A method for accessing a vehicle system and navigating across systems by a user device

Abstract

The application provides a method for cross-system virtual access of user equipment to a vehicle-mounted system and navigation, comprising: in the cross-system virtual access process, converting operation requests and data transmission instructions of the user equipment into a format recognizable by the vehicle-mounted system through virtualization middleware technology, and encrypting and integrity checking the transmission content to ensure compatibility and security of communication; establishing a heartbeat detection mechanism between the vehicle-mounted system and the user equipment, evaluating network connection quality in real time by regularly sending probe packets and monitoring response time, automatically switching to a backup link if timeout or high packet loss rate is found to ensure seamless connection and service continuity; analyzing historical load data of the vehicle-mounted system through a machine learning algorithm, constructing a load prediction model, and dynamically scheduling and reserving computing resources and network bandwidth in advance according to the spatio-temporal distribution characteristics of navigation requests to avoid resource bottlenecks and service congestion.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of information technology, and in particular to a method for virtually accessing a vehicle-mounted system by a user device across systems and navigation. BACKGROUND

[0002] In the scenario of virtually accessing a vehicle-mounted system by a user device across systems and using a navigation function, there is a key technical contradiction: how to dynamically adjust and accurately match the computing resources and network bandwidth of the vehicle-mounted system according to the real-time behaviors and differentiated needs of different users. The root cause of this contradiction lies in the heterogeneity between the user device and the vehicle-mounted system in terms of software and hardware architecture, operating system, network protocol, etc., and the use habits, operation behaviors, and data transmission needs of different users are also different and changeable at any time. The vehicle-mounted navigation system must realize real-time perception, intelligent prediction, and dynamic adaptation of user behaviors and resource needs, and continuously optimize strategies and upgrade algorithms in complex and variable application scenarios, in order to balance the real-time performance, smoothness, and safety of services. This puts forward very high requirements for the adaptation of heterogeneous environments across systems, the collection and analysis of user behavior data, the formulation and execution of resource scheduling strategies, and the compatibility and collaboration of system components. At the same time, the particularity of the vehicle-mounted scenario must be fully considered, the relationship between navigation performance and driving safety must be accurately balanced, different levels of quality of service must be provided, and sudden extreme situations must be properly handled. These all bring many technical challenges to the realization of virtually accessing a vehicle-mounted navigation system across systems, and require in-depth exploration and continuous innovation. SUMMARY

[0003] The present application provides a method for virtually accessing a vehicle-mounted system by a user device across systems and navigation, mainly including:

[0004] By collecting data indicators such as CPU occupancy, memory usage, and network throughput of the user device, combining historical operation records and application use habits of the user, and using machine learning algorithms such as support vector machines or random forests, the user behavior patterns are modeled and predicted to obtain the real-time demand feature vector of the user;

[0005] According to the real-time demand feature vector of the user, a heuristic optimization algorithm such as genetic algorithm is used to dynamically generate the optimal computing resource and network bandwidth allocation strategy by considering factors such as the hardware and software performance, network status, and load balancing of the vehicle-mounted system, and to determine the priority and resource occupancy ratio of different user requests;

[0006] In the process of virtually accessing across systems, the operation request and data transmission instruction of the user device are converted into a format recognizable by the vehicle-mounted system through virtualization middleware technology, and the transmission content is encrypted and integrity checked to ensure the compatibility and security of communication;

[0007] To address the high real-time requirements of navigation services, edge computing technology is adopted to deploy navigation data caching and computing modules on local nodes of the in-vehicle system. By quickly matching real-time traffic information with user location, low-latency navigation route planning and voice guidance services are provided.

[0008] Establish a heartbeat detection mechanism between the vehicle system and user equipment. By periodically sending probe packets and monitoring response time, the network connection quality is evaluated in real time. If a timeout or excessive packet loss rate is detected, the system automatically switches to a backup link to ensure seamless connection and service continuity.

[0009] In the resource management module of the vehicle system, a dynamic task priority adjustment mechanism is set up. Based on the urgency of user requests and the quality feedback of navigation services, the computing resources and network bandwidth usage ratios of different tasks are adjusted in real time to ensure priority response and processing of critical services.

[0010] By analyzing historical load data of the vehicle system through machine learning algorithms, a load prediction model is built. Based on the spatiotemporal distribution characteristics of navigation requests, computing resources and network bandwidth are dynamically scheduled and reserved in advance to avoid resource bottlenecks and service blockages.

[0011] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0012] This invention discloses a method for virtually accessing a vehicle system across systems for navigation, solving the problem of unreliable service quality caused by diverse user needs, limited vehicle system resources, and unstable network connections. First, this invention models user behavior patterns using machine learning algorithms to obtain real-time user demand feature vectors. Then, combining factors such as network conditions, vehicle system hardware and software performance, and load balancing, it uses a genetic algorithm to dynamically generate optimal computing resources and network bandwidth allocation strategies. A dynamic task priority adjustment mechanism is established to adjust resource allocation in real time based on the urgency of user requests and navigation service quality feedback, ensuring priority for critical services. Simultaneously, to address the real-time requirements of navigation services, edge computing technology is employed to deploy navigation data and computing modules locally, achieving low-latency navigation. Furthermore, virtualization middleware technology enables cross-system virtual access, ensuring communication compatibility and security. Finally, a heartbeat detection mechanism and load prediction model are established to ensure network connection stability and effective resource allocation. This invention ultimately achieves dynamic optimization of vehicle system resource allocation, improving the user service experience, especially in critical application scenarios such as navigation services, ensuring service real-time performance and stability. Attached Figure Description

[0013] Fig. 1 This is a flowchart of a method for virtually connecting user equipment across systems to a vehicle system and navigating, according to the present invention.

