A mobile platform-oriented oil-driven unmanned helicopter composite landing guidance method

Abstract

The application provides a mobile platform-oriented oil-driven unmanned helicopter composite landing guidance method, which comprises the following steps: a far-near integrated guidance stage: collecting the positions of the helicopter and the mobile platform, combining a mixed path search with a reverse tracing closed-loop planning to a three-dimensional waypoint of an exit point behind the platform. The exit point is pre-planned as an end point to obtain a transition entry heading; then, the heading is reversed to construct a smooth flight path; finally, the transition point heading is used as a constraint for quadratic programming to generate a complete waypoint sequence. A near-distance accurate guidance stage: two independent vector channels are used to solve the position vector of the helicopter gravity center relative to the landing target point in parallel. One channel relies on differential satellite positioning, and the other channel relies on airborne laser ranging. The master and standby vectors are switched to output, the end waypoint of the previous waypoint sequence is corrected, and the flight control guidance command is formed. The method can dynamically compensate for the six-degree-of-freedom motion of the platform and the navigation drift, ensure the centimeter-level positioning of the landing end segment, and realize the closed-loop unification of continuous flight path and high-frequency end adjustment.

Description

Technical Field

[0001] This invention belongs to the field of unmanned aerial vehicle (UAV) navigation technology, and particularly relates to a combined landing guidance method for gasoline-powered unmanned helicopters on mobile platforms. Background Technology

[0002] Unmanned helicopters, with their vertical takeoff and landing, hovering, and low-speed maneuverability, have become key equipment for aerial operations on mobile platforms (such as ground vehicles and offshore platforms), capable of performing tasks such as reconnaissance, material transport, and communication relay. Landing and recovery on a mobile platform is the riskiest part of the entire mission chain, requiring the guidance system to possess autonomous decision-making capabilities and high reliability throughout the entire process.

[0003] Landing guidance for mobile platforms faces three core technical challenges. First, gasoline-powered unmanned helicopters are constrained by a large minimum turning radius, limited climb / dive rates, and minimum flight segment lengths, requiring their flight trajectories to meet strict kinematic constraints. Meanwhile, the mobile platform itself operates in six degrees of freedom, with its position and heading changing in real time, demanding that the landing path meet specified entry azimuth and glide slope angles at the terminal point. How to integrate long-distance guidance and terminal transition into a feasible trajectory planning problem that satisfies initiation and termination boundary conditions, kinematic constraints, and airspace constraints, and to provide an engineering-solvable closed-loop method, remains a difficult challenge that current technologies have not yet addressed.

[0004] Second, during the close-range landing phase, the guidance system needs to acquire the three-dimensional position vector of the aircraft's center of gravity relative to the platform's landing target point in real time with centimeter-level accuracy. Existing differential positioning methods based on a single global satellite navigation system are prone to failure under platform structural obstruction, multipath effects, or electromagnetic interference; while monocular vision-based ranging methods are sensitive to lighting, weather, and target features, limiting their accuracy and reliability. Some solutions employ multi-sensor data fusion, but this typically involves loosely or tightly coupled filtering at the data level. The sensor channels are not completely independent in terms of physical principles and information sources, making it difficult to form true heterogeneous redundancy. Once a common mode failure occurs, guidance may still be interrupted.

[0005] Third, most existing phased guidance methods rely on preset distance thresholds to switch between different strategies, with each phase's algorithm operating independently and lacking a unified planning model. For example, long-distance guidance only considers simple waypoint generation, the transition phase depends on geometric rules, and no planning loop is formed between the two phases. This fragmented architecture makes it difficult to guarantee the kinematic continuity and terminal accuracy throughout the entire process from long distance to short distance.

[0006] Therefore, there is an urgent need for a composite landing guidance method that can incorporate long-distance global planning and transition route construction into a unified closed-loop framework and provide physical heterogeneous dual vector synthesis redundancy at close range, so as to meet the stringent requirements of accuracy, reliability and kinematic feasibility for autonomous landing of oil-powered unmanned helicopter mobile platforms in high-dynamic environments. Summary of the Invention

[0007] Objective of the Invention: The technical problem to be solved by this invention is to address the shortcomings of existing technologies by providing a composite landing guidance method for nitrogen-powered unmanned helicopters operating on mobile platforms. Existing methods, when faced with motion constraints such as the large minimum turning radius and limited climb / dive rates of nitrogen-powered unmanned helicopters, struggle to simultaneously satisfy global optimization for long-range guidance and precise closure of the terminal approach azimuth angle. In the near-range phase, they rely on single sensors or homogeneous fusion architectures, lacking heterogeneous redundancy guarantees at the physical principle level. Furthermore, the three phases of long-range guidance, transitional route construction, and near-end precise guidance are fragmented, lacking a unified planning model.

