Method and system for generating 5G base station navigation map and driving path and intelligent driving device

Abstract

The invention discloses a method and system for generating a 5G base station navigation map and a driving path and an intelligent driving device, and the method comprises the steps: obtaining the position, coverage range, signal intensity and other data of a 5G base station, integrating the data with a high-precision map, and marking the position and coverage area of the base station; generating a network coverage heat map, and identifying a signal strength area; preferably selecting the network quality as the weight of path planning; planning a driving path by using a multi-objective optimization algorithm; dynamically adjusting the path in real time; using a generative AI optimization path planning algorithm to plan a standby path; and generating a visual 5G base station navigation driving map. The method realizes close combination of network coverage analysis and path planning, introduces generative AI learning training to further improve the accuracy and real-time adjustment capability of path planning, ensures that the unmanned vehicle always keeps stable 5G network connection and an efficient communication environment in the driving process, has the characteristics of high efficiency, stability and safety, and is suitable for popularization and application. The method is suitable for various complex traffic environments.

Description

Technical Field

[0001] The present invention relates to the field of unmanned driving technology, and in particular to a method, system, and intelligent driving device for generating a 5G base station navigation map and driving path. Background Art

[0002] With the continuous development of autonomous driving technology, especially long-distance autonomous driving, the demand for stable 5G network connections and high-precision maps is increasing. Existing path planning methods are mostly based on traditional road network data and lack consideration of 5G base station network coverage. As a result, the communication stability and reliability of autonomous vehicles are affected in areas with poor 5G network signals.

[0003] Therefore, it is of great significance to provide a method that integrates 5G base station distribution data, high-precision maps and C-V2X technology to achieve a close combination of network coverage analysis and unmanned driving AI path planning, ensure that unmanned vehicles always maintain a stable 5G network connection during driving, and improve the operating efficiency and safety of unmanned vehicles in complex network environments. Summary of the Invention

[0004] In view of this, the present invention provides a method, system and intelligent driving device for generating a 5G base station navigation map and driving path.

[0005] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0006] The method for generating a 5G base station navigation map and driving path includes the following steps:

[0007] S1: Obtain 5G base station data information and download high-precision map data at the same time;

[0008] S2: Perform data cleaning, format unification and coordinate alignment preprocessing operations;

[0009] S3: Calculates signal strength based on 5G base station power and distance attenuation. Generates a network coverage heat map based on 5G base station network signal strength, marking areas with strong and weak signals. The generated heat map is input into the path planning algorithm, which dynamically adjusts the path using real-time data, detecting and marking areas with no or weak signal coverage.

[0010] S4: Initialize path weights based on 5G base station network quality and road conditions;

[0011] S5: The user inputs the starting and ending locations of the route planning. Based on the shortest path, network quality, and traffic conditions, the optimization goal of the route planning is determined. Multi-objective route planning is performed and the route is planned by using the A* / Dijkstra algorithm.

[0012] S6: The NPU unit dynamically adjusts the path weight based on the network heat map:

[0013] If the signal strength of the candidate path is lower than the threshold, its weight is reduced;

[0014] If the candidate path has a high signal strength, then that path is selected first;

[0015] S7: Output balances 5G network quality and driving distance, continuously optimizing the optimal driving path;

[0016] S8: Superimpose the optimal path onto the high-precision map to generate a visual optimal 5G base station driving map;

[0017] S9: To cope with sudden network outages, plan alternative routes to 5G base stations based on secondary network weights and traffic conditions; monitor base station status in real time, such as base station failures or congestion; and obtain real-time traffic data, including congestion and accident information.

[0018] S10: Triggers dynamic rerouting in the event of signal loss or traffic congestion, checking whether the route meets safety standards, such as avoiding signal blind spots;

[0019] S11: Mark the signal strength and risk areas on the map and output the path data to the unmanned driving control system.

[0020] Preferably, in step S1, the 5G base station data information obtained includes: location, coverage and signal strength; the high-precision map data includes: roads, obstacles and traffic rules;

[0021] It obtains 5G base station location, coverage, and signal strength data from the three major operators, integrates it with high-precision maps, and marks the base station location and coverage area. At the same time, it collects C-V2X device distribution data and incorporates it into the path planning system.

