Intelligent highway control system for low-altitude aircraft and vehicles

Abstract

The present invention discloses an intelligent highway control system for low-altitude aircraft and vehicles in collaboration, relating to the field of intelligent transportation technology. The system includes: a heterogeneous network fusion unit, which is used to establish a unified communication protocol framework based on the needs of the aircraft and vehicle communication networks, and formulate common communication rules and data formats to achieve protocol conversion and adaptation between different networks; a positioning perception and recognition unit, which is used to integrate multiple positioning technologies, fuse raw data from different sensors, identify target behavior attributes through a hierarchical uncertainty modeling and transmission mechanism, and establish a target tracking algorithm to track the position and motion status of aircraft and vehicles in real time; and a collaborative decision-making control unit, which is used to fuse aircraft and vehicle traffic data, plan aircraft and vehicle paths, and design a collaborative emergency event processing mechanism through sensor and data fusion technology. The present invention realizes efficient collaborative control of low-altitude aircraft and vehicles.

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation technology, and more specifically, to an intelligent highway control system for the coordination of low-altitude aircraft and vehicles. Background Art

[0002] With the rapid development of the economy and society and the continued growth in transportation demand, highways, as vital national transportation infrastructure, face unprecedented challenges in their management and control. Traditional highway control systems primarily rely on fixed ground sensors and monitoring equipment. This single-source control model is no longer sufficient to meet the increasingly complex demands of traffic management. In recent years, low-altitude aircraft, with their advantages of maneuverability, wide field of view, and rapid response, have demonstrated tremendous potential in traffic management, providing new technical means and development directions for building a new generation of intelligent highway control systems.

[0003] Currently, technological innovations in highway management and control systems are primarily focused on communication and data interaction, perception and positioning, and collaborative decision-making and control. In the area of ​​communication and data interaction, research focuses on heterogeneous network integration, low-latency, high-bandwidth communications, and data security and privacy protection. In the area of ​​perception and positioning, efforts are focused on addressing technical challenges such as high-precision positioning integration, environmental perception, and target recognition. In the area of ​​collaborative decision-making and control, efforts are focused on achieving functions such as coordinated traffic flow optimization and coordinated emergency response. While some dispatch and management systems based on collaborative drones and vehicles have been applied to the material distribution sector, implementing functions such as material management, route planning, and delivery management, these systems are primarily targeted at logistics and distribution scenarios and are unable to meet the complex and ever-changing control needs of highways.

[0004] However, existing technologies still face numerous limitations in practical applications. First, in terms of communication technology, due to the complex highway environment, traditional civilian communication networks face problems such as frequent signal switching, attenuation, and interference. Furthermore, the communication frequency bands of low-altitude aircraft and vehicles are inconsistent, lacking a unified coordination mechanism. Second, in terms of perception and positioning, existing technologies overly rely on a single positioning method, which can lead to a significant drop in positioning accuracy in special terrain environments such as mountainous areas and tunnels. Furthermore, different sensors have their own limitations. For example, millimeter-wave radar's detection performance degrades in inclement weather, while the perception systems of vehicles and low-altitude aircraft lack effective coordination mechanisms. Finally, in terms of collaborative decision-making and control, existing technologies lack precise traffic flow coordination models and comprehensive emergency response mechanisms, making it difficult to achieve optimal traffic flow allocation and rapid response to emergencies, thus restricting the overall effectiveness of smart highway management and control systems.

[0005] Currently, no effective solutions have been proposed for the problems in related technologies. Summary of the Invention

[0006] (1) Technical problems solved

[0007] In response to the shortcomings of the existing technology, the present invention provides an intelligent high-speed control system for the collaboration of low-altitude aircraft and vehicles, which has the advantages of heterogeneous network fusion, multi-source perception and positioning, and collaborative decision-making and control, thereby solving the problems of inconsistent communication networks, inaccurate positioning perception, and inefficient collaborative control in the existing technology.

[0008] (2) Technical solution

[0009] To achieve the advantages of heterogeneous network integration, multi-source sensing and positioning, and collaborative decision-making and control, the specific technical solutions adopted by the present invention are as follows:

[0010] The intelligent highway control system for low-altitude aircraft and vehicles in collaboration includes:

[0011] Heterogeneous network fusion unit, which is used to establish a unified communication protocol framework based on the communication network requirements of aircraft and vehicles, and formulate common communication rules and data formats to achieve protocol conversion and adaptation between different networks;

[0012] The positioning perception and recognition unit integrates multiple positioning technologies, fuses raw data from different sensors, identifies target behavior attributes through a hierarchical uncertainty modeling and transfer mechanism, and establishes a target tracking algorithm to track the position and motion status of aircraft and vehicles in real time.

[0013] The collaborative decision-making control unit is used to integrate the traffic data of aircraft and vehicles, plan the paths of aircraft and vehicles, and quickly detect emergencies through sensor and data fusion technology, and design a collaborative emergency processing mechanism.

[0014] Furthermore, the heterogeneous network fusion unit includes an intelligent gateway hardware module and an intelligent gateway software module;

[0015] An intelligent gateway software module, used to connect different types of networks and perform protocol conversion, converting the aircraft's communication protocol into a protocol that can be received by vehicles or highway infrastructure;

[0016] Among them, the intelligent gateway software module includes:

[0017] The protocol conversion module is used to implement protocol conversion and adaptation between different networks by adopting adaptive communication protocols and communication protocol stack optimization technology and building a communication performance evaluation model;

[0018] Data filtering module, which is used to filter and screen communication data and send the data to the correct destination according to different application requirements;

[0019] Routing module, used to select the optimal communication path and improve data transmission efficiency;

[0020] The security protection module is used to protect network security and privacy and prevent data from being stolen or tampered with.

[0021] Furthermore, the protocol conversion module adopts adaptive communication protocol and communication protocol stack optimization technology, and builds a communication performance evaluation model to achieve protocol conversion and adaptation between different networks, including:

[0022] Real-time assessment of network status through network detection tools and signal indicators fed back by the receiving end;

[0023] Develop adaptive protocol parameter adjustment strategies based on network status and communication requirements;

[0024] Build a multi-dimensional performance evaluation indicator system based on network, application and user experience levels;

[0025] The analytic hierarchy process is used to determine the weights of the evaluation indicators at each level in the multi-dimensional performance evaluation index system, and the measured data are mapped to different evaluation levels through the membership function to construct a communication performance evaluation model;

[0026] The communication performance evaluation model is trained and optimized based on historical data, and the accuracy and efficiency of protocol conversion are improved by adjusting parameters.

[0027] Furthermore, the analytic hierarchy process is used to determine the weights of the evaluation indicators at each level in the multi-dimensional performance evaluation index system, and the measured data are mapped to different evaluation levels through the membership function. The performance evaluation model is constructed including:

[0028] Set evaluation indicators and determine the evaluation level based on actual application scenarios and needs;

[0029] According to the characteristics and actual situation of the evaluation indicators, the membership function corresponding to each indicator is constructed, and the membership vector of each evaluation indicator for different evaluation levels is calculated;

[0030] The fuzzy relationship matrix is ​​constructed by using the membership vectors of each evaluation index to different evaluation levels;

[0031] The weight vector is determined by the hierarchical analysis method and synthesized with the fuzzy relationship matrix to obtain the comprehensive evaluation result vector. The maximum membership principle is used to determine the final evaluation level.

[0032] Furthermore, the expression of the membership function is:

[0033] ;

[0034] Where, is the membership degree of the i-th evaluation index to the j-th evaluation level; represents the actual measured value of the i-th evaluation index; a and b represent the upper limit constant and lower limit constant of the j-th evaluation level respectively;

[0035] The expression of the fuzzy relationship matrix is:

[0036] ;

[0037] Where R represents the fuzzy relationship matrix; the i-th row of R represents the membership of the i-th evaluation indicator to each evaluation level; the j-th column of R represents the membership of all evaluation indicators to the j-th evaluation level.

