Intelligent driving vehicle driving control methods, systems, equipment and media

Abstract

This invention discloses a method, system, device, and medium for controlling the driving of intelligent vehicles. The method includes: real-time acquisition of vehicle driving information and network quality information of base stations near the vehicle's location; prediction of vehicle network connection failures within a preset future time period using a pre-trained time series prediction model based on the vehicle driving information, the network quality information, and the vehicle's historical driving status along similar historical driving paths, to obtain the probability of vehicle network connection failures; determination of the vehicle's driving risk level based on the vehicle network connection failures, their probabilities, and the vehicle's driving task; selection and generation of an optimal driving strategy that includes network resource scheduling and vehicle behavior guidance using a pre-trained machine learning model based on the vehicle network connection failures, the vehicle's driving task, and the driving risk level; and control the corresponding vehicles to execute the optimal driving strategy according to the priority order of the vehicle's driving services.

Description

Technical Field

[0001] This invention relates to the field of intelligent driving technology, and more specifically to an intelligent driving vehicle driving control method, system, device and medium. Background Technology

[0002] New energy and intelligent driving are two major trends in the automotive industry. As a key area for high-tech development, major automakers have begun to invest in the field of intelligent driving. Currently, Level 4 intelligent driving systems are becoming increasingly common in passenger vehicles, enabling vehicles to complete all driving tasks within a specific Design Operating Domain (ODD) without human driver intervention.

[0003] Level 4 autonomous driving relies on continuous and reliable data exchange between the vehicle and the cloud platform, roadside facilities, and other vehicles. However, current cellular network (such as 4G / 5G) management mainly adopts passive or reactive fault handling mechanisms. When vehicle driving is interrupted, data transmission rate drops sharply, or connection is interrupted, existing vehicle driving control methods usually trigger alarms based on preset static network signal thresholds (such as signal strength below -110dBm). After the alarm is generated, a predefined and fixed recovery process is usually executed, such as attempting to reconnect the vehicle to the nearest base station with the strongest signal, or having maintenance personnel intervene to analyze the situation. However, because the aforementioned driving control methods only respond after a fault occurs, there is a delay in the processing. For Level 4 autonomous driving, even a communication interruption of hundreds of milliseconds during critical operations (such as lane changes on highways or unprotected left turns at intersections) can lead to decision-making errors and serious safety accidents. Furthermore, existing vehicle driving control methods adopt a one-size-fits-all recovery strategy for all vehicles that trigger alarms, without considering the vehicle's current driving task and business priority. For example, existing control uses the same network recovery strategy for a vehicle performing high-precision positioning data differential calculation and a vehicle that is only uploading logs, resulting in the failure to prioritize critical business operations. This fails to meet the extreme reliability and continuity requirements of Level 4 autonomous driving for network connectivity, thus posing a safety hazard. Summary of the Invention

[0004] This invention provides a method, system, device, and medium for intelligent driving vehicle driving control, in order to solve the technical problem that the response lag and recovery non-adaptability of existing vehicle driving control cannot meet the extreme reliability and continuity requirements of L4 autonomous driving network connectivity, thus causing safety hazards.

[0005] Firstly, a method for controlling the driving of an intelligent driving vehicle is provided, including:

[0006] Real-time collection of vehicle driving information and network quality information of base stations near the vehicle's location;

[0007] Based on the vehicle driving information, the network quality information, and the vehicle's historical driving status on similar historical driving paths, a pre-trained time series prediction model is used to predict vehicle network connection failures within a preset time period in the future, thereby obtaining the probability of vehicle network connection failures.

[0008] The vehicle's driving risk level is determined based on the vehicle network connection failure and its probability, as well as the vehicle's driving task.

[0009] Based on the vehicle network connection failure, the vehicle's driving task, and the driving risk level, a pre-trained machine learning model is used to select and generate an optimal driving strategy that includes network resource scheduling and vehicle behavior guidance. The corresponding vehicles are then controlled to execute the optimal driving strategy according to the priority order of the vehicle's driving services.

[0010] Secondly, an intelligent driving vehicle driving control system is provided, including a unit for executing the above-described intelligent driving vehicle driving control method.

[0011] Thirdly, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described intelligent driving vehicle driving control method.

[0012] Fourthly, a computer-readable storage medium is provided, which stores a computer program that, when executed by a processor, implements the steps of the above-described intelligent driving vehicle driving control method.