[0014] Fig. 2 This is a schematic diagram of a method for virtually connecting a user device to a vehicle system for navigation across systems according to the present invention.

[0015] Fig. 3 This is another schematic diagram of a method for virtually accessing a user device across systems and navigating a vehicle system according to the present invention. Detailed Implementation

[0016] To further understand the content of this invention, a detailed description of the invention is provided in conjunction with the accompanying drawings and embodiments. The specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention. It should also be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0017] like Figs. 1-3 This embodiment of a method for virtually accessing a user device across systems to a vehicle system and providing navigation may specifically include:

[0018] S101. By collecting data indicators such as CPU utilization, memory usage, and network throughput of user devices, and combining them with users' historical operation records and application usage habits, machine learning algorithms such as support vector machines or random forests are used to model and predict user behavior patterns to obtain the user's real-time demand feature vector.

[0019] The system acquires performance metrics data such as CPU utilization, memory usage, and network throughput of user devices, and categorizes and stores these metrics according to their type. Simultaneously, it analyzes users' historical operation records to obtain application usage pattern data, such as frequency and duration of use for different applications. The acquired performance metrics data and application usage pattern data are then fused to generate a user behavior dataset. This dataset undergoes data cleaning and preprocessing to remove noise and outliers, ensuring data quality. Feature engineering is performed on the preprocessed dataset to extract key features such as application usage frequency, usage duration, and device performance metrics. Feature selection and dimensionality reduction are then performed to obtain an optimized user behavior feature set. A logistic regression algorithm is used to train the optimized user behavior feature set, establishing a user behavior prediction model. Cross-validation is used to evaluate the model, calculating its prediction accuracy. If the prediction accuracy falls below a preset threshold (e.g., 80%), a decision tree algorithm is used to retrain the model to improve prediction performance. The real-time collected user device performance metrics and application usage data undergo the same data preprocessing and feature extraction process, and are then input into a trained user behavior prediction model to obtain real-time user demand prediction results. Based on these real-time demand prediction results and combined with historical user preference data, a user demand feature vector is generated. This feature vector is then transmitted to a collaborative filtering-based recommendation engine, which uses user similarity and item similarity calculations to provide users with personalized application recommendations and services. User feedback on the recommendation results is continuously tracked, and user behavior data such as clicks, downloads, and usage are collected to optimize the recommendation model and improve recommendation quality. Simultaneously, the user behavior prediction model is regularly updated to adapt to changes in user needs.

[0020] For example, performance metrics data such as CPU utilization, memory usage, and network throughput of user devices are acquired and categorized for storage based on different types of performance metrics. For instance, CPU utilization can be stored as a time series, recorded every minute; memory usage can be divided into used memory and available memory; and network throughput can be stored as upload speed and download speed. Categorized storage facilitates subsequent data analysis and processing. By analyzing users' historical operation records, application usage pattern data such as frequency and duration of use for different applications can be obtained. For example, user A uses WeChat for 2 hours a day, 10 times a day; user B uses Douyin for 1 hour a day, 5 times a day. This data reflects different application usage habits. The acquired performance metric data and application usage pattern data are then fused to generate a user behavior dataset. For example, user A's CPU utilization data, memory usage data, and network throughput data are fused with their WeChat usage time and frequency data to form user A's behavior dataset. The purpose of data fusion is to integrate data from different dimensions to gain a more comprehensive understanding of user behavior. The generated user behavior dataset undergoes data cleaning and preprocessing to remove noisy data and outliers, ensuring data quality. For example, if CPU utilization suddenly spikes to 100% for a short period, it can be considered noise and removed. Data cleaning improves data quality and avoids the impact of outliers on model training. Feature engineering is then performed on the preprocessed user behavior dataset to extract key features such as application usage frequency, usage duration, and device performance metrics. Feature selection and dimensionality reduction are then performed to obtain an optimized set of user behavior features. For example, the duration and frequency of a user's WeChat usage can be used as features, along with average CPU utilization and average network throughput. The purpose of feature engineering is to transform the raw data into features that the model can understand and to select the most effective subset of features. A logistic regression algorithm is then used to train the optimized set of user behavior features to build a user behavior prediction model. Logistic regression is a commonly used classification algorithm that can predict whether a user will perform a certain action. For example, it can predict whether a user will use a particular application within the next hour based on their characteristics. Cross-validation is used to evaluate the model and calculate its prediction accuracy. Cross-validation can avoid overfitting and more accurately evaluate the model's generalization ability. For example, the dataset can be divided into five parts, with four parts used to train the model and one part used for testing each time. The average of the five test results is then used as the model's prediction accuracy. If the prediction accuracy is lower than a preset threshold (e.g., 80%), the model is retrained using a decision tree algorithm to improve prediction performance. Decision trees are a non-linear classification algorithm that may be more suitable than logistic regression for certain datasets. Choosing an appropriate algorithm can improve the model's prediction accuracy.The real-time collected user device performance metrics and application usage data undergo the same data preprocessing and feature extraction process, and are then input into a trained user behavior prediction model to obtain real-time user demand predictions. For example, if a user currently has high CPU usage and memory consumption, and is using a video editing application, the model might predict that the user needs more computing resources or storage space. Based on the real-time user demand predictions and combined with the user's historical preference data, a user demand feature vector is generated. For example, if the model predicts that the user needs more storage space, and the user has frequently used cloud storage services in the past, then "needs storage space" and "prefers cloud storage" can be included as part of the feature vector. The feature vector can more comprehensively describe the user's needs. The feature vector is then transmitted to a collaborative filtering-based recommendation engine, which provides personalized application recommendations and services to users through user similarity calculations and item similarity calculations. For example, if user A and user B have similar feature vectors, and user B frequently uses a certain cloud storage application, then that application can be recommended to user A. Collaborative filtering can make recommendations based on user similarity and item similarity. Continuously track user feedback on recommendation results, collecting user behavior data such as clicks, downloads, and usage to optimize the recommendation model and improve recommendation quality. For example, if users frequently click on recommended apps, the recommendations are considered relatively accurate, allowing for further model learning. User feedback helps the model continuously optimize and improve recommendation accuracy. Regularly update the user behavior prediction model to adapt to changes in user needs. For example, users' app usage habits may change over time, requiring regular model updates to capture these changes. Regular model updates ensure the model's effectiveness.

