Driving suggestion generation method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention provides a driving suggestion generation method and device, electronic equipment and a storage medium, and belongs to the technical field of traffic. The method comprises the following steps: acquiring macroscopic traffic data and microscopic individual data, and fusing the macroscopic traffic data and the microscopic individual data by adopting a multi-modal data fusion technology to obtain a current traffic situation model; predicting the future driving track of each regional vehicle based on the current traffic situation model through a machine learning algorithm and a traffic flow theory to obtain a vehicle track prediction result; based on the vehicle track prediction result, conflict detection is carried out on regional vehicles, and a conflict pool is generated; and performing conflict negotiation on the conflicting vehicles to generate driving sequence suggestions of the conflicting vehicles. According to the embodiment of the invention, the method can achieve the real-time monitoring, analysis and decision making of the traffic condition, facilitates the improvement of the adaptability and effectiveness of decision making, improves the passing efficiency and safety, and reduces the vehicle waiting time and energy consumption.

Description

Technical Field

[0001] This application relates to the field of transportation technology, and in particular to a method and apparatus for generating driving suggestions, an electronic device, and a storage medium. Background Technology

[0002] With the continuous increase in the number of vehicles in cities, traffic congestion has become increasingly serious, which not only reduces road traffic efficiency but also increases the risk of traffic accidents.

[0003] Traditional methods for alleviating traffic congestion fall into two categories: one is to alter the original road shape by changing "hard infrastructure," such as building overpasses or modifying pavement design; the other is to add "soft conditions" to the existing road network, such as controlling traffic light timings, setting up tidal flow lanes, and improving traffic rules. Both methods have drawbacks, including high construction and maintenance costs, poor road flexibility and adaptability, and increased difficulty and safety hazards for drivers.

[0004] In summary, the technical problems existing in the relevant technologies need to be improved. Summary of the Invention

[0005] The main objective of this application is to provide a method and apparatus for generating driving suggestions, an electronic device and a storage medium, so as to realize real-time monitoring, analysis and decision-making of traffic conditions, and improve traffic efficiency and safety.

[0006] To achieve the above objectives, one aspect of this application proposes a method for generating driving suggestions, the method comprising: Macro-level traffic data and micro-level individual data are acquired, and multi-modal data fusion technology is used to fuse the macro-level traffic data and the micro-level individual data to obtain a current traffic situation model. The macro-level traffic data is collected by roadside equipment, and the micro-level individual data is collected by regional vehicles. Based on the current traffic situation model, using machine learning algorithms and traffic flow theory, the future driving trajectory of each vehicle in the area is predicted, and the vehicle trajectory prediction results are obtained. Based on the vehicle trajectory prediction results, collision detection is performed on vehicles in the area to generate a collision pool. The collision pool consists of multiple collision points, and each collision point includes the collision coordinates, information of the conflicting vehicles, and the arrival time of the collision point. Based on the conflict pool, macro-level traffic data, and micro-level individual data, and using preset negotiation rules, conflict negotiation is conducted on the conflicting vehicles to generate a suggested driving order for the conflicting vehicles.

[0007] In some embodiments, the step of performing collision detection on vehicles in the area based on the vehicle trajectory prediction results and generating a collision pool includes: Based on the vehicle trajectory prediction results, determine whether the predicted trajectories of vehicles in the area intersect. If so, determine the intersecting vehicles, the intersection location, and the intersection time point. The intersecting vehicles are associated with the intersection time point. The intersection time difference is calculated based on the intersection time point. If the intersection time difference is less than the safety threshold, a conflict point is generated based on the intersecting vehicles, the intersection position, and the intersection time point, and the conflict point is added to the conflict pool.

[0008] In some embodiments, the step of calculating the intersection time difference based on the intersection time point, and if the intersection time difference is less than a safety threshold, generating a conflict point based on the intersecting vehicles, the intersection location, and the intersection time point, and adding the conflict point to a conflict pool, includes: The intersection time difference is calculated based on the intersection time point. If the intersection time difference is less than the safety threshold, the intersecting vehicles are identified as conflicting vehicles, the intersection time point is identified as the arrival time of the conflict point, and the intersection position is identified as the conflict coordinate position. Add the conflict points to the conflict pool.

[0009] In some embodiments, the vehicle trajectory prediction result includes route priority, and the step of negotiating the conflicting vehicles based on the conflict pool, the macroscopic traffic data, and the microscopic individual data using preset negotiation rules to generate a driving order suggestion for the conflicting vehicles includes: The distance to the conflict point is determined based on the conflict pool, the macro-level traffic data, and the micro-level individual data. The priority order of conflicting vehicles is determined by a preset negotiation rule based on the distance to the conflict point, the vehicle speed, and the route priority. A driving order suggestion for the conflicting vehicles is generated based on the priority order of the conflicting vehicles.

[0010] In some embodiments, the method further comprises: The vehicle trajectory prediction results for the area are updated according to the driving order suggestion, and the vehicle trajectory prediction results for the area include the vehicle trajectory prediction results for the conflicting vehicles. Conflict detection is performed based on the updated vehicle trajectory prediction results for the area to determine whether a conflict exists. If a conflict exists, a conflict point is generated and added to the conflict pool. The system returns to the conflict pool based on preset negotiation rules, the macro traffic data, and the micro individual data to conduct conflict negotiation on the conflicting vehicles, generating suggested driving order steps for the conflicting vehicles until there is no conflict.

[0011] In some embodiments, the method further comprises: Based on the vehicle trajectory prediction results of vehicles in the region, the driving order suggestions of conflicting vehicles, the macro traffic data and the micro individual data, the driving behavior decisions of vehicles in the region are made by the large language model, and driving decision suggestions for vehicles in the region are generated. In response to the first instruction, the driving decision suggestion is displayed on the in-vehicle human-machine interface of the vehicle in the area.

[0012] To achieve the above objectives, another aspect of this application provides a driving suggestion generation apparatus, the apparatus comprising: The data fusion module is used to acquire macro-level traffic data and micro-level individual data. It uses multimodal data fusion technology to fuse the macro-level traffic data and the micro-level individual data to obtain the current traffic situation model. The macro-level traffic data is collected by roadside equipment, and the micro-level individual data is collected by regional vehicles. The prediction module is used to predict the future driving trajectory of each vehicle in the area based on the current traffic situation model using machine learning algorithms and traffic flow theory, and to obtain the vehicle trajectory prediction result. The conflict detection module is used to perform conflict detection on vehicles in the area based on the vehicle trajectory prediction results and generate a conflict pool. The conflict pool consists of multiple conflict points, and each conflict point includes the conflict coordinates, information of the conflicting vehicles, and the arrival time of the conflict point. The conflict negotiation module is used to conduct conflict negotiation on the conflicting vehicles based on the conflict pool, the macro traffic data and the micro individual data through preset negotiation rules, and generate a driving order suggestion for the conflicting vehicles.

