Digital flight rule-based manned aerial vehicle and unmanned aerial vehicle cooperative flight method

Abstract

The invention discloses a manned aerial vehicle and unmanned aerial vehicle cooperative flight method based on a digital flight rule, and the method comprises the steps: S01, task-driven multi-level airspace dynamic division, S02, digital operation parameter configuration based on DFR, S03, an air-ground cooperative closed-loop decision-making mechanism, S04, fusing ADS-B, radar, meteorological radar and unmanned aerial vehicle visual perception data, and S05, carrying out the task-driven multi-level airspace dynamic division. S05, collecting flight parameters, control instructions and abnormal events in real time, generating an encrypted data chain with a timestamp, analyzing historical conflict cases through a machine learning algorithm, and automatically optimizing interval control logic and obstacle avoidance rules; the method has the advantages that a dynamic channel distribution mechanism is adopted, horizontal track channels and vertical height zones are dynamically divided according to dynamic characteristics of different aircrafts, and air route conflicts caused by performance differences of the aircrafts are avoided.

Description

Technical Field

[0001] The present invention relates to the field of unmanned aerial vehicle (UAV) control technology, and in particular to a manned-unmanned aerial vehicle (UAV) collaborative flight method based on digital flight rules. Background Art

[0002] With the development of aviation technology, there is an increasing demand for the application of coordinated flight between manned and unmanned aircraft in military reconnaissance, emergency rescue, air transportation and other fields. Existing coordinated flight technologies mainly face the following technical bottlenecks: airspace management relies on static division and cannot be dynamically adjusted according to mission scenarios; multi-aircraft coordination lacks differentiated trajectory planning and does not fully consider the dynamic characteristics of different aircraft models; collision avoidance and separation management rely on a single sensor, and the dynamic priority adjustment capability is insufficient. Summary of the Invention

[0003] The purpose of the present invention is to provide a manned-unmanned aerial vehicle collaborative flight method based on digital flight rules in order to solve the problems that airspace management relies on static division, cannot be dynamically adjusted according to mission scenarios, and multi-aircraft collaboration lacks differentiated trajectory planning.

[0004] The object of the present invention can be achieved by the following technical solution: A method for coordinated flight of a manned aircraft and a UAV based on digital flight rules, comprising the following steps:

[0005] S01: Mission-driven multi-level dynamic airspace demarcation;

[0006] S02: Based on DFR digital operation parameter configuration, manned aircraft submit a 4D plan including take-off and landing points, mission nodes, and speed range. UAVs upload collaborative plans in batches through the cluster controller. Manned aircraft broadcast in real time, and UAVs perform local situation updates via UWB. Heterogeneous data is integrated through a federated learning algorithm to set safety intervals between manned aircraft, between manned aircraft and UAVs, and within the UAV cluster. Manned aircraft are defined as the highest priority, followed by fixed-wing UAVs, and finally rotary-wing UAVs.

[0007] S03: Air-ground collaborative closed-loop decision-making mechanism. The ground control station and the airborne terminal interact in real time via 5G / satellite communications. When a track conflict is detected, the GCS generates an avoidance plan, which is pushed to the manned aircraft cockpit HUD and the UAV cluster controller. After confirmation by the pilot / ground operator, the plan is automatically executed.

[0008] S04: Integrate ADS-B, radar, weather radar, and drone visual perception data to construct a multidimensional state space containing position, speed, heading, and weather parameters, and use the extended Kalman filter algorithm to predict the position of each aircraft within the next 30 seconds;

[0009] S05: Collect flight parameters, control instructions, and abnormal events in real time, generate encrypted data links with timestamps, analyze historical conflict cases through machine learning algorithms, and automatically optimize interval control logic and obstacle avoidance rules.