[0021] S102. Based on the user's real-time demand feature vector, a heuristic optimization algorithm such as a genetic algorithm is used to dynamically generate the optimal computing resources and network bandwidth allocation strategy by comprehensively considering factors such as the hardware and software performance of the vehicle system, network conditions, and load balancing, and to determine the priority and resource usage ratio of different user requests.

[0022] Step 1: Obtain the user demand feature vector. Parse the user request content, extract key parameters, quantify user needs, and form a multi-dimensional feature vector. Each dimension represents a different aspect of the user's needs, such as required computing resources, network bandwidth, latency requirements, etc.

[0023] Step 2: Assess the status of the vehicle system. Monitor key indicators such as CPU utilization, memory usage, available bandwidth, and number of connected users in real time to comprehensively evaluate the overall hardware and software performance and network status of the system.

[0024] Step 3: Calculate the system load balancing index. Based on the current system status and user requests, calculate the system load balancing index. This index reflects the allocation of system resources and the degree of load balancing. If the system load balancing index is lower than a preset threshold, proceed to Step 4; if the system load balancing index is higher than the preset threshold, proceed to Step 5.

[0025] Step 4: Dynamically Adjust Resource Allocation Strategy. Using a genetic algorithm, with the goal of maximizing the system load balancing index, the optimal allocation strategy for computing resources and network bandwidth is searched. The fitness function of the genetic algorithm integrates information such as the system load balancing index, user demand feature vectors, and system state. Through iterative calculation using the genetic algorithm, the optimal resource allocation strategy is obtained.

[0026] Step 5: Determine user request priorities. Based on the user demand feature vector and the current system resource availability, calculate the priority of each user request. Higher-priority user requests will receive resource allocation first.

[0027] Step Six: Allocate Resource Ratios. Based on the priority of user requests and the optimal resource allocation strategy, determine the proportion of computing resources and network bandwidth usage for each user request.

[0028] Step 7: Issue resource allocation instructions. Instructions for allocating computing resources and network bandwidth are issued to the vehicle system to achieve dynamic resource allocation and priority processing of user requests.

[0029] For example, obtaining a user demand feature vector means transforming the user's request into data that the computer can understand and process. For instance, if a user wants to search for nearby restaurants on an in-vehicle map, this request can be broken down into several key parameters: search type (restaurant), geographic location (current location), and search range (nearby). These parameters can then be quantified into a multi-dimensional feature vector, for example: [search type = 2 (assuming the restaurant corresponds to the number 2), longitude = 120.123, latitude = 30.456, search radius = 5 kilometers]. Each dimension represents a different aspect of the user's demand, allowing the system to analyze and process it. The benefit of this is that the system can more accurately understand user needs, thereby providing services that better meet user expectations. Assessing the status of an in-vehicle system is like a doctor performing a physical examination. It requires real-time monitoring of various key indicators of the in-vehicle system, such as CPU utilization, memory usage, available bandwidth, and the number of connected users. For example, the current CPU utilization is 60%, memory usage is 40%, available bandwidth is 10Mbps, and the number of connected users is 2. These indicators collectively reflect the overall health of the system, that is, the system's hardware and software performance and network status. These metrics help determine whether the system can meet user needs. Calculating the system load balancing index measures the rationality of system resource allocation. This index acts like a balance scale, calculating the degree of load balancing based on the current system state and user requests. Assuming the system load balancing index ranges from 0 to 1, a value closer to 1 indicates a more balanced load. If the current index is 0.8, it means the system resource allocation is relatively balanced and can handle user requests normally; if the index is below a preset threshold, such as 0.2, it indicates an unbalanced system load, and resource allocation strategies may need adjustment. Dynamically adjusting resource allocation strategies aims to make more efficient use of system resources. When the system load balancing index falls below the preset threshold, resource allocation strategies need to be adjusted. At this point, a genetic algorithm can be used to find the optimal resource allocation scheme. The genetic algorithm acts like an intelligent resource scheduler; it continuously tries different resource allocation schemes based on information such as the system load balancing index, user demand feature vectors, and system state, ultimately finding the optimal solution. For example, it might allocate more computing resources to higher-priority user requests or transfer some user requests to servers with lower load. Prioritizing user requests is crucial for satisfying more important user requests when resources are limited. Each user request may have varying importance, therefore, the priority of each request needs to be calculated based on the user's demand characteristic vector and the current availability of system resources. For example, an emergency call for help will certainly have a higher priority than a regular navigation request. Assuming we use a number from 1 to 10 to represent priority, with higher numbers indicating higher priority, then an emergency call for help might have a priority of 10, while a navigation request might have a priority of 5.Resource allocation aims to rationally distribute limited resources among different user requests. Based on the priority of user requests and the optimal resource allocation strategy, the proportion of computing resources and network bandwidth used for each user request can be determined. For example, an emergency call request with a priority of 10 might consume 80% of computing resources and 90% of available bandwidth, while a navigation request with a priority of 5 might only consume 10% of computing resources and 5% of available bandwidth. Issuing resource allocation instructions is to put the resource allocation scheme into practice. After determining the resource proportion for each user request, the system will issue computing resource and network bandwidth allocation instructions to the vehicle system. For example, it might allocate more CPU resources to the process handling emergency call requests and allocate more network bandwidth to the channel transmitting emergency call data. This achieves dynamic resource allocation and priority processing of user requests.