[0013] To achieve the above objectives, another aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method described above.

[0014] To achieve the above objectives, another aspect of the embodiments of this application proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the methods described above.

[0015] To achieve the above objectives, another aspect of the embodiments of this application proposes a computer program product, including a computer program that, when executed by a processor, implements the aforementioned method.

[0016] The embodiments of this application include at least the following beneficial effects: This application provides a method and apparatus for generating driving suggestions, an electronic device, and a storage medium. This solution acquires macroscopic traffic data and microscopic individual data, and uses multimodal data fusion technology to fuse the macroscopic traffic data and microscopic individual data to obtain a current traffic situation model. Based on the current traffic situation model, machine learning algorithms and traffic flow theory are used to predict the future driving trajectory of each vehicle in the area, obtaining vehicle trajectory prediction results. Based on the vehicle trajectory prediction results, conflict detection is performed on vehicles in the area to generate a conflict pool, which can automatically identify potential vehicle collision risks and improve the safety of vehicle passage. Based on the conflict pool, macroscopic traffic data, and microscopic individual data, conflict negotiation is performed on conflicting vehicles through preset negotiation rules to generate driving order suggestions for conflicting vehicles, realizing real-time monitoring, analysis, and decision-making of traffic conditions. This is conducive to improving the adaptability and effectiveness of decision-making, improving traffic efficiency and safety, reducing vehicle waiting time and energy consumption, and also reducing the risk of traffic accidents. Attached Figure Description

[0017] Figure 1 This is a flowchart of the driving suggestion generation method provided in the embodiments of this application; Figure 2 yes Figure 1 The flowchart of step S103 in the process; Figure 3 yes Figure 1 The flowchart of step S104 in the process; Figure 4 This is a flowchart of the conflict verification steps of the driving suggestion generation method provided in this application embodiment; Figure 5 This is a flowchart of the decision generation steps of the driving suggestion generation method provided in the embodiments of this application; Figure 6 This is a flowchart illustrating a specific implementation of the driving suggestion generation method provided in this application when applied to an intersection traffic system; Figure 7 This is a schematic diagram of the driving suggestion generation device provided in the embodiments of this application; Figure 8 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit it. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with those of this application; they are merely examples of apparatuses and methods consistent with some aspects of the embodiments of this application as detailed in the appended claims.

[0019] It is understood that the terms “first,” “second,” etc., used in this application may be used herein to describe various concepts, but unless otherwise stated, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of the embodiments of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the words “if,” “when,” or “in response to a determination” as used herein may be interpreted as “when…” or “when…” or “in response to a determination.”

[0020] As used in this application, the terms "at least one", "multiple", "each", "any", etc., "at least one" includes one, two or more, "multiple" includes two or more, "each" refers to each of the corresponding multiples, and "any" refers to any one of the multiples.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0022] Before providing a detailed description of the embodiments of this application, some of the nouns and terms involved in the embodiments of this application will be explained first. The nouns and terms involved in the embodiments of this application are subject to the following interpretations.

[0023] 1) The in-vehicle human-machine interface (HMI) is an interface in the in-vehicle information system that allows drivers and passengers to interact with various functions of the vehicle. The HMI helps users control the vehicle's entertainment system, navigation, air conditioning, in-vehicle communication, and other functions through various input methods such as displays, touchscreens, buttons, and voice control.

[0024] With the continuous increase in the number of vehicles in cities, traffic congestion at intersections is becoming increasingly serious, not only reducing road efficiency but also increasing the risk of traffic accidents. Traditional methods to alleviate traffic congestion at intersections include building overpasses, modifying road surface design, controlling traffic light timings, setting up tidal flow lanes, and improving traffic rules to adjust vehicle flow and alleviate congestion.

[0025] However, the traditional methods also have obvious drawbacks. Building elevated highways requires huge investments, significant capital investment, and has a long construction period, from planning and design to completion, significantly impacting the lives and traffic of surrounding residents. During construction, roads may be occupied or closed, exacerbating traffic congestion and causing inconvenience. The construction of elevated highways also alters the urban spatial layout and landscape, and vehicles traveling on them generate noise pollution. In emergencies such as traffic accidents or fires, evacuating vehicles from elevated highways is relatively difficult.

[0026] Modifying the design of intersection surfaces, such as adding safety zones and parking areas, not only occupies road space and reduces road flexibility, but also reduces the actual usable lane area, altering vehicle trajectories and flow, and disrupting normal traffic flow. The new surface design may also increase the difficulty for drivers, requiring them to anticipate whether to enter a safety zone and precisely control speed and distance. Large-scale road reconstruction requires significant investment of manpower, materials, and financial resources, resulting in high construction costs. Furthermore, the newly added road infrastructure is prone to damage during use, requiring frequent maintenance and repairs, increasing maintenance costs.

[0027] While controlling traffic light timings to alleviate traffic congestion at intersections can be effective, it also has drawbacks. Traditional traffic light timing methods are often based on fixed time intervals or historical data, making it difficult to accurately reflect dynamic changes in traffic flow in real time. They also cannot quickly adapt to differences in traffic flow at different times and in different directions, leading to low traffic efficiency in some directions. Adjusting traffic light timings requires considering not only the traffic conditions at individual intersections but also the impact on traffic flow distribution and efficiency at surrounding intersections to ensure smooth traffic flow throughout the entire area. However, such regional coordination and optimization often requires substantial data support and complex model analysis, making practical implementation challenging.

[0028] Setting up tidal flow lanes often requires detailed road planning and design, including lane width, length, and the placement of signs and markings. It also necessitates the installation of appropriate traffic signs, traffic lights, and monitoring equipment to ensure the safe and effective operation of the lanes. This increases the setup and management costs for traffic management departments. The traffic rules for tidal flow lanes are relatively unique, and some drivers may be unfamiliar with their usage, potentially leading to lane errors, illegal lane changes, and other issues that affect the efficiency and safety of the lanes.

[0029] Improving traffic rules at intersections to alleviate traffic congestion requires time for drivers and pedestrians to learn and understand. If the rules are not clear and unambiguous, it can lead to confusion and misunderstanding among drivers and pedestrians, affecting their effectiveness. Simultaneously, traffic enforcement departments need to strengthen the promotion and enforcement of the new rules to ensure their effective implementation, which also increases management costs. The formulation of traffic rules typically requires a certain procedure and time, making it difficult to quickly respond to real-time changes in traffic flow and special circumstances. In some complex traffic scenarios, fixed traffic rules may not meet actual traffic needs, requiring on-site command and adjustments by traffic management personnel, increasing the difficulty and workload of management.