[0013] Compared to existing technologies, this invention integrates real-time driving information from the vehicle and network quality information from the network end. It uses a time-series prediction model to probabilistically predict network connection failures within a preset timeframe. Unlike existing technologies that rely solely on static thresholds for post-event judgment, this invention can predict vehicle network connection failures, enabling intelligent driving vehicles to shift from passive response to active defense, facilitating subsequent avoidance measures. After predicting a failure, it can also determine the vehicle's driving risk level based on the network connection failure, its probability, and the vehicle's driving task to assess the impact of the network connection failure on the driving task. Finally, based on the vehicle network connection failure, the vehicle's driving task, and the driving risk level, pre-trained data is used... The machine learning model selects and generates the optimal driving strategy, which includes network resource scheduling and vehicle behavior guidance. Based on the priority order of the vehicle's driving services, it controls the corresponding vehicles to execute the optimal driving strategy. This ensures that among all vehicles predicted to fail, those performing the most critical driving services receive the highest priority network protection, achieving adaptive, differentiated, and refined fault recovery. This avoids the inefficiency and resource waste of traditional "one-size-fits-all" strategies. Therefore, the intelligent driving vehicle driving control method of this invention can accurately predict and proactively intervene before network connection failures occur, avoiding a potential network connection interruption, ensuring the continuity and reliability of L4 vehicle autonomous driving, and ensuring that vehicles performing the most critical driving services receive the highest priority network protection. Attached Figure Description

[0014] Figure 1 This is a flowchart illustrating the intelligent driving vehicle driving control method provided in the first embodiment of the present invention;

[0015] Figure 2 for Figure 1 A schematic diagram of a specific implementation of step S120;

[0016] Figure 3 for Figure 1 A schematic diagram of a specific implementation of step S130;

[0017] Figure 4 For Figure 1 A schematic diagram of a specific implementation of step S140;

[0018] Figure 5 This is a flowchart illustrating the intelligent driving vehicle driving control method provided in the second embodiment of the present invention;

[0019] Figure 6 This is a schematic diagram of the intelligent driving vehicle driving control system provided in the first embodiment of the present invention;

[0020] Figure 7This is a schematic diagram of the intelligent driving vehicle driving control system provided in the second embodiment of the present invention;

[0021] Figure 8 This is a schematic diagram of the structure of a computer device according to an embodiment of the present invention. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] It should be understood that, when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0024] It should also be understood that the term “and / or” as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0025] Please see Figure 1 , Figure 1 This is a schematic flowchart illustrating the intelligent driving vehicle driving control method provided in the first embodiment of the present invention. In the embodiment shown in the figures, the intelligent driving vehicle driving control method includes the following steps S110-140:

[0026] S110: Real-time collection of vehicle driving information and network quality information of base stations near the vehicle's location.

[0027] In this invention, the vehicle driving information may include the vehicle's real-time location, speed, acceleration, planned driving path, and driving service type (such as high-precision positioning, environmental perception data upload, etc.); the network quality information of the base stations near the vehicle's location may include the network quality information of the base station at the vehicle's real-time location and neighboring base stations. The network quality information may include the base station's radio link quality indicators (RSRP (Reference Signal Receiving Power), RSRQ (Reference Signal Receiving Quality), SINR (Signal to Interference plus Noise Ratio), cell load, uplink / downlink rates, etc.

[0028] S120. Based on the vehicle driving information, the network quality information, and the vehicle's historical driving status on similar historical driving paths, a pre-trained time series prediction model is used to predict vehicle network connection failures within a preset future time period, thereby obtaining the probability of vehicle network connection failures.

[0029] In this step, the time series prediction model uses the acquired network quality information and vehicle driving information as input, and combines this with the vehicle's historical driving status along similar historical driving paths to continuously predict the probability of vehicle network connection failure within a future period (e.g., 5-30 seconds). Vehicle network connection failures can include network connection / switching failures, and historical driving status can include whether the vehicle network connection has dropped.

[0030] like Figure 2 As shown, step S120 may specifically include steps S121-S122:

[0031] S121. Using an LSTM-based time series prediction model, continuously predict the network quality of the base station network used by the vehicle within a preset time period based on the vehicle's driving information, including the vehicle's driving direction, speed, and driving path, as well as the network quality information of base stations near the vehicle's location.

[0032] In this embodiment, a pre-trained time series prediction model based on LSTM (Long Short-Term Memory) is used to probabilistically predict vehicle network connection failures within a preset time period.

[0033] In this step, the area the vehicle will reach within a preset time period can be continuously predicted based on the vehicle's driving direction, speed, and driving path in the vehicle's driving information. The network quality of the base station network used by the vehicle within the preset time period can also be predicted based on the network quality information of the base station and neighboring base stations at the vehicle's real-time location.

[0034] S122. Based on the network quality and the vehicle's disconnection records in historical driving states of similar historical driving paths, predict the probability of vehicle network connection failure within a preset time period in the future.