[0038] Furthermore, the positioning perception and recognition unit includes:

[0039] The sensor fusion module is used to perform preliminary analysis of the collected data using vehicles, low-altitude aircraft, and fixed sensors. Based on the transmission bandwidth, the preliminary analysis results and some key raw data are sent to the central processing module.

[0040] The central processing module is used to integrate the raw data from different sensors and the preliminary analysis results, and identify the target behavior attributes through a hierarchical uncertainty modeling and transmission mechanism;

[0041] The target tracking module is used to track the position and motion status of aircraft and vehicles in real time based on the target tracking algorithm through the three-dimensional covariance matrix and multi-factor covariance prediction equation.

[0042] Furthermore, the sensor fusion module includes:

[0043] The network monitoring module is used to obtain the current network available transmission bandwidth and bandwidth utilization information through network monitoring tools;

[0044] A preliminary analysis module is used to extract and fuse simple features based on vehicles, low-altitude aircraft, and fixed sensors to determine the data volume and transmission priority of key raw data;

[0045] A transmission allocation module is used to allocate transmission resources according to transmission bandwidth conditions and data characteristics;

[0046] Dynamic adjustment module, used to continuously monitor bandwidth changes and data transmission progress, and dynamically adjust transmission strategies;

[0047] The verification and optimization module is used to check data integrity and accuracy, and optimize transmission strategies and data processing methods based on the inspection results.

[0048] Furthermore, the central processing module includes:

[0049] The data acquisition module is used to analyze the impact of sensor accuracy and environmental noise on the collected data, select the corresponding probability distribution according to the characteristics of different sensors, and establish the probability density function of the collected data;

[0050] The feature extraction module is used to build an uncertainty model for feature extraction based on the quantitative results of algorithm limitations and data ambiguity;

[0051] The information fusion module is used to fuse the collected data from different sensors based on the uncertainty model and the weighted average information fusion algorithm.

[0052] Furthermore, the target tracking module includes:

[0053] The state initialization module is used to determine the initial state vector and three-dimensional covariance matrix of the target based on prior knowledge and initial measurement values;

[0054] The state prediction module is used to obtain the target's measurement values ​​based on different sensors, establish a state prediction equation, and use the state transition matrix to predict the target state at the current moment based on the state estimate at the previous moment;

[0055] The gain calculation module is used to establish a multi-factor covariance prediction equation using observation residuals, dynamically adjust the noise covariance, and calculate the Kalman gain based on the prediction covariance and the measurement noise covariance;

[0056] The state update module is used to update the state estimate through the Kalman gain based on the state update equation, combined with the predicted value and the measured value, and to update the covariance matrix to determine whether to continue tracking the target according to actual needs.

[0057] Furthermore, the expression of the three-dimensional covariance matrix is:

[0058] ;

[0059] Where, Represents the variance of position x, position y, and position z respectively; σ xy , σ xz , σ yz Represent the covariance between position x and position y, position x and position z, position y and position z respectively; Represents position x and velocity v in the x direction x The covariance between represents the velocity v in the x direction x and the velocity v in the y direction y The covariance between

[0060] The expression of the multi-factor covariance prediction equation is:

[0061] ;

[0062] Where, Represents the prior estimated covariance matrix at the current moment; Represents the posterior estimated covariance matrix of the previous moment; represents the process noise covariance matrix; represents the forgetting factor; represents the attenuation factor; represents the expansion factor; represents the Kalman gain, represents the observation residual.

[0063] (3) Beneficial effects

[0064] Compared with the existing technology, the present invention provides an intelligent high-speed control system for low-altitude aircraft and vehicles, which has the following beneficial effects:

[0065] (1) The present invention designs a heterogeneous network fusion and intelligent gateway system architecture through a heterogeneous network fusion unit, establishes a unified communication protocol framework, and can achieve seamless integration of different communication networks. It uses intelligent gateway devices to perform protocol conversion, data filtering and routing to adapt to the different needs of aircraft and vehicles. At the same time, it adopts adaptive communication protocols and optimization technologies to automatically adjust communication protocols according to different communication scenarios and needs, optimizes the communication protocol stack, ensures the communication quality between aircraft and vehicles, and between them and the traffic management center, and achieves low-latency, high-bandwidth data transmission. It also protects the security and privacy of data interaction in all directions through a reliable security encryption and identity authentication system, effectively guarantees the efficiency and reliability of data transmission, and provides solid technical support for achieving efficient collaborative management and control.

[0066] (2) The present invention integrates multiple positioning technologies such as satellite positioning and inertial navigation through a positioning perception and recognition unit using a high-precision positioning fusion algorithm, and fuses and processes information through filtering and other technologies. It can accurately determine the position of vehicles and low-altitude aircraft regardless of the terrain and environment. At the same time, it comprehensively utilizes radars, cameras, lidars and other sensors on aircraft and vehicles, combined with fixed sensors on roads and airspace, such as meteorological sensors and traffic flow monitoring sensors, to achieve comprehensive perception of the environment, including each other's position, speed, direction and weather conditions. The raw data of different sensors are pre-processed and then transmitted to the central processing module for fusion processing. The features extracted by different sensors are fused using methods such as principal component analysis, thereby improving the accuracy of target detection and recognition and discovering potential dangers in advance. In addition, the target tracking algorithm of the present invention effectively utilizes the target's prior motion knowledge and real-time measurement data through two stages of state prediction and measurement update, and continuously corrects the tracking results. Even in the presence of noise and uncertainty, it can track the target's position and motion state in real time more accurately to ensure traffic safety.

[0067] (3) The present invention integrates the aircraft and vehicle operation data through a collaborative decision-making control unit, and uses a simulated annealing algorithm to collaboratively plan flight paths and driving routes, thereby achieving goals such as minimizing total driving time and reducing traffic congestion, implementing dynamic traffic control measures, adjusting traffic flow according to real-time conditions, improving the efficiency of road and airspace use, and achieving scientific optimization of traffic flow; at the same time, an emergency event monitoring system is established, which uses sensors and data fusion technology to quickly detect emergencies and classify and grade them, and through the designed collaborative processing mechanism and emergency response process, resources are reasonably dispatched. Once an emergency occurs, resources such as aircraft and vehicles can be quickly coordinated to guide rescue vehicles to quickly arrive at the accident scene, realizing collaborative operations of aircraft and vehicles, effectively improving the collaborative processing capabilities of the entire high-speed scene to respond to emergencies, ensuring the safety and smoothness of high-speed traffic, and reducing accidents and delays caused by poor communication. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0069] Figure 1 This is a principle block diagram of an intelligent highway control system for collaboration between low-altitude aircraft and vehicles according to an embodiment of the present invention;

[0070] Figure 2This is a principle block diagram of a heterogeneous network fusion unit in an intelligent highway control system for low-altitude aircraft and vehicle collaboration according to an embodiment of the present invention;

[0071] Figure 3 This is a flow chart illustrating the use of adaptive communication protocols and communication protocol stack optimization technology by a protocol conversion module in an intelligent highway control system for low-altitude aircraft and vehicle collaboration according to an embodiment of the present invention;

[0072] Figure 4 2. It is a schematic diagram of a sensor fusion architecture composed of a central processing module in a smart highway control system for low-altitude aircraft and vehicle collaboration according to an embodiment of the present invention;

[0073] Figure 5 This is a flow chart of a target tracking algorithm of a target tracking module in a smart highway control system for low-altitude aircraft and vehicle collaboration according to an embodiment of the present invention;

[0074] Figure 6 It is a principle block diagram of a collaborative decision-making control unit in an intelligent highway control system for collaboration between low-altitude aircraft and vehicles according to an embodiment of the present invention. DETAILED DESCRIPTION

[0075] To further illustrate each embodiment, the present invention provides drawings, which are part of the disclosure of the present invention. They are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. By referring to these contents, ordinary technicians in this field should be able to understand other possible implementation methods and advantages of the present invention. The components in the figures are not drawn to scale, and similar component symbols are generally used to represent similar components.

