Low-delay communication control system and method for analog machine in fixed flight training route

Abstract

The invention relates to the technical field of space flight and aviation, in particular to a low-delay communication control system and method for an analog machine in a fixed flight training course. The control system comprises a multi-source sensor fusion module, a data simplification processing module, a dynamic time delay optimization communication framework, a distributed intelligent decision-making terminal, a virtual-real mapping synchronous controller, a milli delay regulation and control module and an intelligent sensing alarm interaction module. According to the invention, 4G, 5G and satellite double links are integrated, an optimal link is dynamically selected by adopting a Q-Learning algorithm, millisecond-level time delay control is realized, transmission reliability is improved through a packet loss rate compensation algorithm, multi-sensor data is fused based on an FPGA hardware accelerator and an EKF algorithm, and a data compression ratio is improved in combination with Huffman coding and Kalman filtering. The bandwidth requirement is greatly reduced, through software and hardware collaborative optimization, the end-to-end time delay is reduced to be within hundreds of milliseconds, the virtual-real synchronization precision is improved to the centimeter level, and the decision response speed is improved by two times.

Description

Technical Field

[0001] This invention relates to the field of aerospace technology, specifically to a low-latency communication control system and method for simulators on fixed flight training routes. Background Technology

[0002] In fixed flight training routes, the simulator, as the core tool for pilot training, needs to interact in real time with the ground control center and digital twin platform through the communication system to achieve two-way transmission of data such as flight attitude, environmental parameters, and equipment status. Existing technologies usually rely on traditional communication protocols (such as TCP / IP) and general sensor fusion algorithms, using a single communication link (such as 4G or satellite communication) for data transmission, and combining simple trajectory prediction models to provide auxiliary decision-making. However, due to the extremely high requirements of flight training for latency, data integrity, and real-time decision-making, the performance bottleneck of existing technologies in complex network environments (such as multi-link switching and highly dynamic scenarios) is gradually becoming apparent, resulting in limited training efficiency and safety.

[0003] The communication protocol is not optimized for flight training scenarios, and is prone to latency jitter and packet loss during multi-link switching. The delay is usually more than 50ms, which affects the real-time performance of decision-making. Sensor data is not efficiently compressed, resulting in excessive bandwidth occupancy and actual utilization of less than 30%. It is especially prone to transmission congestion under low-bandwidth networks. The trajectory prediction model relies on a single historical data source, which is difficult to adapt to sudden environmental changes. Furthermore, the virtual-real synchronization delay (usually >300ms) causes the virtual sandbox to deviate from the actual flight state by more than 1 meter, affecting training accuracy. Summary of the Invention

[0004] The purpose of this invention is to provide a low-latency communication control system and method for simulators in fixed flight training routes, which solves the problem of high latency in current simulators.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a low-latency communication control system for a simulator on a fixed flight training route, the control system comprising a multi-source sensor fusion module, a data simplification and processing module, a dynamic latency optimization communication framework, a distributed intelligent decision terminal, a virtual-real mapping synchronization controller, a millisecond delay control module, and an intelligent perception alarm interaction module.

[0006] Preferably, the multi-source sensor fusion module consists of a high-precision inertial measurement unit, differential GPS, barometric altimeter, and lidar, and the data simplification processing module incorporates a Kalman filter algorithm and a Huffman coding compression chip.

[0007] Preferably, the dynamic latency optimization communication framework includes 4G, 5G and satellite link modules, the distributed intelligent decision terminal has a built-in trajectory prediction model, and the intelligent sensing alarm interaction module consists of an AR device and a touch screen.

[0008] A low-latency communication control method for simulators operating on fixed flight training routes, comprising the following steps: Step S1: During the flight simulation, flight data is collected in real time through the multi-source sensor fusion module. The sensor fusion algorithm based on extended Kalman filter is applied to eliminate the pose error between GPS and IMU, and the six-degree-of-freedom pose data in the flight simulation is fused. The collected data is prioritized and fragmented using the QoS tagging method to generate compressed data packets with timestamps and check codes. Step S2: The compressed data packets are transmitted through a dynamic latency optimization communication framework. The system adopts a link decision model based on Q-Learning to implement communication, establishes a reward function, selects the optimal communication link through weighted calculation, and during the data transmission process, the data first enters the LSTM trajectory prediction model to output the future simulated path. Step S3: Forward prediction compensation, using historical latency distribution to predict the effective time of the current instruction, using hash chain to verify the consistency of data synchronization status, triggering resynchronization when the difference exceeds the limit, using electronic sand table to dynamically mark conflict hotspots on VR device, and displaying the hotspots of each conflicting data based on Delaunay triangulation algorithm; Step S4: Based on lightweight blockchain technology, the operation log is stored and the simulated path is recorded. When an abnormal operation occurs, the system triggers a three-level alarm and provides voice prompts and warnings on the operation interface. At the same time, the system uses a multi-objective optimization algorithm to generate alternative solutions and projects alternative paths on the AR interface. The operator confirms the execution of the optimal solution through voice and completes the simulated flight and data transmission.

