Vehicle control function dynamic management method and system based on driving scene safety level

Abstract

The invention provides a vehicle control function dynamic management method and system based on a driving scene safety level, and relates to the technical field of vehicles, and the method comprises the steps: obtaining historical traffic data, real-time traffic data and environment data of a traffic congestion region and an accident-prone road section; based on the historical traffic data and the real-time traffic data, generating risk level information by using a traffic data prediction model; generating a three-dimensional perception result of the surrounding environment of the vehicle by utilizing a three-dimensional reconstruction model based on the environment data, and determining a target safety level and a level change trend of the vehicle in a preset time period by combining the risk level information so as to generate a management instruction of each vehicle control function; according to the method, the management instruction is executed through the pre-constructed cooperative computing framework, cooperative operation information fed back by the traffic participants and the roadside equipment in the traffic congestion area and the accident-prone road section is received, the management instruction is optimized, the target management instruction is obtained, and the safety, comfort and passing efficiency of the vehicle in the high-risk area are improved.

Description

Technical Field

[0001] This application relates to the field of vehicle technology, and in particular to a method and system for dynamic management of vehicle control functions based on driving scenario safety levels. Background Technology

[0002] With the rapid development of intelligent connected vehicles, vehicles need to achieve refined management of vehicle control functions in complex and ever-changing road environments. Especially in traffic-congested areas or accident-prone sections, the system must comprehensively consider the road risk situation and the surrounding dynamic environment, and adjust control strategies such as automatic emergency braking and lane keeping assist in real time to improve driving safety and traffic efficiency.

[0003] Current solutions construct risk maps by integrating historical accident statistics with real-time traffic flow information and use onboard sensors to identify surrounding obstacles, thereby setting fixed thresholds for activating vehicle control functions. However, these solutions lack the ability to dynamically model the evolution of safety levels, making it difficult to adapt to continuous changes in traffic conditions. Furthermore, their decision-making process does not fully incorporate collaborative feedback from roadside units and other traffic participants, resulting in delayed or overly conservative functional adjustments, failing to balance safety with driving comfort and traffic efficiency. Summary of the Invention

[0004] The purpose of this application is to provide a method and system for dynamic management of vehicle control functions based on the safety level of driving scenarios, so as to solve the problem that the adjustment of vehicle control functions is lagging or overly conservative due to the lack of dynamic modeling and collaborative feedback mechanism for safety situation in the prior art.

[0005] To address the aforementioned technical problems, in a first aspect, this application provides a method for dynamic management of vehicle control functions based on driving scenario safety levels, including: Obtain historical traffic data, real-time traffic data, and environmental data for traffic congestion areas and accident-prone road sections; Based on the historical and real-time traffic data, risk level information is generated using a pre-built traffic data prediction model. Based on the environmental data, a three-dimensional perception result of the vehicle's surrounding environment is generated using a pre-built three-dimensional reconstruction model. Based on the risk level information and the three-dimensional perception results, the target safety level and level change trend of the vehicle within a preset time period are determined. Based on the target safety level and the trend of level change, management instructions for each vehicle control function are generated. The management instructions are executed through a pre-built collaborative computing framework, and collaborative operation information from traffic participants and roadside equipment in the traffic congestion area and accident-prone road sections is received through the collaborative computing framework to optimize and adjust the management instructions and obtain the target management instructions.

[0006] Optionally, based on the historical traffic data and real-time traffic data, risk level information is generated using a pre-built traffic data prediction model, including: Based on the road segmentation standards for traffic congestion areas and accident-prone road sections, and the time cycle of urban commuting, the historical traffic data and real-time traffic data are processed by road segment matching and time period segmentation to obtain historical traffic sub-data and real-time traffic sub-data for each sub-road segment in different time periods. Based on historical and real-time traffic data, the traffic operation status of each sub-road segment is inferred through a pre-built traffic data prediction model, and the traffic trend information of each sub-road segment within a preset time period is obtained. Risk assessment is performed on the traffic trend information of each sub-road segment to obtain the risk level information of each sub-road segment within a preset time period.

[0007] Optionally, based on the environmental data, a three-dimensional perception result of the vehicle's surrounding environment is generated using a pre-built three-dimensional reconstruction model, including: Based on the spatial orientation around the vehicle, the environmental data is divided into different data categories. Based on each classification data group, the spatial positional relationship, relative distance and object type of each object around the vehicle are determined by a pre-built 3D reconstruction model. Based on the spatial positional relationship, relative distance and object type, the spatial area around the vehicle is divided into the vehicle driving area, the surrounding object area and the safety buffer area, and the motion state information of each object in each area is obtained. By integrating the spatial positional relationships, relative distances, object types, and motion state information, a three-dimensional perception result of the vehicle's surrounding environment is generated.

[0008] Optionally, based on the risk level information and the three-dimensional perception results, the target safety level and level change trend of the vehicle within a preset time period are determined, including: The risk level information and the information in the three-dimensional perception results are compared with the preset safety level determination rules to determine the initial safety level of the vehicle at each time node within the preset time period. Based on the risk level information, the fluctuation range and duration of each sub-segment within a preset time period, combined with the speed change range, movement offset trend and motion stability of the object motion state information in the three-dimensional perception results, the logic of the change of the initial safety level of each adjacent time node within the preset time period is analyzed. Based on the change logic of each adjacent time node, the initial security level of each time node is adjusted to obtain the target security level and generate the level change trend.

[0009] Optionally, based on the target safety level and the trend of level change, management instructions for each vehicle control function are generated, including: Based on the adjustable range, operation priority, and target safety level of each vehicle control function, determine the vehicle control function parameters at each time point; Based on the trend characteristics, fluctuation amplitude, and fluctuation duration of the aforementioned level changes, the adjustment parameters for each vehicle control function are set. Based on the operation priority and the adjustment parameters of each vehicle control function, the vehicle control function parameters at each time point are adapted and adjusted to generate operation instructions for each vehicle control function. Identify abrupt change nodes in the trend of the level change, and supplement emergency operation instructions into the operation instructions corresponding to the abrupt change nodes to form management instructions for each vehicle control function.

[0010] Optionally, based on the adjustable range, operational priority, and target safety level of each vehicle control function, the vehicle control function parameters at each time point are determined, including: Determine the adjustable range of each vehicle control function based on its type; Based on the degree of impact of each vehicle control function on vehicle driving safety, an operation priority is assigned to each vehicle control function. Based on the vehicle control function type, the adjustable range of each vehicle control function and the operation priority, multiple driving safety levels and corresponding vehicle control function parameters are set for each level. Match the target safety level corresponding to each time node within a preset time period to determine the vehicle control function parameters for each time node; Based on the risk level information and the three-dimensional perception results, the vehicle control function parameters at each time point are adjusted to obtain the vehicle control function parameters.

[0011] Optionally, the management instructions are executed through a pre-built collaborative computing framework, and collaborative operation information from various traffic participants and roadside equipment in the traffic congestion area and accident-prone road sections is received through the collaborative computing framework to optimize and adjust the management instructions, resulting in target management instructions, including: Establish information exchange channels among traffic participants in traffic congestion areas and accident-prone road sections by using a pre-built collaborative computing framework. The collaborative computing framework distributes management instructions for each vehicle control function to the corresponding vehicle control module, thereby driving each vehicle control module to perform vehicle control operations according to the corresponding management instructions, and receives collaborative operation information from traffic participants and roadside equipment through the information interaction channel. The target information in the collaborative operation information is compared and analyzed with the target security level and level change trend at the corresponding time node to determine the degree of influence of each collaborative operation information on each management instruction. Based on the degree of impact and the operational priority of the vehicle control function, conflict optimization is performed on the vehicle control function parameters and execution time points in each management instruction to form the target management instruction.