[0035] In this step, the probability of a vehicle network connection failure within the future preset time period is predicted by combining the vehicle's disconnection records on similar historical driving routes. If there have been multiple instances of intelligent driving vehicles disconnecting in the area they reach within the predicted future preset time period, the probability of a vehicle network connection failure is relatively high.

[0036] Based on steps S121-S122 above, the LSTM-based time series prediction model can be pre-trained using a fault prediction dataset containing past vehicle network connection failures, corresponding vehicle driving information, network quality information of base stations near the vehicle's location, and historical driving states of the vehicle along similar historical driving paths. That is, the LSTM-based time series prediction model can also combine historical driving states of vehicles along similar historical driving paths to further improve the accuracy of the model's predictions. For example, based on the speed and direction of a vehicle (vehicle A), the high user load of the target cell B it arrives at, and historical dropout records along similar paths, the LSTM-based time series prediction model predicts that vehicle A will experience a network handover failure (i.e., the probability of a vehicle network connection failure) exceeding 95% after 10 seconds due to entering the weak coverage edge of a neighboring cell (cell B).

[0037] S130. Determine the vehicle's driving risk level based on the vehicle network connection failure and its probability, and the vehicle's driving task.

[0038] Specifically, such as Figure 3 As shown, step S130 may include steps S131-S132:

[0039] S131. If the probability of the vehicle network connection failure is greater than a preset threshold, the root cause of the vehicle network connection failure shall be determined by a root cause diagnosis method.

[0040] In this step, when the probability of a vehicle network connection failure is greater than a preset threshold (e.g., 60%), it indicates that the failure will occur within a preset time period in the future. In this invention, a root cause diagnosis method is also used to determine the root cause of the vehicle network connection failure. For example, in the above example of handover failure, this step can confirm that the root cause of the failure is "handover failure caused by target cell congestion".

[0041] Preferably, this step can also be performed using a root cause diagnosis method based on an LSTM-based time series prediction model to determine the root cause of the fault.

[0042] S132. Based on the root cause of the vehicle network connection failure and the vehicle's driving task, assess the impact of the vehicle network connection failure on the vehicle's driving task, thereby determining the vehicle's driving risk level.

[0043] In this step, the impact of the vehicle network connectivity failure on the vehicle's driving tasks (such as lane changing, cruising, and parking) can be assessed based on a rule engine or a deep learning model used for risk level determination, according to the root cause of the vehicle network connectivity failure and the dependence level of the network function on the driving tasks that the vehicle is currently or about to perform (such as lane changing, cruising, and parking), thereby determining the driving risk level (e.g., low, medium, high, and very high).

[0044] Understandably, a knowledge base or rule base can be predefined to define the level of dependence on network functions (e.g., the level of network latency requirements) for different driving tasks. For example, if the driving task is "following the car in front for cruise control", the network latency requirement is medium. Corresponding to the example of failure to enter the neighboring cell after 10 seconds, the rule engine or the deep learning model used for risk level determination can assess the driving risk level of this potential failure as medium.

[0045] S140. Based on the vehicle network connection failure, the vehicle's driving task, and the driving risk level, a pre-trained machine learning model is used to select and generate an optimal driving strategy that includes network resource scheduling and vehicle behavior guidance. The corresponding vehicles are then controlled to execute the optimal driving strategy according to the priority order of the vehicle's driving services.

[0046] In this invention, the machine learning model can be a reinforcement learning-based model. After generating the driving strategy, it can also select and control the corresponding vehicles to execute the optimal driving strategy according to the priority order determined by the driving service types (such as high-precision positioning, environmental perception data uploading, etc.) of all vehicles predicted to fail within a preset time in the future. This adaptively determines the control execution order, which can improve the utilization of network resources and ensure that the most critical driving services receive the highest priority network protection and are the first to recover from the fault. For example, if the priority of the driving service type performing high-precision positioning is higher than that of the driving service type performing environmental perception data uploading, then the vehicles performing the driving service type performing high-precision positioning are given priority in executing the optimal driving strategy for fault recovery.

[0047] In this embodiment, as Figure 4 As shown, step S140, which selects and generates an optimal driving strategy that includes network resource scheduling and vehicle behavior guidance based on the vehicle network connection failure, the vehicle's driving task, and the driving risk level using a pre-trained machine learning model, may specifically include steps S141-S142:

[0048] S141. Using a pre-trained machine learning model, generate multiple alternative driving strategies and corresponding reward functions that include network resource scheduling and vehicle behavior guidance based on the vehicle network connection failure, the vehicle's driving task, and driving risk level.