[0022] In step S2, a big data platform is used to perform data cleaning, format unification, and coordinate alignment to ensure data accuracy and consistency.

[0023] Preferably, in step S3, the base station data is used to establish a signal strength model, evaluate the network quality of each area, generate a network coverage heat map, and identify areas with strong and weak signals; combined with the C-V2X communication range data, the effectiveness of V2V and V2I communications is analyzed to form a comprehensive communication quality map.

[0024] Preferably, in step S4, the network quality 5G signal strength and C-V2X communication quality are used as weights for path planning, and areas with strong signals and good V2X communication are given priority; a multi-objective optimization algorithm is used to plan the path in combination with the shortest path, network quality and traffic conditions.

[0025] Preferably, in step S5, the base station status, C-V2X device status and traffic data are accessed in real time, the path is dynamically adjusted, and the driving path is continuously optimized according to the real-time network conditions and traffic information; the real-time update mechanism ensures the real-time and accuracy of path planning through the stream data processing framework.

[0026] Preferably, in step S8, the optimal path is visualized for reference by the driver or the system, and a WebGIS framework is used to display heat maps and paths, collect user feedback, and continuously optimize the algorithm and path planning.

[0027] Preferably, in step S9, an alternative path is planned to ensure that the vehicle can still travel safely when the network is interrupted or the C-V2X communication fails, and the path safety is regularly evaluated to ensure that it meets the safety standards for unmanned driving.

[0028] Preferably, in step S10, generative AI technology is used to collect and analyze historical path planning data and actual driving feedback to establish a learning model; the path planning algorithm is optimized through training to improve path selection accuracy and real-time adjustment capabilities;

[0029] The specific implementation methods and steps are as follows:

[0030] (1) Data collection and cleaning: Collect a large amount of historical path planning data and actual driving feedback of unmanned vehicles, including driving paths, network connection status, and traffic information; use data cleaning technology to remove invalid data and noise data to ensure data accuracy and completeness;

[0031] (2) Feature extraction and annotation: Extract key features from the collected data, such as path length, network signal strength, and traffic congestion; annotate the data to indicate which path selections are successful and which are unsatisfactory;

[0032] (3) Model selection and training: Select an appropriate generative AI model, such as generative adversarial networks (GANs) or variational autoencoders (VAEs), to learn the underlying distribution of historical path planning data. Input the labeled dataset into the model for training. During the training process, continuously adjust the model parameters to enable it to generate path planning solutions that meet the requirements.

[0033] (4) Model evaluation and optimization: Use the test data set to evaluate the trained model and calculate the accuracy and robustness of the model; based on the evaluation results, optimize the model, such as adjusting hyperparameters and increasing training data;

[0034] (5) Path generation and verification: Generate new path planning solutions using the trained generative AI model; compare and verify the generated paths with the actual driving conditions to evaluate their effectiveness;

[0035] If the generated path can improve the stability of network connections and optimize driving efficiency, it will be applied to actual autonomous driving systems;

[0036] (6) Real-time adjustment and feedback: During the actual driving process, network conditions and traffic information are collected in real time, and the generative AI model is used to dynamically adjust the path; the adjusted path is compared with the actual driving results, and user feedback is collected; based on the feedback data, the generative AI model is further optimized to improve its path selection and real-time adjustment capabilities.

[0037] The system for generating 5G base station navigation maps and driving routes includes:

[0038] Data acquisition module, data preprocessing module, network coverage modeling and analysis module, path planning initialization module, multi-target path planning module, real-time dynamic adjustment module, safety and redundancy processing module, visualization and output module;

[0039] The data acquisition module is used to obtain 5G base station data information and download high-precision map data information;

[0040] The data preprocessing module is used to perform data cleaning, format unification and coordinate alignment preprocessing operations;

[0041] The network coverage modeling and analysis module is used to calculate the base station signal strength model, generate the network coverage heat map, and detect and mark the signal coverage blind spots;

[0042] The path planning initialization module is used to input the path planning starting and ending positions, define the optimization target, and initialize the path weight;