[0008] To address the aforementioned issues, this invention abstracts the landing guidance process into an optimal trajectory planning problem constrained by initial pose, terminal position and heading, kinematic constraints, and airspace constraints. In the long-range to transition phase, a closed-loop solution strategy of "hybrid A* pre-planning-backtracking-quadratic planning" is employed to generate a three-dimensional waypoint sequence satisfying all constraints, from the current position of the unmanned helicopter to the exit point behind the mobile platform. In the short-range phase, a first vector synthesis channel based on the superposition of antenna vectors from a differential global satellite navigation system and a second vector synthesis channel based on the superposition of laser ranging and cooperation identifier vectors from the airborne pod are constructed. The position vector of the aircraft's center of gravity relative to the landing target point of the mobile platform is calculated in real time using a hot-swap primary / backup method. The vector is then used to make online corrections to the final waypoints of the S1 stage waypoint sequence, providing continuous and highly reliable guidance commands to the flight control system.

[0009] The method includes the following steps:

[0010] Step 1, Integrated Long-Range Guidance Phase: Obtain the current position of the unmanned helicopter and the position of the mobile platform, and use a hybrid A* algorithm and reverse tracing method to plan a closed-loop path from the current position of the unmanned helicopter to the exit point behind the mobile platform. The three-dimensional waypoint sequence; the closed-loop planning process is as follows: first, using the exit point... Pre-planning is performed for the destination but not for the terminal heading, and the transition heading is extracted. ; and then For entry angle, for exit point To construct a smooth transition route, trace back to the endpoint; finally, use the reverse-tracing route at the transition entry point. The heading at the point is used as the terminal heading constraint for secondary planning to obtain a complete flight waypoint sequence;

[0011] Step 2, Close-range Precision Guidance Phase: When the distance D between the unmanned helicopter and the mobile platform is less than the distance threshold... At the same time, the position vector of the unmanned helicopter's center of gravity relative to the landing target point of the mobile platform is calculated in parallel using the first vector synthesis channel and the second vector synthesis channel. The position vector is used to correct the final waypoint in the waypoint sequence generated during the near-far integrated guidance phase, and the corrected target waypoint is used as the guidance command of the flight control system. The output of the position vector is provided in a hot-swap mode. The first vector synthesis channel obtains relative vectors based on the differential global satellite navigation system (GNSS) and performs algebraic superposition, while the second vector synthesis channel obtains relative vectors based on airborne pod laser ranging and performs algebraic superposition. The two channels are independent of each other in terms of physical medium and information source.

[0012] In step 1, the reverse tracing method in the far-near integrated guidance phase specifically includes:

[0013] Exit point Starting point, exit point Expected course Current course of the mobile platform Consistent;

[0014] From the exit point Initially, with a fixed change in heading. Add waypoints incrementally in reverse order, ensuring that the turning radius between adjacent waypoints is greater than the minimum turning radius each time a waypoint is added. And the first Distance between waypoints Greater than twice the minimum turning radius subscript Number the waypoints;

[0015] For each waypoint added, calculate the change in altitude between the added waypoint and the previous waypoint. And verify whether the rate of climb or rate of dive is within the maximum rate of climb. and maximum dive rate Within the range;

[0016] Continue building until waypoint ( The expected heading (the waypoint number added during the reverse tracing process). Transition into course The difference is less than the preset heading difference threshold, and the waypoints With transition entry point The distance between them must be greater than twice the minimum turning radius. At that time, the transition entry point will be... With waypoints Connect them to complete the reverse tracing route.

[0017] In step 1, the exit point Located directly behind the mobile platform, the exit point The position depends on the speed of the mobile platform and the minimum turning radius. Dynamic adjustment.

[0018] In step 1, the hybrid A* algorithm performs node search by maintaining an Open table and a Close table. The Open table stores nodes to be checked, and the Close table stores nodes that have already been processed. The cost function of the algorithm is:

[0019] ,

[0020] in, Indicates starting from the initial state via the current state Reaching the target state The estimated total cost; Indicates starting from the initial state To the current state They have already paid the price. Indicates starting from the current state To the target state Heuristic cost estimation;

[0021] Suppose the dynamic model of the unmanned helicopter at node state X is a state-space equation:

[0022] ,

[0023] in, Representing state The derivative with respect to time, Here is the state transition matrix. To control the input matrix, To control the input vector;

[0024] Discretizing the dynamic model, then during the sampling period Internally, from the parent node state To child node state The state transition process is described by the following discretized state equations:

[0025] ,

[0026] in, The sampling period is d, where d is the differential symbol. Represents the natural constant; For the integration variable, in the range of 0 to... Integral over the interval; For the first A discrete control input, that is, a continuous control input Discretize the value range into a set of optional values, each This corresponds to one type of operation.

[0027] In step 1, during the closed-loop planning phase, terminal heading constraints are not applied in the pre-planning stage to verify the transition heading. The accessibility is assessed, and in the reverse tracing phase, geometric constraints (such as turning radius and segment spacing) are used to construct the terminal transition route. In the secondary planning phase, the verified accessibility of the transition route is incorporated into the course. As a hard constraint, it ensures the closure of the terminal heading constraint and the kinematic constraint.

