Flight model machine and control tower dialogue generation system based on artificial intelligence

Abstract

The invention relates to the technical field of aircraft communication, in particular to an artificial intelligence-based flight model machine and control tower dialogue generation system, which comprises a multi-modal data acquisition module, a data processing module, a data processing module, a data processing module, a data processing module, a data processing module, a data processing module and a data processing module, wherein the multi-modal data acquisition module is used for collecting various sensor data from a flight model machine and meteorological conditions and airspace state external information; sensor data comprises a positioning sensor, a flight speed sensor and a flight height sensor, the system integrates multiple technologies such as multi-modal data acquisition, a deep learning model and context awareness, efficient, accurate and natural language communication between an aircraft and a ground control center is achieved, the authenticity and interactivity of flight training are improved, and the system is suitable for popularization and application. Through automatic and intelligent dialogue generation, the risk of human errors is remarkably reduced, the flight safety and reliability are improved, the system is suitable for various flight simulation training occasions and actual flight operation, and a training environment closer to an actual combat is provided for a pilot.

Description

Technical Field

[0001] The present invention relates to the field of aircraft communication technology, and in particular to a flight model aircraft and control tower dialogue generation system based on artificial intelligence. Background Art

[0002] In traditional flight training and actual operations, the communication between pilots and the tower mainly relies on preset scripts or manual guidance. However, this communication method has many significant shortcomings, which constitute an important background for the development of an artificial intelligence-based flight model aircraft and tower dialogue generation system.

[0003] First, the traditional method is inefficient. Due to the complex and changeable flight environment, preset scripts often cannot cover all possible communication scenarios, resulting in the need to frequently switch scripts or seek manual guidance in actual operations, thus reducing communication efficiency. This inefficient communication method not only affects the progress of flight training, but may also cause unnecessary delays or risks in actual flights.

[0004] Secondly, the flexibility of traditional methods is poor. When faced with emergencies, preset scripts often lack the ability to be flexibly adjusted and cannot make timely and effective responses based on actual conditions. This increases the difficulty of response for pilots and tower commanders and may also pose a threat to flight safety. In contrast, a dialogue generation system that can flexibly adapt to different flight scenarios and emergencies will have higher practical value1.

[0005] In addition, traditional methods also have the problem of high risk of human error. Due to reliance on manual judgment and operation, pilots and tower controllers are prone to misjudgment or operational errors during communication. These errors may lead to misunderstanding or improper execution of flight instructions, which in turn has a serious impact on flight safety. Therefore, reducing the risk of human error is the key to improving flight safety.

[0006] Finally, traditional methods also have shortcomings in multi-language support. In cross-border flight scenarios, communication barriers between different languages ​​limit the widespread application of the system. Pilots and tower controllers may need to master multiple languages ​​to communicate effectively, which increases their burden and training costs. A multi-language dialogue generation system will be able to break this limitation and promote international flight coordination and cooperation. Summary of the invention

[0007] 1. Technical issues to be solved

[0008] In view of the deficiencies of the prior art, the present invention provides a flight model aircraft and control tower dialogue generation system based on artificial intelligence.

[0009] (II) Technical solution

[0010] To achieve the above-mentioned purpose, the present invention provides the following technical solution: The artificial intelligence-based flight model aircraft and tower dialogue generation system of the present invention comprises:

[0011] Multimodal data acquisition: used to collect various sensor data from the flight model aircraft and external information on meteorological conditions and airspace status provided by the tower. The sensor data includes positioning sensors, flight speed sensors, and flight altitude sensors.