[0009] Preferably, in step S1, during the compression process, the Kalman filter algorithm built into the data simplification module and the Huffman coding compression chip work together to maintain the integrity and accuracy of the data, as shown in the adaptive quantization formula below: ; In the formula: Q(x) is the quantized value, x is the original value, and q s It is the quantization step size; The sensor fusion algorithm based on extended Kalman filtering has the following fusion formula: ; In the formula: Xk|k is the state estimate, K k It is the Kalman gain, k is the observation value, and H is the Kalman gain. k It is the observation matrix.

[0010] Preferably, in step S2, when the predicted path deviates from the training route by ≥3 meters, the system automatically generates a correction instruction.

[0011] Preferably, in step S2, the reward function of the Q-Learning link decision model is as follows: ; In the formula: R(s,a) is the reward value, B(a) is the link bandwidth, D(a) is the latency, C(a) is the cost, and α, β, and γ are the weights.

[0012] Preferably, in step S2, the LSTM trajectory prediction model formula is as follows: ; Where: h t It is in a hidden state, x t It is the input, W h and b h These are weights and biases.

[0013] Preferably, in step S3, the instruction issuance time = current delay + predicted compensation amount, and the predicted compensation amount is a sliding window mean filter.

[0014] Preferably, in step S4, the alternative scheme executes commands such as detour, deceleration, or emergency landing.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention integrates 4G, 5G, and satellite dual links, employs the Q-Learning algorithm to dynamically select the optimal link, achieves millisecond-level latency control, and improves transmission reliability through a packet loss rate compensation algorithm. Based on an FPGA-based hardware accelerator and the EKF algorithm, it fuses multi-sensor data, combining Huffman coding and Kalman filtering to improve data compression ratio and significantly reduce bandwidth requirements. Through hardware and software co-optimization, end-to-end latency is reduced to less than 100 milliseconds, virtual-real synchronization accuracy is improved to centimeter level, and decision response speed is doubled. Simultaneously, through data compression and link adaptive technology, network bandwidth requirements are significantly reduced, providing a highly reliable, low-latency, and highly intelligent communication and control solution for flight training. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the method of the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] A low-latency communication control system for simulators operating on fixed flight training routes is presented. This system comprises a multi-source sensor fusion module, a data simplification and processing module, a dynamic latency-optimized communication framework, a distributed intelligent decision-making terminal, a virtual-real mapping synchronization controller, a millisecond latency control module, and an intelligent sensing and alarm interaction module. The multi-source sensor fusion module consists of a high-precision inertial measurement unit (IMU), differential GPS (DGPS), barometric altimeter, and lidar, which collects flight attitude, position, velocity, and environmental data in real time at a sampling rate ≥1kHz. Through an FPGA-based hardware accelerator, it achieves rapid data acquisition and preliminary processing. The data simplification and processing module uses a built-in Kalman filter algorithm and Huffman coding compression chip to perform noise reduction and compression on the collected data, achieving a compression ratio ≥5:1, effectively reducing transmission bandwidth requirements. The dynamic latency-optimized communication framework includes 4G / 5G and satellite link modules, supports a dynamic priority scheduling protocol, and automatically switches links based on network status. With latency jitter ≤10ms and packet loss rate compensation algorithm to ensure data transmission reliability, the distributed intelligent decision terminal utilizes an AI-powered chip and a built-in trajectory prediction model to generate flight path correction commands in real time with a response time ≤50ms, providing intelligent decision support for simulated flight. The virtual-real mapping synchronization controller, based on a WebSocket protocol API gateway, interacts bidirectionally with the cloud-based digital twin platform, synchronizing the aircraft status to the virtual sandbox every 200ms with an error tolerance ≤0.1m, achieving precise mapping between virtual and reality. It features a built-in timestamp synchronization chip and latency compensation algorithm (PID controller), ensuring end-to-end latency control ≤100ms and dynamically adjusting data stream priority to ensure priority transmission of critical data. The intelligent perception alarm interaction module consists of an AR device and a touchscreen, integrating a voice alarm module to visualize flight status and conflict alarms in real time, and providing voice feedback for operation commands in the AR device, enhancing the human-computer interaction experience.