[0043] The multi-objective path planning module is used to generate candidate paths, dynamically adjust path weights, and output a preliminary optimal path;

[0044] The real-time dynamic adjustment module is used to obtain base station status and traffic data in real time and dynamically re-plan the route;

[0045] The safety and redundancy processing module is used to check whether the multi-target path meets the safety standards, such as avoiding signal blind spots; to cope with sudden network interruptions, generate backup paths;

[0046] The visualization and output module is used to superimpose the optimal path on a high-precision map, mark the signal strength and risk areas on the map, and output the path data: the path data is output to the unmanned driving control system.

[0047] The intelligent driving device that generates 5G base station navigation maps and driving paths includes:

[0048] 5G+C-V2X full network communication module OBU vehicle communication unit, built-in GPS positioning module, IMU sensor, intelligent cockpit domain and intelligent driving domain control integrated control device;

[0049] Among them, the 5G+C-V2X full network communication module OBU vehicle communication unit is used to collect 5G base station network signal data and download navigation map data;

[0050] The intelligent cockpit domain and intelligent driving domain control integrated control device is used to detect the driving path track;

[0051] The intelligent cockpit domain and intelligent driving domain control integrated control device mainly includes:

[0052] Central control screen and instrument panel unit, multi-core heterogeneous AP processor, built-in GPU processor, high-computing-power NPU neural network processor, and large-capacity memory for running system programs and storage for data;

[0053] Among them, the central control screen and instrument panel unit are used for multi-core heterogeneous AP processors of smart cockpit applications to directly drive the external central control screen and digital instrument panel display through the MIPIDSI display interface. The central control screen is used to display navigation maps, multimedia entertainment and control of various ECU domain control units on the vehicle. The instrument panel is used to display mileage, power or fuel consumption status of unmanned vehicles;

[0054] When the AP processor executes the program, it implements the steps of generating a 5G base station navigation map and a method for generating a driving path of an unmanned vehicle;

[0055] The NPU neural network processor is used to execute the path planning related algorithms in the method.

[0056] Compared with the prior art, the present invention has achieved the following technical effects:

[0057] (1) By integrating 5G base station data, network coverage analysis, and path planning algorithms, and combining the learning and training functions of generative AI with C-V2X technology, the present invention can generate an optimal 5G network channel driving path map, ensuring the stability and reliability of the network connection of unmanned vehicles during driving, and improving operational efficiency and safety;

[0058] (2) The present invention is applicable to various complex traffic environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 A flow chart of the method for generating a 5G base station navigation map and driving route planning for the present invention;

[0060] Figure 2 A schematic diagram of a system for generating a 5G base station navigation map and driving route planning for the present invention;

[0061] Figure 3 To generate a 5G base station navigation map and driving path planning device architecture diagram for the present invention. DETAILED DESCRIPTION

[0062] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0063] like Figure 1 As shown, the present invention discloses a method for generating a 5G base station navigation map and a driving path, comprising the following steps:

[0064] S1: Obtain 5G base station data information, including location, coverage, and signal strength; and download high-precision map data, including roads, obstacles, and traffic rules;

[0065] It obtains 5G base station location, coverage, and signal strength data from the three major operators, integrates it with high-precision maps, and marks the base station location and coverage area. At the same time, it collects C-V2X device distribution data and incorporates it into the path planning system.

[0066] S2: Perform data cleaning, format unification and coordinate alignment preprocessing operations;

[0067] It uses big data platforms such as Hadoop and Spark for data cleaning, format unification and coordinate alignment to ensure data accuracy and consistency;

[0068] S3: Calculates signal strength based on 5G base station power and distance attenuation. Generates a network coverage heat map based on 5G base station network signal strength, marking areas with strong and weak signals. The generated heat map is input into the path planning algorithm, which dynamically adjusts the path using real-time data, detecting and marking areas with no or weak signal coverage.

[0069] It uses base station data to build a signal strength model, evaluate network quality in each area, generate a network coverage heat map, and identify areas of strong and weak signals. It also combines C-V2X communication range data to analyze the effectiveness of V2V and V2I communications and form a comprehensive communication quality map.