[0012] Secondly, this application provides a dynamic management system for vehicle control functions based on driving scenario safety levels, including: The acquisition module is used to acquire historical traffic data, real-time traffic data, and environmental data for traffic congestion areas and accident-prone road sections. The first generation module is used to generate risk level information based on the historical traffic data and real-time traffic data, using a pre-built traffic data prediction model. The second generation module is used to generate a three-dimensional perception result of the vehicle's surrounding environment based on the environmental data and using a pre-built three-dimensional reconstruction model. The determination module is used to determine the target safety level and level change trend of the vehicle within a preset time period based on the risk level information and the three-dimensional perception results. The third generation module is used to generate management instructions for each vehicle control function based on the target safety level and the level change trend. The adjustment module is used to execute the management instructions through a pre-built collaborative computing framework, and to receive collaborative operation information from traffic participants and roadside equipment in the traffic congestion area and accident-prone road section through the collaborative computing framework, so as to optimize and adjust the management instructions to obtain the target management instructions.

[0013] Thirdly, this application provides a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are to be invoked and executed by the processing component to implement the steps of the dynamic management method for vehicle control functions based on driving scenario safety levels as described in the first aspect above.

[0014] Fourthly, this application provides a computer storage medium storing a computer program, which, when executed by a computer, implements the steps of the vehicle control function dynamic management method based on driving scenario safety levels as described in the first aspect above.

[0015] The vehicle control function dynamic management method based on driving scenario safety levels provided in this application acquires historical and real-time traffic data and environmental data from congested areas and accident-prone road sections. It then uses a traffic data prediction model to generate risk level information and combines this with a 3D reconstruction model to construct a three-dimensional perception of the vehicle's surrounding environment. This allows for a comprehensive assessment of the vehicle's target safety level and its changing trends over future periods. Based on this, it dynamically generates appropriate vehicle control function management instructions and, while executing these instructions using a collaborative computing framework, integrates feedback information from other traffic participants and roadside equipment to continuously optimize and adjust the instructions. Ultimately, this achieves accurate, forward-looking, and collaborative management of vehicle control functions in complex, high-risk scenarios. Therefore, this application overcomes the problem of rigid functional response caused by static threshold settings and isolated decision-making mechanisms in traditional solutions, improving the system's adaptability to dynamic traffic environments and the rationality of its control strategies.

[0016] Furthermore, this application performs spatiotemporal matching and segmentation processing on historical and real-time traffic data according to road segment division standards and urban commuting cycles to form refined sub-road segment traffic sub-data. Based on this, a predictive model is used to infer the future traffic operation status of each sub-road segment, thereby completing the risk assessment. This makes the risk assessment not only accurate in spatial granularity, but also reflects the evolution characteristics of traffic conditions at different times, thus providing a more timely and scenario-specific safety level basis for vehicle control functions. This makes up for the shortcomings of existing technologies that are difficult to support refined and differentiated functional control due to coarse-grained risk modeling. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A flowchart illustrating the dynamic management method for vehicle control functions based on driving scenario safety levels provided in this application embodiment; Figure 2 A schematic diagram illustrating a specific implementation of the vehicle control function dynamic management method based on driving scenario safety levels provided in this application embodiment; Figure 3 This is a schematic diagram of the structure of the vehicle control function dynamic management system based on driving scenario safety level provided in the embodiments of this application. Detailed Implementation

[0019] To address the problem that existing solutions struggle to dynamically adapt to safety requirements in complex traffic scenarios due to their reliance on static thresholds and isolated perception, this application provides a dynamic management method for vehicle control functions based on driving scenario safety levels. The core idea of ​​this method is to generate risk levels with temporal evolution characteristics by integrating historical and real-time traffic data, and combine this with high-dimensional environment reconstruction results to jointly characterize the vehicle's safety status and its direction of change in future time periods, using this as the basis for adjusting vehicle control functions. Simultaneously, a multi-party collaborative feedback channel is introduced to continuously absorb interactive information from roadside facilities and other traffic participants during command execution, enabling online optimization of management strategies. This overcomes the limitations of traditional methods in terms of response lag, environmental adaptability, and control granularity.

[0020] To enable those skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are merely some embodiments of the present application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0021] The core of this application is to provide a dynamic management method for vehicle control functions based on driving scenario safety levels. A flowchart of one specific implementation is shown below. Figure 1 As shown, the method includes: Step 101: Obtain historical traffic data, real-time traffic data, and environmental data for traffic congestion areas and accident-prone road sections.

[0022] In this step, historical traffic data refers to a dataset obtained from past traffic operation records of traffic congestion areas and accident-prone road sections. This data includes traffic flow, vehicle speed, road section travel time, accident frequency, and traffic conditions at the time of accidents at different times.

[0023] Real-time traffic data refers to a dynamic dataset collected based on the current traffic conditions in congested areas and accident-prone road sections. This data includes information such as the number of vehicles on the current road section, vehicle trajectories, lane occupancy, and following distance.

[0024] Environmental data refers to datasets collected based on the natural environment and surrounding environment of traffic congestion areas and accident-prone road sections. This data includes information on light intensity, visibility, temperature and humidity, road surface conditions, and the distribution of objects around vehicles.

[0025] In this embodiment, the geographical scope of traffic congestion areas and accident-prone road sections is first delineated. Then, historical traffic data of the corresponding road sections is retrieved from the historical database of traffic management according to the delineated geographical scope. Real-time traffic data within the geographical scope is collected in real time through sensing devices such as roadside cameras and microwave radar deployed on the above-mentioned road sections. Environmental data is obtained by simultaneously collecting natural environmental data and vehicle surrounding environment sensing data within the above-mentioned road sections through vehicle-mounted sensors and roadside environmental sensing devices.

[0026] Step 102: Based on the historical traffic data and real-time traffic data, generate risk level information using a pre-built traffic data prediction model.

[0027] In this step, the risk level information refers to the time-based risk classification results of the sub-road segment obtained after risk assessment. This information includes three levels: high risk, medium risk, and low risk, and also marks the risk factors corresponding to each level, such as traffic saturation and excessive speed fluctuations.

[0028] In the embodiments of this application, such as Figure 2 As shown, step 102 specifically includes the following steps: Step 201: Based on the road segmentation standards for traffic congestion areas and accident-prone road sections and the time cycle of urban commuting, perform road segment matching and time period segmentation processing on the historical traffic data and real-time traffic data respectively to obtain historical traffic sub-data and real-time traffic sub-data of each sub-road segment in different time periods.

[0029] In this step, the road segmentation criteria refer to the criteria for classifying traffic congestion areas and accident-prone road segments. These criteria include road grade, number of lanes, intersection distribution, and road segment length.

[0030] The time period of urban commuting refers to the time interval divided based on the travel patterns of urban residents. This time period includes the morning peak, off-peak, evening peak, and nighttime off-peak.

[0031] Historical traffic sub-data refers to the set of historical traffic data assigned to a specific sub-segment and time period after being matched with road segments and divided into time periods. This data includes historical records such as traffic flow and driving speed within the corresponding sub-segment and time period.

[0032] Real-time traffic sub-data refers to the set of real-time traffic data assigned to a specific sub-segment and time period after being matched with road segments and divided into time periods. This data includes dynamic data such as the current number of vehicles and lane occupancy within the corresponding sub-segment and time period.