[0030] Therefore, traditional methods fall into two categories: one is to change the shape of the original intersection by modifying the "hard infrastructure," such as building overpasses or altering road surface design; the other is to add "soft conditions" to the existing intersection, such as controlling traffic light timings, setting up tidal flow lanes, and improving traffic rules. Both methods have drawbacks such as high construction and maintenance costs, poor road flexibility and adaptability, and increased difficulty for drivers and safety hazards, and cannot fundamentally solve the problem of insufficient road capacity.

[0031] In view of this, this application provides a method, apparatus, device, and medium for generating driving suggestions. This solution acquires macroscopic traffic data and microscopic individual data, and uses multimodal data fusion technology to fuse the macroscopic traffic data and microscopic individual data to obtain a current traffic situation model. Based on the current traffic situation model, machine learning algorithms and traffic flow theory are used to predict the future driving trajectory of each vehicle in the area, obtaining vehicle trajectory prediction results. Based on the vehicle trajectory prediction results, conflict detection is performed on vehicles in the area to generate a conflict pool, which can automatically identify potential vehicle collision risks and improve vehicle traffic safety. Based on the conflict pool, macroscopic traffic data, and microscopic individual data, preset negotiation rules are used to negotiate conflicts between vehicles, generating driving order suggestions for conflicting vehicles. This enables real-time monitoring, analysis, and decision-making of traffic conditions, improving the adaptability and effectiveness of decisions, enhancing traffic efficiency and safety, reducing vehicle waiting time and energy consumption, and also reducing the risk of traffic accidents.

[0032] The driving suggestion generation method provided in this application relates to the field of transportation technology. This method can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, or in-vehicle terminal, but is not limited to these. The server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The server can also be a node server in a blockchain network. The software can be an application implementing the driving suggestion generation method, but is not limited to the above forms.

[0033] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0034] It should be noted that in all specific embodiments of this application, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards. In addition, when embodiments of this application require access to sensitive personal information of users, separate permission or consent from the user is obtained through pop-ups or redirection to confirmation pages. Only after obtaining the user's separate permission or consent is the necessary user-related data required for the proper functioning of these embodiments acquired.

[0035] Figure 1 This is an optional flowchart of the driving suggestion generation method provided in the embodiments of this application. Figure 1 The method may include, but is not limited to, steps S101 to S104: Step S101: Obtain macro-level traffic data and micro-level individual data, and use multimodal data fusion technology to fuse the macro-level traffic data and micro-level individual data to obtain the current traffic situation model.

[0036] Specifically, macro-level traffic data is collected by roadside equipment, while micro-level individual data is collected by vehicles in the region.

[0037] It should be noted that "regional vehicles" refers to vehicles within the target area where driving recommendations are required.

[0038] This application does not impose specific limitations on the target area, which can be flexibly selected based on actual detection needs. For example, the target area can be determined manually, or the system can select areas where congestion or accidents may occur based on traffic conditions.

[0039] In some embodiments, vehicle driving data and traffic condition information are collected in real time by roadside equipment (such as cameras, radar, vehicle-to-infrastructure communication units), and the vehicle itself (through onboard sensors and communication modules) collects its own data (such as navigation information, vehicle speed, vehicle position, acceleration, etc.) in real time and transmits these data to the cloud.

[0040] Optionally, vehicles can be assigned ID numbers based on information such as coordinates to facilitate subsequent transmission.

[0041] Furthermore, the data uploaded from the roadside and vehicles is preprocessed and integrated with macro and micro data to construct a current traffic situation model.

[0042] In this embodiment, macro-level traffic data and micro-level individual data are acquired, and multimodal data fusion technology is used to fuse the macro-level traffic data and micro-level individual data to obtain a current traffic situation model. This is beneficial for fully understanding the traffic situation and prepares for generating driving suggestions in the future.

[0043] Step S102: Based on the current traffic situation model, using machine learning algorithms and traffic flow theory, predict the future driving trajectory of each vehicle in the area to obtain the vehicle trajectory prediction result.

[0044] In some embodiments, the large language model uses machine learning algorithms and traffic flow theory to predict future vehicle trajectories based on fused data and integrates multiple factors.

[0045] Optionally, the prediction process comprehensively considers multiple factors such as the vehicle's current state (position, speed, acceleration), driver's intention (inferred from navigation information), road geometry, and traffic rules to generate trajectory prediction results that include the vehicle's possible travel path and speed changes.

[0046] In this embodiment, the future driving trajectory of each vehicle in the area is predicted based on the current traffic situation model using machine learning algorithms and traffic flow theory, which is beneficial for subsequent conflict detection.

[0047] Step S103: Based on the vehicle trajectory prediction results, perform collision detection on vehicles in the area and generate a collision pool.

[0048] Specifically, the conflict pool consists of multiple conflict points, each including the conflict coordinates, information about the conflicting vehicles, and their arrival times.

[0049] In some embodiments, based on the vehicle trajectory prediction results, the large language model analyzes parameters such as the intersection of the predicted trajectories, relative positions, and arrival times at conflict points to perform conflict detection on the future trajectories of vehicles within the intersection one by one, and records the information of vehicles with conflicts in a conflict pool containing the coordinates of the conflict point, vehicle information, and arrival time.

[0050] Optionally, based on the vehicle trajectory prediction results, it is determined whether the predicted trajectories of vehicles in the area intersect. If so, the intersecting vehicles, intersection locations, and intersection times are identified. The intersection time difference is calculated based on the intersection time point. If the intersection time difference is less than a safety threshold, a conflict point is generated based on the intersecting vehicles, intersection locations, and intersection times, and the conflict point is added to the conflict pool. Among these, the intersecting vehicles are associated with the intersection time point.

[0051] In this embodiment, based on the vehicle trajectory prediction results, collision detection is performed on vehicles in the area to generate a collision pool, which can automatically identify potential vehicle collision risks and improve the safety of vehicle passage.

[0052] Step S104: Based on the conflict pool, macro traffic data and micro individual data, and using preset negotiation rules, conflict negotiation is carried out on conflicting vehicles to generate a driving order suggestion for conflicting vehicles.

[0053] In some embodiments, the large language model determines the distance to the conflict point based on the conflict pool, macro-level traffic data, and micro-level individual data; determines the priority order of conflicting vehicles based on the distance to the conflict point, vehicle speed, and route priority through preset negotiation rules; and generates a driving order suggestion for conflicting vehicles based on the priority order of conflicting vehicles.

[0054] Optionally, for each vehicle in the conflict pool, conflict negotiation is conducted one by one based on factors such as speed, distance, and route priority, and driving priority order is assigned. Then, the vehicle driving order suggestions are adjusted and optimized through simulation and deduction until all vehicles drive without conflict, ensuring safety.

[0055] It should be noted that the vehicle trajectory prediction results of the area are updated according to the driving order suggestion. The vehicle trajectory prediction results of the area include the vehicle trajectory prediction results of conflicting vehicles. Conflict detection is performed based on the updated vehicle trajectory prediction results of the area to determine whether a conflict exists. If a conflict exists, a conflict point is generated and added to the conflict pool. The process then returns to the previous steps, which involve conflict negotiation based on the conflict pool, macro traffic data, and micro individual data, using preset negotiation rules, to generate a driving order suggestion for the conflicting vehicles, until no conflict exists.