[0070] S4: Initialize path weights based on 5G base station network quality and road conditions;

[0071] It uses network quality, 5G signal strength, and C-V2X communication quality as weights for path planning, prioritizing areas with strong signals and good V2X communication. It uses multi-objective optimization algorithms, such as A* and Dijkstra, to plan paths, combining the shortest path, network quality, and traffic conditions. The algorithm can dynamically adjust weight parameters to accommodate different optimization objectives, such as path length, network signal strength, and traffic congestion.

[0072] S5: The user inputs the starting and ending locations of the route planning. Based on the shortest path, network quality, and traffic conditions, the optimization goal of the route planning is determined. Multi-objective route planning is performed and the route is planned by using the A* / Dijkstra algorithm.

[0073] It accesses base station status, C-V2X device status, and traffic data in real time, dynamically adjusts routes, and continuously optimizes driving routes based on real-time network conditions and traffic information. The real-time update mechanism can be implemented through a streaming data processing framework such as Apache Flink to ensure the real-time and accuracy of route planning.

[0074] S6: The NPU unit dynamically adjusts the path weight based on the network heat map:

[0075] If the signal strength of the candidate path is lower than the threshold, its weight is reduced;

[0076] If the candidate path has a high signal strength, then that path is selected first;

[0077] S7: Output balances 5G network quality and driving distance, continuously optimizing the optimal driving path;

[0078] S8: Superimpose the optimal path onto the high-precision map to generate a visual optimal 5G base station driving map;

[0079] It visualizes the best path for drivers or system reference, uses WebGIS frameworks such as Leaflet and Mapbox to display heat maps and paths, collects user feedback, and continuously optimizes algorithms and path planning;

[0080] S9: To cope with sudden network outages, plan alternative routes to 5G base stations based on secondary network weights and traffic conditions; monitor base station status in real time, such as base station failures or congestion; and obtain real-time traffic data, including congestion and accident information.

[0081] It plans backup routes to ensure safe driving in the event of network outages or C-V2X communication failures, and regularly assesses route safety to ensure compliance with autonomous driving safety standards.

[0082] S10: Triggers dynamic rerouting in the event of signal loss or traffic congestion, checking whether the route meets safety standards, such as avoiding signal blind spots;

[0083] It uses generative AI technology to collect and analyze historical path planning data and actual driving feedback to build a learning model; through training and optimizing the path planning algorithm, it improves path selection accuracy and real-time adjustment capabilities;

[0084] The specific implementation methods and steps are as follows:

[0085] (1) Data collection and cleaning: Collect a large amount of historical path planning data and actual driving feedback of unmanned vehicles, including driving paths, network connection status, and traffic information; use data cleaning technology to remove invalid data and noise data to ensure data accuracy and completeness;

[0086] (2) Feature extraction and annotation: Extract key features from the collected data, such as path length, network signal strength, and traffic congestion; annotate the data to indicate which path selections are successful and which are unsatisfactory;

[0087] (3) Model selection and training: Select an appropriate generative AI model, such as generative adversarial networks (GANs) or variational autoencoders (VAEs), to learn the underlying distribution of historical path planning data. Input the labeled dataset into the model for training. During the training process, continuously adjust the model parameters to enable it to generate path planning solutions that meet the requirements.

[0088] (4) Model evaluation and optimization: Use the test data set to evaluate the trained model and calculate the accuracy and robustness of the model; based on the evaluation results, optimize the model, such as adjusting hyperparameters and increasing training data;

[0089] (5) Path generation and verification: Generate new path planning solutions using the trained generative AI model; compare and verify the generated paths with the actual driving conditions to evaluate their effectiveness;

[0090] If the generated path can improve the stability of network connections and optimize driving efficiency, it will be applied to actual autonomous driving systems;

[0091] (6) Real-time adjustment and feedback: During the actual driving process, network conditions and traffic information are collected in real time, and the generative AI model is used to dynamically adjust the route; the adjusted route is compared with the actual driving results and user feedback is collected; based on the feedback data, the generative AI model is further optimized to improve its route selection and real-time adjustment capabilities;

[0092] S11: Mark the signal strength and risk areas on the map and output the path data to the unmanned driving control system.