[0030] S103. During cross-system virtual access, virtualization middleware technology is used to convert user equipment operation requests and data transmission instructions into a format recognizable by the vehicle system, and the transmitted content is encrypted and its integrity is verified to ensure communication compatibility and security.

[0031] According to a pre-established cross-system virtual access protocol, the system acquires operation requests and data transmission instructions from user equipment, and parses and extracts key information. Using virtualization middleware technology, the extracted key information is matched and converted against a preset in-vehicle system-recognizable format to generate operation requests and data transmission instructions that meet the in-vehicle system's requirements. The converted operation requests and data transmission instructions are then encrypted using an encryption algorithm to ensure the confidentiality and tamper-proof nature of the transmitted content. Based on this encryption, an integrity verification algorithm calculates a checksum for the transmitted content and sends it along with the encrypted content to the in-vehicle system. Upon receiving the encrypted content and checksum, the in-vehicle system first decrypts the received content to reconstruct the original operation requests and data transmission instructions. It then performs an integrity verification on the decrypted content by recalculating the checksum and comparing it with the received checksum to determine if the transmitted content is complete and error-free. If the integrity verification passes, the decrypted operation requests and data transmission instructions are transmitted to the corresponding modules of the in-vehicle system for processing and execution; if the integrity verification fails, an error message is sent to the user equipment, requesting a resend of the operation requests and data transmission instructions.

[0032] For example, a cross-system virtual access protocol (CVR) acts like a universal translator, understanding the "language" of different devices and translating it into commands that the in-vehicle system can understand. For instance, a user plays a song using a music app on their phone, sending the playback command to the in-vehicle system via Bluetooth. This command might be in a specific data format on the phone, but the CVR converts it into a command format that the in-vehicle system can recognize, such as "Play song XXX". Parsing and extracting key information is like extracting key content from a letter. For example, if a user sends a navigation command on their phone, "Navigate to the nearest gas station," the CVR parses this command and extracts key information: navigation and destination (nearest gas station). Virtualization middleware technology acts like an adapter, converting information in different formats into a unified format. For example, the format of navigation commands sent by a mobile navigation application might differ from the format required by the in-vehicle navigation system. The virtualization middleware converts the mobile navigation command into a format that the in-vehicle navigation system can recognize, such as converting "Navigate to the nearest gas station" into a standardized data format containing destination coordinates, navigation route, and other information. To protect user information security, the converted commands need to be encrypted. For example, the command "Play song XXX" is encrypted using the AES encryption algorithm, converting it into a ciphertext to prevent the command from being stolen or tampered with during transmission. Integrity verification is like a seal on an envelope, ensuring that information has not been altered during transmission. For instance, calculating a checksum for the encrypted command "Play song XXX" is like stamping the envelope. When the vehicle system receives the command, it recalculates the checksum and compares it with the received checksum. If the two values ​​match, the command has not been tampered with. After receiving the encrypted command and checksum, the vehicle system first needs to decrypt it, like opening an envelope to restore the ciphertext to the original command "Play song XXX". After decryption, the vehicle system performs an integrity check, like checking if the seal on the envelope is intact. If the checksums match, the command has not been tampered with during transmission and can be executed safely. If the integrity check fails, it indicates that the command may have been tampered with or corrupted during transmission. The vehicle system will send an error message to the user device, requesting a resend of the command, much like notifying a sender that a letter has been damaged and needs to be resent. For example, if the checksum calculated by the vehicle system does not match the received checksum, it will send an error message to the phone, requesting a resend of the navigation command. Finally, if the integrity check passes, the vehicle system will pass the command to the appropriate module for processing and execution. For example, passing the command "Play song XXX" to the multimedia module will start playing the song.

[0033] S104. To address the high real-time requirements of navigation services, edge computing technology is adopted to deploy navigation data caching and computing modules on local nodes of the vehicle system. Through rapid matching of real-time traffic information and user location, low-latency navigation route planning and voice guidance services are provided.

[0034] The system acquires the vehicle's geographic location and driving direction information as initial input data. Based on the vehicle's geographic location information, it retrieves road network data of the surrounding area from the navigation data cache of the local node. Real-time traffic information for the area, including traffic congestion and accidents, is obtained from the traffic management center via a wireless communication module. The road network data and real-time traffic information are fused to construct a dynamic road network model for subsequent route planning. An improved Dijkstra algorithm is used to search for paths in the dynamic road network model, resulting in multiple alternative routes from the origin to the destination. The alternative routes are scored and ranked according to user preferences, and the optimal route is selected by comprehensively considering factors such as route length, estimated travel time, and road comfort. The navigation information of the optimal route is converted into voice prompts and provided to the user via the vehicle's audio system, with the navigation information dynamically updated based on the driving progress.