[0012] Deep learning model construction:

[0013] Build a deep neural network (DNN)-based dialogue generation model that is trained on a large number of real flight logs and voice recordings to generate powerful semantic understanding and generation capabilities;

[0014] Use reinforcement learning to optimize model parameters to ensure high accuracy in complex environments;

[0015] Situational awareness module: dynamically adjusts the dialogue strategy according to the current flight mission and environmental conditions for different flight phases and emergencies;

[0016] Natural language processing unit: used to convert the generated text commands into clear and understandable voice output, and supports several language options;

[0017] Feedback mechanism: Collect feedback information through user interaction to continuously improve the quality and accuracy of conversations;

[0018] Security verification layer: All generated conversation content is subject to strict security checks to ensure compliance with international aviation regulations and operating specifications;

[0019] Cloud Collaboration Platform: Allows flight model aircraft to connect to remote servers, utilizing the powerful computing resources and storage capabilities of cloud services to further improve the speed and quality of dialogue generation.

[0020] Preferably, the multimodal data acquisition device further comprises:

[0021] Visual sensors: including cameras and infrared cameras, used to capture visual information around the aircraft;

[0022] Acoustic sensor: used to monitor the surrounding sound environment and identify potential threats;

[0023] Airflow sensor: used to monitor meteorological parameters such as wind speed and direction to assist flight decision-making;

[0024] Radio receiver: used to receive radio signals from the tower and other aircraft.

[0025] Further preferably, the deep learning model also includes:

[0026] Pre-training module: Use large-scale public datasets for preliminary training to improve the basic performance of the model;

[0027] Adaptive learning module: continuously updates model parameters based on actual flight data to fit specific application scenarios;

[0028] Scenario simulation module: By simulating dialogue scenarios under different flight conditions, the model's ability to deal with emergencies is enhanced.

[0029] Again preferably, the situational awareness module is capable of identifying the following flight phases and adjusting the dialogue strategy accordingly:

[0030] Pre-takeoff preparation phase;

[0031] Take-off process phase;

[0032] Cruise phase;

[0033] Approach and landing phase;

[0034] Emergency response phase.

[0035] Preferably, the natural language processing unit also has the following functions:

[0036] Text-to-speech conversion: converting generated text instructions into natural and fluent speech;

[0037] Speech-to-text conversion: Parsing voice input from a tower or other source and converting it into a processable text format;

[0038] Sentiment Analysis: Assess the tone and sentiment in a conversation so you can respond more humanely.

[0039] Further preferably, the feedback mechanism includes:

[0040] Real-time monitoring system: allows operators to instantly view the generated conversation content and its effects;

[0041] Manual Correction Programmer: allows users to directly modify inaccurate or inappropriate conversation content and add it to the model as new training samples;

[0042] Automatic update mechanism: Regularly summarize user feedback, automatically adjust model parameters, and optimize conversation quality.

[0043] Again preferably, the security verification layer comprises:

[0044] Rule base: a predefined set of standards and specifications used to check whether the generated dialogue violates aviation regulations;

[0045] Anomaly detection algorithm: used to identify conversation content that may cause security risks and trigger alarms;

[0046] Human review channel: When necessary, key conversations are submitted to professionals for final review.

[0047] Preferably, the cloud collaboration platform provides the following services:

[0048] Distributed computing resources: Allocate multiple server nodes to speed up the conversation generation process;

[0049] Data backup and recovery: ensure safe storage and fast recovery of flight data;

[0050] Cross-regional collaboration: Supporting multilingual communication and cross-border cooperation around the world;

[0051] Intelligent scheduling system: Rationally arrange computing resources according to network conditions and task priorities.

[0052] Further preferably, the communication interface supports the following communication protocols:

[0053] Very high frequency (VHF) and ultra high frequency (UHF) radio communications: used for direct contact with the tower and other aircraft over short distances;

[0054] Satellite communications: used for data transmission over long distances or in remote areas;

[0055] Wi-Fi and cellular networks: for convenient connection between ground stations and flight models;

[0056] Bluetooth and Zigbee: For simple communication between devices over short distances.

[0057] Again preferably, a user training module is also included,

[0058] Virtual reality (VR) training environment: provides an immersive flight simulation experience to help novice pilots become familiar with system operations;

[0059] Online courses and tutorials: provide rich learning materials and technical support through the Internet platform;

[0060] Interactive Practice Mode: Allows users to practice conversational skills in a controlled environment and gradually improve proficiency.