[0093] like Figure 2As shown, the present invention also discloses a system for generating a 5G base station navigation map and a driving path, including:

[0094] Data acquisition module, data preprocessing module, network coverage modeling and analysis module, path planning initialization module, multi-target path planning module, real-time dynamic adjustment module, safety and redundancy processing module, visualization and output module;

[0095] The data acquisition module is used to obtain 5G base station data information and download high-precision map data information;

[0096] The data preprocessing module is used to perform data cleaning, format unification and coordinate alignment preprocessing operations;

[0097] The network coverage modeling and analysis module is used to calculate base station signal strength models, generate network coverage heat maps, and detect and mark signal coverage blind spots;

[0098] The path planning initialization module is used to input the starting and ending positions of the path planning, define the optimization target, and initialize the path weight;

[0099] The multi-objective path planning module is used to generate candidate paths, dynamically adjust path weights, and output the preliminary optimal path;

[0100] The real-time dynamic adjustment module is used to obtain base station status and traffic data in real time and dynamically re-plan the route;

[0101] The safety and redundancy processing module is used to check whether the multi-target paths meet safety standards, such as avoiding signal blind spots; and to generate backup paths to deal with sudden network interruptions;

[0102] The visualization and output module is used to overlay the optimal path onto a high-precision map, mark the signal strength and risk areas on the map, and output the path data: the path data is output to the unmanned driving control system.

[0103] like Figure 3 As shown, the present invention also discloses an intelligent driving device for generating a 5G base station navigation map and a driving path, including:

[0104] 5G+C-V2X full network communication module OBU vehicle communication unit, built-in GPS positioning module, IMU sensor, intelligent cockpit domain and intelligent driving domain control integrated control device;

[0105] Among them, the 5G+C-V2X full-network communication module OBU vehicle communication unit is used to collect 5G base station network signal data and download navigation map data;

[0106] The integrated control device of intelligent cockpit domain and intelligent driving domain control is used to detect driving path patterns;

[0107] The integrated control device for the intelligent cockpit domain and intelligent driving domain control mainly includes:

[0108] Central control screen and instrument panel unit, multi-core heterogeneous AP processor, built-in GPU processor, high-computing-power NPU neural network processor, and large-capacity memory for running system programs and storage for data;

[0109] Among them, the central control screen and instrument panel unit are used for multi-core heterogeneous AP processors in smart cockpit applications, which directly drive the external central control screen and digital instrument panel display through the MIPIDSI display interface. The central control screen is used to display navigation maps, multimedia entertainment and control of various ECU domain control units on the vehicle. The instrument panel is used to display mileage, power or fuel consumption status in unmanned vehicles.

[0110] When executing the program, the AP processor implements the steps of generating a 5G base station navigation map and a method for generating a driving path for an unmanned vehicle;

[0111] The NPU neural network processor is used to execute the path planning related algorithms in the method.

[0112] Working principle of the present invention:

[0113] When the system starts, the data collection module is initialized to obtain the latest 5G base station data from the API interfaces of the three major operators, including location, coverage and signal strength information; the acquired base station data is transmitted to the big data processing platform, and Hadoop or Spark is used to clean the data and standardize the format to ensure data consistency and accuracy; the processed base station data is matched with the high-precision map, and the location and coverage area of each base station are marked to generate comprehensive map data; the network coverage analysis module is started, and a signal strength model is established based on the base station data to generate a network coverage heat map and identify areas with strong and weak signals; the effectiveness of V2V and V2I communications is analyzed in combination with C-V2X communication range data to form a comprehensive communication quality map to provide a basis for path planning; the path planning algorithm module is started, the start and end point information is input, and a preliminary path is generated in combination with network quality, traffic conditions and the shortest path algorithm (such as A* or Dijkstra); the weight of the preliminary path is adjusted to give priority to areas with strong signals and V2X Generate the best path in areas with good communication; visualize the best path and present it to the driver or system through a WebGIS framework (such as Leaflet or Mapbox); access base station status, C-V2X device status and traffic data in real time to dynamically adjust the path to ensure the real-time and accuracy of the path; plan alternative paths and switch to alternative paths when the network is interrupted or C-V2X communication fails to ensure safe driving of the vehicle; regularly evaluate path safety to ensure that all paths meet the safety standards of unmanned driving; start the generative AI learning and training module, collect historical path planning data and actual driving feedback, and perform data cleaning and feature extraction; select a suitable generative AI model (such as GANs or VAEs), perform model training, and optimize the path planning algorithm; use the test data set to evaluate and optimize the model, adjust hyperparameters or increase training data; apply the optimized generative AI model to actual path generation, adjust the path in real time and collect user feedback to further optimize the model.