[0056] Furthermore, based on the predicted vehicle trajectories, suggested driving order of conflicting vehicles, macroscopic traffic data, and microscopic individual data, a large language model is used to make driving behavior decisions for vehicles in the region, generating driving decision suggestions for the region's vehicles. In response to a first command, the driving decision suggestions are displayed on the vehicle's in-vehicle human-machine interface. Specifically, the first command is triggered when the vehicle has received the driving decision suggestions and the in-vehicle human-machine interface is active, and is used to display the driving decision suggestions on the in-vehicle human-machine interface.

[0057] The system utilizes a large language model combined with all vehicle information and traffic rules to first determine the driving order for each vehicle through conflict negotiation. Then, it comprehensively considers global factors to formulate and verify the optimal driving decision suggestions, ensuring safe, efficient, and low-energy traffic. Finally, the suggestions are transmitted to each vehicle via wireless communication.

[0058] Optionally, the memory in the cloud system records relevant traffic scene information, operation records, etc., and compares the human driver's operation behavior with driving decision suggestions. If the human operation is better, it is incorporated into the large language model as a knowledge base for model optimization.

[0059] In this embodiment, conflict negotiation is conducted on conflicting vehicles based on a conflict pool, macro-level traffic data, and micro-level individual data through preset negotiation rules. This generates a driving order suggestion for conflicting vehicles, enabling real-time monitoring, analysis, and decision-making of traffic conditions. This improves the adaptability and effectiveness of decision-making, enhances traffic efficiency and safety, reduces vehicle waiting time and energy consumption, and also reduces the risk of traffic accidents.

[0060] Steps S101 to S104 as illustrated in this embodiment involve acquiring macroscopic traffic data and microscopic individual data, fusing them using multimodal data fusion technology to obtain a current traffic situation model, predicting the future trajectory of each vehicle in the area based on the current traffic situation model using machine learning algorithms and traffic flow theory, and obtaining vehicle trajectory prediction results. Based on the vehicle trajectory prediction results, conflict detection is performed on vehicles in the area to generate a conflict pool, which can automatically identify potential vehicle collision risks and improve vehicle traffic safety. Based on the conflict pool, macroscopic traffic data, and microscopic individual data, conflict negotiation is conducted on conflicting vehicles using preset negotiation rules to generate a driving order suggestion for conflicting vehicles. This enables real-time monitoring, analysis, and decision-making of traffic conditions, which helps improve the adaptability and effectiveness of decision-making, enhances traffic efficiency and safety, reduces vehicle waiting time and energy consumption, and also reduces the risk of traffic accidents.

[0061] Please see Figure 2 In some embodiments, step S103 may include, but is not limited to, steps S201 to S202: Step S201: Determine whether the predicted trajectories of vehicles in the area intersect based on the vehicle trajectory prediction results. If so, determine the intersecting vehicles, the intersection location, and the intersection time point.

[0062] Specifically, the intersection of vehicles is associated with the time point of intersection.

[0063] In step S201 of some embodiments, for all vehicles in the target area at the same time, it is determined whether they intersect based on the vehicle trajectory prediction results. If they intersect, the intersecting vehicles, the intersection location, and the intersection time point are determined.

[0064] The intersection point refers to the time when the intersecting vehicles arrive at the intersection location.

[0065] Optionally, based on the vehicle trajectory prediction results, it can be determined whether the predicted trajectories of vehicles in the area intersect within a certain time period. If so, a conflict point is generated and added to the conflict pool. Here, the time period is the intersection safety threshold.

[0066] In this embodiment, the predicted trajectories of vehicles in the area are determined based on the vehicle trajectory prediction results. If they intersect, the intersecting vehicles, the intersection location, and the intersection time point are determined. This is beneficial for intersection prediction and helps with subsequent decision-making and preventive measures.

[0067] Step S202: Calculate the intersection time difference based on the intersection time point. If the intersection time difference is less than the safety threshold, generate a conflict point based on the intersecting vehicles, intersection positions, and intersection time points, and add the conflict point to the conflict pool.

[0068] In step S202 of some embodiments, the intersection time difference is calculated based on the intersection time point, and it is determined whether the intersection time difference is less than a safety threshold.

[0069] Optionally, if the intersection time difference is less than the safety threshold, a conflict point is generated based on the intersecting vehicles, the intersection location, and the intersection time point, and the conflict point is added to the conflict pool.

[0070] If the time difference between the intersections is greater than or equal to the safety threshold, the intersecting vehicles will not collide.

[0071] It is understandable that intersecting vehicles can be two vehicles or three or more vehicles.

[0072] If there are three intersecting vehicles, the intersection time difference between each pair is calculated. If one of the intersection time differences for the intersecting vehicles is less than a safety threshold, it is identified as a conflicting vehicle. Otherwise, if none of the intersection time differences for the intersecting vehicles are less than the safety threshold, the vehicles will not conflict.

[0073] In this embodiment, the intersection time difference is calculated based on the intersection time point. If the intersection time difference is less than the safety threshold, a conflict point is generated based on the intersecting vehicles, intersection positions, and intersection time points. The conflict point is added to the conflict pool, which can detect vehicle conflicts in real time and help improve road safety.

[0074] Please see Figure 3 In some embodiments, step S104 may include, but is not limited to, steps S301 to S303: Step S301: Determine the distance to the conflict point based on the conflict pool, macro-level traffic data, and micro-level individual data.

[0075] In step S301 of some embodiments, the distance from the conflicting vehicle to the corresponding conflict point is determined by the conflict pool, macro traffic data and micro individual data.

[0076] Among them, Euclidean distance and Manhattan distance algorithms can be selected.

[0077] Step S302: Determine the priority order of conflicting vehicles based on the distance to the conflict point, vehicle speed and route priority through preset negotiation rules.

[0078] In step S302 of some embodiments, the priority score of the conflicting vehicles is calculated based on the distance to the conflict point, vehicle speed and route priority by a preset negotiation rule, and the priority order of the conflicting vehicles is determined according to the priority score from largest to smallest.

[0079] The priority of conflicting vehicles is determined by comprehensively considering the distance to the conflict point, vehicle speed, and route priority.

[0080] For example, vehicles closer to the conflict point have higher priority scores; faster vehicles have higher priority scores; and vehicles with higher route priority have higher priority scores.

[0081] The route priority can be determined based on the vehicle trajectory prediction results or the current road of the conflicting vehicles.

[0082] Step S303: Generate a driving order suggestion for the conflicting vehicles based on their priority order.

[0083] In step S303 of some embodiments, a driving order suggestion is generated based on the priority order of conflicting vehicles.