[0035] For example, the system acquires the vehicle's geographic location and driving direction information as initial input data. For instance, the vehicle's GPS module might determine that the vehicle is currently located at 31.2297°N, 121.4737°E, and driving due north. This information forms the basis for route planning. Obtaining the driving direction helps the system more accurately determine the vehicle's current road location and driving intention, avoiding misguided directions at complex intersections. Based on the vehicle's geographic location information, the system retrieves surrounding road network data from the local node's navigation data cache. For example, after acquiring the vehicle's current location, the system extracts all road information within a 1-kilometer radius of the current location from the navigation data cache stored in the vehicle's system, including road names, road classifications, road directions, connections, speed limits, etc. This reduces reliance on cloud servers and improves navigation response speed. Assuming the vehicle is near People's Square in Shanghai, the system would extract information on surrounding roads such as People's Avenue and Nanjing West Road. Real-time traffic information for the area, including traffic congestion and accidents, is obtained from the traffic management center via a wireless communication module. For example, through 4G / 5G networks, the vehicle-mounted system can receive real-time traffic information released by the traffic management center, such as: congestion on Renmin Avenue heading east to west, with an average speed of 15 km / h; a traffic accident on Nanjing West Road, with an estimated processing time of 30 minutes. Obtaining real-time traffic information helps the system avoid congested and accident-prone sections. By fusing road network data with real-time traffic information, a dynamic road network model is constructed for subsequent route planning. For instance, the system combines road information for Renmin Avenue heading east to west with congestion information from real-time traffic updates, marking it as congested and increasing the estimated travel time. Meanwhile, accident information on Nanjing West Road will cause that section to be temporarily marked as impassable. In this way, the dynamic road network model can reflect the real-time traffic conditions. An improved Dijkstra algorithm is used to perform path search on the dynamic road network model, obtaining multiple alternative paths from the origin to the destination. For example, assuming the vehicle's destination is Shanghai Railway Station, the system will use an improved Dijkstra algorithm based on a dynamic road network model to search for multiple alternative routes, such as those via Yan'an Elevated Road, North-South Elevated Road, and surface roads. The improved Dijkstra algorithm incorporates real-time traffic information into the route search process, thus finding a more suitable route. The alternative routes are scored and ranked according to user preferences, taking into account factors such as route length, estimated travel time, and road comfort to select the optimal route. For example, if the user's preference is "fastest arrival," the system will prioritize the route with the shortest estimated travel time; if the user's preference is "avoid highways," the system will prioritize routes that do not include highways.Assuming the estimated travel time via the Yan'an Elevated Road is 20 minutes, the estimated travel time via the North-South Elevated Road is 25 minutes, and the estimated travel time via ground-level roads is 30 minutes, the system will select the route via the Yan'an Elevated Road based on the user's "fastest arrival" preference. The navigation information for the optimal route will be converted into voice prompts, providing real-time voice guidance through the in-car audio system, and dynamically updating the navigation information based on the driving progress. For example, the system will convert the selected route via the Yan'an Elevated Road into a series of voice prompts, such as: "Turn right onto the Yan'an Elevated Road in 500 meters," "Drive 2 kilometers on the Yan'an Elevated Road," and "Exit at the Yan'an Elevated Road exit." Simultaneously, the system will dynamically update the navigation information based on the vehicle's progress and real-time traffic conditions, and will replan the route when necessary. For example, if a sudden traffic jam occurs ahead, the system will replan the route and inform the user via voice prompts.

[0036] S105. Establish a heartbeat detection mechanism between the vehicle system and user equipment. By periodically sending probe packets and monitoring response time, the network connection quality is evaluated in real time. If a timeout or excessive packet loss rate is detected, the system automatically switches to a backup link to ensure seamless connection and service continuity.

[0037] A bidirectional communication link is established between the vehicle-mounted system and the user equipment, and a heartbeat detection protocol is agreed upon, specifying the probe packet format, transmission frequency, and timeout threshold. The vehicle-mounted system periodically sends probe packets to the user equipment, while simultaneously starting a timer to record the transmission time. Upon receiving a probe packet, the user equipment immediately replies with an acknowledgment packet. The vehicle-mounted system receives the acknowledgment packet, calculates the round-trip time, and updates the average response time. If the vehicle-mounted system fails to receive an acknowledgment packet after sending a probe packet and the preset timeout threshold has elapsed, it is considered packet loss, and the packet loss rate is updated. Based on the calculated average response time and packet loss rate, the vehicle-mounted system uses a weighted average algorithm to evaluate the current network quality in real time. If the network quality evaluation result is lower than the preset quality threshold and the duration exceeds the predetermined tolerance period, the primary link is deemed unavailable, and a link switching process is initiated. During link switching, the vehicle-mounted system selects the best-quality backup link from the backup link list, establishes a new communication connection, and resumes data transmission to ensure uninterrupted service.

[0038] For example, a bidirectional communication link is established between the vehicle-mounted system and the user equipment, and a heartbeat detection protocol is agreed upon. The probe packet format is specified as JSON, the sending frequency is once every 5 seconds, and the timeout threshold is set to 200 milliseconds. The vehicle-mounted system periodically sends probe packets to the user equipment and simultaneously starts a timer to record the sending time. After receiving the probe packet, the user equipment immediately replies with an acknowledgment packet containing the reception timestamp. Upon receiving the acknowledgment packet, the vehicle-mounted system calculates the round-trip time (RTT) and updates the average response time of the last 10 RTTs. If the vehicle-mounted system does not receive an acknowledgment packet within 200 milliseconds after sending the probe packet, it is considered a packet loss, and the packet loss rate of the last 100 packets is updated. The vehicle-mounted system uses a weighted average algorithm, with a weight allocation of 70% for the average response time and 30% for the packet loss rate, to calculate the current network quality score in real time. If the network quality score is lower than 8, and the quality assessment results for three consecutive periods are all lower than 8, the main link is determined to be unavailable, and a link switching process is initiated. During link switching, the vehicle system selects the backup link with the highest current network quality score from three backup links, establishes a new TCP connection via Socket, and resumes data transmission. Data transmission adopts the breakpoint resume mode, continuing transmission from the last interrupted position to ensure uninterrupted service.

[0039] S106. In the resource management module of the vehicle system, a dynamic task priority adjustment mechanism is set up. Based on the urgency of user requests and the quality feedback of navigation services, the computing resources and network bandwidth usage ratios of different tasks are adjusted in real time to ensure priority response and processing of critical services.