[0033] In this embodiment of the application, firstly, based on the road segmentation standard, the traffic congestion area and accident-prone road segment are divided into several sub-segments according to the attributes of main roads and secondary roads, combined with the number of lanes, the distance between intersections and the length of the road segment; at the same time, the time period of urban commuting is divided. For example, one way to divide it is to divide the morning peak into 7:00-9:00, the off-peak into 9:00-17:00, the evening peak into 17:00-19:00, and the nighttime off-peak into 19:00-7:00 the next day.

[0034] Subsequently, historical traffic data and real-time traffic data are associated with the divided sub-road segments. By matching the road segment numbers, it is ensured that each piece of data corresponds to a unique sub-road segment. Then, the two types of data after matching are split according to the four time periods mentioned above, and the historical traffic data and real-time traffic data corresponding to each sub-road segment in different time periods are filtered out and integrated into structured historical traffic sub-data and real-time traffic sub-data respectively.

[0035] Step 202: Based on historical traffic sub-data and real-time traffic sub-data, the traffic operation status of each sub-road segment is inferred through the pre-built traffic data prediction model to obtain the traffic trend information of each sub-road segment within a preset time period.

[0036] In this step, traffic operation status refers to the dynamic performance of traffic flow within the sub-segment. This traffic operation status includes changes in traffic volume, fluctuations in vehicle speed, changes in lane occupancy, and the distribution of following distance.

[0037] The preset time period refers to the short-term forecasting needs of traffic operation in traffic congestion areas and accident-prone road sections. The time interval for the model to predict the future traffic status of sub-road sections is set in advance, and can be set to a short-term time range of 1-2 hours according to the traffic flow change characteristics of the road section.

[0038] Traffic trend information refers to the changes in traffic operation status of sub-road segments within a preset time period, derived from current and historical traffic data through model extrapolation. This information includes the changing trends of traffic parameters in each time period, the time of peak occurrence, and the amplitude of fluctuations.

[0039] In this embodiment, historical traffic data and real-time traffic data of each sub-road segment at different time periods are first input into a pre-built traffic data prediction model. The core features of the two types of data are extracted by the model encoding module and temporal correlation calculation is performed. Then, the model decoding module infers the traffic operation status of each sub-road segment within a preset 1-2 hour period based on the features that have completed the temporal correlation. Finally, the traffic flow increase / decrease trend, driving speed fluctuation range, lane occupancy rate peak and peak occurrence time are integrated to obtain the traffic trend information of each sub-road segment within a preset time period.

[0040] Among them, the traffic data prediction model can adopt the time series prediction model type. The model structure is designed as an improved encoder-decoder structure. The input layer parameters correspond to the feature dimensions of historical traffic sub-data and real-time traffic sub-data, specifically including three core feature dimensions: traffic flow, driving speed, and lane occupancy rate. The hidden layer is set with two layers of time series memory units, and the number of neurons in each time series memory unit is set to 64, which adapts to the volume of traffic data while taking into account computational efficiency and prediction accuracy. The output layer parameters correspond to the three core feature values ​​of traffic operation status. The model training process uses historical traffic data and real-time traffic data from the past three years for traffic congestion areas and accident-prone road sections as training sets, and uses the actual traffic operation data of the corresponding road sections during the same period as labels. The gradient descent algorithm is used to iteratively adjust the internal parameters of the model until the error between the model's predicted value and the actual value is lower than the preset threshold, thus completing the pre-construction training of the model.

[0041] The structural design, parameter design, and training process of the traffic data prediction model described above are exemplary. This application does not limit these aspects and can be configured accordingly based on actual circumstances.

[0042] Step 203: Assess the risk of traffic trends in each sub-segment to obtain the risk level information of each sub-segment within a preset time period.

[0043] In this step, the risk level information refers to the time-based risk classification results of the sub-segment obtained after risk assessment. This information includes three levels: high risk, medium risk, and low risk, and also marks the risk causes corresponding to each level.

[0044] In this embodiment, a risk assessment threshold standard is first preset, setting grading thresholds based on three core dimensions in traffic trend information: traffic flow saturation rate, duration of driving speed below the safe threshold, and peak lane occupancy rate. Then, the traffic trend information of each sub-segment is compared with the preset thresholds one by one. If two or more dimensions exceed the corresponding threshold, the corresponding time period of the sub-segment is judged as high-risk; if only one dimension exceeds the threshold or two dimensions are close to the threshold, the corresponding time period of the sub-segment is judged as medium-risk; if all dimensions are within the safe threshold range, the corresponding time period of the sub-segment is judged as low-risk. Simultaneously, the risk level and corresponding risk triggers for different time periods of each sub-segment are labeled. Finally, the time-based risk grading results and trigger labels of all sub-segments are integrated to obtain the risk level information of each sub-segment within a preset time period.

[0045] This application's embodiments combine hierarchical data processing with model extrapolation to ensure that the risk assessment results are adapted to the actual scenarios of traffic congestion and accident-prone road sections. This provides a reliable basis for subsequent target safety level determination and vehicle control command generation, and makes up for the risk prediction bias caused by fuzzy model design and disordered data processing in traditional technologies.

[0046] Step 103: Based on the environmental data, generate a three-dimensional perception result of the vehicle's surrounding environment using a pre-built three-dimensional reconstruction model.

[0047] In this step, the pre-built 3D reconstruction model refers to the type of 3D reconstruction model that combines vision and point cloud. The structure design can be a three-layer architecture that includes a feature extraction layer, a data fusion layer, and a 3D modeling layer. The input layer parameters correspond to the feature dimensions of the environmental data. The feature extraction layer has 4 convolutional modules, each with a 3×3 kernel size. The data fusion layer integrates multi-source data using a feature stitching method. The 3D modeling layer outputs 3D point cloud data based on the Cartesian coordinate system. The training process uses environmental data from traffic congestion areas and accident-prone road sections as the training set, and actual 3D spatial data of the corresponding scenarios as labels. A gradient descent algorithm is used to iteratively adjust model parameters until the deviation between the 3D modeling result and the actual scene is below a preset threshold, thus completing the model pre-construction. The above structural design, parameter design, and training process of the 3D reconstruction model are exemplary, and this application does not limit them; appropriate settings can be made according to actual circumstances.

[0048] The three-dimensional perception result refers to a structured three-dimensional data set that integrates the spatial parameters, region division, and motion status of objects around the vehicle. This result includes the three-dimensional coordinates, type, motion parameters, and region affiliation of each object.

[0049] In this embodiment of the application, step 103 specifically includes the following steps: Step 301: Divide the environmental data into different classification data groups according to the spatial orientation around the vehicle.

[0050] In this step, spatial orientation refers to the orientation of the surrounding area centered on the vehicle, divided in the horizontal and vertical directions. This spatial orientation can include eight directions: front, rear, left, right, left front, right front, left rear, and right rear.

[0051] A categorized data set refers to a set of environmental data obtained by splitting it according to spatial orientation. Each set corresponds to a spatial orientation and includes visual data, distance data, and environmental parameter data collected in that orientation.

[0052] In this embodiment, the vehicle's center of gravity is used as the origin, and a perception range of 0° to 360° in the horizontal direction and -10° to 30° in the vertical direction is defined. The area boundaries are divided according to eight spatial orientations, and the angle range of each orientation is clearly defined. Subsequently, the collection location identifier is extracted from the environmental data, and the visual data, distance data, temperature and humidity data, and road condition data of the corresponding orientation are assigned to the corresponding area, splitting them into eight classification data groups. Each data group is labeled with the corresponding spatial orientation to ensure that the data corresponds accurately with the collection location.

[0053] Step 302: Based on each classification data group, determine the spatial positional relationship, relative distance and object type between objects around the vehicle using a pre-built 3D reconstruction model.