[0076] According to an embodiment of the present invention, an intelligent highway control system for collaboration between low-altitude aircraft and vehicles is provided.

[0077] The present invention will now be further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, the intelligent highway control system for low-altitude aircraft and vehicles in collaboration according to an embodiment of the present invention includes:

[0078] Heterogeneous network fusion unit 1 is used to establish a unified communication protocol framework based on the needs of low-altitude aircraft communication networks and vehicle communication networks, and formulate common communication rules and data formats to achieve protocol conversion and adaptation between different networks;

[0079] Specifically, the heterogeneous network fusion unit 1 adopts a heterogeneous network fusion and intelligent gateway system architecture design.

[0080] Specifically, current highway communication networks are diverse, including specialized networks used by aircraft, on-board vehicle networks, and the inherent communication networks of highway infrastructure. Establishing a unified communication protocol framework is fundamental to achieving heterogeneous network convergence. By establishing common communication rules and data formats, these different networks can understand and exchange information. In this embodiment, the aircraft communication network takes into account its high mobility and communication range requirements, while the vehicle communication network prioritizes real-time performance and reliability. Through a unified protocol framework, these diverse requirements are integrated to achieve seamless information transmission.

[0081] Specifically, the key to achieving heterogeneous network convergence is the intelligent gateway device. It can connect different types of networks and perform protocol conversion, converting information from one network into a format understandable by another. In this embodiment, when information sent by an aircraft needs to be transmitted to a vehicle or highway infrastructure, the intelligent gateway can convert the aircraft's communication protocol into a protocol that the vehicle or highway infrastructure can receive. The intelligent gateway can also implement data filtering and routing functions, sending information to the correct destination based on different needs. In this embodiment, in an emergency, the intelligent gateway can prioritize sending rescue information to relevant rescue vehicles and command centers. The intelligent gateway device contains a high-performance processor and communication module, with powerful computing and data processing capabilities. It can support multiple communication protocols, such as Wi-Fi, Bluetooth, ZigBee, 4G / 5G, etc., and can perform protocol conversion and data routing as needed.

[0082] Specifically, adaptive communication protocols and communication protocol stack optimization technologies are adopted to optimize the communication protocol stack by optimizing data compression algorithms, reducing the size of data packets, and improving the reliability of data transmission, thereby reducing data transmission overhead. At the same time, protocol parameters are automatically adjusted according to network conditions and communication needs to achieve the best communication effect.

[0083] Positioning perception and recognition unit 2 is used to integrate multiple positioning technologies, fuse raw data from different sensors, identify target behavior attributes through a hierarchical uncertainty modeling and transmission mechanism, and establish a target tracking algorithm to track the position and motion status of aircraft and vehicles in real time;

[0084] Specifically, the positioning perception and recognition unit 2 adopts high-precision positioning fusion and environment perception and target recognition technology.

[0085] Specifically, a high-precision positioning fusion algorithm is used to integrate multiple positioning technologies, such as GPS, Beidou satellite positioning, inertial navigation, and base station positioning. Kalman filtering, extended Kalman filtering, particle filtering, and other technologies are used to fuse information from different positioning data sources. In this embodiment, the Kalman filter can obtain the optimal estimate by weighted averaging the measurement values ​​from different positioning data sources; the extended Kalman filter can handle nonlinear positioning problems; and the particle filter can provide more accurate positioning estimates in complex environments. A model-based fusion algorithm is used to establish a motion model of the aircraft or vehicle, and positioning fusion is achieved by estimating the model parameters.

[0086] Specifically, to achieve omnidirectional perception of the highway environment and target recognition, multiple sensors, such as cameras, lidar, and millimeter-wave radar, need to be deployed. Different sensors have different characteristics and advantages, and by fusing their information, the accuracy of environmental perception and target recognition can be improved. In this embodiment, cameras can provide rich visual information, but their performance may be affected in inclement weather conditions. Lidar can provide high-precision three-dimensional spatial information, but at a higher cost. Millimeter-wave radar can operate in adverse weather conditions and has high range and velocity resolution.

[0087] Specifically, a target tracking algorithm is established to track the position and motion of targets such as aircraft and vehicles in real time. A model-based tracking algorithm establishes a target motion model and achieves target tracking by estimating model parameters. In this embodiment, a Kalman filter-based tracking algorithm is used to track vehicles in a highway environment. Target tracking is achieved by predicting and updating the vehicle's position, speed, and other states.

[0088] The collaborative decision control unit 3 is used to integrate the traffic data of aircraft and vehicles, plan the paths of aircraft and vehicles, and quickly detect emergencies through sensor and data fusion technology, and design a collaborative emergency processing mechanism.

[0089] Specifically, the collaborative decision control unit 3 implements a collaborative decision and control solution.

[0090] Specifically, the system integrates aircraft and vehicle operational data and utilizes a simulated annealing algorithm to collaboratively plan aircraft flight paths and vehicle travel routes, minimizing total travel time and reducing traffic congestion. Dynamic traffic control measures and a coordinated emergency response mechanism are also implemented. Traffic data from highways and low-altitude areas is collected, and a traffic flow model is developed to understand the spatiotemporal distribution of traffic. By implementing dynamic traffic control measures, traffic flows are adjusted based on real-time traffic conditions, improving the efficiency of road and airspace utilization.

[0091] It's important to note that the traffic flow model employed by the collaborative decision-making control unit 3 is an existing technology based on fluid dynamics theory. This technology constructs a network-level traffic flow evolution framework using continuity equations and conservation laws. Its core model, the Lighthill-Whitham-Richards (LWR), describes the propagation of traffic waves through density-flow relationships. Based on the analysis results of this model, the collaborative decision-making control unit 3 can accurately grasp the spatiotemporal evolution of road network congestion, providing a basis for decision-making regarding the coordinated scheduling of aircraft and vehicles.

[0092] Specifically, an emergency monitoring system is established, deploying sensors and data fusion technology to rapidly detect and categorize emergencies. A collaborative emergency response mechanism is designed, along with a detailed emergency response process. Rescue resources are rationally dispatched, enabling collaborative operations between aircraft and vehicles. In this embodiment, when an accident occurs, an optimal route is planned for rescue vehicles, traffic signal priority is set, and aircraft are used to guide rescue operations.

[0093] In one embodiment, the heterogeneous network fusion unit 1 includes an intelligent gateway hardware module and an intelligent gateway software module;

[0094] An intelligent gateway software module, used to connect different types of networks and perform protocol conversion, converting the aircraft's communication protocol into a protocol that can be received by vehicles or highway infrastructure;

[0095] Among them, the intelligent gateway software module includes:

[0096] The protocol conversion module is used to implement protocol conversion and adaptation between different networks by adopting adaptive communication protocols and communication protocol stack optimization technology and building a communication performance evaluation model;

[0097] Data filtering module, which is used to filter and screen communication data and send the data to the correct destination according to different application requirements;

[0098] Routing module, used to select the optimal communication path and improve data transmission efficiency;

[0099] The security protection module is used to protect network security and privacy and prevent data from being stolen or tampered with.

[0100] Specifically, the intelligent gateway hardware module is used to implement data filtering and routing functions, and send information to the correct destination according to different needs;

[0101] Among them, the intelligent gateway hardware modules include:

[0102] Processor module, used to process communication data and control the operation of the intelligent gateway;

[0103] Communication module, used to realize communication between different networks;

[0104] A storage module, used for storing communication data and program codes;

[0105] The power module is used to provide power for the smart gateway.