[0040] Based on a pre-established user request urgency assessment model, the urgency score of the current user request is obtained. If the urgency score exceeds a preset threshold, the request is marked as a high-priority request. Real-time quality feedback data of the navigation service is obtained, and a quality assessment algorithm is used to determine whether the current navigation service quality meets preset standards. If not, the priority of navigation service-related tasks is dynamically increased. Based on the task priority ranking results, a resource allocation algorithm is used to dynamically adjust the proportion of computing resources and network bandwidth used by each task, ensuring that high-priority tasks receive more resources and improving the response speed of critical services. During the resource allocation process, a resource prediction model is used to estimate the computing resource and network bandwidth requirements of different priority tasks in the future, and resources are reserved and scheduled in advance to avoid critical services being affected by resource constraints. Insufficient resources can lead to delayed responses. The system continuously monitors resource usage and task execution progress. If the response time of critical services exceeds a preset threshold, an emergency resource scheduling mechanism is triggered to preempt resources from other low-priority tasks, ensuring the timely completion of critical services. Task execution logs and performance analysis models are established to regularly evaluate the system's resource management and task scheduling, identifying issues such as low resource utilization and slow task response. Resource management strategies and task priority settings are automatically adjusted to achieve continuous optimization of system performance. User feedback and service quality evaluation data are collected, and machine learning algorithms are used to continuously optimize the user request urgency assessment model and navigation service quality assessment algorithm, improving the accuracy of dynamic resource management and task priority adjustments, and providing users with a more intelligent in-vehicle service experience.

[0041] For example, the system uses a decision tree-based machine learning model to assess the urgency of user requests. Requests with an urgency score exceeding 8 are marked as high priority. The navigation service quality assessment algorithm collects feedback data such as GPS positioning error and path planning time every 5 seconds. When the positioning error exceeds 10 meters or the planning time exceeds 5 seconds, the navigation task priority is increased by 20%. During resource allocation, a min-heap algorithm is used to sort task priorities, and a weighted round-robin algorithm allocates 80% of CPU time and bandwidth to the top 20% of high-priority tasks. The system uses an LSTM neural network to predict resource demand within the next minute. If the estimated resource utilization is higher than 90%, reserved resources are initiated 5 seconds in advance. When the response time of critical services exceeds 1 second, resource preemption is triggered, and 10% of resources are reclaimed from low-priority tasks. The system performs a performance evaluation every 10 minutes, using an association rule mining algorithm to identify tasks with resource utilization below 50% and automatically lower their priority. Finally, the system uses a collaborative filtering algorithm to analyze user feedback, continuously improving the urgency assessment and service quality assessment models to achieve intelligent in-vehicle service management.

[0042] S107. Analyze the historical load data of the vehicle system through machine learning algorithms, build a load prediction model, and dynamically schedule and reserve computing resources and network bandwidth in advance according to the spatiotemporal distribution characteristics of navigation requests to avoid resource bottlenecks and service blockages.

[0043] Historical load data of the vehicle system, including indicators such as CPU utilization, memory usage, and network traffic, is acquired to construct a time-series dataset. A Long Short-Term Memory (LSTM) neural network algorithm is used to train the historical load data, establishing a load prediction model to predict future load trends based on past load conditions. The spatiotemporal distribution characteristics of navigation requests are obtained, including the time, location, and frequency of requests, and analyzed using density clustering algorithms to identify high-demand areas and peak periods. The load prediction results are combined with the request distribution characteristics to estimate resource demands under different spatiotemporal conditions and formulate dynamic resource scheduling strategies. Sufficient computing resources and network bandwidth are reserved in advance for high-demand areas and peak periods to avoid resource bottlenecks. Idle resources are released promptly for sparsely requested areas and low-demand periods to improve utilization efficiency. During resource scheduling, the system load level is monitored in real time. When there is a significant deviation between the actual load and the predicted load, incremental training and parameter updates of the prediction model are triggered to improve prediction accuracy. System operation data, including indicators such as resource utilization and request response time, are collected to evaluate scheduling performance and optimize the scheduling algorithm, continuously improving the system's service quality.

[0044] For example, historical load data such as CPU usage, memory consumption, and network traffic are collected through the vehicle system, and a time-series dataset is constructed at 5-minute intervals. An LSTM neural network model is built using the TensorFlow framework, with 3 hidden layers, 128 hidden units, and a learning rate of 0.1, trained for 1000 epochs. Based on the load data of the past hour, the model predicts the load change trend within the next 30 minutes, with an average prediction error controlled within 10%. Spatiotemporal distribution data of navigation requests, including latitude, longitude, and timestamps, is obtained and analyzed using the DBSCAN density clustering algorithm. The radius parameter eps is set to 500 meters, and the minimum sample size minPts is set to 50, identifying high-demand areas (such as commercial centers and transportation hubs) and peak times (such as morning and evening rush hours). Combining the load prediction results with the request distribution characteristics, heuristic algorithms such as simulated annealing are used to optimize resource scheduling strategies. In high-demand areas and peak hours, 20% of redundant resources are reserved in advance through an elastic scaling mechanism; in sparsely requested areas and off-peak hours, resource allocation is dynamically reduced to keep the idle rate below 5%. System load levels are monitored in real time; when the deviation between the actual load and the predicted load exceeds 20%, incremental training is triggered to fine-tune the LSTM model parameters using the latest monitoring data. System operation data such as CPU utilization and average response time are collected to calculate key indicators such as resource utilization efficiency and request timeout rate. Scheduling performance is displayed through a visual dashboard, continuously optimizing the scheduling algorithm and parameter configuration to reduce the request timeout rate to below 1% and continuously improve service quality.

[0045] The above-disclosed embodiments are merely preferred embodiments of the present invention and should not be construed as limiting the scope of the invention. Those skilled in the art will understand that implementing all or part of the above-described embodiments and making equivalent changes in accordance with the claims of the present invention are still within the scope of the invention.