[0019] A low-latency communication control method for simulators operating on fixed flight training routes, comprising the following steps: Step S1: During the simulated flight, the system first collects flight data in real time through a multi-source sensor fusion module. A sensor fusion algorithm based on Extended Kalman Filter (EKF) is applied to eliminate GPS and IMU pose errors and fuse six-DOF pose data from the simulated flight. This six-DOF pose data includes various parameters in the coordinate system (x, y, z, roll, pitch, yaw). Subsequently, the collected data is prioritized and fragmented using the QoS (Quality-of-Service) tagging method to generate compressed data packets with timestamps and checksums. During compression, the Kalman filter algorithm built into the data simplification module and the Huffman coding compression chip work together to ensure a data compression ratio ≥ 5:1 while maintaining data integrity and accuracy. The adaptive quantization formula is shown below: ; In the formula: Q(x) is the quantized value, x is the original value, and q s It is the quantization step size; The sensor fusion algorithm based on extended Kalman filtering has the following fusion formula: ; In the formula: Xk|k is the state estimate, K k It is the Kalman gain, k is the observation value, and H is the Kalman gain. k It is the observation matrix; Step S2: The compressed data packets are transmitted through a dynamic latency optimization communication framework. The system uses a link decision model based on Q-Learning for communication and establishes a reward function. This function comprehensively considers factors such as bandwidth, latency, and cost, and selects the optimal communication link through weighted calculation, such as a 4G or 5G main link, or satellite backup. During data transmission, the data first enters the LSTM (Long Short-Term-Memory) trajectory prediction model. This model uses historical simulated flight trajectories and simulated environmental data to output the simulated path for the next 5 seconds. When the predicted path deviates from the training route by ≥3 meters, the system automatically generates a correction command to ensure the accuracy of the flight path. The reward function of the Q-Learning link decision model is shown below: ; In the formula: R(s,a) is the reward value, B(a) is the link bandwidth, D(a) is the latency, C(a) is the cost, and α, β, and γ are the weights; The formula for the LSTM trajectory prediction model is as follows: ; Where: h t It is in a hidden state, x t It is the input, W h and bh These are weights and biases; Step S3: Forward prediction compensation. The system uses historical latency distribution to predict the effective time of the current instruction. Instruction issuance time = current latency + predicted compensation amount. The predicted compensation amount is calculated using a sliding window mean filter. A hash chain is used to verify the consistency of data synchronization. Resynchronization is triggered when the difference exceeds the limit. On the VR device, an electronic sand table dynamically marks conflict hotspots. Based on the Delaunay triangulation algorithm, the hotspots of each conflicting data are displayed. Before the instruction is issued, the system uses a forward prediction compensation algorithm to predict the effective time of the current instruction using historical latency distribution. The instruction issuance time equals the current latency plus the predicted compensation amount, which is calculated using a sliding window mean filter algorithm. Simultaneously, the system uses a hash chain to verify the consistency of data synchronization. When the difference exceeds the limit, a resynchronization mechanism is triggered. On the VR device, the system uses an electronic sand table to dynamically mark conflict hotspots and displays the hotspots of each conflicting data based on the Delaunay triangulation algorithm, providing operators with intuitive conflict warnings. Step S4: To ensure the safety of the simulated flight, the system uses lightweight blockchain technology to store operation logs. All simulated path records are securely stored and support post-event traceability. When abnormal operations occur, such as command tampering or significant deviation of the simulated course, the system triggers a level three alarm and provides voice prompts and warnings on the operation interface. At the same time, the system uses a multi-objective optimization algorithm to generate alternative solutions and executes commands such as detour, deceleration, and emergency landing. Alternative paths are projected on the AR interface, and the operator can confirm the execution of the optimal solution via voice, completing the simulated flight and data transmission.