[0114] Example 1: City Expressway Navigation

[0115] On an urban expressway, a driverless vehicle needs to travel from starting point A to end point B;

[0116] Using this method, the system first collects 5G base station data along the expressway and generates a network coverage heat map. It finds that the network signal between sections C and D is weak, so the path planning algorithm prioritizes sections E to F with stronger signals. The final driving path is determined to be AEFDB, avoiding signal blind spots and ensuring stable communication and efficient driving on the expressway. At the same time, using C-V2X technology, vehicles can receive real-time traffic information from surrounding vehicles and infrastructure, further improving driving safety.

[0117] Example 2: Passing through a complex intersection

[0118] At a complex intersection, autonomous vehicles need to accurately judge and choose their driving path.

[0119] The system dynamically adjusts routes based on real-time updates of base station status and traffic data. In this case, due to a temporary outage in the north-south base station signal, the system quickly planned an alternate route, bypassing the intersection in the east-west direction, ensuring the vehicle could safely and smoothly pass through the intersection. Furthermore, C-V2X technology enables vehicles to communicate with other vehicles and traffic lights at the intersection, adjusting driving speed and safe distance in real time to avoid potential collisions.

[0120] Example 3: Highway Driving

[0121] On highways, autonomous vehicles need to maintain a stable network connection for a long time to ensure real-time data transmission and remote control.

[0122] Using this method, the system pre-plans a highway route with strong 5G signal coverage and makes fine-tuning adjustments during driving based on real-time network conditions. If weather conditions weaken the signal from some base stations on a certain section of highway, the system promptly adjusts the route, selecting a nearby parallel section with stronger signal, ensuring communication stability and driving safety. Furthermore, using C-V2X technology, vehicles can receive real-time speed and position information from vehicles ahead, enabling them to proactively slow down or change lanes, improving driving smoothness and safety.

[0123] Furthermore, numerous successful autonomous driving cases demonstrate the application value of this method. For example, Waymo, a Google subsidiary, has conducted extensive road tests with its autonomous vehicles in cities across the United States, accumulating over millions of miles. During these tests, Waymo's autonomous vehicles, leveraging advanced sensor technology and precise path planning, were able to navigate complex urban traffic environments safely and smoothly. Specifically, the 5G base station navigation map-based path generation technology provided by this method provides Waymo's autonomous vehicles with stable network connectivity and reliable path planning, effectively improving their operational efficiency and safety.

[0124] For example, China's Baidu Apollo autonomous driving project is also making continuous breakthroughs. Baidu Apollo has already launched autonomous taxi services in cities such as Beijing and Changsha, accumulating extensive operational experience. In these services, Baidu Apollo's autonomous vehicles utilize a variety of advanced technologies, including this method, to achieve precise environmental perception and efficient path planning. This method generates optimal driving paths using 5G network channels, providing stable network support for Baidu Apollo's autonomous vehicles and ensuring safe and smooth operation in various traffic conditions.

[0125] In summary, this method can flexibly generate the optimal 5G network channel driving path based on different traffic environments and network conditions, providing stable and reliable network connections and efficient path planning for unmanned vehicles, ensuring the stability and reliability of network connections during driving, improving operational efficiency and safety; and is applicable to various complex traffic environments.