[0106] Specifically, the heterogeneous network fusion unit 1 ( Figure 2 The intelligent gateway is a key component of heterogeneous network convergence. First, it must implement protocol conversion and adaptation between different networks to ensure smooth communication. Second, it must provide data filtering, routing, and forwarding capabilities to send data to the correct destination based on different application requirements. Based on the functional requirements of the intelligent gateway, the heterogeneous network convergence unit 1 is divided into an intelligent gateway hardware module and an intelligent gateway software module.

[0107] Specifically, the intelligent gateway hardware module ( Figure 2 The hardware modules in the smart gateway include a processor module, communication module, storage module, and power module. The processor module is the core component of the smart gateway, responsible for processing communication data and controlling the operation of the smart gateway. The communication module, including satellite communication modules, mobile communication modules, and dedicated short-range communication modules, facilitates communication between different networks. The storage module stores communication data and program code. The power module provides power to the smart gateway. To meet future system expansion requirements, the hardware architecture of the smart gateway needs to be highly scalable. For example, a modular design can facilitate the addition of new hardware modules, and certain expansion interfaces can be reserved to facilitate the connection of external devices.

[0108] Specifically, the intelligent gateway software module ( Figure 2 The software modules in the smart gateway include a protocol conversion module, a data filtering module, a routing module, and a security protection module. The protocol conversion module is responsible for implementing protocol conversion and adaptation between different networks. The data filtering module is responsible for filtering and screening communication data, sending data to the correct destination based on different application requirements. The routing module is responsible for selecting the optimal communication path to improve data transmission efficiency. The security protection module is responsible for protecting network security and privacy, preventing data theft or tampering. To improve the maintainability and scalability of the software, the software architecture of the smart gateway needs to adopt technologies such as modular design and object-oriented programming. Modular design divides the software system into multiple independent modules, facilitating maintenance and upgrades. Object-oriented programming can improve the scalability and maintainability of the software, making it easier to add new features and modify existing ones.

[0109] In one embodiment, the protocol conversion module adopts adaptive communication protocol and communication protocol stack optimization technology to achieve protocol conversion and adaptation between different networks by building a communication performance evaluation model, including:

[0110] Real-time assessment of network status through network detection tools and signal indicators fed back by the receiving end;

[0111] Develop adaptive protocol parameter adjustment strategies based on network status and communication requirements;

[0112] Build a multi-dimensional performance evaluation indicator system based on network, application and user experience levels;

[0113] The analytic hierarchy process is used to determine the weights of the evaluation indicators at each level in the multi-dimensional performance evaluation index system, and the measured data are mapped to different evaluation levels through the membership function to construct a communication performance evaluation model;

[0114] The communication performance evaluation model is trained and optimized based on historical data, and the accuracy and efficiency of protocol conversion are improved by adjusting parameters.

[0115] In one embodiment, the weights of the evaluation indicators at each level in the multi-dimensional performance evaluation index system are determined using the analytic hierarchy process, and the measured data are mapped to different evaluation levels through the membership function. The performance evaluation model is constructed including:

[0116] Set evaluation indicators and determine the evaluation level based on actual application scenarios and needs;

[0117] According to the characteristics and actual situation of the evaluation indicators, the membership function corresponding to each indicator is constructed, and the membership vector of each evaluation indicator for different evaluation levels is calculated;

[0118] The fuzzy relationship matrix is ​​constructed by using the membership vectors of each evaluation index to different evaluation levels;

[0119] The weight vector is determined by the hierarchical analysis method and synthesized with the fuzzy relationship matrix to obtain the comprehensive evaluation result vector. The maximum membership principle is used to determine the final evaluation level.

[0120] Specifically, if Figure 3 As shown, the implementation scheme of the protocol conversion module using the adaptive communication protocol and communication protocol stack optimization technology includes:

[0121] ① Network status monitoring and application demand analysis. Using network detection tools and receiving-end feedback on indicators such as signal strength, packet loss rate, and latency, we can monitor network bandwidth, stability, and congestion in real time. In this embodiment, network detection packets are sent periodically, and network status is assessed based on data such as round-trip time (RTT) and packet loss rate.

[0122] ② Adaptive adjustment strategy formulation. An adaptive protocol parameter adjustment strategy is formulated based on network conditions and communication requirements. In this embodiment, when network bandwidth is low (downlink <5Mbps and uplink <2Mbps), the data transmission rate is reduced (to 50%-70% of the baseline value) and the compression ratio is increased (4:1-6:1 for lossless compression; 8:1-12:1 for lossy compression). When latency requirements are high (latency >100ms), a low-latency communication protocol (such as QUIC) is selected or the packet size (256-512 bytes) and transmission frequency are adjusted. The compression algorithm's compression level is dynamically adjusted based on network conditions and communication requirements. In this embodiment, when network bandwidth is low, zlib levels 7-9 are enabled; when network conditions are good (downlink 5-20Mbps or uplink 2-5Mbps), zlib levels 4-6 are enabled; and when network conditions are excellent (downlink >20Mbps and uplink >5Mbps), zlib levels 1-3 are enabled.

[0123] ③ Build a multi-dimensional performance evaluation indicator system covering three levels: network, application, and user experience. At the network layer, we focus on throughput, using specialized software to capture packets and analyze the amount of data transmitted during specific periods. We also focus on latency, using multiple timestamp probe packets to measure the average round-trip latency. At the application layer, we quantify video quality using algorithms such as PSNR and SSIM for video applications. For interactive applications, we embed a timing module to measure user operation response time. At the user experience level, we collect satisfaction scores through regular online questionnaires with multiple levels of options and subjective feedback. We also measure the frequency of negative experiences using error reporting rates.

[0124] ④ Build a multi-dimensional communication performance evaluation model. This method uses an organic combination of the analytic hierarchy process (AHP) and the fuzzy comprehensive evaluation method. Using AHP, we invite authoritative experts in network communications, senior application developers, and representative users to participate in a detailed pairwise comparison of the importance of evaluation indicators at different levels, construct a scientific judgment matrix, and then accurately calculate the weight distribution of indicators at each level of network, application, and user experience relative to the overall performance. On this basis, with the help of the fuzzy comprehensive evaluation method, the actual measured values ​​of each indicator are mapped to the corresponding evaluation level, such as excellent, good, medium, and poor, through a carefully designed membership function, forming a clear fuzzy relationship matrix. The specific steps are as follows:

[0125] (1) Determine the evaluation indicators and evaluation levels. Clarify various indicators (including communication delay, packet loss rate, throughput, bandwidth utilization, energy consumption, etc.). The expression of the evaluation indicator set is:

[0126] ;

[0127] Where, represents the i-th evaluation index, and n represents the total number of evaluation indicators. In this embodiment, n=5, which represent communication delay, packet loss rate, throughput, bandwidth utilization, and energy consumption respectively.

[0128] (2) Set the evaluation level. Determine the evaluation level based on the actual application scenario and requirements. The expression of the evaluation level set V is:

[0129] ;

[0130] Where, represents the jth evaluation grade, m represents the total number of evaluation grades, in this embodiment m=4, represented by excellent, good, medium, and poor.

[0131] (3) Construct membership function. Design membership function for each indicator according to the characteristics of the indicator and the actual situation. ,If the indicator value range corresponding to the evaluation level "excellent" is [a, b], the expression of its membership function belonging to "excellent" is:

[0132] ;

[0133] Where, is the membership degree of the i-th evaluation index to the j-th evaluation level; represents the actual measured value of the i-th evaluation index; a and b represent the upper limit constant and lower limit constant of the j-th evaluation level respectively;

[0134] (4) Calculate the membership degree. The actual measured value of each indicator Substitute the corresponding membership function to calculate the membership of each indicator for each evaluation level and obtain the membership vector The expression is:

[0135] ;

[0136] (5) Forming a fuzzy relationship matrix. The membership vectors of all indicators are combined to form a fuzzy relationship matrix R. The expression of the fuzzy relationship matrix is:

[0137] ;

[0138] Where R represents the fuzzy relationship matrix; the i-th row of R represents the membership of the i-th evaluation indicator to each evaluation level; the j-th column of R represents the membership of all evaluation indicators to the j-th evaluation level.