[0049] In this invention, the pre-trained machine learning model can achieve network resource scheduling from aspects such as network rerouting, network resource redistribution, and / or primary / backup switching based on vehicle network connection failures. It generates alternative driving strategies that include network resource scheduling and vehicle behavior guidance by combining the current or upcoming driving task and driving risk level. The reward function includes a reward function determined by the degree of impact of executing the alternative driving strategy on the current driving task. The lower the degree of impact on the current driving task, the higher the reward function value.

[0050] In this embodiment, the generated alternative driving strategies may include: not interfering with the current vehicle driving strategy; or, forcing the vehicle to switch signal connection paths in advance to connect to a neighboring base station with a stable signal; or, reducing the vehicle speed and delaying entry into the area where the vehicle network connection failure occurs. For example, corresponding to the example of entering the neighboring cell after 10 seconds and failing to switch, if the strategy of not interfering is executed, the vehicle will lose connection and the driving level will be degraded. The reward function value of the alternative driving strategy generated by the model is approximately -100. If the strategy of forcing the vehicle to switch signal connection paths in advance to connect to a non-optimal but stable neighboring base station is executed, since only the channel connection is switched, the vehicle speed and direction remain unchanged, so the network is not interrupted. Although the bandwidth is slightly lower, it does not affect the current driving task. The reward function value is approximately +80. If the strategy of reducing the vehicle speed and delaying entry into the area where the vehicle network connection failure occurs is executed, it may affect the smoothness of the driving task. The reward function value is approximately +20.

[0051] S142. Select the alternative driving strategy with the largest reward function value as the optimal driving strategy.

[0052] In this embodiment, since the lower the impact on the current driving task, the higher the reward function value, the alternative driving strategy with the largest reward function value is selected as the optimal driving strategy. This avoids network connection failures while ensuring the smooth completion of the driving task, thereby improving the overall driving safety and continuity of intelligent driving.

[0053] Understandably, the intelligent driving vehicle driving control method of the present invention can be applied to a server. Multiple vehicles can communicate with the server through the 5G core network to exchange data. Corresponding to the example of failure to enter the neighbor cell handover after 10 seconds, if the strategy of forcing vehicle (vehicle A) to switch the signal connection path in advance and connect to a non-optimal but signal-stable neighbor cell base station (such as the base station of cell C next to cell B) is selected, the specific implementation steps are as follows:

[0054] The server uses Netconf (an interface protocol that guarantees the atomicity of transaction operations) to establish a secure connection channel via the SSH protocol, and connects to the designated port of AMF (Access and Mobility Management Function) to exchange policy commands (forced handover commands), including the SUPI of vehicle A and the physical cell identifier of the target cell C.

[0055] AMF calls the Nudm_SDM (User Data Management Service) service to obtain the user's (vehicle A's) subscription data from the UDM (Unified Data Management Function) and verifies the slice access rights of the target cell C (whether S-NSSAI matches, etc.);

[0056] AMF sends a HandoverRequest (cell handover control signaling) message to the source base station (the base station at the current location of vehicle A) via NG-AP (signaling protocol between the core network and the base station), carrying the physical cell identifier, QoS flow parameters and handover reason of the target cell C. The source base station sends a HandoverRequest to the target base station (the base station of the target cell C) through the Xn interface (interface between base stations) to request resource allocation (beamforming configuration of the target cell, etc.).

[0057] The source base station sends a handover command to vehicle A via an RRC reconfiguration message. The handover command includes parameters such as the SSB frequency point (synchronization signal block, used for vehicle-cell interface) and random access configuration (physical random access channel resource index) of the target cell C. Vehicle A performs non-contention-based random access. After completing access in the target cell C, it sends an RRC Reconfiguration Complete confirmation. The source base station then sends a Path Switch Request (path switching signaling message) to the AMF via NG-AP to complete the signal connection path handover.

[0058] As can be seen, in the above scheme, the intelligent driving vehicle driving control method of the present invention integrates real-time driving information from the vehicle and network quality information from the network end, and uses a time series prediction model to probabilistically predict vehicle network connection failures within a preset time period. Compared with the existing technology that only relies on static thresholds on the network side for ex-post judgment, the present invention can predict network connection failures, enabling intelligent driving vehicles to shift from passive response to active defense, which is beneficial for taking subsequent avoidance measures. After predicting the failure, the driving risk level of the vehicle can be determined based on the vehicle network connection failure and its probability and the vehicle's driving task, so as to assess the impact of the vehicle network connection failure on the vehicle's driving task. Then, based on the vehicle network connection failure, The vehicle's driving task and driving risk level are selected using a pre-trained machine learning model to generate an optimal driving strategy that includes network resource scheduling and vehicle behavior guidance. The optimal driving strategy is then executed by the corresponding vehicles according to the priority order of the vehicle's driving business. This ensures that the vehicle performing the most critical driving business among all vehicles predicted to have a fault receives the highest priority network protection, achieving adaptive, differentiated, and refined fault recovery. It can be seen that the intelligent driving vehicle driving control method of the present invention can accurately predict and actively intervene before network connection failure occurs, avoiding a potential network connection interruption. Without affecting driving safety, it seamlessly resolves potential network connection failures, ensuring the continuity and reliability of L4 vehicle autonomous driving.