[0126] The above description is merely a preferred embodiment of the present invention and does not limit the technical scope of the present invention. Therefore, any minor modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A method for generating a 5G base station navigation map and a driving path, characterized in that: The steps include: S1: Obtain 5G base station data information and download high-precision map data at the same time; S2: Perform data cleaning, format unification and coordinate alignment preprocessing operations; S3: Calculates signal strength based on 5G base station power and distance attenuation. Generates a network coverage heat map based on 5G base station network signal strength, marking areas with strong and weak signals. The generated heat map is input into the path planning algorithm, which dynamically adjusts the path using real-time data, detecting and marking areas with no or weak signal coverage. S4: Initialize path weights based on 5G base station network quality and road conditions; S5: The user inputs the starting and ending locations of the route planning. Based on the shortest path, network quality, and traffic conditions, the optimization goal of the route planning is determined. Multi-objective route planning is performed and the route is planned by using the A* / Dijkstra algorithm. S6: The NPU unit dynamically adjusts the path weight based on the network heat map: If the signal strength of the candidate path is lower than the threshold, its weight is reduced; If the candidate path has a high signal strength, then that path is selected first; S7: Output balances 5G network quality and driving distance, continuously optimizing the optimal driving path; S8: Superimpose the optimal path onto the high-precision map to generate a visual optimal 5G base station driving map; S9: To cope with sudden network outages, plan alternative routes to 5G base stations based on secondary network weights and traffic conditions; monitor base station status in real time, such as base station failures or congestion; and obtain real-time traffic data, including congestion and accident information. S10: Triggers dynamic rerouting when signal loss or traffic congestion occurs, checking whether the route meets safety standards, such as avoiding signal blind spots; S11: Mark the signal strength and risk areas on the map and output the path data to the unmanned driving control system.

2. The method for generating a 5G base station navigation map and driving path according to claim 1, characterized in that: In step S1, the 5G base station data information obtained includes: location, coverage and signal strength; the high-precision map data includes: roads, obstacles and traffic rules; It obtains 5G base station location, coverage, and signal strength data from the three major operators, integrates it with high-precision maps, and marks the base station location and coverage area. At the same time, it collects C-V2X device distribution data and incorporates it into the path planning system. In step S2, a big data platform is used to perform data cleaning, format unification, and coordinate alignment to ensure data accuracy and consistency.

3. The method for generating a 5G base station navigation map and driving path according to claim 1, characterized in that: In step S3, a signal strength model is established using base station data to evaluate the network quality of each area, generate a network coverage heat map, and identify areas with strong and weak signals. Combined with C-V2X communication range data, the effectiveness of V2V and V2I communications is analyzed to form a comprehensive communication quality map.

4. The method for generating a 5G base station navigation map and driving path according to claim 1, characterized in that: In step S4, the network quality 5G signal strength and C-V2X communication quality are used as weights for path planning, and areas with strong signals and good V2X communication are given priority. A multi-objective optimization algorithm is used to plan the path based on the shortest path, network quality, and traffic conditions.

5. The method for generating a 5G base station navigation map and driving path according to claim 1, characterized in that: In step S5, the base station status, C-V2X device status, and traffic data are accessed in real time to dynamically adjust the route and continuously optimize the driving route based on the real-time network conditions and traffic information. The real-time update mechanism ensures the real-time and accuracy of route planning through the streaming data processing framework.

6. The method for generating a 5G base station navigation map and driving path according to claim 1, characterized in that: In step S8, the optimal path is visualized for reference by the driver or the system, and a WebGIS framework is used to display heat maps and paths, collect user feedback, and continuously optimize the algorithm and path planning.

7. The method for generating a 5G base station navigation map and driving path according to claim 1, characterized in that: In step S9, an alternative route is planned to ensure that the vehicle can still travel safely when the network is interrupted or the C-V2X communication fails. The path safety is regularly evaluated to ensure that it meets the safety standards for unmanned driving.