Claims

1. A method for virtually connecting user equipment to a vehicle system for navigation across systems, characterized in that, The method includes: By collecting data on user device CPU utilization, memory usage, and network throughput, and combining this with users' historical operation records and application usage habits, machine learning algorithms are used to model and predict user behavior patterns, thereby obtaining a real-time demand feature vector for users. Based on the user's real-time demand feature vector, heuristic optimization is adopted, taking into account the hardware and software performance of the vehicle system, network conditions, and load balancing factors, to dynamically generate the optimal computing resources and network bandwidth allocation strategy, and determine the priority and resource usage ratio of different user requests. During cross-system virtual access, virtualization middleware technology is used to convert user equipment operation requests and data transmission instructions into a format recognizable by the vehicle system, and to encrypt and verify the integrity of the transmitted content to ensure communication compatibility and security. To address the high real-time requirements of navigation services, edge computing technology is adopted to deploy navigation data caching and computing modules on local nodes of the in-vehicle system. By quickly matching real-time traffic information with user location, low-latency navigation route planning and voice guidance services are provided. Establish a heartbeat detection mechanism between the vehicle system and user equipment. By periodically sending probe packets and monitoring response time, the network connection quality is evaluated in real time. If a timeout or excessive packet loss rate is detected, the system automatically switches to a backup link to ensure seamless connection and service continuity. In the resource management module of the vehicle system, a dynamic task priority adjustment mechanism is set up. Based on the urgency of user requests and the quality feedback of navigation services, the computing resources and network bandwidth usage ratios of different tasks are adjusted in real time to ensure priority response and processing of critical services. By analyzing historical load data of the vehicle system through machine learning algorithms, a load prediction model is built. Based on the spatiotemporal distribution characteristics of navigation requests, computing resources and network bandwidth are dynamically scheduled and reserved in advance to avoid resource bottlenecks and service blockages.

2. The method for virtually accessing a user device across systems and navigating a vehicle system according to claim 1, characterized in that, The process involves collecting data on user device CPU utilization, memory usage, and network throughput, combining this data with the user's historical operation records and application usage habits, and using machine learning algorithms to model and predict user behavior patterns to obtain a real-time demand feature vector for the user, including: Acquire performance metrics data such as CPU utilization, memory usage, and network throughput of user devices, and classify and store different types of performance metrics data. At the same time, by analyzing users' historical operation records, we can obtain data on the frequency and duration of users' use of different applications and their application usage patterns. The acquired performance metric data and application usage pattern data are fused together to generate a user behavior dataset. The generated user behavior dataset is cleaned and preprocessed to remove noisy data and outliers, ensuring data quality. Feature engineering is performed on the preprocessed user behavior dataset to extract key features, and feature selection and feature dimensionality reduction are performed to obtain an optimized set of user behavior features. A user behavior prediction model is established by training the optimized user behavior feature set using the logistic regression algorithm. The model is evaluated using cross-validation, and the model's prediction accuracy is calculated. If the prediction accuracy is lower than a preset threshold, the decision tree algorithm will be used to retrain the model in order to improve the prediction performance. The real-time collected user device performance index data and application usage data undergo the same data preprocessing and feature extraction process, and are then input into the trained user behavior prediction model to obtain real-time user demand prediction results. Based on the real-time user demand prediction results and combined with the user's historical preference data, a user demand feature vector is generated. The feature vectors are transmitted to a collaborative filtering-based recommendation engine, which provides users with personalized application recommendations and services through user similarity calculation and item similarity calculation. Continuously track user feedback on recommendation results and collect user click, download, and usage behavior data to optimize recommendation models and improve recommendation quality; At the same time, user behavior prediction models are updated regularly to adapt to changes in user needs.

3. The method for virtually accessing a user device across systems and navigating a vehicle system according to claim 1, characterized in that, Based on the user's real-time demand feature vector, a heuristic optimization algorithm is used to dynamically generate the optimal computing resource and network bandwidth allocation strategy, taking into account the vehicle system's hardware and software performance, network conditions, and load balancing factors, to determine the priority and resource usage ratio of different user requests. This includes: Obtain the feature vector of user needs; Analyze user request content, extract key parameters, quantify user needs, and form a multi-dimensional feature vector; Each dimension represents a different aspect of user needs.

4. The method for virtually accessing and navigating a user equipment across systems using a vehicle system according to claim 1, characterized in that, During the cross-system virtual access process, virtualization middleware technology is used to convert user equipment operation requests and data transmission instructions into a format recognizable by the vehicle system, and to encrypt and verify the integrity of the transmitted content to ensure communication compatibility and security, including: Based on the pre-established cross-system virtual access protocol, the system obtains operation requests and data transmission instructions issued by user equipment, and parses and extracts key information from them. By using virtualization middleware technology, the extracted key information is matched and converted with the preset vehicle system recognizable format to generate operation requests and data transmission instructions that meet the requirements of the vehicle system. The converted operation requests and data transmission instructions are encrypted using an encryption algorithm to ensure the confidentiality and tamper-proof nature of the transmitted content. Based on encryption, an integrity verification algorithm is used to calculate the verification value of the transmitted content, and this value is sent to the vehicle system along with the encrypted transmitted content. After receiving the encrypted transmission content and check value, the vehicle system first decrypts the received content to restore the original operation request and data transmission instructions. The integrity of the decrypted transmitted content is verified by recalculating the check value and comparing it with the received check value to determine whether the transmitted content is complete and error-free. If the integrity verification passes, the decrypted operation request and data transmission instructions are transmitted to the corresponding module of the vehicle system for processing and execution. If the integrity check fails, an error message is sent to the user equipment, requesting a resend of the operation request and data transmission instructions.