[0061] (III) Beneficial effects

[0062] Compared with the prior art, the present invention provides a flight model aircraft and tower dialogue generation system based on artificial intelligence, which has the following beneficial effects:

[0063] Efficient, accurate and natural language communication:

[0064] The system integrates multiple technologies such as multimodal data acquisition, deep learning models, and situational awareness, enabling efficient, accurate, and natural language communication between the aircraft and the ground control center, and improving the authenticity and interactivity of flight training.

[0065] Reduce the risk of human error:

[0066] Through automated and intelligent dialogue generation, the risk of human error is significantly reduced, improving flight safety and reliability.

[0067] Wide applicability:

[0068] The system is suitable for various flight simulation training scenarios and actual flight operations, providing pilots with a training environment that is closer to actual combat.

[0069] Multimodal data collection improves environmental perception:

[0070] Through positioning sensors, visual sensors, acoustic sensors and other sensors, the system can comprehensively capture flight environment information and assist in environmental perception and decision-making.

[0071] Deep learning models enhance adaptability:

[0072] The pre-training module and adaptive learning module enable the system to quickly adapt to new environments and maintain high accuracy and stability under different conditions.

[0073] The context-aware module optimizes the dialogue strategy:

[0074] Dynamically adjust dialogue strategies according to flight phases and emergencies to ensure smooth and effective communication.

[0075] Natural language processing improves communication efficiency:

[0076] Text-to-speech (TTS) and speech-to-text (STT) support multilingual communication, and sentiment analysis makes responses more humane and appropriate.

[0077] Feedback mechanisms promote continuous optimization:

[0078] A real-time monitoring system, a manual correction programmer and an automatic update mechanism together ensure continuous optimization and improvement of the conversation quality.

[0079] Safety verification layer ensures flight safety:

[0080] The rule base, anomaly detection algorithm and manual review channel together form a safety barrier to ensure that all instructions are legal and compliant and prevent accidents.

[0081] The cloud collaboration platform provides strong support:

[0082] Distributed computing resources, data backup and recovery, cross-regional collaboration and intelligent scheduling systems provide strong support for the efficient operation of the system.

[0083] A variety of communication interfaces to meet different needs:

[0084] The system supports multiple communication methods such as VHF, UHF, satellite communication, Wi-Fi, cellular network, Bluetooth and Zigbee to meet the communication needs in different scenarios.

[0085] User training modules shorten training cycles:

[0086] Virtual reality training environment, online courses and tutorials, and interactive practice mode provide users with rich training resources to help them quickly master system operation and improve their skill level. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] Figure 1 It is a schematic diagram of the system module of the present invention;

[0088] Figure 2 It is a detailed working process schematic diagram of the present invention; DETAILED DESCRIPTION

[0089] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0090] See also Figure 1-2 The artificial intelligence-based flight model aircraft and tower dialogue generation system of the present invention aims to achieve efficient, accurate and natural language communication between the aircraft and the ground control center by integrating multiple components such as multimodal data acquisition, deep learning model, situational awareness module, natural language processing unit, feedback mechanism, safety verification layer and cloud collaboration platform. This system not only improves the authenticity and interactivity of flight training, but also significantly reduces the risk of human error. It is suitable for various flight simulation training occasions and actual flight operations.

[0091] Working principles of each preferred technical solution

[0092] Multimodal data acquisition

[0093] Positioning sensor, flight speed sensor and flight altitude sensor: used to obtain basic parameters of flight status to ensure accurate flight path planning.

[0094] Visual sensor (camera, infrared camera): captures visual information around the aircraft, assists in environmental perception, and identifies obstacles and other aircraft.

[0095] Acoustic sensor: monitors the surrounding sound environment and identifies potential threats such as other aircraft or abnormal noise to enhance flight safety.

[0096] Airflow sensor: monitors meteorological parameters such as wind speed and direction to assist flight decisions and optimize flight paths.

[0097] Radio receiver: Receives radio signals from the tower and other aircraft to ensure smooth communication and support real-time command and dispatch.