[0054] In this step, spatial positional relationship refers to the three-dimensional arrangement of objects around the vehicle relative to the vehicle and to each other.

[0055] Relative distance refers to the straight-line distance between each object and the center of gravity of the vehicle, as well as the straight-line distance between objects.

[0056] Object type refers to the object attributes determined based on environmental data characteristics. It can be divided into four categories: motor vehicles, non-motor vehicles, pedestrians, and fixed obstacles, which are used to distinguish the motion and safety impact characteristics of different objects.

[0057] In this embodiment, each category of data is first input into a pre-built 3D reconstruction model. The convolutional module of the model's feature extraction layer extracts visual features, distance features, and environmental auxiliary features from each data group. A data fusion layer then integrates multiple features from the same location, eliminating data redundancy and errors. Based on the fused features, the 3D modeling layer marks the preliminary 3D coordinates of each object in a Cartesian coordinate system. Coordinate calculations determine the spatial relationships between each object and the vehicle, as well as the corresponding relative distances. Simultaneously, the model's built-in feature matching module compares the extracted object contours and texture features with a pre-defined object type feature library to determine the specific type of each object, outputting 3D basic data including spatial relationships, relative distances, and object types.

[0058] Step 303: Based on the spatial positional relationship, relative distance and object type, divide the space around the vehicle into the vehicle driving area, the surrounding object area and the safety buffer area, and obtain the motion state information of each object in each area.

[0059] In this step, the vehicle driving area refers to the three-dimensional spatial range in which the vehicle can drive normally, based on the current lane range, vehicle size, and driving trajectory planning. This area includes the spatial area corresponding to the lane width and driving route.

[0060] The surrounding object area refers to the three-dimensional space occupied by each object itself, as well as the area that the object's trajectory may cover, and is defined according to the object type and size.

[0061] A safety buffer zone is a safe space reserved between the vehicle driving area and the surrounding object area to deal with emergencies. Its width is dynamically set according to the type of object and the relative distance.

[0062] Motion state information refers to the dynamic motion parameters of each object, including speed, direction of motion, and acceleration.

[0063] In this embodiment, firstly, based on spatial relationships and relative distances, the horizontal range of the vehicle's driving area is defined as 1.2 times the vehicle's width, and the vertical range is defined as the vehicle's height, using the current driving lane as a reference. Then, based on the type and size of each object, the surrounding object area for each object is defined to ensure coverage of the object itself and its potential range of motion. Next, a safety buffer zone width is set between the vehicle's driving area and the surrounding object area, according to the object type. Simultaneously, based on the three-dimensional base data corresponding to continuous frame environmental data, the coordinate changes of each object per unit time are calculated to obtain the driving speed and direction of motion. Acceleration is calculated using the speed change, and finally, the information obtained above is integrated to form the motion state information of objects within each area.

[0064] Step 304: Integrate the spatial positional relationships, relative distances, object types, and motion state information to generate a three-dimensional perception result of the vehicle's surrounding environment.

[0065] In this step, the three-dimensional perception result refers to the structured and integrated three-dimensional data set of the vehicle's surrounding environment. This result includes the three-dimensional coordinates of each object in the Cartesian coordinate system, spatial positional relationships, relative distances, object types, motion state information, and the division results of each region.

[0066] In this embodiment, a Cartesian coordinate system with the vehicle's center of mass as the origin is first used as a reference. Spatial positional relationships and relative distances are converted into precise three-dimensional coordinates for each object and region, and the corresponding object types and motion state information are associated to construct a three-dimensional data matrix. This data matrix is ​​then structured and organized, with data categorized by region. The parameters of all objects within each region and the spatial relationships between objects are labeled. At the same time, auxiliary information such as lighting and road conditions from the environmental data is supplemented to correct errors in the three-dimensional coordinates and motion state information, generating a complete three-dimensional perception result of the vehicle's surrounding environment.

[0067] This application embodiment accurately reconstructs the surrounding environment of the vehicle, providing comprehensive environmental support for subsequent target safety level determination, and making up for the environmental perception deviation caused by the disordered data splitting and insufficient object recognition accuracy of traditional 3D modeling.

[0068] Step 104: Based on the risk level information and the three-dimensional perception results, determine the target safety level and level change trend of the vehicle within a preset time period.

[0069] In this step, the target safety level refers to the safety classification result that matches the actual risk status of the vehicle at each time point within a preset time period after the initial safety level is adjusted. It can be divided into three levels: high risk, medium risk, and low risk.

[0070] The level change trend refers to the continuous change pattern of the target security level at each time node within a preset time period. This trend includes the level rise and fall direction, fluctuation frequency, and abrupt change nodes.

[0071] In this embodiment of the application, step 104 specifically includes the following steps: Step 401: Compare the risk level information and the information in the three-dimensional perception results with the preset safety level determination rules to determine the initial safety level of the vehicle at each time node within the preset time period.

[0072] In this step, the preset safety level determination rule refers to a pre-defined safety classification standard that integrates road segment risk and surrounding environmental characteristics for traffic congestion areas and accident-prone road sections. This rule covers road segment risk thresholds corresponding to risk level information, object safety distance thresholds corresponding to 3D perception results, and motion state thresholds. It should be noted that this embodiment does not limit the size of each threshold and can be set according to actual conditions.

[0073] A time node refers to a discrete time point in which a preset time period is divided at fixed intervals. The interval can be set to 1 minute. In this embodiment of the application, the size of the interval is not limited and can be set according to the actual situation.

[0074] The initial security level refers to the security classification results of each time point based on preset rules. It does not consider the risk continuity between adjacent time points, and provides a basis for subsequent adjustments.

[0075] In this embodiment, the preset safety level determination rules are first defined, and the determination dimensions are divided according to road segment risk and environmental situation. Specifically: the road segment risk dimension corresponds to risk level information, high-risk sub-road segments correspond to safety level downgrading thresholds, medium-risk sub-road segments correspond to baseline thresholds, and low-risk sub-road segments correspond to upgrading thresholds; the environmental situation dimension corresponds to the three-dimensional perception results, and the relative distance between objects and vehicles is less than the width of the safety buffer zone, and the movement speed exceeds the preset range, all of which correspond to a safety level downgrading.

[0076] Then, the risk level information and 3D perception results are broken down by time node, and the two pieces of information at each time node are compared with the preset rules one by one. If the road segment risk is high risk or there is an object intruding into the safety buffer zone, it is initially judged as high risk; if the road segment risk is medium risk and the object's movement is stable, it is initially judged as medium risk; if the road segment risk is low risk and the object is far away from the driving area, it is initially judged as low risk. Finally, the initial safety level of each time node within the preset time period is obtained.

[0077] Step 402: Based on the risk level information, the fluctuation range and duration of each sub-segment within a preset time period, combined with the speed change range, movement offset trend and motion stability of the object motion state information in the three-dimensional perception results, analyze the change logic of the initial safety level of each adjacent time node within the preset time period.

[0078] In this step, the risk level fluctuation range refers to the difference in risk level between each sub-segment within a preset time period, that is, the difference between the highest and lowest risk levels, which is used to reflect the severity of risk fluctuations in the segment.

[0079] The duration of risk level refers to the continuous time during which the same risk level is maintained across different sub-segments, reflecting the stability of risk in the segment.

[0080] The magnitude of velocity change refers to the difference in the speed of an object per unit time, which is used to reflect the degree of drastic change in the object's motion state.

[0081] The movement offset trend refers to the direction and degree of deviation of an object's trajectory from its initial trajectory, and is used to predict whether an object may intrude into the vehicle's driving area.