[0084] Optionally, algorithms (such as shortest path algorithms or scheduling algorithms) can be used to determine the driving order for conflicting vehicles, so that vehicles with higher priority can pass through before the conflict point to avoid conflict.

[0085] In this embodiment, a driving order suggestion for conflicting vehicles is generated based on their priority order, which helps to reduce traffic congestion and accidents and improve traffic efficiency.

[0086] Please see Figure 4 In some embodiments, the driving suggestion generation method provided in this application further includes a conflict verification step, which may include, but is not limited to, steps S401 to S404: Step S401: Update the vehicle trajectory prediction results for the area based on the driving sequence.

[0087] Specifically, the vehicle trajectory prediction results for regional vehicles include the vehicle trajectory prediction results for conflicting vehicles.

[0088] In step S401 of some embodiments, the vehicle trajectory prediction results of conflicting vehicles are updated according to the driving sequence.

[0089] When vehicles involved in a conflict pass through the conflict point, it is recommended to proceed in the order of travel.

[0090] Step S402: Based on the updated vehicle trajectory prediction results of regional vehicles, perform conflict detection to determine whether a conflict exists.

[0091] In step S402 of some embodiments, the large language model performs another simulation based on the vehicle driving order after conflict negotiation, combined with all vehicle information at the current intersection.

[0092] Step S403: If a conflict exists, a conflict point is generated and added to the conflict pool.

[0093] In step S403 of some embodiments, if new conflicts still arise after deduction, the new conflict points are added to the conflict pool. Step S404: Return to the process of negotiating the conflicting vehicles based on the conflict pool, macro traffic data and micro individual data according to the preset negotiation rules, and generate a suggested driving order for the conflicting vehicles until there is no conflict.

[0094] In step S404 of some embodiments, the large language model is used again to negotiate vehicle conflicts in the manner described above, and driving order suggestions are generated until the driving order suggestions of each vehicle do not conflict with any other vehicle, thereby ensuring vehicle driving safety.

[0095] In this embodiment, the conflict negotiation is performed on conflicting vehicles based on a conflict pool, macro-level traffic data, and micro-level individual data using preset negotiation rules. This generates a suggested driving order for the conflicting vehicles until no conflict exists. This ensures that after the conflicting vehicles drive in this order, they will not cause any further conflict or new conflicts with other vehicles at the intersection.

[0096] Please see Figure 5 In some embodiments, the driving suggestion generation method provided in this application further includes a decision generation step, which may include, but is not limited to, steps S501 to S502: Step S501: Based on the vehicle trajectory prediction results of regional vehicles, the driving order suggestions of conflicting vehicles, macro traffic data and micro individual data, the large language model makes driving behavior decisions for regional vehicles and generates driving decision suggestions for regional vehicles.

[0097] In step S501 of some embodiments, based on the regional vehicle information, traffic information (including information such as the vehicle's current lane and available lanes), traffic rules, precautions, and vehicle spacing safety assessment of the target area, combined with the vehicle driving order determined through conflict negotiation, the large language model makes decisions on the driving behavior of each vehicle at the current intersection, and outputs driving decision suggestions for all vehicles in the current scene in real time. The decision content includes specific driving operation suggestions such as maintaining the current speed, accelerating, decelerating, and changing lanes left or right.

[0098] The decision-making process must take into account special vehicles, such as fire trucks, ambulances, and buses, which require priority passage. Other vehicles must give way. At the same time, the relationship between the vehicle and other road users, such as electric bikes, bicycles, and pedestrians, must also be considered. Vehicles must obey traffic rules, pay attention to pedestrians, and avoid collisions.

[0099] Furthermore, the large language model needs to reflect on and extrapolate the proposed decision suggestions to determine whether the current vehicle decision is the globally optimal solution. Otherwise, it needs to be reconsidered until the optimal driving strategy for all vehicles in the current scenario is found. The globally optimal solution needs to consider three aspects: maximum safety for all vehicles (no collisions), minimum energy consumption, and minimum travel time (maximum traffic efficiency). For example, if calculations show that both trucks and cars can achieve the same traffic requirements after deceleration, the heavier vehicle (truck) takes longer to decelerate and consumes more energy; the lighter vehicle (car) decelerates faster and consumes less energy. Therefore, priority should be given to allowing the lighter vehicle to decelerate first to avoid the high energy consumption caused by the truck's deceleration.

[0100] In step S502, in response to the first instruction, driving decision suggestions are displayed on the in-vehicle human-machine interface of the regional vehicle.

[0101] Specifically, the first instruction is triggered when the vehicle receives a driving decision suggestion, and is used to display the driving decision suggestion on the in-vehicle human-machine interface.

[0102] Optionally, the analytical thinking and logical reasoning capabilities of large language models can be leveraged to provide safe, fast, and low-energy decision-making suggestions for each vehicle. Then, the cloud-based large language model server, through a wireless communication network, accurately sends these detailed driving decision suggestions to the corresponding vehicles based on their vehicle IDs.

[0103] In step S502 of some embodiments, driving decision suggestions are displayed on the HMI (human-machine interface) of the vehicle display, and the driving decision suggestions are read aloud.

[0104] Taking the intersection traffic system as an example, Figure 6 This is a flowchart illustrating a specific implementation of the driving suggestion generation method provided in this application when applied to an intersection traffic system. Figure 6 The methods may include, but are not limited to, the following steps: Step 1: Data collection and uploading.

[0105] In some embodiments, at intersections, roadside equipment (such as cameras, radar, vehicle-to-infrastructure communication units) and the vehicles themselves (through onboard sensors and communication modules) collect vehicle driving data and traffic condition information, as well as vehicle data (such as navigation information, vehicle speed, vehicle position, acceleration, etc.) in real time, and transmit these data to the cloud-based large language model. The vehicles are also assigned ID numbers based on coordinates and other information to facilitate subsequent transmission.

[0106] Optionally, roadside equipment deployed at intersections, such as cameras, radar, and vehicle-to-infrastructure (V2I) communication units, can collect real-time road segment data, including vehicle driving data and traffic condition information. Vehicle driving data includes information such as the number of vehicles, vehicle location, speed, and vehicle type. This data is then transmitted to a large language model in the cloud.

[0107] Within the intersection area, all vehicles collect their own data in real time through onboard sensors and communication modules. This data includes navigation information (destination, route), speed, vehicle position, and acceleration, and is uploaded to a large language model in the cloud. All vehicles in the current scenario can be assigned IDs based on their coordinates and other information for easier subsequent information transmission.

[0108] Step 2: Multimodal data fusion and trajectory prediction.

[0109] In some embodiments, multimodal data fusion and trajectory prediction encompasses the preprocessing of roadside and vehicle-uploaded data using a cloud-based large language model, the fusion of macro and micro data to construct a traffic situation model, and the prediction of future vehicle trajectories based on the fused data using machine learning algorithms and traffic flow theory, taking into account multiple factors.

