is determined that the larger the cargo load data of the target automatic vehicle is, the greater the deviation degree of the target automatic vehicle in the high-frequency deviation area is, then determine that there is a positive correlation between the deviation degree of the high-frequency deviation area and the cargo load data;

[0136] The objective influencing factor analysis unit 303 is used to analyze whether the objective influencing factor of the high-frequency deviation area is only caused by the curve path change after determining that there is a positive correlation;

[0137] A longitudinal gradient change determination unit 304 is used to obtain the inclination angle data of the target vehicle in the high-frequency deviation area according to the historical operation log after determining that there is a positive correlation, and determine whether there is a longitudinal gradient change of the target vehicle in the high-frequency deviation area according to the inclination angle data;

[0138] A driving track drawing unit 305 is used to draw a driving track of the target automatic vehicle in the high-frequency deviation area according to the GPS data of the target automatic vehicle in the high-frequency deviation area if it is determined that there is no longitudinal slope change in the high-frequency deviation area of ​​the target automatic vehicle;

[0139] The driving trajectory analysis unit 306 is used to analyze whether the driving trajectory of the target automatic vehicle in the high-frequency deviation area is a curved trajectory. If so, it is determined that the objective influencing factor of the high-frequency deviation area is only caused by the change of the curved path.

[0140] In an embodiment of the present invention, when analyzing different designated operation records of the target automated vehicle, by comparing the cargo load data and the degree of deviation of the vehicle in the high-frequency deviation area, the relationship between the two can be evaluated. If it is observed that as the cargo load increases, the degree of deviation of the vehicle in the high-frequency deviation area also increases, it can be determined that the degree of deviation is positively correlated with the cargo load data.

[0141] The basis for this positive correlation judgment lies in the correlation analysis in statistics, that is, when one variable (cargo load) changes, another variable (deviation degree) also shows a trend of regular change. In this context, a larger cargo load may affect the vehicle's stability, braking distance, and turning radius, thus resulting in more significant deviations. The judgment process usually adopts methods such as regression analysis and correlation coefficient calculation, and determines whether there is a significant positive trend by verifying the numerical relationship between the load and deviation data.

[0142] The basis for this judgment also includes physical and vehicle dynamics principles: as the load increases, the inertial force of the vehicle increases, and the response ability to turning and trajectory control decreases, so the deviation degree increases. The manifestation of this physical effect in actual operation reflects the positive correlation between the load and the deviation degree. Therefore, based on the combination of statistical analysis results and actual dynamics principles, the existence of this positive correlation can be reasonably judged.

[0143] To determine whether there is a longitudinal slope change in the high-frequency deviation area of the target vehicle, it can be achieved by analyzing the vehicle's tilt angle data (such as pitch angle). These data are usually collected by the vehicle's inertial measurement unit (IMU) or other sensors, reflecting the longitudinal attitude change of the vehicle during driving. The specific implementation steps include: First, extract the tilt angle data of the vehicle in the high-frequency deviation area, perform time series analysis on these data, and calculate key statistical indicators such as the average value, maximum value, and minimum value of the tilt angle. Then, by setting a slope change threshold (such as the pitch angle exceeding a certain degree), determine whether the vehicle has experienced an obvious longitudinal slope change. If the tilt angle remains in a stable state or the change is not significant, it can be considered that there is no longitudinal slope change in this area.

[0144] To analyze whether the vehicle's driving trajectory is a curved trajectory, it can be achieved by calculating the path curvature and analyzing the trajectory shape of the vehicle's GPS data. First, extract the GPS coordinate data in the high-frequency deviation area from the vehicle's historical operation log, and use these data to draw the actual driving trajectory. Then, judge the degree of curvature of the trajectory by calculating the curvature. The higher the curvature, the greater the degree of curvature of the trajectory. The curvature formula or trajectory fitting algorithm (such as polynomial fitting) can be used to analyze the trajectory and evaluate whether the trajectory shape presents a curved feature. If the curvature value reaches a certain standard, the trajectory can be confirmed as a curved trajectory.

[0145] It is of great significance to determine that the deviation in the high-frequency deviation area is only caused by the change of the curved path. This conclusion helps to rule out other potential influencing factors (such as changes in longitudinal slope, road obstacles, environmental conditions, etc.) and attribute the deviation phenomenon to the curvature characteristics of the road design itself. This means that the deviation is not caused by vehicle failure, operating error or external interference, but is a normal reaction of the vehicle in response to changes in the curved path of the road. This understanding helps to optimize the control strategy of the vehicle in the curved area in a targeted manner, such as improving driving performance by adjusting the turning curvature, optimizing speed control, etc., thereby improving the vehicle's operating stability and safety. This conclusion provides a clear guiding direction for the optimization of vehicle path planning and autonomous driving control systems, and helps to reduce unnecessary deviations.