[0139] (6) Finally, the communication performance evaluation model is constructed by combining the weights determined by AHP to obtain the comprehensive evaluation results. The specific steps are as follows:

[0140] a) Weight vector obtained by AHP calculation:

[0141] ;

[0142] Where, represents the weight of the i-th evaluation index, and .

[0143] b) Perform fuzzy synthesis operation. Given the fuzzy relationship matrix R, perform fuzzy synthesis operation on the weight vector W and the fuzzy relationship matrix R to obtain the comprehensive evaluation result vector A:

[0144] ;

[0145] Where, It represents the membership of the comprehensive evaluation result to the jth evaluation level, , .

[0146] c) Obtaining a comprehensive evaluation result. Analyze and process the comprehensive evaluation result vector A, and adopt the maximum membership principle to select the evaluation grade corresponding to the maximum membership value in A as the comprehensive evaluation result. In this embodiment: , where the maximum membership degree is 0.35, and its corresponding evaluation level is selected as the comprehensive evaluation result.

[0147] ⑤ Based on the massive amount of historical data collected, the training and test sets are divided into a scientific ratio of 70% and 30%. The model is repeatedly trained using the training set, and model parameters, such as membership function parameters and weight coefficients, are fine-tuned based on key metrics such as accuracy and recall until the model performance meets the target. After the model is built, a real-time monitoring and visualization platform is established, presenting the model output with intuitive dashboards and line charts. Once key performance indicators fall below preset thresholds, such as throughput falling below 80% of the rated value or user satisfaction falling below 60%, an alert is immediately triggered to notify operations and maintenance personnel. A detailed description of the problem and relevant data reports are also accurately pushed to the development team. Based on this feedback, the development team quickly develops targeted optimization strategies, such as dynamically adjusting the traffic scheduling algorithm in the event of network congestion and promptly updating the coding library in the event of low application-layer coding efficiency. After the optimization is implemented, performance indicators are continuously monitored and newly collected data is used to dynamically update and calibrate the communication performance evaluation model, enabling protocol conversion and adaptation between different networks, forming an efficient closed-loop optimization process.

[0148] In one embodiment, the positioning awareness and recognition unit 2 includes:

[0149] The sensor fusion module is used to perform preliminary analysis of the collected data using vehicles, low-altitude aircraft, and fixed sensors. Based on the transmission bandwidth, the preliminary analysis results and some key raw data are sent to the central processing module.

[0150] The central processing module is used to integrate the raw data from different sensors and the preliminary analysis results, and identify the target behavior attributes through a hierarchical uncertainty modeling and transmission mechanism;

[0151] The target tracking module is used to track the position and motion status of aircraft and vehicles in real time based on the target tracking algorithm through the three-dimensional covariance matrix and multi-factor covariance prediction equation.

[0152] In one embodiment, the sensor fusion module includes:

[0153] The network monitoring module is used to obtain the current network available transmission bandwidth and bandwidth utilization information through network monitoring tools;

[0154] A preliminary analysis module is used to extract and fuse simple features based on vehicles, low-altitude aircraft, and fixed sensors to determine the data volume and transmission priority of key raw data;

[0155] A transmission allocation module is used to allocate transmission resources according to transmission bandwidth conditions and data characteristics;

[0156] Dynamic adjustment module, used to continuously monitor bandwidth changes and data transmission progress, and dynamically adjust transmission strategies;

[0157] The verification and optimization module is used to check data integrity and accuracy, and optimize transmission strategies and data processing methods based on the inspection results.

[0158] Specifically, the sensor fusion module implements environmental perception and target recognition based on the sensor fusion architecture, using a hybrid fusion architecture that combines the advantages of centralized and distributed fusion. First, the computing power of fixed sensors on vehicles, low-altitude aircraft, and roads and airspace is used to perform a preliminary analysis of the collected data, extracting and fusing simple features such as the approximate location and direction of movement of the target. The on-board camera locks the position of the object in front and the radar measures the target's movement trend. The preliminary processing results and some key raw data (including the camera's uncompressed image frames, radar echo signals, and lidar point cloud data in this embodiment) are then sent to the central processing module after comprehensive consideration of factors such as transmission bandwidth. The specific steps are as follows:

[0159] ① Evaluate transmission bandwidth conditions. Monitor the current network transmission bandwidth in real time. Use network monitoring tools to obtain information such as the currently available transmission bandwidth and bandwidth utilization. In this embodiment, network performance monitoring software is used to periodically query bandwidth usage data of network interfaces to understand the current network load.

[0160] ② Analyze data characteristics and requirements. Conduct a detailed analysis of the preliminary processing results and raw data. Clarify the data volume and data type (e.g., the format and size of simple features such as target location and direction of movement) of the preliminary processing results. Determine the data volume and transmission priority of this key raw data.

[0161] Specifically, simple features such as the approximate position and direction of movement of the target are fused, and the multi-sensor data benchmark is unified through time synchronization and spatial alignment. Based on the computing power of edge devices such as cameras and radars, the target position, speed, contour, category and other features are extracted respectively. In this embodiment, single sensor data is processed for spatiotemporal consistency using algorithms such as weighted averaging, Kalman filtering or Bayesian networks to generate a simplified feature vector containing position, motion trend and confidence, while retaining key original data fragments. Finally, after completing low-complexity fusion locally, the feature package is transmitted to the central module to achieve distributed real-time processing and data dimensionality reduction.

[0162] ③ Analyze the data transmission model based on bandwidth availability and data characteristics. Allocate transmission resources to ensure that important data is transmitted to the central processing module in a timely and accurate manner. Prioritize critical raw data and important preliminary processing results for transmission.

[0163] ④ Dynamically adjust the transmission strategy. During data transmission, continuously monitor bandwidth changes and data transmission progress. If bandwidth fluctuations, data transmission delays, or packet loss are detected, the transmission strategy is dynamically adjusted based on the transmission model and current conditions. In this embodiment, when bandwidth becomes narrow (transmission speed suddenly drops below 128 kbps), the transmission rate of some non-critical data is appropriately reduced (by 20%-50%) to ensure the transmission quality of critical data.

[0164] 5. Verification and Optimization. Data transmitted to the central processing module is verified for integrity and accuracy. Based on the verification results, the transmission strategy and data processing methods are optimized to continuously improve the efficiency and quality of data transmission. In this embodiment, data integrity and accuracy are determined by comparing the checksum of the received data with the original transmitted data, or by checking whether the data conforms to a specific format and range. If any problems are found, the cause is analyzed and the transmission parameters or data processing methods are adjusted.

[0165] In one embodiment, the central processing module includes:

[0166] The data acquisition module is used to analyze the impact of sensor accuracy and environmental noise on the collected data, select the corresponding probability distribution according to the characteristics of different sensors, and establish the probability density function of the collected data;

[0167] The feature extraction module is used to build an uncertainty model for feature extraction based on the quantitative results of algorithm limitations and data ambiguity;

[0168] The information fusion module is used to fuse the collected data from different sensors based on the uncertainty model and the weighted average information fusion algorithm.