[0059] It should be noted that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0060] Reference Figure 5 , Figure 5 This is a flowchart illustrating the intelligent driving vehicle driving control method provided in the second embodiment of the present invention. As shown in the figure, the method includes the following steps S200-S290:

[0061] S200 collects real-time vehicle driving information and network quality information of base stations near the vehicle's location.

[0062] Initially, vehicle driving information and network quality information of base stations near the vehicle's location are acquired in real time. Based on the information acquired in real time, vehicle network connection failures are predicted within a future preset time period or hardware failures are detected at key intersections along the driving path. Steps S210 and S250 are executed respectively. Since step S200 is the same as or similar to step S110, it will not be described again here.

[0063] S210. Based on the vehicle driving information, the network quality information, and the vehicle's historical driving status on similar historical driving paths, a pre-trained time series prediction model is used to predict vehicle network connection failures within a preset future time period, thereby obtaining the probability of vehicle network connection failures.

[0064] This step is the same as or similar to step S120, and will not be described again here.

[0065] S220. Determine the vehicle's driving risk level based on the vehicle network connection failure and its probability, and the vehicle's driving task.

[0066] This step is the same as or similar to step S130, and will not be repeated here.

[0067] S230. Based on the vehicle network connection failure, the vehicle's driving task, and the driving risk level, a pre-trained machine learning model is used to select and generate an optimal driving strategy that includes network resource scheduling and vehicle behavior guidance. The corresponding vehicles are then controlled to execute the optimal driving strategy according to the priority order of the vehicle's driving services.

[0068] This step is the same as or similar to step S140, and will not be described again here.

[0069] S240. Continuously monitor the network connection status of the vehicle to obtain information about the stability of the vehicle's network connection.

[0070] In this step, after the vehicle executes the optimal driving strategy, the network connection status of the vehicle is continuously monitored to ensure the stability of the network connection after execution, thus guaranteeing the reliability and continuity of vehicle driving. It can also verify whether the executed optimal driving strategy is effective.

[0071] S250. Detect whether the RSUs at key intersections in the driving path are offline based on the driving path in the vehicle driving information.

[0072] In this invention, the vehicle obtains traffic light information and pedestrian warnings at the intersection ahead through the RSU (Roadside Unit) used for V2I communication. If the RSU fails, it may lead to a driving accident.

[0073] In this embodiment, the heartbeat detection mechanism can be used to detect whether the RSU at key intersections in the vehicle's travel path is offline. If it is offline, a fault alarm can be triggered immediately. The key intersection can be a crossroads or an intersection with high pedestrian traffic.

[0074] S260. If an RSU is offline, identify the service range of the RSU, search for vehicles that are about to enter the service range based on the service range of the RSU and the vehicle driving information, and determine that the driving risk level of the vehicles that are about to enter the service range is high risk.

[0075] Because the RSU is offline, intelligent driving vehicles are prone to running red lights or colliding due to the lack of RSU information at key intersections, which may lead to major traffic accidents. Therefore, the driving risk level is directly judged as high risk.

[0076] S270. Based on the RSU offline status and the vehicle's driving risk level, generate a driving strategy that includes source switching and vehicle speed reduction, so as to send the traffic light information of the key intersection released by the city traffic management center to the corresponding vehicles through the cellular network, control the vehicles to reduce the maximum speed limit, and improve the target detection sensitivity of the on-board sensors.

[0077] In this step, the RSU offline status includes whether the RSU is offline and its specific location. Since the RSU is offline, traffic light and pedestrian information cannot be obtained from the RSU, so it is necessary to switch the information source. In this embodiment, the urban traffic management center serves as the backup information source. The urban traffic management center is the management center of the intelligent transportation system in urban areas (such as some intelligent connected vehicle demonstration zones and pilot zones). It can monitor, manage and optimize the traffic flow of the entire city or region in real time, integrating data aggregation, processing, decision-making and command. It sends the corresponding traffic light information to vehicles that are about to enter the service range of the offline RSU through the 5G cellular network, realizing seamless replacement of key information. In addition, in order to make up for the lack of V2P (vehicle-to-pedestrian) information, the intelligent driving vehicle driving control method in this embodiment can also issue a cautious driving command to control the vehicle to slow down, while improving the target detection sensitivity of on-board sensors (such as cameras and lidar) to further improve the driving reliability of intelligent driving vehicles.