[0098] Deep learning model construction

[0099] Pre-training module: Use large-scale public data sets for preliminary training to improve the basic performance of the model, enable it to have basic semantic understanding and generation capabilities, and quickly adapt to new environments.

[0100] Adaptive learning module: Continuously updates model parameters based on actual flight data to make the model more suitable for specific application scenarios, enhance its adaptability under different conditions, and ensure long-term stable operation.

[0101] Scenario simulation module: By simulating dialogue scenarios under different flight conditions, the model's ability to respond to emergencies is improved, ensuring high accuracy and providing reliable instructions in complex environments.

[0102] Situational Awareness Module

[0103] Flight phase identification:

[0104] Preparation before takeoff: Check whether all equipment is normal and confirm takeoff conditions;

[0105] Take-off process: monitor various parameters during take-off to ensure smooth lift-off;

[0106] Cruise phase: Maintain stable flight and adjust flight routes to cope with weather changes;

[0107] Approach and landing: accurately navigate to the target location to ensure safe landing;

[0108] Emergency handling: Identify and respond to emergencies and take emergency measures to ensure safety.

[0109] Dynamically adjust dialogue strategies: According to the current flight mission and environmental conditions, select the most appropriate dialogue template to adapt to different flight phases and emergencies to ensure smooth and effective communication.

[0110] Natural Language Processing Unit

[0111] Text-to-speech (TTS): Converts generated text instructions into natural and fluent speech output, supports multiple language options, and ensures barrier-free cross-border communication.

[0112] Speech-to-Text (STT): Parses voice input from the tower or other sources and converts it into a processable text format for further analysis and response, improving communication efficiency.

[0113] Sentiment analysis: Evaluate the tone and emotions in a conversation to make responses more humane and appropriate, enhancing the user experience.

[0114] Feedback Mechanism

[0115] Real-time monitoring system: allows operators to instantly view the generated conversation content and its effects, ensure the effectiveness of communication, identify problems in a timely manner and make adjustments.

[0116] Manual Correction Programmer: Allows users to directly modify inaccurate or inappropriate conversation content and add it to the model as new training samples to continuously optimize the quality of conversations and ensure continuous improvement of the system.

[0117] Automatic update mechanism: Regularly summarize user feedback and automatically adjust model parameters to ensure continuous improvement of the system and adapt to new needs, maintaining a leading technical level.

[0118] Security verification layer

[0119] Rule base: Contains a set of predefined standards and specifications for checking whether the generated dialogue complies with international aviation regulations and operating specifications, ensuring that all instructions are legal and compliant.

[0120] Anomaly detection algorithm: used to identify conversation content that may cause safety hazards and trigger alarms to remind relevant personnel to take measures to prevent accidents.

[0121] Manual review channel: When necessary, key conversations will be submitted to professionals for final review to ensure the security and legality of all instructions and ensure flight safety.

[0122] Cloud Collaboration Platform

[0123] Distributed computing resources: Allocate multiple server nodes to speed up the conversation generation process, provide powerful computing power support, and ensure efficient operation of the system.

[0124] Data backup and recovery: Ensure the safe storage and rapid recovery of flight data, prevent data loss or damage, and ensure the integrity of flight records.

[0125] Cross-regional collaboration: Support multilingual communication and cross-border cooperation around the world, promote international flight coordination, and expand the scope of application.

[0126] Intelligent scheduling system: Rationally arrange computing resources according to network conditions and task priorities to ensure efficient operation of the system and optimize resource utilization.

[0127] Communication interface

[0128] Very High Frequency (VHF) and Ultra High Frequency (UHF) radio communications: used for direct contact with the tower and other aircraft over short distances, ensuring real-time communications and supporting rapid response.

[0129] Satellite communications: used for data transmission over long distances or in remote areas, covering a wider geographical range and supporting remote control.

[0130] Wi-Fi and cellular networks: used for convenient connection between ground station and flight model aircraft, providing flexible communication methods and facilitating ground control.