[0010] Furthermore, the step S01 is specifically as follows:

[0011] Build a three-dimensional airspace model, collect geographic information, meteorological conditions, and mission objectives of the mission scenario, and use geographic information systems and computer graphics technology to convert two-dimensional map data into a three-dimensional spatial model with height dimensions, while marking static airspace constraints;

[0012] Delineate the core mission area and determine the core operation range based on the mission type. If the mission type is reconnaissance, the target area is the center; if the mission type is transportation, the material drop point is the benchmark;

[0013] The collaborative operation area is set up to expand the scope outside the core mission area to form a collaborative operation area. The scope of the collaborative operation area is expanded outward by 1-2 kilometers based on the radius of the core mission area, serving as a buffer space for information exchange, formation change and route adjustment between manned and unmanned aircraft;

[0014] Safety buffer zone planning: Based on the maximum out-of-control distance of the aircraft and the emergency response time for faults, a safety buffer zone is constructed outside the collaborative operation area. A virtual fence is set up to prevent aircraft from crossing the boundary, and flight safety is ensured through real-time monitoring and dynamic early warning mechanisms.

[0015] Exclusive horizontal track channels and vertical altitude bands are allocated to manned aircraft, fixed-wing UAVs, and rotary-wing UAVs respectively, and the channel parameters are dynamically adjusted according to the dynamic characteristics of the aircraft model.

[0016] Furthermore, the specific method of planning the safety buffer zone is as follows:

[0017] Calculate the maximum out-of-control distance D based on the aircraft's power system performance and maximum flight speed. max , combined with the fault emergency response time T response With the average flight speed V avg Calculate the emergency response distance D response =T response ×V avg , safety buffer radius R=max(D max ,D response )×1.2;

[0018] A circular virtual boundary is generated around the collaborative work area based on the geographic information system. A no-fly zone is set using electronic fencing technology. The real-time position data of the aircraft is compared with the coordinates of the virtual fence. When the aircraft enters the threshold range of the fence boundary, a level 1 warning is triggered, and the automatic hovering or return-to-home procedure is initiated when entering the boundary.

[0019] A three-dimensional perception network is built through millimeter-wave radar, ADS-B system and visual sensors, which updates aircraft position data up to 20 times per second. The flight trajectory is predicted using machine learning algorithms. When the predicted trajectory reaches the safety buffer zone, an audible and visual warning is pushed to the pilot through the onboard terminal, and an early warning information is sent to the ground control center simultaneously.

[0020] Furthermore, the channel parameters are dynamically adjusted according to the dynamic characteristics of the aircraft model in a specific manner:

[0021] The horizontal track channel width is set based on the maximum turning radius of the aircraft type. The calculation formula is channel width = maximum turning radius × safety factor × 2. The calculated result is rounded down. When the aircraft position data is updated up to 20 times per second, the speed difference between adjacent aircraft types is analyzed. If the speed difference exceeds 30 kilometers per hour, the channel width is increased by an additional 20% to avoid wake interference.

[0022] The vertical altitude band interval is determined based on the maximum climb rate and descent rate of the aircraft model. The formula is altitude interval = (maximum climb / descent rate × safety reaction time) + equipment error compensation value. The result is rounded up and a compensation value is added based on the airspace congestion situation. When the aircraft density in a specific airspace exceeds 15 aircraft per square kilometer within the specified time, or there are more than three flight routes in progress in the airspace at the same time, and the probability of track intersection exceeds 50%, it is determined to be airspace congestion, and the compensation value is increased to 15 meters.

[0023] Furthermore, the avoidance plan generated by the GCS includes altitude adjustment, heading deviation, and speed increase or decrease.

[0024] Furthermore, the specific steps of S05 are:

[0025] Flight parameters, control instructions, and abnormal event data from historical conflict cases are cleaned and uniformly formatted into feature vectors. A supervised learning algorithm uses historical conflict data as input and conflict outcomes as labels to train a model capable of predicting potential conflict risks.

[0026] Based on the trained model, the key features and decision boundaries that affect the occurrence of conflicts are analyzed, and the interval control logic and obstacle avoidance rules applicable to different scenarios are extracted;

[0027] New flight data is continuously fed into the model to evaluate the effectiveness of existing rules. When the model predicts an increased risk of conflict, it automatically adjusts the thresholds of separation control and obstacle avoidance rules.