[0082] Motion stability refers to whether the continuous changes in the velocity and direction of an object are within a preset range; the smaller the change, the higher the stability.

[0083] The change logic refers to the reasons and correlations between the initial security levels at adjacent time points, which is used to clarify the rationality and inevitability of the level changes.

[0084] In this embodiment, the risk level information first extracts the risk level fluctuation range and duration of each sub-segment, marking periods of severe risk fluctuation or sustained high risk. Then, the velocity change range, movement offset trend, and motion stability of objects are extracted from the 3D perception results. For two adjacent time points, the change logic is analyzed based on the above parameters. If the risk level of a road segment remains high and fluctuates significantly in the previous period, and the object's speed changes drastically and it shifts towards the driving area, then the logic for the initial safety level to remain high or rise to an even higher risk is that the road segment is high-risk plus the risk of object intrusion. If the road segment risk level fluctuates slightly and remains medium-risk, and the object's movement is stable and far from the driving area, then the logic for the initial safety level to drop from medium-risk to low risk is that the road segment risk is stable plus the environmental situation is safe. By analyzing all adjacent time points one by one, the complete logic for the change of the initial safety level can be identified.

[0085] Step 403: Based on the change logic of each adjacent time node, adjust the initial security level of each time node to obtain the target security level and generate the level change trend.

[0086] In this step, the target security level refers to the final security classification result after correcting the initial security level deviation and taking into account the risk continuity between adjacent time points.

[0087] The level change trend refers to the continuous change curve and characteristic description formed by integrating the target security levels at various time points. This trend includes the level rise and fall range, fluctuation frequency, and location of abrupt change nodes.

[0088] In this embodiment, based on the change logic of each adjacent time node, the initial security level of each time node is verified and adjusted one by one. Specifically: if the initial security level of an adjacent time node changes abruptly, but the change logic shows only a brief and slight object shift and stable road segment risk, the initial security level of the abrupt node is lowered by one level to medium risk to avoid misjudgment; if the change logic shows that the road segment risk continues to escalate and the risk of object intrusion intensifies, the initial security level remains unchanged or is increased according to logic to ensure that the level matches the actual risk. After the adjustment is completed, the target security levels of each time node are integrated, arranged in chronological order according to a preset time period, and the trend of level rise and fall, fluctuation frequency, and abrupt nodes are marked to generate a complete level change trend.

[0089] The embodiments of this application realize the time-series and precise determination of safety levels; it integrates the macro-risk of road segments with the micro-environment of the vehicle's surroundings, and also takes into account the risk continuity of adjacent time nodes, avoiding the one-sidedness of determination at a single time node, and making up for the problems of the discreteness of traditional safety level determination and insufficient adaptation to dynamic risks.

[0090] Step 105: Generate management instructions for each vehicle control function based on the target safety level and the level change trend.

[0091] In this step, vehicle control functions refer to the core functions used to adjust the vehicle's driving status and ensure driving safety. These functions include four categories: vehicle speed adjustment, distance control, braking response, and steering operation limitation. Each category of vehicle control functions corresponds to 2-3 core vehicle control function parameters, and the vehicle control function parameters are dynamically adjusted according to the safety level and changing trends.

[0092] Management instructions refer to a structured set of instructions that includes integrated operation instructions and emergency operation instructions. These instructions include vehicle control function parameters, adjustment rules, and emergency triggering conditions at each time point.

[0093] In this embodiment of the application, step 105 specifically includes the following steps: Step 501: Determine the vehicle control function parameters for each time point based on the adjustable range, operation priority, and target safety level of each vehicle control function.

[0094] In this step, vehicle control function type refers to the function category divided according to the vehicle driving adjustment dimension, namely, four categories: vehicle speed adjustment, distance control, braking response, and steering operation restriction.

[0095] Adjustable range refers to the range of values ​​for parameters corresponding to each vehicle control function. It is set by vehicle hardware performance, road safety standards and scenario requirements to limit the parameter adjustment boundary and avoid exceeding the hardware's capacity.

[0096] Vehicle control function parameters refer to specific indicators used to quantify the degree of adjustment of vehicle control functions. Vehicle speed adjustment corresponds to target vehicle speed, acceleration rate, and deceleration rate; vehicle distance control corresponds to safe vehicle distance and vehicle distance maintenance accuracy; braking response corresponds to braking trigger threshold and braking force; and steering operation limitation corresponds to steering angle upper limit and steering response delay.

[0097] In this embodiment of the application, step 501 specifically includes the following steps: Step 511: Determine the adjustable range of each vehicle control function according to its type.

[0098] In this embodiment, the adjustable range is defined one by one according to the vehicle control function type. For example, one way to define it is to set the target speed range of vehicle speed adjustment to 0-60 km / h to adapt to congested and accident-prone road sections, and the speed range of acceleration rate and deceleration rate to 0.5-2 m / s squared; the safe distance range of distance control is 1-5 meters, and the distance maintenance accuracy is ±0.3 meters; the braking trigger threshold range of braking response is 0.5-2 seconds, and the braking force range is 30%-100%; the upper limit of steering angle of steering operation restriction is ±30°, and the steering response delay range is 0.1-0.3 seconds, thereby clarifying the adjustment boundaries of each parameter.

[0099] Step 512: Assign operation priority to each vehicle control function based on its impact on vehicle driving safety.

[0100] In this step, the operation priority refers to the execution order when multiple vehicle control functions are adjusted simultaneously. They are sorted from high to low according to their impact on driving safety, with core safety functions being adjusted first.

[0101] Safety impact refers to the probability and severity of safety accidents caused by abnormal vehicle control functions. Braking and distance control are directly related to collision risk and have the highest impact.

[0102] In this embodiment of the application, the operation priority is allocated according to the degree of safety impact. For example, one allocation method may be: the first priority is braking response and distance control, the second priority is vehicle speed adjustment, and the third priority is steering operation restriction, thereby ensuring that high-priority functional parameters are adjusted first and avoiding safety hazards caused by conflicts in the adjustment of multiple parameters.

[0103] Step 513: Based on the vehicle control function type, the adjustable range of each vehicle control function, and the operation priority, set multiple driving safety levels and the corresponding vehicle control function parameters for each level.

[0104] In this step, the driving safety level refers to the detailed classification of the target safety level, namely, high risk, medium risk, and low risk, which correspond one-to-one with the target safety level.

[0105] The vehicle control function parameters corresponding to each level refer to the benchmark parameters adapted to the corresponding risk level. The parameter values ​​are within the adjustable range and conform to the risk prevention and control requirements of that level.

[0106] In this embodiment of the application, three driving safety levels and corresponding parameters are first set. For example, one setting method is as follows: under the high-risk level, the braking response parameters are a trigger threshold of 0.5 seconds and a braking force of 100%, the vehicle distance control parameters are a safe vehicle distance of 5 meters and a holding accuracy of ±0.2 meters, the vehicle speed adjustment parameters are a target vehicle speed ≤20 km / h, an acceleration rate, and a deceleration rate ≤0.5 m / s squared, and the upper limit of the steering angle is ±15°. Under the medium risk level, the braking trigger threshold is set in seconds, the braking force is 70%, the safe distance is 3 meters, the target speed is 20-40 km / h, and the upper limit of the steering angle is ±20°. Under the low-risk level, the braking trigger threshold is 2 seconds, the braking force is 30%, the safe distance is 1 meter, the target speed is 40-60 km / h, and the upper limit of the steering angle is ±30°. All of the above parameters are within the preset adjustable range and are matched with the first priority function.

[0107] Step 514: Match the target safety level corresponding to each time node within the preset time period to determine the vehicle control function parameters for each time node.