[0146] Furthermore, the autonomous driving vehicle detection system further includes:

[0147] The turning curvature adjustment module 400 is used to generate a correlation curve diagram between load and deviation degree according to the deviation degree and cargo load data in different specified operation records if it is determined that the objective influencing factor of the high-frequency deviation area is only caused by the change of the curve path, and adjust the original predetermined turning curvature of the target automatic vehicle in the high-frequency deviation area according to the correlation curve diagram.

[0148] In the embodiment of the present invention, after determining that the objective influencing factor of the high-frequency deviation area is only caused by the change of the curved path, it is necessary to collect the deviation degree and the corresponding cargo load data in different specified operation records, and sort these data in ascending order. The sorted data is used to draw a correlation curve chart between load and deviation degree to analyze the influence of load on deviation degree.

[0149] These data should be coherent and consistent when analyzed, especially the gap between load data should not be too large. Adjacent load data should be kept within a reasonable range to avoid extreme values ​​or abnormal values ​​that deviate significantly from other data. If the cargo load data differ too much or there are extremely large data points, the curve chart may lose coherence and fail to accurately reflect the true relationship between load and deviation. Such abnormal data may be caused by specific operating errors or rare situations and are not universal. Therefore, they should be handled or excluded with caution when drawing and analyzing to ensure the accuracy and practical significance of the correlation analysis. Maintaining a reasonable interval and continuity of the data can more truly reflect the deviation characteristics of the vehicle under different load conditions and provide a reliable basis for optimization.

[0150] The original predetermined turning curvature in the high-frequency deviation area refers to the ideal or planned turning curvature for a specific high-frequency deviation area set in the vehicle planning and navigation system. This is usually preset based on factors such as vehicle design parameters, road design, and traffic regulations, with the purpose of guiding the vehicle to drive safely along the predetermined path in this area without affecting safety and efficiency. This curvature takes into account the curve of the road, as well as the operability and stability of the vehicle, and is the basic guidance in the vehicle control system on how to turn in this area.

[0151] The benefit of multiplying the original predetermined turning curvature with the overall slope of the correlation curve to obtain the optimized turning curvature is mainly reflected in the dynamic adjustment of vehicle behavior to adapt to the actual driving needs under different load conditions. The original predetermined turning curvature may not fully consider the changes in vehicle behavior under different loads, especially when the load is large, the vehicle's handling performance and dynamic response will be different. By analyzing the correlation between load and deviation degree, a slope indicator that describes this change trend can be obtained.

[0152] This slope reflects the rate of change of the deviation degree as the load increases, that is, the proportion of the deviation degree increasing for each unit increase in load. By multiplying this slope with the original curvature, the original curvature can be adjusted to make it closer to the actual driving situation. This adjustment helps to improve the driving accuracy and safety of the vehicle under high load conditions. The adjusted curvature can better reflect the actual vehicle dynamic characteristics, optimize the driving trajectory, reduce deviation accidents caused by improper driving on curves, and enhance the adaptability and stability of autonomous driving vehicles in complex road conditions.

[0153] The basis of this optimization measure is that the correlation results obtained through analysis of actual driving data are based on actual driving conditions and are more valuable as references than a single theoretical model or preset parameters, and can significantly improve navigation accuracy and driving safety.

[0154] In summary, by carefully analyzing the historical operation data of autonomous vehicles, especially the correlation between load and deviation, we can significantly improve the driving accuracy and safety of vehicles under various load conditions. This method enables autonomous vehicles to automatically adjust the curvature according to different loads and pass through winding roads in the best way, thereby reducing the risk of deviation and improving driving stability.

[0155] More specifically, the optimized turning curvature is not only suitable for regular driving conditions, but also significantly improves the vehicle's responsiveness and control accuracy in emergency avoidance situations. This optimization measure effectively enhances the vehicle's ability to respond to emergencies by keeping the vehicle's deviation level low when quick avoidance is required, thereby greatly improving the safety of passengers and pedestrians, while enhancing the public's trust and acceptance of autonomous vehicle technology.