[0028] Compared with the prior art, the present invention has the following beneficial effects:

[0029] 1. Dynamic channel allocation mechanism: dynamically divides horizontal track channels and vertical altitude bands according to the dynamic characteristics of different aircraft models to avoid route conflicts caused by differences in aircraft performance.

[0030] 2. Multi-area hierarchical control ensures the execution accuracy of core tasks such as reconnaissance and transportation by demarcating high-precision operating ranges, reducing mission errors, supporting dynamic information exchange, formation reorganization and route adjustment between manned and unmanned aircraft, and improving the flexibility and adaptability of collaborative operations of multiple aircraft types. The safety buffer zone has a radius scientifically calculated based on dynamic parameters, and through virtual fences and multi-level early warning mechanisms, it achieves automatic protection in abnormal conditions such as loss of control and failure, thereby reducing the risk of flight accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] To facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0032] Figure 1 The present invention provides a flow chart of a method for coordinated flight of manned and unmanned aircraft based on digital flight rules. DETAILED DESCRIPTION

[0033] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0034] See also Figure 1 As shown, a method for coordinated flight of a manned aircraft and a UAV based on digital flight rules includes the following steps:

[0035] S01: Mission-driven multi-level dynamic airspace demarcation;

[0036] S02: Based on DFR digital operation parameter configuration, manned aircraft submit a 4D plan including take-off and landing points, mission nodes, and speed range. UAVs upload collaborative plans in batches through the cluster controller. Manned aircraft broadcast in real time, and UAVs perform local situation updates via UWB. Heterogeneous data is integrated through a federated learning algorithm to set safety intervals between manned aircraft, between manned aircraft and UAVs, and within the UAV cluster. Manned aircraft are defined as the highest priority, followed by fixed-wing UAVs, and finally rotary-wing UAVs.

[0037] S03: Air-ground collaborative closed-loop decision-making mechanism. The ground control station and the airborne terminal interact in real time via 5G / satellite communications. When a track conflict is detected, the GCS generates an avoidance plan. The plan includes altitude adjustment, heading deviation, and speed increase or decrease. The plan is pushed to the manned aircraft cockpit HUD and the UAV cluster controller, and is automatically executed after confirmation by the pilot / ground operator.

[0038] S04: Integrate ADS-B, radar, weather radar, and drone visual perception data to construct a multidimensional state space containing position, speed, heading, and weather parameters, and use the extended Kalman filter algorithm to predict the position of each aircraft within the next 30 seconds;

[0039] S05: Real-time collection of flight parameters, control instructions, and abnormal events, generating a time-stamped encrypted data link, analyzing historical conflict cases through machine learning algorithms, and automatically optimizing separation control logic and obstacle avoidance rules;

[0040] By building a three-dimensional airspace model, integrating geographic, meteorological and mission data, marking static airspace constraints, dividing the core mission area, collaborative operation area, and safety buffer zone, clarifying the functions and boundaries of each area, and allocating exclusive horizontal track channels and vertical altitude bands for different aircraft models, the parameters are dynamically adjusted according to the dynamic characteristics of the aircraft model and the airspace congestion situation, standardizing the submission format of manned and unmanned aircraft flight plans, relying on 5G / satellite communications, building a closed-loop process from situational awareness to command execution, integrating multi-source data to build a multi-dimensional state space, recording flight data throughout the process, building an unalterable digital evidence chain, and optimizing system rules through machine learning to form a continuous improvement closed loop. The steps of constructing a multidimensional state space from source data are to decode and filter the data, remove abnormal position and speed information, suppress clutter and associate radar data with targets, extract wind speed, wind direction, air pressure, and precipitation intensity parameters from meteorological radar data, and use computer vision algorithms to identify obstacle positions and other aircraft outline information from drone visual perception data. Based on a unified timestamp and geographic coordinate system, data from different sources are mapped to the same time and space reference, and motion features of position, speed, and heading angle are extracted from each data source. Multidimensional vectors are constructed in combination with meteorological parameters, and redundant data are fused through the Kalman filter algorithm to eliminate conflicting information and generate a unified state space model.