[0108] In this embodiment, the target safety level for each time node within a preset time period is extracted and matched one by one to the corresponding driving safety level. For example, if the target safety level for a time node is high-risk, the braking, distance, speed, and steering parameters corresponding to the high-risk level are retrieved; if it is medium-risk or low-risk, the corresponding baseline parameters are retrieved. The matched baseline parameters are organized according to the time node, and the vehicle control functions and operation priorities corresponding to each parameter are labeled to form the vehicle control function parameters for each time node.

[0109] Step 515: Based on the risk level information and the three-dimensional perception results, adjust the vehicle control function parameters at each time point to obtain the vehicle control function parameters.

[0110] In this embodiment, the initial parameters at each time point are first fine-tuned based on the risk level information and the three-dimensional perception results. For example, one adjustment method is as follows: if the sub-segment is high-risk and there are non-motorized vehicles following closely, the vehicle distance control parameter is adjusted from 5 meters to 6 meters, and the braking trigger threshold is maintained at 0.5 seconds; if the object in the medium-risk segment is stable and there is no risk of intrusion, the vehicle speed adjustment parameter is increased from 30 km / h to 35 km / h; if the road surface of the low-risk segment is slippery, the braking force is increased from 30% to 50%. Through the above adjustments, the parameters are adapted to the real-time scenario, and the vehicle control function parameters at each time point are obtained.

[0111] Step 502: Based on the trend characteristics, fluctuation amplitude, and fluctuation duration of the level change, set the adjustment parameters for each vehicle control function parameter.

[0112] In this step, the trend characteristic refers to the direction of the rise and fall of the level change trend, which includes three categories: rising, falling, and stable.

[0113] Fluctuation amplitude refers to the difference in the level fluctuation of the target security level in the trend of level change.

[0114] Fluctuation duration refers to the duration of the same fluctuation cycle.

[0115] Adjustment parameters refer to indicators used to quantify the adjustment rate and magnitude of vehicle control function parameters. These parameters include parameter adjustment step size, adjustment interval, and dynamic setting of adaptation level change trend.

[0116] In this embodiment, adjustment parameters are set according to the trend characteristics of risk level changes. For example, one setting method is as follows: if the trend characteristic is upward, the fluctuation amplitude is ≥1, and the fluctuation duration is ≤5 minutes, the adjustment step size for the first-priority function is set to 20% of the baseline value with an adjustment interval of 10 seconds, and the adjustment step size for the second-priority function is 15% with an interval of 15 seconds; if the trend characteristic is downward, the fluctuation amplitude is 1, and the fluctuation duration is ≥10 minutes, the adjustment step size is set to 10% with an adjustment interval of 30 seconds; if the trend characteristic is stable and the fluctuation amplitude is 0, the adjustment step size is 5% with an interval of 60 seconds. For example, when the risk level decreases from high risk to medium risk, the braking force is adjusted from 100% to 70% in increments of 10% with an interval of 30 seconds, ensuring that the parameter adjustment is synchronized with the risk change.

[0117] Step 503: According to the operation priority and the adjustment parameters of each vehicle control function, adapt and adjust the vehicle control function parameters at each time point to generate operation instructions for each vehicle control function.

[0118] In this step, the operation command refers to the command for a single vehicle control function. This command includes specific parameter values ​​and adjustment rules, and only covers normal risk scenarios, excluding emergency operations.

[0119] In this embodiment, adaptation adjustments are performed in order of priority from level one to level three. Specifically: first, braking response and distance control parameters are adjusted; then, parameters at each time point are corrected according to the set adjustment step size and interval. For example, if the risk increases, the braking trigger threshold is first adjusted from 1 second to 0.8 seconds, and then the safe distance is adjusted from 3 meters to 4 meters. Next, vehicle speed adjustment parameters are adjusted, and finally, steering operation restriction parameters are adjusted. After the adjustments are completed, the parameters at each time point and the adjustment rules are integrated according to the vehicle control function type to generate operation commands for vehicle speed adjustment, distance control, braking response, and steering operation restrictions.

[0120] Step 504: Identify abrupt change nodes in the level change trend, and supplement emergency operation instructions into the operation instructions corresponding to the abrupt change nodes to form management instructions for each vehicle control function.

[0121] In this step, the mutation node refers to the time point in the trend of level change where the target security level changes across levels, which is often caused by sudden risks.

[0122] Emergency operation instructions refer to specific instructions for sudden risks at mutation nodes, corresponding to emergency parameters of vehicle control functions, and have a higher priority than regular operation instructions.

[0123] Management instructions refer to the final instructions that integrate routine operation instructions and emergency operation instructions. These instructions include adjustments to routine scenario parameters and emergency response plans for sudden scenarios.

[0124] In this embodiment, the grade change trend is first traversed to identify nodes where grades change across levels at adjacent time points, and the causes of the changes are marked. Then, for each change node, emergency operation instructions are added to the corresponding operation instructions. Specifically: if a high-risk change is triggered by a pedestrian entering, emergency braking parameters are added to the braking response instructions, and parameters for immediately reducing the speed to 0 km / h are added to the vehicle speed adjustment instructions; if a change is triggered by a sudden vehicle stop, emergency parameters for urgently increasing the distance to 8 meters and applying 100% braking force are added. The various vehicle control function instructions after adding emergency instructions are integrated to form management instructions that include both routine operations and emergency responses, ensuring that the risks of change nodes are quickly controlled.

[0125] This application's embodiments realize the dynamic and scenario-based generation of management instructions; clarify the core parameters and adjustment logic corresponding to each type of vehicle control function, solve the problem of ambiguous correspondence between vehicle control functions and parameters, and at the same time take into account both conventional and sudden risks, ensuring that instructions are adapted to the target safety level and changing trends, providing a precise basis for subsequent collaborative computing framework to execute instructions and optimize adjustments, and making up for the shortcomings of traditional vehicle control instructions lacking emergency adaptation and insufficient adaptation of parameters and safety levels.

[0126] Step 106: Execute the management command through the pre-built collaborative computing framework, and receive collaborative operation information from traffic participants and roadside equipment in the traffic congestion area and accident-prone road section through the collaborative computing framework, so as to optimize and adjust the management command to obtain the target management command.

[0127] In this step, the pre-built collaborative computing framework refers to a distributed computing framework built to meet the vehicle-road collaboration needs of traffic congestion areas and accident-prone road sections. It integrates communication interaction, instruction distribution, data processing, and instruction optimization functions, and adapts to the multi-source data interaction and instruction collaborative execution needs of traffic participants and roadside equipment.

[0128] Collaborative operation information refers to a collection of data, including the operating status of other traffic participants and roadside equipment within traffic congestion areas and accident-prone road sections, as well as the dynamic traffic conditions of the road section.

[0129] Target management instructions refer to the final vehicle control management instructions adapted to vehicle-road cooperative scenarios after collaborative operation information analysis and conflict optimization processing, taking into account both the vehicle's own risk prevention and control and the operational coordination of surrounding traffic participants.

[0130] In this embodiment of the application, step 106 specifically includes the following steps: Step 601: Establish information exchange channels among traffic participants in traffic congestion areas and accident-prone road sections through a pre-built collaborative computing framework.

[0131] In this step, the information exchange channel refers to the communication link built on the collaborative computing framework, which enables bidirectional data transmission between traffic participants and roadside equipment in traffic congestion areas and accident-prone road sections. This channel includes point-to-point direct connection channels and regional broadcast channels to ensure the real-time performance and coverage of data transmission.