[0110] The multimodal data fusion process includes: receiving data from roadside equipment and vehicles via a large cloud-based language model, preprocessing the data to remove noise and outliers; and then using multimodal data fusion technology to combine macroscopic traffic data collected by roadside equipment with microscopic individual data uploaded by vehicles to construct a real-time traffic situation model for the intersection.

[0111] Furthermore, trajectory prediction includes: based on the fused data, the large language model uses machine learning algorithms and traffic flow theory to predict the future trajectory of each vehicle within the intersection. The prediction comprehensively considers multiple factors such as the vehicle's current state (position, speed, acceleration), driver intent (inferred from navigation information), road geometry, and traffic rules, generating a trajectory prediction result that includes the vehicle's possible travel path and speed changes in the next few seconds.

[0112] Step 3: Conflict detection and conflict pool formation.

[0113] In some embodiments, based on the vehicle trajectory prediction results, the large language model analyzes parameters such as the intersection of the predicted trajectories, relative positions, and arrival times at conflict points to perform conflict detection on the future trajectories of vehicles within the intersection one by one, and records the information of vehicles with conflicts in a conflict pool containing the coordinates of the conflict point, vehicle information, and arrival time.

[0114] Specifically, based on vehicle trajectory prediction results, the large language model performs conflict detection on the future trajectories of vehicles within the intersection one by one. By analyzing parameters such as the intersection of different vehicle prediction trajectories, the relative positional relationship between different vehicle prediction trajectories, and the time when vehicles arrive at the conflict point, it determines whether vehicles will meet at the same spatial location during their journey, i.e., a conflict will occur. For example, if the predicted trajectories of two vehicles intersect at a certain moment, and the time difference between the vehicles arriving at the intersection point is less than a safety threshold, then it is determined that the two vehicles have a conflict, and the relevant conflict vehicle information is recorded in the conflict pool. The conflict pool consists of multiple conflict points, each containing the coordinates of the conflict point, information on each vehicle that will collide at that conflict point, and the time of arrival at that conflict point.

[0115] Step 4: Conflict negotiation and driving sequence suggestions.

[0116] In some embodiments, the large language model negotiates conflicts and assigns driving priority order for each vehicle in the conflict pool, taking into account factors such as speed, distance, and route priority. Then, it adjusts and optimizes the vehicle driving order suggestions through simulation and deduction until all vehicles drive without conflict, ensuring safety.

[0117] Specifically, for all conflicting vehicles at each conflict point in the conflict pool, the large language model negotiates the conflict for each vehicle individually. Based on preset negotiation rules, and taking into account factors such as vehicle speed (high-speed vehicles may have greater kinetic energy and inertia), distance from the conflict point (vehicles closer to the conflict point may have priority to reduce braking loss), and route priority, the model assigns a driving priority order to each conflicting vehicle, determines their order of passage through the intersection, and generates corresponding vehicle driving order suggestions.

[0118] Furthermore, the large language model, based on the vehicle driving order after conflict negotiation and combined with information on all vehicles at the intersection, performs another simulation to ensure that conflicting vehicles will not clash with each other or create new conflicts with other vehicles at the intersection if they proceed in this order. If new conflicts still arise after the simulation, the new conflict points are added to the conflict pool, and the large language model is used again to negotiate vehicle conflicts and generate driving order suggestions, until the generated driving order suggestions for each vehicle do not conflict with any other vehicle, thus ensuring driving safety.

[0119] Step 5: Decision-making and issuance of driving recommendations.

[0120] In some embodiments, a large language model is used in conjunction with information on all vehicles at the intersection, traffic rules, etc., to first determine the driving order for each vehicle through conflict negotiation, then to formulate and verify the optimal driving decision suggestions by comprehensively considering global factors, so as to ensure safe, efficient and low-energy traffic, and finally to send them to each vehicle via wireless communication.

[0121] Based on all vehicle information at the current intersection, traffic information (including current and available lanes), traffic rules, precautions, and vehicle spacing safety assessments, combined with the vehicle driving order determined through conflict negotiation, the large language model makes decisions on the driving behavior of each vehicle at the current intersection, outputting driving decision suggestions for all vehicles in the current scenario in real time. The decisions include specific driving operation suggestions such as maintaining current speed, accelerating, decelerating, and changing lanes left or right.

[0122] First, the decision-making process must take into account special vehicles, such as fire trucks, ambulances, and buses, which require priority passage. Other vehicles must give way. At the same time, the relationship between the vehicle and other road users, such as electric bikes, bicycles, and pedestrians, must also be considered. Vehicles must obey traffic rules, pay attention to pedestrians, and avoid collisions.

[0123] Secondly, the large language model needs to reflect on and extrapolate the proposed decision suggestions to determine whether the current vehicle decision is the globally optimal solution. Otherwise, it needs to be reconsidered until the optimal driving strategy for all vehicles in the current scenario is found. The globally optimal solution needs to consider three aspects: maximum safety for all vehicles (no collisions), minimum energy consumption, and minimum travel time (maximum traffic efficiency). For example, if calculations show that both trucks and cars can achieve the same traffic requirements after deceleration, the heavier vehicle (truck) takes longer to decelerate and consumes more energy; the lighter vehicle (car) decelerates faster and consumes less energy. Therefore, priority should be given to allowing the lighter vehicle to decelerate first to avoid the high energy consumption caused by the truck's deceleration.

[0124] Specifically, leveraging the analytical thinking and logical reasoning capabilities of large language models, safe, fast, and low-energy decision-making suggestions are provided for each vehicle. Then, the cloud-based large language model server, through a wireless communication network, accurately sends these detailed driving decision suggestions to the corresponding vehicles based on their vehicle IDs.

[0125] Step 6: Vehicle execution and passage are realized.

[0126] In some embodiments, autonomous vehicles receive instructions from a cloud-based big data model, which are then displayed on an HDMI monitor via a TBOX. The onboard control system then adjusts the power, braking, and steering systems in real time to complete the intersection maneuver. Human-driven vehicles, on the other hand, receive driving suggestions via the TBOX displayed on an HDMI monitor. The driver then proceeds according to these suggestions until leaving the area. If the driver fails to follow the prompts, the vehicle records and uploads the driving behavior. Throughout this process, the big data model coordinates all vehicles to ensure they travel in the adjusted manner, improving intersection efficiency and reducing waiting time, energy consumption, and accident risks.

[0127] The vehicles are categorized into autonomous vehicles and human-driven vehicles. The autonomous vehicle receives driving suggestions from the cloud-based big data model, and TOBX receives the communication information and displays the instructions on an HDMI display. Simultaneously, the onboard control system performs real-time control of the vehicle's powertrain, braking system, and steering system based on the suggestions. For example, it controls engine output power for acceleration or deceleration, operates the brakes for smooth stopping, and controls the steering mechanism for lane changes, enabling passage through intersections.