[0131] Bluetooth and Zigbee: used for simple communication between short-range devices, facilitating the connection of small devices or accessories and expanding system functions.

[0132] User training modules

[0133] Virtual reality (VR) training environment: provides an immersive flight simulation experience to help novice pilots become familiar with system operations, improve their skills, and shorten the training cycle.

[0134] Online courses and tutorials: Provide rich learning materials and technical support through the Internet platform, so that users can obtain the required knowledge at any time and improve the training effect.

[0135] Interactive Practice Mode: Allows users to practice conversational skills in a controlled environment, gradually improve proficiency, enhance practical operation capabilities, and ensure training quality.

[0136] Detailed workflow

[0137] Data collection

[0138] Various sensors on the flight model aircraft (such as GPS, IMU, barometer, camera, infrared camera, etc.) collect flight status data in real time.

[0139] External information such as meteorological conditions and airspace status provided by the tower is also incorporated into the system to ensure a comprehensive understanding of the flight environment.

[0140] Central Processing

[0141] The data acquisition subsystem transmits the collected data to the central processing unit (CPU), which parses the data and calls the corresponding dialogue generation model to ensure the accuracy of information processing.

[0142] Dialogue Generation

[0143] The deep learning model generates appropriate dialogue content based on the received data and the evaluation results of the context awareness module.

[0144] Use reinforcement learning to optimize model parameters to ensure high accuracy and provide reliable instructions in complex environments.

[0145] Situational Adjustment

[0146] The situational awareness module dynamically adjusts the dialogue strategy according to the current flight mission and environmental conditions to adapt to different flight phases and emergencies, ensuring the flexibility and effectiveness of communication.

[0147] Voice conversion

[0148] The natural language processing unit converts the generated text commands into clear and easy-to-understand voice output, supports multiple language options, ensures accurate information transmission and promotes cross-border communication.

[0149] User Interaction

[0150] The feedback mechanism collects user feedback information through the user interface, including real-time monitoring, manual correction and automatic update functions, continuously improving the quality and accuracy of conversations and ensuring continuous optimization of the system.

[0151] Security Check

[0152] The security verification layer performs strict security checks on all generated conversation content to ensure that it complies with international aviation regulations and operating specifications, ensure flight safety, and prevent potential risks.

[0153] Cloud support

[0154] When necessary, the flight model aircraft can access the powerful computing resources and data backup of the remote server through the cloud collaboration platform, further improving the speed and quality of dialogue generation and ensuring the efficient operation of the system.

[0155] Communication transmission

[0156] Through a variety of communication protocols (such as VHF, UHF, satellite communication, Wi-Fi, cellular network, Bluetooth, Zigbee, etc.), reliable data transmission between the flight model aircraft and the tower and other relevant parties is ensured, supporting real-time communication and remote control.

[0157] Training and Education

[0158] The user training module provides a virtual reality training environment, online courses and tutorials, and an interactive practice mode to help pilots become familiar with system operations, improve their skill levels, ensure training effectiveness, and shorten the learning curve.

[0159] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. The artificial intelligence-based flight model aircraft and tower dialogue generation system is characterized by: include: Multimodal data acquisition: used to collect various sensor data from the flight model aircraft and external information on meteorological conditions and airspace status provided by the tower. The sensor data includes positioning sensors, flight speed sensors, and flight altitude sensors. Deep learning model construction: Build a deep neural network (DNN)-based dialogue generation model that is trained on a large number of real flight logs and voice recordings to generate powerful semantic understanding and generation capabilities; Use reinforcement learning to optimize model parameters to ensure high accuracy in complex environments; Situational awareness module: dynamically adjusts the dialogue strategy according to the current flight mission and environmental conditions for different flight phases and emergencies; Natural language processing unit: used to convert the generated text commands into clear and understandable voice output, and supports several language options; Feedback mechanism: Collect feedback information through user interaction to continuously improve the quality and accuracy of conversations; Security verification layer: All generated conversation content is subject to strict security checks to ensure compliance with international aviation regulations and operating specifications; Cloud Collaboration Platform: Allows flight model aircraft to connect to remote servers, utilizing the powerful computing resources and storage capabilities of cloud services to further improve the speed and quality of dialogue generation.