[0132] In this embodiment, a pre-built collaborative computing framework equipped with a vehicle-road cooperative communication protocol is used to allocate dedicated communication frequency bands for traffic congestion areas and accident-prone road sections. Specifically: First, a main interaction channel is established between the vehicle and roadside equipment. Then, using the roadside equipment as relay nodes, a point-to-point direct connection channel is established between the vehicle and other traffic participants such as motor vehicles and non-motor vehicles in the area. At the same time, a regional broadcast channel is opened to receive dynamic traffic information across the entire area released by the roadside equipment. A data transmission verification mechanism is set for all interaction channels to ensure that the information transmitted through the channels is without delay or loss. Finally, an information interaction channel is established between various traffic participants in traffic congestion areas and accident-prone road sections.

[0133] It should be noted that this embodiment does not limit the structure used in the collaborative computing framework, and can be configured accordingly based on the actual situation.

[0134] Step 602: Distribute the management instructions of each vehicle control function to the corresponding vehicle control module through the collaborative computing framework, so as to drive each vehicle control module to perform vehicle control operations according to the corresponding management instructions, and receive collaborative operation information fed back by each traffic participant and roadside equipment through the information interaction channel.

[0135] In this step, the vehicle control module refers to the hardware control unit on the vehicle that corresponds one-to-one with each vehicle control function. This module includes the vehicle speed adjustment module, the distance control module, the braking response module, and the steering operation restriction module.

[0136] Vehicle control operation refers to the actions taken by the vehicle control module to adjust the vehicle's driving status in real time according to the parameter values ​​in the management instructions. It is the hardware execution link for implementing management instructions.

[0137] In this embodiment, the collaborative computing framework first identifies the type of each vehicle control function in the management command, and distributes the vehicle speed adjustment command to the vehicle speed adjustment module and the braking response command to the braking response module according to the one-to-one correspondence rule. The other vehicle control function commands are distributed in the same way. After receiving the command, each vehicle control module performs the vehicle control operation according to the parameter value and execution time point in the command. For example, the braking response module performs braking adjustment according to the braking trigger threshold and braking force, and the vehicle speed adjustment module performs vehicle speed control according to the target vehicle speed. Meanwhile, the collaborative computing framework receives real-time feedback from other traffic participants in the area regarding their own vehicle control operation status and changes in driving parameters, as well as data such as changes in road traffic flow and sudden risk warnings from roadside equipment, through information exchange channels, and integrates all received feedback data into structured collaborative operation information.

[0138] Step 603: Compare and analyze the target information in the collaborative operation information with the target security level and level change trend at the corresponding time node to determine the degree of influence of each collaborative operation information on each management instruction; In this step, the target information refers to the core data extracted from the collaborative operation information that is directly related to the vehicle's control operation and driving safety. This information eliminates redundant and invalid data and includes other traffic participants' emergency braking and lane changing operations, roadside equipment's information on sudden road risks and traffic control, etc.

[0139] The degree of impact refers to the degree of interference or assistance that each piece of target information provides to the normal execution of the vehicle management instructions. It can be divided into three levels: high, medium, and low, according to the magnitude of the impact.

[0140] In this embodiment, the collaborative operation information is first filtered to extract target information directly related to the vehicle's driving safety, while redundant data such as traffic participants driving at a constant speed and risk-free road segment broadcasts from roadside equipment are removed. Then, the extracted target information is compared and analyzed one by one with the target safety level and its changing trend at the corresponding time points. Specifically: If the target information is other vehicles suddenly braking, roadside equipment issuing a pedestrian intrusion warning, and the target safety level at the corresponding time point is high risk with an upward trend, then the impact of this information on management instructions is determined to be high; if the target information is other vehicles routinely changing lanes, roadside equipment issuing a slight change in traffic flow on the road section, and the target safety level at the corresponding time point is medium risk with a stable trend, then the impact is determined to be medium; if the target information is traffic participants driving at a constant speed, roadside equipment issuing no risk warnings for the road section, and the target safety level at the corresponding time point is low risk, then the impact is determined to be low.

[0141] Step 604: Based on the degree of impact and the operation priority of the vehicle control function, perform conflict optimization processing on the vehicle control function parameters and execution time points in each management instruction to form the target management instruction.

[0142] In this step, the target management instruction refers to the final vehicle control management instruction after conflict optimization, which takes into account its own risk prevention and control, vehicle control operation execution logic and surrounding traffic coordination situation. It is the final execution basis for each vehicle control module.

[0143] In this embodiment, the management instructions are first optimized based on the dual rules of the degree of impact from high to low and the priority of vehicle control function operation from level one to level three. Specifically, the vehicle control function instructions corresponding to the target information with a high degree of impact are processed first, and the braking response and distance control parameters of level one priority are adjusted first. Then, the level two and level three priority parameters are adjusted. For example, for high impact information of other vehicles braking suddenly, the braking trigger threshold is further shortened from 0.5 seconds to 0.3 seconds and the safe distance is adjusted from 5 meters to 7 meters. At the same time, the execution time of the relevant parameters is advanced by 1 second. For target information with a medium impact level, only minor adjustments are made to the corresponding vehicle control function parameters. For example, for information about other vehicles' routine lane changes, the upper limit of the steering angle is slightly adjusted from ±15° to ±10°, while the execution time remains unchanged. For target information with a low impact level, the parameters and execution time of the management command are not adjusted, and the original command content is maintained. All optimization operations are performed within the adjustable range of each vehicle control function. After optimization, all adjusted vehicle control function commands are integrated, and the parameter values, execution time, and emergency trigger conditions of each command are labeled to form a target management command, which drives each vehicle control module to execute vehicle control operations according to the command.

[0144] This application's embodiments address the problem that traditional vehicle control commands only consider the vehicle's own risks and lack coordination with the operations of surrounding traffic participants. Through vehicle-road data interaction and dynamic command optimization, it effectively avoids safety hazards caused by multi-vehicle operation conflicts, improves the safety and coordination of vehicle driving in traffic congestion areas and accident-prone road sections, and makes the execution of vehicle control commands more in line with actual traffic management scenarios.

[0145] Figure 3 This is a schematic diagram of a specific implementation of the vehicle control function dynamic management system based on driving scenario safety levels provided in this application embodiment, with reference to... Figure 3 The system may include: The acquisition module 31 is used to acquire historical traffic data, real-time traffic data, and environmental data of traffic congestion areas and accident-prone road sections; The first generation module 32 is used to generate risk level information based on the historical traffic data and real-time traffic data, using a pre-built traffic data prediction model. The second generation module 33 is used to generate a three-dimensional perception result of the vehicle's surrounding environment based on the environmental data and using a pre-built three-dimensional reconstruction model. The determination module 34 is used to determine the target safety level and level change trend of the vehicle within a preset time period based on the risk level information and the three-dimensional perception results. The third generation module 35 is used to generate management instructions for each vehicle control function based on the target safety level and the level change trend. The adjustment module 36 is used to execute the management instructions through a pre-built collaborative computing framework, and to receive collaborative operation information from traffic participants and roadside equipment in the traffic congestion area and accident-prone road section through the collaborative computing framework, so as to optimize and adjust the management instructions to obtain the target management instructions.

[0146] The vehicle control function dynamic management system based on driving scenario safety level in this application embodiment is used to implement the aforementioned vehicle control function dynamic management method based on driving scenario safety level. Therefore, the specific implementation of the vehicle control function dynamic management system based on driving scenario safety level can be found in the embodiment section of the vehicle control function dynamic management method based on driving scenario safety level above. The specific implementation can be referred to the description of the corresponding embodiments, which will not be repeated here.

[0147] This application also provides a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are to be invoked and executed by the processing component to implement the steps of the above-described method for dynamic management of vehicle control functions based on driving scenario safety levels.