5. A method for virtually accessing a user device across systems and navigating a vehicle system according to claim 1, characterized in that, To address the high real-time requirements of navigation services, edge computing technology is employed. Navigation data caching and computation modules are deployed on local nodes of the in-vehicle system. Through rapid matching of real-time traffic information and user location, low-latency navigation route planning and voice guidance services are provided, including: Obtain the geographical location and driving direction information of the vehicle system and use them as initial input data; Based on the geographical location information of the vehicle system, obtain the road network data around the location from the navigation data cache of the local node; Real-time traffic information for this location, including road congestion and traffic accidents, is obtained from the traffic management center via a wireless communication module. By integrating road network data with real-time traffic information, a dynamic road network model is constructed for subsequent route planning; An improved Dijkstra algorithm is used to perform path search on a dynamic road network model, resulting in multiple alternative paths from the origin to the destination. The alternative routes are scored and ranked according to user preferences, and the optimal route is selected by comprehensively considering factors such as route length, estimated travel time and road condition comfort. The system converts the optimal route navigation information into voice prompts, provides real-time voice guidance to users through the in-vehicle audio system, and dynamically updates the navigation information based on the driving progress.

6. A method for virtually accessing a user equipment across systems and navigating a vehicle system according to claim 1, characterized in that, The established heartbeat detection mechanism between the vehicle system and user equipment, by periodically sending probe packets and monitoring response times, assesses network connection quality in real time. If a timeout or excessively high packet loss rate is detected, it automatically switches to a backup link to ensure seamless connection and service continuity, including: A two-way communication link is established between the vehicle system and the user equipment, and a heartbeat detection protocol is agreed upon, specifying the detection packet format, transmission frequency, and timeout threshold; The vehicle-mounted system periodically sends probe packets to the user equipment and simultaneously starts a timer to record the sending time. After receiving the probe packet, the user equipment immediately replies with an acknowledgment packet. Upon receiving the acknowledgment packet, the vehicle system calculates the round-trip time and updates the average response time. If the vehicle system sends a probe packet but does not receive an acknowledgment packet after a preset timeout threshold has elapsed, it is determined to be a packet loss, and the packet loss rate is updated. The vehicle system uses a weighted average algorithm to evaluate the current network quality in real time based on the calculated average response time and packet loss rate. If the network quality assessment result is lower than the preset quality threshold and the duration exceeds the predetermined tolerance period, the main link is determined to be unavailable and the link switching process is initiated. During link switching, the vehicle system selects the best-quality backup link from the backup link list, establishes a new communication connection, and resumes data transmission to ensure uninterrupted service.

7. A method for virtually accessing a user equipment across systems and navigating a vehicle system according to claim 1, characterized in that, The resource management module of the in-vehicle system includes a dynamic task priority adjustment mechanism. Based on the urgency of user requests and the quality feedback of navigation services, the mechanism adjusts the computing resources and network bandwidth usage ratios for different tasks in real time to ensure priority response and processing of critical services, including: Based on a pre-established user request urgency assessment model, the urgency score of the current user request is obtained. If the urgency score exceeds a preset threshold, the request is marked as a high-priority request. Obtain real-time quality feedback data of the navigation service, and use a quality assessment algorithm to determine whether the current navigation service quality meets the preset standards. If it does not meet the standards, dynamically increase the priority of navigation service-related tasks. Based on the task priority ranking results, a resource allocation algorithm is used to dynamically adjust the proportion of computing resources and network bandwidth used by each task, ensuring that high-priority tasks receive more resources and improving the response speed of critical services. During the resource allocation process, the resource prediction model is used to estimate the computing resources and network bandwidth requirements of different priority tasks in the future, and to reserve and schedule resources in advance to avoid delays in critical services due to insufficient resources. The system continuously monitors resource usage and task execution progress. If the response time of critical services exceeds a preset threshold, an emergency resource scheduling mechanism is triggered to preempt some resources from other low-priority tasks to ensure the timely completion of critical services. Establish task execution logs and performance analysis models, regularly evaluate the system's resource management and task scheduling, identify problems such as low resource utilization and slow task response, and automatically adjust resource management strategies and task priority settings to achieve continuous optimization of system performance; By collecting user feedback and service quality evaluation data, and continuously optimizing the user request urgency assessment model and navigation service quality assessment algorithm through machine learning algorithms, we can improve the accuracy of resource management and dynamic adjustment of task priorities, and provide users with a more intelligent in-vehicle service experience.

8. A method for virtually accessing a user equipment across systems and navigating a vehicle system according to claim 1, characterized in that, The process involves analyzing historical load data of the vehicle system using machine learning algorithms to construct a load prediction model. Based on the spatiotemporal distribution characteristics of navigation requests, it dynamically schedules and reserves computing resources and network bandwidth in advance to avoid resource bottlenecks and service blockages. This includes: Acquire historical load data of the vehicle system, including CPU utilization, memory usage, and network traffic metrics, and construct a time-series dataset; The Long Short-Term Memory (LSTM) neural network algorithm is used to train historical load data to establish a load prediction model, which predicts the load change trend in the future based on the load situation in the past period. Obtain the spatiotemporal distribution characteristics of navigation requests, including the time, location, and frequency of the requests, and analyze them using density clustering algorithms to identify high-incidence areas and peak periods of requests; By combining load forecasting results with request distribution characteristics, resource demand under different spatiotemporal conditions is estimated, and dynamic resource scheduling strategies are formulated. For areas with high request volume and peak hours, reserve sufficient computing resources and network bandwidth in advance to avoid resource bottlenecks; For regions with sparse requests and during off-peak hours, release idle resources in a timely manner to improve utilization efficiency; During resource scheduling, the system load level is monitored in real time. When there is a large deviation between the actual load and the predicted load, incremental training and parameter updates of the prediction model are triggered to improve prediction accuracy. To assess the effectiveness of scheduling strategies, system operation data, including resource utilization and request response time metrics, are collected to evaluate scheduling performance, optimize scheduling algorithms, and continuously improve the system's service quality.

Images