[0128] Specifically, after receiving driving suggestions from the cloud, the driver transmits this information to an HDMI display via a TBOX. The driver then follows the prompts on the HDMI display until the vehicle leaves the intersection area. If the driver fails to follow the prompts, the vehicle records the driver's actions and uploads this record to the cloud system.

[0129] It should be noted that, whether it is an autonomous vehicle or a human-driven vehicle, all vehicles drive according to the adjusted driving mode. Under the overall coordination of the large language model, multiple vehicles in the intersection can pass through in an orderly and efficient manner, which greatly improves the traffic efficiency of the intersection, reduces vehicle waiting time and energy consumption, and also reduces the risk of traffic accidents.

[0130] Step 7: Comparative evaluation and model optimization.

[0131] In some embodiments, a memory database set up in the cloud system records traffic scene information, human driver operation records, and driving suggestions issued by a large language model. The travel time of the human driver and the model suggestion are compared. If the human operation is better and is globally optimal, it is included in the model knowledge base to optimize subsequent decision suggestions.

[0132] It's important to note that the cloud system includes a memory database that records current traffic scenario information, all human driver actions, vehicle driving suggestions from the large language model, vehicle trajectories at intersections, time, and speeds, etc. The system compares human driver actions with the large language model's suggestions. It compares the time a vehicle takes to cross the intersection, checking if the time taken by the human driver is less than the time taken by the large language model's suggestions. If the human driver's action is faster and the action is globally optimal for the current scenario, then the large language model is required to issue vehicle suggestions based on this action in similar scenarios in the future. The high-quality driving actions of human drivers (those that achieve a faster passage time than the large language model's suggestions while ensuring safe, collision-free passage) are used as the knowledge base of the large language model. This knowledge serves as context for the large language model, aiding in model optimization and providing higher-quality, more accurate decision-making suggestions.

[0133] In this embodiment, the intersection traffic system includes the following steps: Data Collection and Upload: Roadside equipment collects road segment data, including vehicle quantity, location, and speed, and transmits the data to the cloud-based big data model. Vehicles within the intersection area also upload their own data in real time, such as navigation information, speed, location, and acceleration, to the cloud-based big data model. Through roadside equipment and vehicle-specific equipment at the intersection, multi-source data, including vehicle driving data and traffic condition information, is collected in real time and transmitted to the cloud-based big data model. Multimodal Data Fusion and Trajectory Prediction: After preprocessing the received data (roadside and vehicle data), the cloud-based big data model uses multimodal data fusion technology to comprehensively understand the traffic conditions at the intersection. Based on the fused data, machine learning algorithms and traffic flow theory are used to predict the future driving trajectories of vehicles. 3. Conflict Detection and Conflict Pool Formation: Based on the predicted vehicle driving trajectories, combined with parameters such as trajectory intersection and vehicle arrival time at conflict points, the big data model performs conflict detection on the future vehicle trajectories one by one, recording the information of vehicles with conflicts in the conflict pool.

[0134] 4. Conflict Negotiation and Driving Order Suggestion: For conflicting vehicles in the conflict pool, the large language model conducts conflict negotiation based on a combination of factors such as vehicle speed and distance from the conflict point to determine the driving order of vehicles. The driving order suggestion is then optimized through simulation and deduction until all vehicles drive without conflict.

[0135] 5. Decision-making and issuance of driving recommendations Based on vehicle information, traffic information, negotiation results, and the driving order determined by conflict negotiation, the large language model formulates driving decisions for each vehicle, such as maintaining speed, accelerating, decelerating, changing lanes, and stopping. It then sends decision suggestions to the vehicles, outputs driving decision suggestions in real time, and accurately sends the suggestions to the corresponding vehicles.

[0136] 6. Vehicle execution and passage: Autonomous vehicles and human-driven vehicles perform corresponding operations based on the received driving suggestions. Under the overall coordination of the large language model, the vehicles pass through the intersection in an orderly and efficient manner according to the adjusted driving mode.

[0137] 7. Comparative Evaluation and Model Optimization: The memory bank in the cloud system records relevant traffic scene information, operation records, etc., and compares the operation behavior of human drivers with the operation suggestions of the large language model. If the human operation is better, it will be incorporated into the large language model as a knowledge base for subsequent model optimization.

[0138] It should be noted that the core innovation of this patent lies in applying a large language model to multi-vehicle cooperative control at intersections, leveraging its powerful data processing and analysis capabilities to achieve real-time monitoring, analysis, and decision-making regarding intersection traffic conditions. A key protected aspect of this patent is how to effectively fuse road segment data collected by roadside equipment with vehicle-uploaded data to comprehensively and accurately grasp the traffic situation at intersections. The method of using the large language model to predict future vehicle trajectories based on fused data, the algorithm for conflict detection and conflict pool formation based on trajectory prediction results, and the negotiation of conflicting vehicles based on preset negotiation rules to determine driving priority are crucial technologies for achieving orderly traffic flow at intersections. For conflicting vehicles, the large language model negotiates conflicts based on factors such as vehicle speed and distance from the conflict point. Based on traffic information and the negotiation results, the large language model formulates reasonable driving decision suggestions, outputs specific operational recommendations, and ensures the global optimality of the decision through simulation. The communication method between the cloud-based large language model and the vehicle, the mechanism for issuing decision suggestions, and how the vehicle executes driving operations based on the received decision suggestions are also among the protected contents of this patent. Different types of vehicles receiving and executing driving suggestions issued from the cloud, and how to achieve orderly traffic flow at intersections under the coordination of the large language model, are also among the protected contents of this patent. By comparing human driver operations with model suggestions, incorporating high-quality driving operations into the knowledge base for the optimization of the large language model and the improvement of subsequent decision suggestions, is also among the protected contents of this patent.

[0139] Please see Figure 7 This application also provides a driving suggestion generation device that can implement the above-described driving suggestion generation method. The device includes: The data fusion module 701 is used to acquire macro-level traffic data and micro-level individual data, and to fuse the macro-level traffic data and the micro-level individual data using multimodal data fusion technology to obtain a current traffic situation model. The macro-level traffic data is collected by roadside equipment, and the micro-level individual data is collected by regional vehicles. Prediction module 702 is used to predict the future driving trajectory of each vehicle in the area based on the current traffic situation model using machine learning algorithms and traffic flow theory, and to obtain the vehicle trajectory prediction result. The conflict detection module 703 is used to perform conflict detection on vehicles in the area based on the vehicle trajectory prediction results and generate a conflict pool. The conflict pool consists of multiple conflict points, and each conflict point includes the conflict coordinates, information of the conflicting vehicles, and the arrival time of the conflict point. The conflict negotiation module 704 is used to conduct conflict negotiation on the conflicting vehicles based on the conflict pool, the macro traffic data and the micro individual data through preset negotiation rules, and generate a driving order suggestion for the conflicting vehicles.