[0148] This application also provides a computer storage medium storing a computer program, which, when executed by a computer, implements the steps of any of the above-described methods for dynamic management of vehicle control functions based on driving scenario safety levels.

[0149] In one exemplary embodiment, the computer storage medium may include, but is not limited to, various media capable of storing computer programs, such as USB flash drives, read-only memory, random access memory, portable hard drives, magnetic disks, or optical disks.

[0150] The embodiments of this application also provide a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the steps in any of the embodiments of the vehicle control function dynamic management method based on driving scenario safety level.

[0151] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0152] The foregoing has provided a detailed description of the vehicle control function dynamic management method and system based on driving scenario safety levels provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are merely for the purpose of helping to understand the method and its core ideas. It should be noted that those skilled in the art can make various improvements and modifications to this application without departing from its principles, and these improvements and modifications also fall within the protection scope of this application.

Claims

1. A method for dynamic management of vehicle control functions based on driving scenario safety levels, characterized in that, include: Obtain historical traffic data, real-time traffic data, and environmental data for traffic congestion areas and accident-prone road sections; Based on the historical and real-time traffic data, risk level information is generated using a pre-built traffic data prediction model. Based on the environmental data, a three-dimensional perception result of the vehicle's surrounding environment is generated using a pre-built three-dimensional reconstruction model. Based on the risk level information and the three-dimensional perception results, the target safety level and level change trend of the vehicle within a preset time period are determined. Based on the target safety level and the trend of level change, management instructions for each vehicle control function are generated. The management instructions are executed through a pre-built collaborative computing framework, and collaborative operation information from traffic participants and roadside equipment in the traffic congestion area and accident-prone road sections is received through the collaborative computing framework to optimize and adjust the management instructions and obtain the target management instructions.

2. The method according to claim 1, characterized in that, Based on the historical and real-time traffic data, risk level information is generated using a pre-built traffic data prediction model, including: Based on the road segmentation standards for traffic congestion areas and accident-prone road sections, and the time cycle of urban commuting, the historical traffic data and real-time traffic data are processed by road segment matching and time period segmentation to obtain historical traffic sub-data and real-time traffic sub-data for each sub-road segment in different time periods. Based on historical and real-time traffic data, the traffic operation status of each sub-road segment is inferred through a pre-built traffic data prediction model, and the traffic trend information of each sub-road segment within a preset time period is obtained. Risk assessment is performed on the traffic trend information of each sub-road segment to obtain the risk level information of each sub-road segment within a preset time period.

3. The method according to claim 1, characterized in that, Based on the environmental data, a pre-built 3D reconstruction model is used to generate 3D perception results of the vehicle's surrounding environment, including: Based on the spatial orientation around the vehicle, the environmental data is divided into different data categories. Based on each classification data group, the spatial positional relationship, relative distance and object type of each object around the vehicle are determined by a pre-built 3D reconstruction model. Based on the spatial positional relationship, relative distance and object type, the spatial area around the vehicle is divided into the vehicle driving area, the surrounding object area and the safety buffer area, and the motion state information of each object in each area is obtained. By integrating the spatial positional relationships, relative distances, object types, and motion state information, a three-dimensional perception result of the vehicle's surrounding environment is generated.

4. The method according to claim 1, characterized in that, Based on the risk level information and the three-dimensional perception results, the target safety level and level change trend of the vehicle within a preset time period are determined, including: The risk level information and the information in the three-dimensional perception results are compared with the preset safety level determination rules to determine the initial safety level of the vehicle at each time node within the preset time period. Based on the risk level information, the fluctuation range and duration of each sub-segment within a preset time period, combined with the speed change range, movement offset trend and motion stability of the object motion state information in the three-dimensional perception results, the logic of the change of the initial safety level of each adjacent time node within the preset time period is analyzed. Based on the change logic of each adjacent time node, the initial security level of each time node is adjusted to obtain the target security level and generate the level change trend.

5. The method according to claim 1, characterized in that, Based on the target safety level and its changing trend, management instructions for each vehicle control function are generated, including: Based on the adjustable range, operation priority, and target safety level of each vehicle control function, determine the vehicle control function parameters at each time point; Based on the trend characteristics, fluctuation amplitude, and fluctuation duration of the aforementioned level changes, the adjustment parameters for each vehicle control function are set. Based on the operation priority and the adjustment parameters of each vehicle control function, the vehicle control function parameters at each time point are adapted and adjusted to generate operation instructions for each vehicle control function. Identify abrupt change nodes in the trend of the level change, and supplement emergency operation instructions into the operation instructions corresponding to the abrupt change nodes to form management instructions for each vehicle control function.

6. The method according to claim 5, characterized in that, Based on the adjustable range, operational priority, and target safety level of each vehicle control function, the vehicle control function parameters for each time point are determined, including: Determine the adjustable range of each vehicle control function based on its type; Based on the degree of impact of each vehicle control function on vehicle driving safety, an operation priority is assigned to each vehicle control function. Based on the vehicle control function type, the adjustable range of each vehicle control function and the operation priority, multiple driving safety levels and corresponding vehicle control function parameters are set for each level. Match the target safety level corresponding to each time node within a preset time period to determine the vehicle control function parameters for each time node; Based on the risk level information and the three-dimensional perception results, the vehicle control function parameters at each time point are adjusted to obtain the vehicle control function parameters.

7. The method according to claim 1, characterized in that, The management instructions are executed through a pre-built collaborative computing framework, and collaborative operation information from various traffic participants and roadside equipment in the traffic congestion area and accident-prone road sections is received through the collaborative computing framework to optimize and adjust the management instructions, resulting in target management instructions, including: Establish information exchange channels among traffic participants in traffic congestion areas and accident-prone road sections by using a pre-built collaborative computing framework. The collaborative computing framework distributes management instructions for each vehicle control function to the corresponding vehicle control module, thereby driving each vehicle control module to perform vehicle control operations according to the corresponding management instructions, and receives collaborative operation information from traffic participants and roadside equipment through the information interaction channel. The target information in the collaborative operation information is compared and analyzed with the target security level and level change trend at the corresponding time node to determine the degree of influence of each collaborative operation information on each management instruction. Based on the degree of impact and the operational priority of the vehicle control function, conflict optimization is performed on the vehicle control function parameters and execution time points in each management instruction to form the target management instruction.

8. A dynamic management system for vehicle control functions based on driving scenario safety levels, characterized in that: include: The acquisition module is used to acquire historical traffic data, real-time traffic data, and environmental data for traffic congestion areas and accident-prone road sections. The first generation module is used to generate risk level information based on the historical traffic data and real-time traffic data, using a pre-built traffic data prediction model. The second generation module is used to generate a three-dimensional perception result of the vehicle's surrounding environment based on the environmental data and using a pre-built three-dimensional reconstruction model. The determination module is used to determine the target safety level and level change trend of the vehicle within a preset time period based on the risk level information and the three-dimensional perception results. The third generation module is used to generate management instructions for each vehicle control function based on the target safety level and the level change trend. The adjustment module is used to execute the management instructions through a pre-built collaborative computing framework, and to receive collaborative operation information from traffic participants and roadside equipment in the traffic congestion area and accident-prone road section through the collaborative computing framework, so as to optimize and adjust the management instructions to obtain the target management instructions.

9. A computing device, characterized in that, It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are invoked and executed by the processing component to implement the vehicle control function dynamic management method based on driving scenario safety level as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that, The system contains a computer program that, when executed by a computer, implements the dynamic management method for vehicle control functions based on driving scenario safety levels as described in any one of claims 1 to 7.

Images