8. The method for generating a 5G base station navigation map and driving path according to claim 1, characterized in that: In step S10, generative AI technology is used to collect and analyze historical path planning data and actual driving feedback to establish a learning model; the path planning algorithm is optimized through training to improve path selection accuracy and real-time adjustment capabilities; The specific implementation methods and steps are as follows: (1) Data collection and cleaning: Collect a large amount of historical path planning data and actual driving feedback of unmanned vehicles, including driving paths, network connection status, and traffic information; use data cleaning technology to remove invalid data and noise data to ensure data accuracy and completeness; (2) Feature extraction and annotation: Extract key features from the collected data, such as path length, network signal strength, and traffic congestion; annotate the data to indicate which path selections are successful and which are unsatisfactory; (3) Model selection and training: Select an appropriate generative AI model, such as generative adversarial networks (GANs) or variational autoencoders (VAEs), to learn the underlying distribution of historical path planning data; input the labeled dataset into the model for training; During the training process, the model parameters are continuously adjusted so that it can generate a path planning solution that meets the requirements; (4) Model evaluation and optimization: Use the test data set to evaluate the trained model and calculate the accuracy and robustness of the model; based on the evaluation results, optimize the model, such as adjusting hyperparameters and increasing training data; (5) Path generation and verification: Generate new path planning solutions using the trained generative AI model; compare and verify the generated paths with the actual driving conditions to evaluate their effectiveness; If the generated path can improve the stability of network connections and optimize driving efficiency, it will be applied to actual autonomous driving systems; (6) Real-time adjustment and feedback: During the actual driving process, network conditions and traffic information are collected in real time, and the generative AI model is used to dynamically adjust the path; the adjusted path is compared with the actual driving results, and user feedback is collected; based on the feedback data, the generative AI model is further optimized to improve its path selection and real-time adjustment capabilities.

9. A system for generating a 5G base station navigation map and driving path, characterized in that: include: Data acquisition module, data preprocessing module, network coverage modeling and analysis module, path planning initialization module, multi-target path planning module, real-time dynamic adjustment module, safety and redundancy processing module, visualization and output module; The data acquisition module is used to obtain 5G base station data information and download high-precision map data information; The data preprocessing module is used to perform data cleaning, format unification and coordinate alignment preprocessing operations; The network coverage modeling and analysis module is used to calculate the base station signal strength model, generate the network coverage heat map, and detect and mark the signal coverage blind spots; The path planning initialization module is used to input the path planning starting and ending positions, define the optimization target, and initialize the path weight; The multi-objective path planning module is used to generate candidate paths, dynamically adjust path weights, and output a preliminary optimal path; The real-time dynamic adjustment module is used to obtain base station status and traffic data in real time and dynamically re-plan the route; The safety and redundancy processing module is used to check whether the multi-target path meets the safety standards, such as avoiding signal blind spots; to cope with sudden network interruptions, generate backup paths; The visualization and output module is used to superimpose the optimal path on a high-precision map, mark the signal strength and risk areas on the map, and output the path data: the path data is output to the unmanned driving control system.

10. An intelligent driving device that generates a 5G base station navigation map and driving path, characterized in that: include: 5G+C-V2X full network communication module OBU vehicle communication unit, built-in GPS positioning module, IMU sensor, intelligent cockpit domain and intelligent driving domain control integrated control device; Among them, the 5G+C-V2X full network communication module OBU vehicle communication unit is used to collect 5G base station network signal data and download navigation map data; The intelligent cockpit domain and intelligent driving domain control integrated control device is used to detect the driving path track; The intelligent cockpit domain and intelligent driving domain control integrated control device mainly includes: Central control screen and instrument panel unit, multi-core heterogeneous AP processor, built-in GPU processor, high-computing-power NPU neural network processor, and large-capacity memory for running system programs and storage for data; Among them, the central control screen and instrument panel unit are used for multi-core heterogeneous AP processors of smart cockpit applications to directly drive the external central control screen and digital instrument panel display through the MIPIDSI display interface. The central control screen is used to display navigation maps, multimedia entertainment and control of various ECU domain control units on the vehicle. The instrument panel is used to display mileage, power or fuel consumption status of unmanned vehicles; When the AP processor executes the program, it implements the steps of generating a 5G base station navigation map and a method for generating a driving path of an unmanned vehicle; The NPU neural network processor is used to execute the path planning related algorithms in the method.

Images