[0169] Specifically, the central processing module ( Figure 4 After receiving the data from the central data processing unit (CPU), the system finely integrates the raw data from different sensors, such as combining the target appearance details of the optical camera with the internal structure information of the millimeter-wave radar to build a complete feature description. On the other hand, it integrates the preliminary analysis results obtained by local preprocessing and combines the local fused target motion direction with the 3D model reconstructed based on all sensors ( Figure 4 The vehicle data processing unit, road and airspace data processing unit, and low-altitude aircraft data processing unit) are used to accurately identify target behavior attributes. In this process, the present invention also proposes a hierarchical uncertainty modeling and transmission mechanism, that is, in the data acquisition stage, the uncertainty model is built for the collected data by considering sensor accuracy, environmental noise, etc. The feature extraction stage quantifies feature uncertainty based on algorithm limitations and data fuzziness. The information fusion stage combines the models of each link and reasonably integrates them according to the uncertainty weight. The specific steps are as follows:

[0170] ① Data acquisition link. First, analyze in detail the impact of sensor accuracy and environmental noise on the collected data. In this embodiment, for optical cameras, it is necessary to consider the impact of parameters such as resolution and sensitivity on the quality of captured images, as well as how environmental factors such as different lighting conditions and weather conditions introduce noise. For millimeter-wave radars, it is necessary to study the limitations of performance indicators such as transmission power and receiving sensitivity on detection results, as well as interference caused by factors such as the surrounding electromagnetic environment and object reflection characteristics. According to the characteristics of uncertainty, select the corresponding probability distribution to describe the uncertainty of the collected data. If the uncertainty of the data presents the characteristics of a Gaussian distribution, that is, most of the data is concentrated near the mean, and the farther away from the mean, the smaller the probability, then a Gaussian distribution can be used for modeling. In this embodiment, it is assumed that the measurement value x of a certain characteristic size of the target object collected by the optical camera obeys the mean Variance is Gaussian distribution , and get the probability density function The expression is:

[0171] ;

[0172] ② Feature extraction stage. Conduct an in-depth analysis of the feature extraction algorithm used to understand its performance under different data conditions. Combined with the quantitative results of the algorithm limitations and data ambiguity, an uncertainty model for feature extraction is constructed. By weightedly combining the algorithm's error rate d and the data's ambiguity index f, the uncertainty measure H for feature extraction is obtained. The uncertainty model is expressed as:

[0173] ;

[0174] Where, and is the weight coefficient determined according to the actual situation, and .

[0175] ③Information fusion link. According to the uncertainty model and weight, the weighted average information fusion algorithm is adopted. Assume that two features are to be fused and , and its corresponding uncertainty weight is and , the fused features .

[0176] Specifically, uncertainty consistency is maintained from the source of data collection, which can not only provide reliable results for target identification, but also provide accurate risk assessment basis for the decision-making system, thereby improving the overall efficiency of the system.

[0177] In one embodiment, the target tracking module includes:

[0178] The state initialization module is used to determine the initial state vector and three-dimensional covariance matrix of the target based on prior knowledge and initial measurement values;

[0179] The state prediction module is used to obtain the target's measurement values ​​based on different sensors, establish a state prediction equation, and use the state transition matrix to predict the target state at the current moment based on the state estimate at the previous moment;

[0180] The gain calculation module is used to establish a multi-factor covariance prediction equation using observation residuals, dynamically adjust the noise covariance, and calculate the Kalman gain based on the prediction covariance and the measurement noise covariance;

[0181] The state update module is used to update the state estimate through the Kalman gain based on the state update equation, combined with the predicted value and the measured value, and to update the covariance matrix to determine whether to continue tracking the target according to actual needs.

[0182] Specifically, if Figure 5 As shown in the figure, the target tracking module uses the Kalman filter target tracking algorithm to track the position and motion status of targets such as aircraft and vehicles in real time. The specific steps are as follows:

[0183] ① Initialize the state vector and covariance matrix. At the beginning of tracking, the initial state vector of the target is determined based on prior knowledge or initial measurement values. and covariance matrix , the expression of the initial state vector is:

[0184] ;

[0185] Where x, y and z represent the position coordinates, and Indicates the speed in the x, y and z directions, k represents the current moment, and k-1 represents the previous moment;

[0186] The expression of the three-dimensional covariance matrix is:

[0187] ;

[0188] Where, Represents the variance of position x, position y, and position z respectively; σ xy , σ xz , σ yz Represent the covariance between position x and position y, position x and position z, position y and position z respectively; Represents position x and velocity v in the x direction x The covariance between represents the velocity v in the x direction x and the velocity v in the y direction y The covariance between .

[0189] And so on, Represents the position x and velocity v in the y direction y The covariance between Represents position x and velocity v in z direction z The covariance between Represents the position y and velocity v in the x direction x The covariance between Represents position y and velocity v in the y direction y The covariance between Represents the position y and velocity v in the z direction z The covariance between Represents position z and velocity v in the x direction x The covariance between Represents the position z and velocity v in the y direction y The covariance between Represents position z and velocity v in the z direction z The covariance between represents the velocity v in the x direction x and the velocity v in the x direction x The covariance between represents the velocity v in the x direction x and the velocity v in the z direction z The covariance between represents the velocity v in the y direction y and the velocity v in the z direction z The covariance between .

[0190] It should be noted that covariance is an indicator of the linear relationship between two variables. In stationary motion, the covariance of position and velocity is 0, but in non-stationary motion, the covariance of position and velocity may not be zero.

[0191] Specifically, the three-dimensional covariance matrix in the present invention reflects the uncertainty of the initial state estimate. If the variance of a variable is large, it means that the initial estimate of this variable is very uncertain; if the covariance between two variables is not zero, it means that there is a certain correlation between the two variables.

[0192] ② Obtain measurement values. Obtain target measurement values, such as location information, from sensors (such as cameras and radars).

[0193] ③ State prediction. Use the state transition matrix to predict the target state at the current moment based on the state estimate at the previous moment. State prediction is performed based on the target's motion model. For the motion of a vehicle or aircraft on a plane, it is usually assumed to be a linear motion model. The state prediction equation is

[0194] ;

[0195] Where, is the state transfer matrix. For the uniform motion model, The expression is:

[0196] ;

[0197] Where, is the time interval between two measurements. The role of this matrix is ​​to estimate the state of the last moment (k-1) To predict the state at the current time (k) For example, if you know the position and speed of a vehicle at the previous moment, you can predict its position and speed at the current moment through this matrix.

[0198] ④ Adaptive covariance prediction. In the prediction stage, the covariance matrix of the state estimate also needs to be updated. Traditional Kalman filtering assumes that the covariance of process noise and observation noise is fixed. However, in actual applications, the noise characteristics may change over time. It is necessary to use the observation residual (the difference between the actual observation value and the predicted observation value) to dynamically adjust the noise covariance. The expression of the multi-factor covariance prediction equation constructed by the present invention is:

[0199] ;

[0200] Where, Represents the prior estimated covariance matrix at the current moment; Represents the posterior estimated covariance matrix of the previous moment; represents the process noise covariance matrix; represents the forgetting factor (in this embodiment, the value is 0.95~0.99); represents the attenuation factor (in this embodiment, the value is greater than 1); represents the expansion factor (in this embodiment, the value is greater than 1); represents the Kalman gain, represents the observation residual.

[0201] Specifically, in this solution, considering the motion characteristics of low-altitude aircraft and vehicles in highway scenarios, the settings of the three factors α, β, and γ are particularly critical. The forgetting factor α is set between 0.95 and 0.99, enabling the system to quickly "forget" the influence of historical data and promptly respond to sudden turns by aircraft or emergency lane changes by vehicles. The attenuation factor β is set greater than 1 to effectively suppress measurement noise caused by high-speed vehicles or airflow disturbances around the aircraft. The expansion factor γ is set greater than 1 to accommodate the increased uncertainty in the state estimation of aircraft and vehicles in various scenarios, such as tunnels, bridges, and inclement weather. The appropriate settings of these three factors enable the system to better adapt to the complexities of coordinated air-ground target motion in smart highway scenarios.