[0078] Understandably, a common machine learning model can be used to generate a driving strategy that includes source switching and vehicle deceleration based on the RSU offline status and the vehicle's driving risk level. The role of RSUs at different key intersections in traffic management is also different. The vehicle deceleration magnitude and the activation of specific on-board sensor accuracy improvements in the driving strategy can be learned by the model based on a training set consisting of the RSU offline status and the vehicle's driving risk level.

[0079] S280. The vehicle driving information collected each time, the network quality information of the base stations near the vehicle location, the predicted faults, the driving risk level, the generated driving strategy, and the network connection status monitored after execution are processed in a unified manner to obtain structured data.

[0080] In this embodiment, the vehicle driving information collected each time, the network quality information of base stations near the vehicle location, the predicted vehicle network connection failures and their probabilities, hardware failures, the determined driving risk level, the generated multiple alternative driving strategies, and the selected optimal driving strategy and the network connection status monitored after execution can be processed in a unified format and added to the training dataset as a set of structured data to optimize the above models.

[0081] S290. The model is continuously iterated and updated based on the structured data.

[0082] In this step, based on the input data (real-time collected data) and the data obtained from the above steps, the LSTM-based time series prediction model for fault probability prediction, the LSTM-based time series prediction model for root cause diagnosis, the rule engine or deep learning model for risk level determination, and the machine learning model for driving strategy generation are continuously iteratively updated to continuously optimize model parameters, enabling each model to evolve. As the running time increases, the output results become more accurate, and the decisions become more intelligent, forming a virtuous cycle. Understandably, the position of each data point in the structured data is fixed, and each data point has a unique identifier. When inputting into the model for iteration, each model only selects the data it needs for its own iterative update.

[0083] As can be seen, in the above scheme, the intelligent driving vehicle driving control method of the present invention can transform from "post-event response" to "pre-event prediction and active intervention". It can dynamically generate multiple alternative driving strategies based on the specific driving intentions (driving tasks and driving services) and vehicle network connection failures, and control the corresponding vehicles to execute the optimal driving strategy according to the priority order of driving services. Thus, without affecting driving safety, it can adaptively generate driving strategies for fault recovery, and seamlessly resolve potential network failures. It can improve the safety, continuity and network resource utilization of L4 autonomous driving. Furthermore, it can iterate the model based on each input and generated data to continuously optimize the model.

[0084] Reference Figure 6 , Figure 6 This is a schematic diagram of the structure of an intelligent driving vehicle driving control system provided in the first embodiment of the present invention. In the embodiment shown in the figure, the intelligent driving vehicle driving control system includes a data acquisition unit 101, a fault prediction unit 102, a diagnosis and risk assessment unit 103, and a strategy generation unit 104. Detailed descriptions of each functional unit are as follows:

[0085] Data acquisition unit 101 is used to collect vehicle driving information and network quality information of base stations near the vehicle's location in real time;

[0086] The fault prediction unit 102 is used to predict the vehicle network connection failure within a preset time period based on the vehicle driving information, the network quality information, and the historical driving status of the vehicle on a similar historical driving path, and to obtain the vehicle network connection failure probability.

[0087] Diagnosis and risk assessment 103 is used to determine the driving risk level of the vehicle based on the vehicle network connection failure and its probability and the vehicle's driving task.

[0088] The strategy generation unit 104 is used to select and generate an optimal driving strategy that includes network resource scheduling and vehicle behavior guidance based on the vehicle network connection failure, the vehicle's driving task and the driving risk level using a pre-trained machine learning model, and to control the corresponding vehicles to execute the optimal driving strategy according to the priority order of the vehicle's driving services.

[0089] In some embodiments, the fault prediction unit 102 is specifically used for:

[0090] Using an LSTM-based time series prediction model, the network quality of the base station network used by the vehicle in the future is continuously predicted based on the vehicle's driving direction, speed, and driving path in the vehicle's driving information and the network quality information of the base stations near the vehicle's location.

[0091] Based on the network quality and the vehicle's disconnection records in historical driving states along similar historical driving paths, the probability of vehicle network connection failure within a preset future time period is predicted.

[0092] In some embodiments, the diagnostic and risk assessment unit 103 is specifically used for:

[0093] If the probability of the vehicle network connection failure is greater than a preset threshold, a root cause diagnosis method is used to determine the root cause of the vehicle network connection failure.