[0156] In short, by optimizing turning curvatures based on this data-based approach, autonomous vehicles can better adapt to complex road environments and demonstrate higher safety and reliability in both daily transportation and in the face of emergencies. This has a positive effect on the advancement of autonomous driving technology and its popularization and acceptance in practical applications.

[0157] It should be understood that, although each step in the flow chart of each embodiment of the present invention is shown in sequence according to the indication of the arrow, these steps are not necessarily performed in sequence according to the order indicated by the arrow. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0158] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0159] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0160] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

[0161] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for detecting an autonomous driving vehicle, characterized in that: The method comprises: Obtain the historical operation logs of the target automatic vehicle on the preset transportation route, and screen out the operation records with the operation trajectory deviation phenomenon, and among these operation records, find out the operation records containing the same deviation area and the number of which exceeds the preset value, set them as designated operation records, and set the same deviation area as the high-frequency deviation area; determining cargo load data of the target automated vehicle for each specified run record and the degree of deviation of the target automated vehicle in the high frequency deviation area; Analyze whether the deviation degree of the high-frequency deviation area in different specified operation records of the target automatic vehicle is positively correlated with the cargo load data. If it is determined that there is a positive correlation, analyze whether the objective influencing factors of the high-frequency deviation area are only caused by the change of the curved path; If it is determined that the objective influencing factors of the high-frequency deviation area are only caused by the change of the curve path, a correlation curve graph between the load and the degree of deviation is generated according to the deviation degree and cargo load data in different specified operation records, and the original predetermined turning curvature of the target automatic vehicle in the high-frequency deviation area is adjusted according to the correlation curve graph.

2. The automatic driving vehicle detection method according to claim 1, characterized in that: The steps of determining cargo load data of the target automated vehicle for each designated operation record and the degree of deviation of the target automated vehicle in the high-frequency deviation area include: Determine cargo load data of the target automated vehicle in each designated operation record based on the historical operation log of the target automated vehicle on the preset transportation route; According to the historical operation log, the GPS data of the target automatic vehicle in the high-frequency deviation area is obtained, and the maximum deviation point of the target automatic vehicle from the preset transportation route is determined according to the GPS data; The straight-line distance from the maximum deviation point to the target transport route is set as the deviation degree of the target automatic vehicle in the high-frequency deviation area.

3. The automatic driving vehicle detection method according to claim 2, characterized in that: The steps of analyzing whether the degree of deviation in the high-frequency deviation area of ​​the target automatic vehicle in different designated operation records is positively correlated with the cargo load data, and if it is determined that there is a positive correlation, analyzing whether the objective influencing factor of the high-frequency deviation area is only caused by the change of the curved path include: Analyze whether the target automatic vehicle has a greater degree of deviation in the high-frequency deviation area when the cargo load data of the target automatic vehicle is greater in different designated operation records of the target automatic vehicle; If it is determined that the larger the cargo load data of the target automatic vehicle is, the greater the deviation degree of the target automatic vehicle in the high-frequency deviation area is, then it is determined that there is a positive correlation between the deviation degree of the high-frequency deviation area and the cargo load data; After confirming the existence of a positive correlation, the objective influencing factors in the high-frequency deviation area were analyzed to see whether they were caused only by the change in the curve path.

4. The automatic driving vehicle detection method according to claim 3, characterized in that: After confirming that there is a positive correlation, the steps to analyze whether the objective influencing factors of the high-frequency deviation area are only caused by the change of the curve path include: After determining that there is a positive correlation, the tilt angle data of the target vehicle in the high-frequency deviation area is obtained according to the historical operation log, and it is determined whether there is a longitudinal slope change of the target vehicle in the high-frequency deviation area according to the tilt angle data; If it is determined that there is no longitudinal slope change in the high-frequency deviation area for the target vehicle, a driving trajectory of the target automatic vehicle in the high-frequency deviation area is plotted based on the GPS data of the target automatic vehicle in the high-frequency deviation area; It is analyzed whether the driving trajectory of the target automatic vehicle in the high-frequency deviation area is a curved trajectory. If so, it is determined that the objective influencing factors of the high-frequency deviation area are only caused by the change of the curved path.

5. The automatic driving vehicle detection method according to claim 1, characterized in that: If it is determined that the objective influencing factor of the high-frequency deviation area is only caused by the change of the curved path, then according to the deviation degree and cargo load data in different designated operation records, a correlation curve diagram between the load and the deviation degree is generated, and the steps of adjusting the original predetermined turning curvature of the target automatic vehicle in the high-frequency deviation area according to the correlation curve diagram include: If it is determined that the objective influencing factor of the high-frequency deviation area is only caused by the change of the curved path, the deviation degree and the corresponding cargo load data in different specified operation records are collected, and these cargo load data are sorted in ascending order; According to the sorted cargo load data and their corresponding deviation degrees, a correlation curve between load and deviation degree is drawn; An originally predetermined turning curvature of the target automated vehicle in the high frequency deviation region is adjusted according to the correlation curve graph.