[0020] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A low-latency communication control system for a simulator on a fixed flight training route, characterized in that: The control system includes a multi-source sensor fusion module, a data simplification and processing module, a dynamic delay optimization communication framework, a distributed intelligent decision-making terminal, a virtual-real mapping synchronization controller, a millisecond delay control module, and an intelligent sensing alarm interaction module.

2. The low-latency communication control system for a simulator on a fixed flight training route according to claim 1, characterized in that: The multi-source sensor fusion module consists of a high-precision inertial measurement unit, differential GPS, barometric altimeter, and lidar. The data simplification and processing module incorporates a Kalman filter algorithm and a Huffman coding compression chip.

3. The low-latency communication control system for a simulator on a fixed flight training route according to claim 1, characterized in that: The dynamic latency optimization communication framework includes 4G, 5G and satellite link modules, the distributed intelligent decision terminal has a built-in trajectory prediction model, and the intelligent perception alarm interaction module consists of an AR device and a touch screen.

4. A low-latency communication control method for a simulator on a fixed flight training route, characterized in that: The method includes the following steps: Step S1: During the flight simulation, flight data is collected in real time through the multi-source sensor fusion module. The sensor fusion algorithm based on extended Kalman filter is applied to eliminate the pose error between GPS and IMU, and the six-degree-of-freedom pose data in the flight simulation is fused. The collected data is prioritized and fragmented using the QoS tagging method to generate compressed data packets with timestamps and check codes. Step S2: The compressed data packets are transmitted through a dynamic latency optimization communication framework. The system adopts a link decision model based on Q-Learning to implement communication, establishes a reward function, selects the optimal communication link through weighted calculation, and during the data transmission process, the data first enters the LSTM trajectory prediction model to output the future simulated path. Step S3: Forward prediction compensation, using historical latency distribution to predict the effective time of the current instruction, using hash chain to verify the consistency of data synchronization status, triggering resynchronization when the difference exceeds the limit, using electronic sand table to dynamically mark conflict hotspots on VR device, and displaying the hotspots of each conflicting data based on Delaunay triangulation algorithm; Step S4: Based on lightweight blockchain technology, the operation log is stored and the simulated path is recorded. When an abnormal operation occurs, the system triggers a three-level alarm and provides voice prompts and warnings on the operation interface. At the same time, the system uses a multi-objective optimization algorithm to generate alternative solutions and projects alternative paths on the AR interface. The operator confirms the execution of the optimal solution through voice and completes the simulated flight and data transmission.

5. The low-latency communication control method for simulators on fixed flight training routes according to claim 4, characterized in that: In step S1, during the compression process, the Kalman filter algorithm built into the data simplification module and the Huffman coding compression chip work together to maintain the integrity and accuracy of the data, as shown in the adaptive quantization formula below: ； In the formula: Q(x) is the quantized value, x is the original value, and q s It is the quantization step size; The sensor fusion algorithm based on extended Kalman filtering has the following fusion formula: ； In the formula: Xk|k is the state estimate, K k It is the Kalman gain, k is the observation value, and H is the Kalman gain. k It is the observation matrix.

6. The low-latency communication control method for a simulator in a fixed flight training route according to claim 1, characterized in that: In step S2, when the predicted path deviates from the training route by ≥3 meters, the system automatically generates a correction instruction.

7. The low-latency communication control method for a simulator in a fixed flight training route according to claim 1, characterized in that: In step S2, the reward function of the Q-Learning link decision model is as follows: ； In the formula: R(s,a) is the reward value, B(a) is the link bandwidth, D(a) is the latency, C(a) is the cost, and α, β, and γ are the weights.

8. The low-latency communication control method for a simulator in a fixed flight training route according to claim 1, characterized in that: In step S2, the formula for the LSTM trajectory prediction model is as follows: ； Where: h t It is in a hidden state, x t It is the input, W h and b h These are weights and biases.

9. The low-latency communication control method for a simulator on a fixed flight training route according to claim 1, characterized in that: In step S3, the instruction issuance time = current delay + predicted compensation amount, and the predicted compensation amount is a sliding window mean filter.

10. The low-latency communication and control method for a simulator in a fixed flight training route according to claim 1, characterized in that: In step S4, the alternative solutions execute commands such as detour, deceleration, and emergency landing.

Images