[0140] It is understood that the content of the above method embodiments is applicable to the present device embodiments. The specific functions implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0141] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described driving suggestion generation method. This electronic device can be any smart terminal, including a tablet computer, an in-vehicle computer, or similar device.

[0142] It is understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0143] Please see Figure 8 , Figure 8 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes: The processor 801 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The memory 802 can be implemented as a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 802 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 802 and is called and executed by the processor 801 using the driving suggestion generation method of the embodiments of this application. The 803 input / output interface is used to implement information input and output. The communication interface 804 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 805 transmits information between various components of the device (e.g., processor 801, memory 802, input / output interface 803, and communication interface 804); The processor 801, memory 802, input / output interface 803, and communication interface 804 are connected to each other within the device via bus 805.

[0144] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described driving suggestion generation method.

[0145] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0146] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0147] The driving suggestion generation method, device, electronic device, and storage medium provided in this application embodiment acquire macroscopic traffic data and microscopic individual data, and use multimodal data fusion technology to fuse the macroscopic traffic data and microscopic individual data to obtain a current traffic situation model. Based on the current traffic situation model, the method uses machine learning algorithms and traffic flow theory to predict the future driving trajectory of each vehicle in the area, obtaining vehicle trajectory prediction results. Based on the vehicle trajectory prediction results, the method performs conflict detection on vehicles in the area, generates a conflict pool, and can automatically identify potential vehicle collision risks, improving vehicle traffic safety. Based on the conflict pool, macroscopic traffic data, and microscopic individual data, the method uses preset negotiation rules to negotiate conflicts between vehicles, generating driving order suggestions for conflicting vehicles. This enables real-time monitoring, analysis, and decision-making of traffic conditions, which helps improve the adaptability and effectiveness of decisions, improves traffic efficiency and safety, reduces vehicle waiting time and energy consumption, and also reduces the risk of traffic accidents.

[0148] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0149] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0150] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0151] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0152] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0153] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0154] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0155] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0156] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0157] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0158] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A method for generating driving suggestions, characterized in that, The method includes: Macro-level traffic data and micro-level individual data are acquired, and multi-modal data fusion technology is used to fuse the macro-level traffic data and the micro-level individual data to obtain a current traffic situation model. The macro-level traffic data is collected by roadside equipment, and the micro-level individual data is collected by regional vehicles. Based on the current traffic situation model, using machine learning algorithms and traffic flow theory, the future driving trajectory of each vehicle in the area is predicted, and the vehicle trajectory prediction results are obtained. Based on the vehicle trajectory prediction results, collision detection is performed on vehicles in the area to generate a collision pool. The collision pool consists of multiple collision points, and each collision point includes the collision coordinates, information of the conflicting vehicles, and the arrival time of the collision point. Based on the conflict pool, macro-level traffic data, and micro-level individual data, and using preset negotiation rules, conflict negotiation is conducted on the conflicting vehicles to generate a suggested driving order for the conflicting vehicles.

2. The method according to claim 1, characterized in that, The step of performing collision detection on vehicles in the area based on the vehicle trajectory prediction results and generating a collision pool includes: Based on the vehicle trajectory prediction results, determine whether the predicted trajectories of vehicles in the area intersect. If so, determine the intersecting vehicles, the intersection location, and the intersection time point. The intersecting vehicles are associated with the intersection time point. The intersection time difference is calculated based on the intersection time point. If the intersection time difference is less than the safety threshold, a conflict point is generated based on the intersecting vehicles, the intersection position, and the intersection time point, and the conflict point is added to the conflict pool.

3. The method according to claim 1, characterized in that, The step of calculating the intersection time difference based on the intersection time point, and if the intersection time difference is less than a safety threshold, then generating a conflict point based on the intersecting vehicles, the intersection location, and the intersection time point, and adding the conflict point to the conflict pool, includes: The intersection time difference is calculated based on the intersection time point. If the intersection time difference is less than the safety threshold, the intersecting vehicles are identified as conflicting vehicles, the intersection time point is identified as the arrival time of the conflict point, and the intersection position is identified as the conflict coordinate position. Add the conflict points to the conflict pool.

4. The method according to claim 1, characterized in that, The vehicle trajectory prediction result includes route priority. The process of negotiating conflict among conflicting vehicles based on the conflict pool, macroscopic traffic data, and microscopic individual data using preset negotiation rules, and generating suggested driving order for the conflicting vehicles, includes: The distance to the conflict point is determined based on the conflict pool, the macro-level traffic data, and the micro-level individual data. The priority order of conflicting vehicles is determined by a preset negotiation rule based on the distance to the conflict point, the vehicle speed, and the route priority. A driving order suggestion for the conflicting vehicles is generated based on the priority order of the conflicting vehicles.

5. The method according to claim 1, characterized in that, The method further includes: The vehicle trajectory prediction results for the area are updated according to the driving order suggestion, and the vehicle trajectory prediction results for the area include the vehicle trajectory prediction results for the conflicting vehicles. Conflict detection is performed based on the updated vehicle trajectory prediction results for the area to determine whether a conflict exists. If a conflict exists, a conflict point is generated and added to the conflict pool. The system returns to the conflict pool based on preset negotiation rules, the macro traffic data, and the micro individual data to conduct conflict negotiation on the conflicting vehicles, generating suggested driving order steps for the conflicting vehicles until there is no conflict.

6. The method according to claim 1, characterized in that, The method further includes: Based on the vehicle trajectory prediction results of vehicles in the region, the driving order suggestions of conflicting vehicles, the macro traffic data and the micro individual data, the driving behavior decisions of vehicles in the region are made by the large language model, and driving decision suggestions for vehicles in the region are generated. In response to the first instruction, the driving decision suggestion is displayed on the in-vehicle human-machine interface of the vehicle in the area.

7. A driving suggestion generation device, characterized in that, The device includes: The data fusion module is used to acquire macro-level traffic data and micro-level individual data. It uses multimodal data fusion technology to fuse the macro-level traffic data and the micro-level individual data to obtain the current traffic situation model. The macro-level traffic data is collected by roadside equipment, and the micro-level individual data is collected by regional vehicles. The prediction module is used to predict the future driving trajectory of each vehicle in the area based on the current traffic situation model using machine learning algorithms and traffic flow theory, and to obtain the vehicle trajectory prediction result. The conflict detection module is used to perform conflict detection on vehicles in the area based on the vehicle trajectory prediction results and generate a conflict pool. The conflict pool consists of multiple conflict points, and each conflict point includes the conflict coordinates, information of the conflicting vehicles, and the arrival time of the conflict point. The conflict negotiation module is used to conduct conflict negotiation on the conflicting vehicles based on the conflict pool, the macro traffic data and the micro individual data through preset negotiation rules, and generate a driving order suggestion for the conflicting vehicles.

8. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method according to any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 6.

Images