[0094] Based on the root cause of the vehicle network connectivity failure and the vehicle's driving task, assess the impact of the vehicle network connectivity failure on the vehicle's driving task, thereby determining the vehicle's driving risk level.

[0095] In some embodiments, the policy generation unit 104 is specifically used for:

[0096] A pre-trained machine learning model is used to generate multiple alternative driving strategies and corresponding reward functions that include network resource scheduling and vehicle behavior guidance based on the vehicle network connection failure, the vehicle's driving task, and driving risk level; wherein, the reward function includes a reward function determined by the degree of impact of executing the alternative driving strategy on the current driving task;

[0097] The alternative driving strategy with the largest reward function value is selected as the optimal driving strategy.

[0098] In some embodiments, the alternative driving strategies generated by the strategy generation unit 104 include:

[0099] No intervention is made in the current vehicle driving strategy; or

[0100] Force the vehicle to switch its signal connection path to connect to a neighboring base station with a stable signal; or

[0101] Reduce vehicle speed and delay entering areas where vehicle network connectivity issues occur.

[0102] As can be seen, this invention provides an intelligent driving vehicle driving control system that can probabilistically predict vehicle network connection failures within a preset time period using a time series prediction model. This allows intelligent driving vehicles to shift from passive response to active defense, facilitating subsequent avoidance measures. After predicting a failure, the system can determine the vehicle's driving risk level based on the vehicle network connection failure, its probability, and the vehicle's driving task to assess the impact of the failure on the driving task. Then, based on the vehicle network connection failure, the driving task, and the driving risk level, a pre-trained machine learning model is used to select and generate an optimal driving strategy that includes network resource scheduling and vehicle behavior guidance. The system then controls the corresponding vehicles to execute the optimal driving strategy according to the priority order of the driving services, ensuring that the vehicles performing the most critical driving services among all vehicles predicted to have failures receive the highest priority network protection. This achieves adaptive, differentiated, and refined fault recovery. Therefore, this invention can accurately predict and actively intervene before network connection failures occur, avoiding a potential network connection interruption, ensuring the continuity and reliability of L4 autonomous driving, and ensuring that vehicles performing the most critical driving services receive the highest priority network protection.

[0103] Reference Figure 7 , Figure 7 This is a schematic block diagram of an intelligent driving vehicle driving control system provided in the second embodiment of the present invention. Compared with the intelligent driving vehicle driving control system provided in the first embodiment, the intelligent driving vehicle driving control system provided in this embodiment adds a monitoring and verification unit 105, a hardware detection unit 106, a search and determination unit 107, a data processing unit 108, and an update unit 109. The detailed descriptions of each added functional unit are as follows:

[0104] The monitoring and verification unit 105 is used to continuously monitor the network connection status of the vehicle after the vehicle executes the optimal driving strategy, so as to know the network connection stability of the vehicle.

[0105] Hardware detection unit 106 is used to detect whether the RSUs at key intersections in the driving path are offline based on the driving path in the vehicle driving information.

[0106] The search and determination unit 107 is used to identify the service range of an RSU when it is offline, search for vehicles that are about to enter the service range based on the service range of the RSU and the vehicle driving information, and determine that the driving risk level of the vehicles that are about to enter the service range is high risk.

[0107] The strategy generation unit 104 is also used to generate a driving strategy including source switching and vehicle speed reduction based on the RSU offline status and the vehicle's driving risk level, so as to send the traffic light information of the key intersection issued by the city traffic management center to the corresponding vehicle through the cellular network, and control the vehicle to reduce the maximum speed limit, thereby improving the target detection sensitivity of the on-board sensors. Understandably, the strategy generation unit 104 may include a machine learning model based on reinforcement learning and a general machine learning model, which generate driving strategies for vehicle network connection failure and hardware failure, respectively.

[0108] The data processing unit 108 is used to perform unified data processing on the vehicle driving information collected each time, the network quality information of the base station near the vehicle location, the predicted faults, the driving risk level, the generated driving strategy, and the network connection status monitored after execution, to obtain structured data.

[0109] The update unit 109 is used to continuously iteratively update the model based on the structured data.

[0110] As can be seen, the intelligent driving vehicle driving control system in this embodiment can also detect hardware faults to ensure the driving safety and continuity of intelligent vehicles from multiple dimensions. Furthermore, it can iterate the model based on each input and generated data to continuously optimize the model and further improve the accuracy of driving strategy generation.

[0111] It should be noted that those skilled in the art can clearly understand that the specific implementation process of the above-mentioned intelligent driving vehicle driving control system and each unit or component can be referred to the corresponding description in the foregoing method embodiments. For the sake of convenience and brevity, it will not be repeated here.