2. The artificial intelligence-based flight model aircraft and tower dialogue generation system according to claim 1, characterized in that: The multimodal data acquisition device also includes: Visual sensors: including cameras and infrared cameras, used to capture visual information around the aircraft; Acoustic sensor: used to monitor the surrounding sound environment and identify potential threats; Airflow sensor: used to monitor meteorological parameters such as wind speed and direction to assist flight decision-making; Radio receiver: used to receive radio signals from the tower and other aircraft.

3. The artificial intelligence-based flight model aircraft and tower dialogue generation system according to claim 2, characterized in that: The deep learning model also includes: Pre-training module: Use large-scale public datasets for preliminary training to improve the basic performance of the model; Adaptive learning module: continuously updates model parameters based on actual flight data to fit specific application scenarios; Scenario simulation module: By simulating dialogue scenarios under different flight conditions, the model's ability to deal with emergencies is enhanced.

4. The artificial intelligence-based flight model aircraft and control tower dialogue generation system according to claim 3, characterized in that: The context awareness module can identify the following flight phases and adjust the dialogue strategy accordingly: Pre-takeoff preparation phase; Take-off process phase; Cruise phase; Approach and landing phase; Emergency response phase.

5. The artificial intelligence-based flight model aircraft and control tower dialogue generation system according to claim 4, characterized in that: The natural language processing unit also has the following functions: Text-to-speech conversion: converting generated text instructions into natural and fluent speech; Speech-to-text conversion: Parsing voice input from a tower or other source and converting it into a processable text format; Sentiment Analysis: Assess the tone and sentiment in a conversation so you can respond more humanely.

6. The artificial intelligence-based flight model aircraft and control tower dialogue generation system according to claim 5, characterized in that: The feedback mechanism includes: Real-time monitoring system: allows operators to instantly view the generated conversation content and its effects; Manual Correction Programmer: allows users to directly modify inaccurate or inappropriate conversation content and add it to the model as new training samples; Automatic update mechanism: Regularly summarize user feedback, automatically adjust model parameters, and optimize conversation quality.

7. The artificial intelligence-based flight model aircraft and control tower dialogue generation system according to claim 6, characterized in that: The security verification layer includes: Rule base: a predefined set of standards and specifications used to check whether the generated dialogue violates aviation regulations; Anomaly detection algorithm: used to identify conversation content that may cause security risks and trigger alarms; Human review channel: When necessary, key conversations are submitted to professionals for final review.

8. The artificial intelligence-based flight model aircraft and control tower dialogue generation system according to claim 7, characterized in that: The cloud collaboration platform provides the following services: Distributed computing resources: Allocate multiple server nodes to speed up the conversation generation process; Data backup and recovery: ensure safe storage and rapid recovery of flight data; Cross-regional collaboration: Supporting multilingual communication and cross-border cooperation around the world; Intelligent scheduling system: Rationally arrange computing resources according to network conditions and task priorities.

9. The artificial intelligence-based flight model aircraft and control tower dialogue generation system according to claim 8, characterized in that: The communication interface supports the following communication protocols: Very high frequency (VHF) and ultra high frequency (UHF) radio communications: used for direct contact with the tower and other aircraft over short distances; Satellite communications: used for data transmission over long distances or in remote areas; Wi-Fi and cellular networks: for convenient connection between ground stations and flight models; Bluetooth and Zigbee: For simple communication between devices over short distances.

10. The artificial intelligence-based flight model aircraft and control tower dialogue generation system according to claim 9, characterized in that: Also includes user training modules, Virtual reality (VR) training environment: provides an immersive flight simulation experience to help novice pilots become familiar with system operations; Online courses and tutorials: provide rich learning materials and technical support through the Internet platform; Interactive Practice Mode: Allows users to practice conversational skills in a controlled environment and gradually improve proficiency.

Images