[0041] The step S01 specifically comprises: constructing a three-dimensional airspace model, collecting geographic information, meteorological conditions, and mission objectives of the mission scene, using geographic information systems and computer graphics technology to convert two-dimensional map data into a three-dimensional space model including height dimensions, and marking static airspace constraint areas;

[0042] Delineate the core mission area and determine the core operation range based on the mission type. If the mission type is reconnaissance, the target area is the center; if the mission type is transportation, the material drop point is the benchmark;

[0043] The collaborative operation area is set up to expand the scope outside the core mission area to form a collaborative operation area. The scope of the collaborative operation area is expanded outward by 1-2 kilometers based on the radius of the core mission area, serving as a buffer space for information exchange, formation change and route adjustment between manned and unmanned aircraft;

[0044] Safety buffer zone planning: Based on the maximum out-of-control distance of the aircraft and the emergency response time for faults, a safety buffer zone is constructed outside the collaborative operation area. A virtual fence is set up to prevent aircraft from crossing the boundary, and flight safety is ensured through real-time monitoring and dynamic early warning mechanisms.

[0045] Dedicated horizontal track channels and vertical altitude bands are allocated to manned aircraft, fixed-wing UAVs, and rotary-wing UAVs respectively, and the channel parameters are dynamically adjusted according to the dynamic characteristics of the aircraft model;

[0046] The specific method of planning the safety buffer zone is as follows: according to the performance of the aircraft power system and the maximum flight speed, the maximum loss of control distance D is calculated. max , combined with the fault emergency response time T response With the average flight speed V avg Calculate the emergency response distance D response =T response ×V avg , safety buffer radius R=max(D max ,D response )×1.2;

[0047] A circular virtual boundary is generated around the collaborative work area based on the geographic information system. A no-fly zone is set using electronic fencing technology. The real-time position data of the aircraft is compared with the coordinates of the virtual fence. When the aircraft enters the threshold range of the fence boundary, a level 1 warning is triggered, and the automatic hovering or return-to-home procedure is initiated when entering the boundary.

[0048] A three-dimensional perception network is built through millimeter-wave radar, ADS-B system and visual sensors, which updates aircraft position data up to 20 times per second. The flight trajectory is predicted using machine learning algorithms. When the predicted trajectory reaches the safety buffer zone, an audible and visual warning is pushed to the pilot through the onboard terminal, and an early warning information is sent to the ground control center simultaneously.

[0049] The channel parameters are dynamically adjusted based on the dynamic characteristics of the aircraft model. Specifically, the horizontal track channel width is set based on the maximum turning radius of the aircraft model. The calculation formula is channel width = maximum turning radius × safety factor × 2. The calculated result is rounded down. The speed difference between adjacent aircraft models is analyzed while updating aircraft position data up to 20 times per second. When the speed difference exceeds 30 kilometers per hour, the channel width is increased by an additional 20% to avoid wake interference.

[0050] Vertical altitude band separation is determined based on the aircraft's maximum climb and descent rates. The formula is altitude separation = (maximum climb / descent rate × safety reaction time) + equipment error compensation value. The result is rounded up and a compensation value is added based on airspace congestion. When the aircraft density in a specific airspace exceeds 15 aircraft per square kilometer within a specified time, or when there are more than three flight paths in progress in the airspace at the same time and the probability of track intersection exceeds 50%, airspace congestion is determined and the compensation value is increased to 15 meters.

[0051] The three-dimensional airspace model integrates geographical, meteorological and mission objective information to achieve digital mapping of complex environments, provide accurate spatial benchmarks for flight path planning, effectively avoid terrain obstacles and no-fly zones, and ensure flight safety. The precise delineation of the core mission area enables aircraft to focus on key operating areas. Combined with high-precision route control, it significantly improves the execution accuracy and efficiency of reconnaissance, transportation and other tasks. The collaborative operation area serves as a buffer space, supporting real-time information exchange and dynamic formation adjustment between manned and unmanned aircraft, enhancing the flexibility and robustness of system collaborative operations, and adapting to sudden changes during the mission. The safety buffer zone uses quantitative calculations to ensure the safety and controllability of flight boundaries. Based on the dynamic characteristics of the aircraft model, exclusive track channels and altitude bands are dynamically allocated to avoid airspace conflicts between different types of aircraft, achieve efficient use of airspace resources, and improve the stability of formation flight, supporting large-scale collaborative operations of multiple aircraft types.