[0202] Specifically, process noise is used to consider the deviation of target motion from the ideal model caused by various factors in actual motion (such as uneven road surface, airflow disturbance, etc.). The value of needs to be adjusted and estimated based on the actual situation. Its diagonal elements can represent the noise intensity of each state variable. For example, the noise intensity of the velocity state variable may be greater than that of the position state variable, which reflects that velocity is more susceptible to external factors.

[0203] ⑤Calculate Kalman gain The Kalman gain is calculated based on the prediction covariance and the measurement noise covariance to balance the weights of the prediction value and the measurement value. The expression is:

[0204] ;

[0205] Where, is the measurement matrix, which maps the state vector to the measurement space; It is a robust factor that is dynamically adjusted according to the size of the observation residual and can correct the Kalman gain when there are abnormal observations. is the covariance adjustment term, which uses the observation residual to dynamically adjust the covariance. For simple position measurement, , indicating that only the position information is measured . is the measurement noise covariance matrix, which is used to describe the uncertainty in the measurement process. For example, different positioning devices with different accuracy will have different measurement noises, and this uncertainty is taken into account by Kalman gain. It balances the weight between the predicted state and the measured value. Its value is between 0 and 1. It determines whether to trust the predicted value or the measured value more based on the relative size of the predicted covariance and the measured covariance.

[0206] ⑥ State update. Combining the predicted value and the measured value, the state estimate is updated through the Kalman gain to obtain a more accurate target state. The expression of the state update equation is:

[0207] ;

[0208] Where, is the measured value at the current moment. The meaning of this equation is to use the predicted value A correction term is added, which is the Kalman gain multiplied by the difference between the measurement and the measurement estimate based on the predicted value. This gives a more accurate state estimate that combines the prediction and measurement. For example, if the predicted vehicle position deviates from the actual measured position, the Kalman gain is used to appropriately adjust the predicted position to make it closer to the actual position.

[0209] ⑦ Covariance update. Update the covariance matrix for the next prediction and update cycle. The covariance update equation is:

[0210] ;

[0211] Where, is the identity matrix. This step updates the uncertainty of the state estimate for the next prediction and update cycle. As measurements are continuously acquired and updated, the covariance matrix gradually converges, and the state estimate becomes increasingly accurate.

[0212] ⑧ Determine whether to continue tracking. Determine whether to continue tracking the target based on actual needs, such as whether the tracking system is shut down or the target is out of tracking range. If so, return to the step of obtaining measurement values; if not, end tracking.

[0213] In one embodiment, the collaborative decision control unit 3 is used to integrate the traffic data of aircraft and vehicles, plan the paths of aircraft and vehicles, and quickly detect emergencies through sensor and data fusion technology, and design a collaborative emergency processing mechanism. Figure 6 As shown in the figure, the specific implementation steps of the collaborative decision-making and control scheme for a smart highway control system that integrates low-altitude aircraft and vehicles are as follows: First, traffic data is collected and a traffic flow model is established. Based on this, traffic flow is collaboratively optimized. Simulated annealing algorithms are used to plan routes for aircraft and vehicles. Dynamic traffic control is implemented and measures are adjusted based on traffic conditions. Simultaneously, an emergency monitoring system is established to detect and classify emergencies and release information. An emergency response process is developed, resources are dispatched for collaborative operations, traffic signal priority is set, and aircraft are used to guide and coordinate. This entire process improves traffic efficiency, ensures safety, and allows for rapid response to emergencies, ultimately achieving smart highway control.

[0214] It should be noted that, in this disclosure, a heterogeneous network refers to a network composed of different types of network nodes and technologies. These nodes can be different base stations, access points, devices, or communication technologies. A smart gateway refers to a device used to connect and manage different types of devices and networks in application scenarios such as the Internet of Things and smart homes. ZigBee is a low-power wireless communication technology used to transmit data between devices. It is a short-range wireless communication technology commonly used in fields such as smart homes, industrial control, and medical equipment. A communication protocol is a set of rules for exchanging data between computer networks or devices. These protocols define the format, sequence, error handling, network management, and other details of data transmission. A communication protocol stack refers to a set of layered protocols used to implement data transmission and processing in computer networks. Each layer is responsible for specific functions and interacts with adjacent layers through clear interfaces. A Kalman filter is a recursive filter used to estimate the state of a dynamic system. It can estimate the current state based on previous measurements and a system model while reducing the impact of noise on the results.

[0215] It should also be noted that, in the present invention, the simulated annealing algorithm is a heuristic optimization algorithm inspired by the annealing process of metal materials. The algorithm is used to find the global optimal solution and is particularly suitable for combinatorial optimization problems and large-scale optimization problems; round-trip time refers to the time required for a data packet to be sent from the sender to the receiver and then returned from the receiver to the sender. In computer networks, RTT is an important indicator for measuring network delay; packet loss rate refers to the proportion of data packets lost due to various reasons during data transmission. In network communication, packet loss rate is an important performance indicator that reflects the health and reliability of the network; network detection package refers to a toolkit used to detect network status, performance or faults; prior estimation of covariance matrix is ​​a statistical method for estimating covariance matrix based on existing information or assumptions; a posteriori estimation of covariance matrix is ​​a statistical method for estimating covariance matrix given data.

[0216] To sum up, with the help of the above-mentioned technical solution of the present invention, through the heterogeneous network fusion unit 1, a heterogeneous network fusion and intelligent gateway system architecture is designed, a unified communication protocol framework is established, and seamless integration of different communication networks is achieved. The intelligent gateway device performs protocol conversion, data filtering and routing to adapt to the different needs of aircraft and vehicles; adaptive communication protocols and optimization technologies are used to automatically adjust the communication protocols according to different communication scenarios and needs, and optimize the communication protocol stack to ensure the communication quality between aircraft and vehicles, and between them and the traffic management center, and achieve low-latency, high-bandwidth data transmission. At the same time, through a reliable security encryption and identity authentication system, the security and privacy of the data interaction process are fully protected, effectively ensuring the efficiency and reliability of data transmission, and providing solid technical support for achieving efficient collaborative management and control. The present invention uses a positioning perception and recognition unit 2 to integrate multiple positioning technologies such as satellite positioning and inertial navigation using a high-precision positioning fusion algorithm, and fuses and processes information through filtering and other technologies. Regardless of the terrain and environment, the position of vehicles and low-altitude aircraft can be accurately determined. At the same time, the radar, camera, lidar and other sensors on aircraft and vehicles are comprehensively utilized, combined with fixed sensors on roads and airspace, such as meteorological sensors and traffic flow monitoring sensors, to achieve comprehensive perception of the environment, including each other's position, speed, direction and weather conditions. The raw data of different sensors are pre-processed and then transmitted to the central processing module for fusion processing. The features extracted by different sensors are fused using methods such as principal component analysis, thereby improving the accuracy of target detection and recognition and discovering potential dangers in advance. The target tracking algorithm based on Kalman filtering effectively utilizes the target's prior motion knowledge and real-time measurement data through two stages: state prediction and measurement update, and continuously corrects the tracking results. Even in the presence of noise and uncertainty, the target's position and motion state can be tracked more accurately in real time to ensure traffic safety. The present invention integrates the aircraft and vehicle operation data through the collaborative decision-making control unit 3, and uses the simulated annealing algorithm to collaboratively plan the flight path and driving route, so as to achieve the goals of minimizing the total driving time and reducing traffic congestion, implement dynamic traffic control measures, adjust the flow according to the real-time situation, improve the efficiency of road and airspace use, and realize the scientific optimization of traffic flow; at the same time, an emergency event monitoring system is established, which uses sensors and data fusion technology to quickly detect emergencies and classify and grade them, and reasonably dispatches resources through the designed collaborative processing mechanism and emergency response process. Once an emergency occurs, it can quickly coordinate resources such as aircraft and vehicles, guide rescue vehicles and others to quickly arrive at the accident scene, realize the collaborative operation of aircraft and vehicles, effectively improve the collaborative processing capability of the entire high-speed scene to deal with emergencies, ensure the safety and smoothness of high-speed traffic, and reduce accidents and delays caused by poor communication.