6. The automatic driving vehicle detection method according to claim 5, characterized in that: The step of adjusting the original predetermined turning curvature of the target automatic vehicle in the high-frequency deviation area according to the correlation curve graph includes: Calculate the overall slope of the correlation curve between load and deviation degree; Extracting the original predetermined turning curvature of the target automated vehicle in the high-frequency deviation area from the historical operation log; The original predetermined turning curvature is multiplied by the overall slope of the correlation curve to obtain the optimized turning curvature.

7. An automatic driving vehicle detection system, characterized in that: The system includes: a data screening module, a data extraction module, a forward correlation analysis module, and a curvature adjustment module, wherein: A data screening module is used to obtain the historical operation logs of the target automatic vehicle on the preset transportation route, and screen out the operation records with the operation trajectory deviation phenomenon, and find out the operation records containing the same deviation area and the number of which exceeds the preset value among these operation records, set them as designated operation records, and set the same deviation area as the high-frequency deviation area; a data extraction module for determining cargo load data of the target automated vehicle for each specified run record and a degree of deviation of the target automated vehicle in the high frequency deviation area; A positive correlation analysis module is used to analyze whether the degree of deviation in the high-frequency deviation area of ​​the target automatic vehicle in different specified operation records is positively correlated with the cargo load data. If it is determined that there is a positive correlation, it is analyzed whether the objective influencing factors of the high-frequency deviation area are only caused by the change of the curved path; The turning curvature adjustment module is used to generate a correlation curve diagram between load and deviation degree according to the deviation degree and cargo load data in different specified operation records if it is determined that the objective influencing factor of the high-frequency deviation area is only caused by the change of the curve path, and adjust the original predetermined turning curvature of the target automatic vehicle in the high-frequency deviation area according to the correlation curve diagram.

8. The automatic driving vehicle detection system according to claim 7, characterized in that: The data extraction module specifically includes: A load data acquisition unit, used to determine the cargo load data of the target automatic vehicle in each designated operation record according to the historical operation log of the target automatic vehicle on the preset transportation route; A GPS data acquisition unit, used to acquire GPS data of the target automatic vehicle in the high-frequency deviation area according to the historical operation log, and determine the maximum deviation point of the target automatic vehicle from the preset transportation route according to the GPS data; The deviation degree determination unit is used to set the straight-line distance from the maximum deviation point to the target transportation route as the deviation degree of the target automatic vehicle in the high-frequency deviation area.

9. The automatic driving vehicle detection system according to claim 8, characterized in that: The forward association analysis module specifically includes: A designated operation record analysis unit is used to analyze whether the target automatic vehicle has a greater degree of deviation in a high-frequency deviation area when the cargo load data of the target automatic vehicle is greater in different designated operation records of the target automatic vehicle; a positive correlation determination unit, for determining that the degree of deviation in the high-frequency deviation area is positively correlated with the cargo load data if it is determined that the greater the cargo load data of the target automatic vehicle, the greater the degree of deviation of the target automatic vehicle in the high-frequency deviation area; The objective influencing factor analysis unit is used to analyze whether the objective influencing factor of the high-frequency deviation area is only caused by the change of the curve path after determining that there is a positive correlation.

10. The automatic driving vehicle detection system according to claim 9, characterized in that: The forward association analysis module also includes: A longitudinal gradient change determination unit is used to obtain the inclination angle data of the target vehicle in the high-frequency deviation area according to the historical operation log after determining that there is a positive correlation, and determine whether there is a longitudinal gradient change of the target vehicle in the high-frequency deviation area according to the inclination angle data; A driving trajectory drawing unit is used to draw a driving trajectory of the target automatic vehicle in the high-frequency deviation area according to the GPS data of the target automatic vehicle in the high-frequency deviation area if it is determined that there is no longitudinal slope change in the high-frequency deviation area of ​​the target vehicle; The driving trajectory analysis unit is used to analyze whether the driving trajectory of the target automatic vehicle in the high-frequency deviation area is a curved trajectory. If so, it is determined that the objective influencing factors of the high-frequency deviation area are only caused by the change of the curved path.

Images