[0052] The specific steps of S05 are: cleaning the flight parameters, control instructions and abnormal event data in historical conflict cases, uniformly formatting the data into feature vectors, using a supervised learning algorithm with historical conflict data as input and conflict results as labels to train a model that can predict potential conflict risks;

[0053] Based on the trained model, the system analyzes the key features and decision boundaries that influence conflict occurrences, extracting separation control logic and obstacle avoidance rules applicable to different scenarios. New flight data is continuously fed into the model to evaluate the effectiveness of existing rules. When the model predicts an increased risk of conflict, it automatically adjusts the separation control and obstacle avoidance rule thresholds.

[0054] By collecting flight parameters, control instructions and abnormal events in real time and generating a time-stamped encrypted data link, the transparency and traceability of the flight process are achieved, providing accurate data support for post-event review and accident investigation, significantly improving the safety of aviation operations and the efficiency of responsibility determination. With the help of machine learning algorithms, in-depth analysis of historical conflict cases is carried out to achieve dynamic iterative optimization of interval control logic and obstacle avoidance rules, fundamentally reducing the risk of flight conflicts and promoting the continuous evolution and performance improvement of the collaborative flight system of manned and unmanned aircraft.

[0055] The preferred embodiments of the present invention disclosed above are intended only to help illustrate the present invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the present invention to the specific embodiments described. Obviously, many modifications and variations are possible based on the content of this specification. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better understand and utilize the present invention. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for coordinated flight of manned and unmanned aircraft based on digital flight rules, characterized in that: The steps include: S01: Mission-driven multi-level dynamic airspace demarcation; S02: Based on DFR digital operation parameter configuration, manned aircraft submit a 4D plan including take-off and landing points, mission nodes, and speed range. UAVs upload collaborative plans in batches through the cluster controller. Manned aircraft broadcast in real time, and UAVs perform local situation updates via UWB. Heterogeneous data is integrated through a federated learning algorithm to set safety intervals between manned aircraft, between manned aircraft and UAVs, and within the UAV cluster. Manned aircraft are defined as the highest priority, followed by fixed-wing UAVs, and finally rotary-wing UAVs. S03: Air-ground collaborative closed-loop decision-making mechanism. The ground control station and the airborne terminal interact in real time via 5G / satellite communications. When a track conflict is detected, the GCS generates an avoidance plan, which is pushed to the manned aircraft cockpit HUD and the UAV cluster controller. After confirmation by the pilot / ground operator, the plan is automatically executed. S04: Integrate ADS-B, radar, weather radar, and drone visual perception data to construct a multidimensional state space containing position, speed, heading, and weather parameters, and use the extended Kalman filter algorithm to predict the position of each aircraft within the next 30 seconds; S05: Collect flight parameters, control instructions, and abnormal events in real time, generate encrypted data links with timestamps, analyze historical conflict cases through machine learning algorithms, and automatically optimize interval control logic and obstacle avoidance rules.