[0217] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. An intelligent high-speed control system for low-altitude aircraft and vehicles, characterized by: include: The heterogeneous network fusion unit is used to establish a unified communication protocol framework based on the communication network requirements of aircraft and vehicles, and formulate common communication rules and data formats to achieve protocol conversion and adaptation between different networks. It specifically includes: an intelligent gateway hardware module for connecting different types of networks and performing protocol conversion, converting the aircraft's communication protocol into a protocol that can be received by vehicles or highway infrastructure; and an intelligent gateway software module. The positioning perception and recognition unit integrates multiple positioning technologies, fuses raw data from different sensors, identifies target behavior attributes through a hierarchical uncertainty modeling and transfer mechanism, and establishes a target tracking algorithm to track the position and motion status of aircraft and vehicles in real time. Collaborative decision-making control unit, used to integrate traffic data from aircraft and vehicles, plan routes for aircraft and vehicles, and use sensor and data fusion technology to quickly detect emergencies and design collaborative emergency response mechanisms; The intelligent gateway software module includes: The protocol conversion module is used to implement protocol conversion and adaptation between different networks by adopting adaptive communication protocols and communication protocol stack optimization technology and building a communication performance evaluation model. Specifically, it evaluates the network status in real time through signal indicators fed back by network detection tools and the receiving end; formulates adaptive protocol parameter adjustment strategies based on network status and communication needs; builds a multi-dimensional performance evaluation index system based on network, application and user experience levels; uses the hierarchical analysis method to determine the weights of evaluation indicators at each level in the multi-dimensional performance evaluation index system, and maps measured data to different evaluation levels through membership functions to build a communication performance evaluation model; trains and optimizes the communication performance evaluation model based on historical data, and improves the accuracy and efficiency of protocol conversion by adjusting parameters; Data filtering module, which is used to filter and screen communication data and send the data to the correct destination according to different application requirements; Routing module, used to select the optimal communication path and improve data transmission efficiency; Security protection module, used to protect network security and privacy and prevent data from being stolen or tampered with; The positioning perception and recognition unit includes: The sensor fusion module is used to perform preliminary analysis of the collected data using vehicles, low-altitude aircraft, and fixed sensors. Based on the transmission bandwidth, the preliminary analysis results and some key raw data are sent to the central processing module. The central processing module is used to fuse the raw data and preliminary analysis results of different sensors and identify the target behavior attributes through a hierarchical uncertainty modeling and transmission mechanism. Specifically, it includes: the data acquisition module is used to analyze the impact of sensor accuracy and environmental noise on the collected data, select the corresponding probability distribution according to the characteristics of different sensors, and establish the probability density function of the collected data; the feature extraction module is used to construct the uncertainty model of feature extraction based on the quantitative results of algorithm limitations and data ambiguity; the information fusion module is used to fuse the collected data of different sensors using a weighted average information fusion algorithm based on the uncertainty model; The target tracking module is used to track the position and motion status of aircraft and vehicles in real time based on the target tracking algorithm through the three-dimensional covariance matrix and multi-factor covariance prediction equation.

2. The intelligent high-speed control system for low-altitude aircraft and vehicles according to claim 1 is characterized in that: The method uses the analytic hierarchy process to determine the weights of the evaluation indicators at each level in the multi-dimensional performance evaluation index system, and maps the measured data to different evaluation levels through the membership function to construct the performance evaluation model, including: Set evaluation indicators and determine the evaluation level based on actual application scenarios and needs; According to the characteristics and actual situation of the evaluation indicators, the membership function corresponding to each indicator is constructed, and the membership vector of each evaluation indicator for different evaluation levels is calculated; The fuzzy relationship matrix is ​​constructed by using the membership vectors of each evaluation index to different evaluation levels; The weight vector is determined by the hierarchical analysis method and synthesized with the fuzzy relationship matrix to obtain the comprehensive evaluation result vector. The maximum membership principle is used to determine the final evaluation level.

3. The intelligent high-speed control system for low-altitude aircraft and vehicles according to claim 2 is characterized in that: The expression of the membership function is: Where μ ij (x i ) is the membership degree of the i-th evaluation index to the j-th evaluation level; x i represents the actual measured value of the i-th evaluation index; a and b represent the upper limit constant and lower limit constant of the j-th evaluation level respectively; The expression of the fuzzy relationship matrix is: Where R represents the fuzzy relationship matrix; the i-th row of R represents the membership of the i-th evaluation indicator to each evaluation level; the j-th column of R represents the membership of all evaluation indicators to the j-th evaluation level.

4. The intelligent high-speed control system for low-altitude aircraft and vehicles according to claim 1 is characterized in that: The sensor fusion module includes: The network monitoring module is used to obtain the current network available transmission bandwidth and bandwidth utilization information through network monitoring tools; A preliminary analysis module is used to extract and fuse simple features based on vehicles, low-altitude aircraft, and fixed sensors to determine the data volume and transmission priority of key raw data; A transmission allocation module is used to allocate transmission resources according to transmission bandwidth conditions and data characteristics; Dynamic adjustment module, used to continuously monitor bandwidth changes and data transmission progress, and dynamically adjust transmission strategies; The verification and optimization module is used to check data integrity and accuracy, and optimize transmission strategies and data processing methods based on the inspection results.

5. The intelligent high-speed control system for low-altitude aircraft and vehicles according to claim 1 is characterized in that: The target tracking module includes: The state initialization module is used to determine the initial state vector and three-dimensional covariance matrix of the target based on prior knowledge and initial measurement values; The state prediction module is used to obtain the target's measurement values ​​based on different sensors, establish a state prediction equation, and use the state transition matrix to predict the target state at the current moment based on the state estimate at the previous moment; The gain calculation module is used to establish a multi-factor covariance prediction equation using observation residuals, dynamically adjust the noise covariance, and calculate the Kalman gain based on the prediction covariance and the measurement noise covariance; The state update module is used to update the state estimate through the Kalman gain based on the state update equation, combined with the predicted value and the measured value, and to update the covariance matrix to determine whether to continue tracking the target according to actual needs.

6. The intelligent high-speed control system for low-altitude aircraft and vehicles according to claim 5 is characterized in that: The expression of the three-dimensional covariance matrix is: Where, Respectively represent the variance of position x, position y, and position z; σ xy , σ xz , σ yz Represent the covariance between position x and position y, position x and position z, position y and position z respectively; Represents position x and velocity v in the x direction x The covariance between represents the velocity v in the x direction x and the velocity v in the y direction y The covariance between Represents the position x and velocity v in the y direction y The covariance between Represents position x and velocity v in z direction z The covariance between Represents the position y and velocity v in the x direction x The covariance between Represents position y and velocity v in the y direction y The covariance between Represents the position y and velocity v in the z direction z The covariance between Represents position z and velocity v in the x direction x The covariance between Represents position z and velocity v in the y direction y The covariance between Represents position z and velocity v in the z direction z The covariance between represents the velocity v in the x direction x and the velocity v in the x direction x The covariance between represents the velocity v in the x direction x and the velocity v in the z direction z The covariance between represents the velocity v in the y direction y and the velocity v in the z direction z The covariance between The expression of the multi-factor covariance prediction equation is: Where, P k|k-1 represents the prior estimated covariance matrix at the current moment; P k-1|k-1 represents the posterior estimated covariance matrix of the previous moment; Q k represents the process noise covariance matrix; α represents the forgetting factor; β represents the attenuation factor; γ represents the expansion factor; K k represents the Kalman gain; represents the observation residual; F k represents the state transition matrix.

Images