[0112] In one embodiment, a computer device is provided, the internal structure of which can be shown as follows: Figure 8As shown, the computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile and / or volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface is used to communicate with external devices via a network connection. Furthermore, the computer device may also include a display screen and input devices (e.g., mouse, keyboard, etc.) for interactive purposes.

[0113] Specifically, when the processor in the computer device executes the computer program, it implements the steps of the intelligent driving vehicle driving control method provided in the first and second embodiments described above.

[0114] In one embodiment, the present invention may also provide a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the intelligent driving vehicle driving control method provided in the first and second embodiments described above.

[0115] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0116] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above.

[0117] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for controlling the driving of an intelligent driving vehicle, characterized in that, The intelligent driving vehicle driving control method includes: Real-time collection of vehicle driving information and network quality information of base stations near the vehicle's location; Based on the vehicle's driving information, network quality information, and historical driving status of the vehicle along similar historical driving paths, a pre-trained time series prediction model is used to predict vehicle network connection failures within a preset future time period, thus obtaining the probability of vehicle network connection failures. Specifically, this includes: using an LSTM-based time series prediction model to continuously predict the network quality of the base station network used by the vehicle within a preset future time period based on the vehicle's driving direction, speed, and driving path in the vehicle's driving information, as well as the network quality information of base stations near the vehicle's location; and predicting vehicle network connection failures based on the network quality and the vehicle's disconnection records in historical driving status along similar historical driving paths, thus obtaining the probability of vehicle network connection failures within a preset future time period. The driving risk level of a vehicle is determined based on the vehicle network connection failure and its probability, and the vehicle's driving task. Specifically, this includes: if the probability of the vehicle network connection failure is greater than a preset threshold, using a root cause diagnosis method to determine the root cause of the vehicle network connection failure; and assessing the impact of the vehicle network connection failure on the vehicle's driving task based on the root cause of the vehicle network connection failure and the vehicle's driving task, thereby determining the driving risk level of the vehicle. Based on the vehicle network connection failure, the vehicle's driving task, and the driving risk level, a pre-trained machine learning model is used to select and generate an optimal driving strategy that includes network resource scheduling and vehicle behavior guidance. The optimal driving strategy is then executed by the corresponding vehicle according to the priority order of the vehicle's driving services. Specifically, this includes: using a pre-trained machine learning model to generate multiple alternative driving strategies and corresponding reward functions based on the vehicle network connection failure, the vehicle's driving task, and the driving risk level; wherein the reward function includes a reward function determined by the degree of impact on the current driving task after executing the alternative driving strategy; and selecting the alternative driving strategy with the largest reward function value as the optimal driving strategy.

2. The intelligent driving vehicle travel control method according to claim 1, characterized by, The alternative driving strategies include: No intervention is made in the current vehicle driving strategy; or Force the vehicle to switch its signal connection path to connect to a neighboring base station with a stable signal; or Reduce vehicle speed and delay entering areas where vehicle network connectivity issues occur.

3. The intelligent driving vehicle travel control method according to claim 1, characterized by, Following the step of controlling the corresponding vehicles to execute the optimal driving strategy according to the priority order of the vehicle's driving operations, the method further includes: Continuously monitor the network connection status of the vehicle to determine the stability of the vehicle's network connection; The vehicle driving information collected each time, the network quality information of base stations near the vehicle location, the predicted faults, the driving risk level, the generated driving strategy, and the network connection status monitored after execution are all processed in a unified manner to obtain structured data. The model is continuously iterated and updated based on the structured data.

4. The intelligent driving vehicle travel control method according to claim 1, characterized by, The intelligent driving vehicle driving control method also includes: Based on the driving path in the vehicle driving information, detect whether the RSU at key intersections in the driving path is offline; If an RSU is offline, identify the service area of ​​that RSU, search for vehicles that are about to enter the service area based on the service area of ​​the RSU and the vehicle driving information, and determine that the driving risk level of the vehicles that are about to enter the service area is high risk. Based on the RSU offline status and the vehicle's driving risk level, a driving strategy including source switching and vehicle speed reduction is generated. This strategy is used to transmit the traffic light information of the key intersection issued by the city traffic management center to the corresponding vehicles via the cellular network, control the vehicles to reduce the maximum speed limit, and improve the target detection sensitivity of the on-board sensors.

5. An intelligent driving vehicle travel control system characterized by comprising: It includes a unit for performing the intelligent driving vehicle driving control method as described in any one of claims 1-4.

6. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the intelligent driving vehicle driving control method as described in any one of claims 1 to 4.

7. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the intelligent driving vehicle driving control method as described in any one of claims 1 to 4.

Images