2. The method for manned-unmanned aerial vehicle (UAV) collaborative flight based on digital flight rules according to claim 1, characterized in that: The steps of S01 are specifically as follows: Build a three-dimensional airspace model, collect geographic information, meteorological conditions, and mission objectives of the mission scenario, and use geographic information systems and computer graphics technology to convert two-dimensional map data into a three-dimensional spatial model with height dimensions, while marking static airspace constraints; Delineate the core mission area and determine the core operation range based on the mission type. If the mission type is reconnaissance, the target area is the center; if the mission type is transportation, the material drop point is the benchmark; The collaborative operation area is set up to expand the scope outside the core mission area to form a collaborative operation area. The scope of the collaborative operation area is expanded outward by 1-2 kilometers based on the radius of the core mission area, serving as a buffer space for information exchange, formation change and route adjustment between manned and unmanned aircraft; Safety buffer zone planning: Based on the maximum out-of-control distance of the aircraft and the emergency response time for faults, a safety buffer zone is constructed outside the collaborative operation area. A virtual fence is set up to prevent aircraft from crossing the boundary, and flight safety is ensured through real-time monitoring and dynamic early warning mechanisms. Exclusive horizontal track channels and vertical altitude bands are allocated to manned aircraft, fixed-wing UAVs, and rotary-wing UAVs respectively, and the channel parameters are dynamically adjusted according to the dynamic characteristics of the aircraft model.

3. The method for manned-unmanned aerial vehicle (UAV) collaborative flight based on digital flight rules according to claim 2, characterized in that: The specific method of planning the safety buffer zone is as follows: Calculate the maximum out-of-control distance D based on the aircraft's power system performance and maximum flight speed. max , combined with the fault emergency response time T response With the average flight speed V avg Calculate the emergency response distance D response =T response ×V avg , safety buffer radius R=max(D max ,D response )×1.2; A circular virtual boundary is generated around the collaborative work area based on the geographic information system. A no-fly zone is set using electronic fencing technology. The real-time position data of the aircraft is compared with the coordinates of the virtual fence. When the aircraft enters the threshold range of the fence boundary, a level 1 warning is triggered, and the automatic hovering or return-to-home procedure is initiated when entering the boundary. A three-dimensional perception network is built through millimeter-wave radar, ADS-B system and visual sensors, which updates aircraft position data up to 20 times per second. The flight trajectory is predicted using machine learning algorithms. When the predicted trajectory reaches the safety buffer zone, an audible and visual warning is pushed to the pilot through the onboard terminal, and an early warning information is sent to the ground control center simultaneously.

4. The method for manned-unmanned aerial vehicle (UAV) collaborative flight based on digital flight rules according to claim 2, characterized in that: The specific method of dynamically adjusting the channel parameters according to the dynamic characteristics of the aircraft model is as follows: The horizontal track channel width is set based on the maximum turning radius of the aircraft type. The calculation formula is channel width = maximum turning radius × safety factor × 2. The calculated result is rounded down. When the aircraft position data is updated up to 20 times per second, the speed difference between adjacent aircraft types is analyzed. If the speed difference exceeds 30 kilometers per hour, the channel width is increased by an additional 20% to avoid wake interference. The vertical altitude band interval is determined based on the maximum climb rate and descent rate of the aircraft model. The formula is altitude interval = (maximum climb / descent rate × safety reaction time) + equipment error compensation value. The result is rounded up and a compensation value is added based on the airspace congestion situation. When the aircraft density in a specific airspace exceeds 15 aircraft per square kilometer within the specified time, or there are more than three flight routes in progress in the airspace at the same time, and the probability of track intersection exceeds 50%, it is determined to be airspace congestion, and the compensation value is increased to 15 meters.

5. The method for manned-unmanned aerial vehicle (UAV) cooperative flight based on digital flight rules according to claim 1, characterized in that: The avoidance plan generated by the GCS includes altitude adjustment, heading deviation, and speed increase or decrease.

6. The method for manned-unmanned aerial vehicle (UAV) cooperative flight based on digital flight rules according to claim 1, characterized in that: The specific steps of S05 are: Flight parameters, control instructions, and abnormal event data from historical conflict cases are cleaned and uniformly formatted into feature vectors. A supervised learning algorithm uses historical conflict data as input and conflict outcomes as labels to train a model capable of predicting potential conflict risks. Based on the trained model, the key features and decision boundaries that affect the occurrence of conflicts are analyzed, and the interval control logic and obstacle avoidance rules applicable to different scenarios are extracted; New flight data is continuously fed into the model to evaluate the effectiveness of existing rules. When the model predicts an increased risk of conflict, it automatically adjusts the thresholds of separation control and obstacle avoidance rules.

Images