[0028] In step 2, the first vector synthesis channel relies on radio data link to transmit differential GNSS information, while the second vector synthesis channel relies on laser ranging and photoelectric imaging to obtain relative positioning information. The first and second vector synthesis channels are completely independent in terms of physical medium and information source, forming heterogeneous redundancy. The first vector synthesis channel is selected as the primary channel by default. When a decrease in the data quality of the first vector synthesis channel is detected, the system seamlessly switches to the second vector synthesis channel within milliseconds.

[0029] In step 2, the calculation formula for the first vector synthesis channel is:

[0030] ,

[0031] in, This represents the position vector from the aircraft's center of gravity to the landing target point, calculated by the first vector synthesis channel. This indicates the three-dimensional position of the antenna on the unmanned helicopter. This indicates the three-dimensional position of the mobile platform's ground antenna. This represents a known, fixed vector from the ground antenna to the landing target point; Represents a known fixed vector from the onboard antenna to the aircraft's center of gravity;

[0032] The calculation formula for the second vector synthesis channel is:

[0033] ,

[0034] in, This represents the position vector from the aircraft's center of gravity to the landing target point, calculated by the second vector synthesis channel. This represents the vector from the airborne pod to the cooperation logo. This represents a known fixed vector from the cooperation marker to the landing target point; This represents a known fixed vector from the aircraft's center of gravity to the pod.

[0035] In step 2, calculate At that time, the straight-line distance from the pod to the cooperation logo is first used in spherical coordinates. The angle of deflection of the pod's optical axis in the horizontal plane relative to the nose direction of the unmanned helicopter. The pitch angle of the pod's optical axis relative to the horizontal plane Determine the three-dimensional position of the cooperation identifier relative to the pod:

[0036] ,

[0037] ,

[0038] ,

[0039] in , , This indicates the three-dimensional position coordinates of the cooperation identifier relative to the pod;

[0040] Then , , The coordinates of the pod are transformed to the coordinates of the unmanned helicopter body through a rotation matrix, and then the current attitude angle of the unmanned helicopter provided by the inertial navigation system (INS) is transformed to the local navigation coordinates through a direction cosine matrix.

[0041] The present invention also provides an electronic device, including a processor and a memory, the memory storing program code that, when executed by the processor, causes the processor to perform the steps of the method.

[0042] The present invention also provides a storage medium storing a computer program or instructions that, when the computer program or instructions are run on a computer, execute the steps of the method described.

[0043] Beneficial effects:

[0044] 1. Closed-loop integrated trajectory planning ensures kinematic feasibility throughout the entire process: This invention integrates long-distance guidance and transition route construction into a closed-loop planning problem. Through a three-step solution strategy of "hybrid A* pre-planning - backtracking - quadratic planning", it completely solves the problem of minimum turning radius for gasoline-powered unmanned helicopters. Maximum climb rate Maximum dive rate Minimum flight segment length The challenge lies in simultaneously satisfying both the initial heading and the final entry azimuth angle under constraints. The pre-planning stage generates a coarse trajectory without constraining the final heading and extracts the transition entry heading. The reverse tracing phase utilizes geometric constraints to construct the transition entry point. To exit point The smooth flight path is obtained; in the secondary planning stage, the verified reachable transition heading is resolved as a hard constraint, so that the final waypoint sequence strictly satisfies the kinematic constraints throughout the flight process, ensuring the physical realizability of the flight trajectory.

[0045] 2. The dual vector synthesis channels provide inherent redundancy in physical heterogeneity: In the near-range phase, the traditional data-level fusion architecture is abandoned, employing two physically independent position vector synthesis channels. The first channel relies on the algebraic superposition of the differential global navigation satellite system (GNSS) signal and the fixed antenna vector, while the second channel relies on the algebraic superposition of the airborne pod's laser ranging and the cooperative identification vector. The two channels are completely independent in terms of physical medium (radio signals and laser signals) and information source (differential global navigation satellite system GNSS and laser ranging combined with inertial navigation system INS), constituting true heterogeneous redundancy. If either channel fails due to signal obstruction, electromagnetic interference, or sensor malfunction, the system can seamlessly switch to the other channel within milliseconds, fundamentally avoiding mission failure caused by single-point failures.

[0046] 3. High-precision real-time calculation overcomes platform motion and navigation errors: In the close-range phase, both vector synthesis channels directly calculate the three-dimensional position vector of the aircraft's center of gravity relative to the landing target point of the moving platform at a high frequency. The system continuously corrects the terminal landing point of the long-distance planned route in real time, preserving the stability of the global path while eliminating accumulated navigation errors. The first vector synthesis channel eliminates measurement errors caused by data link transmission delays through high-precision timestamp alignment; the second vector synthesis channel obtains real-time relative geometric relationships through pod servo tracking and attitude synthesis. This real-time solution mechanism can dynamically compensate for the six degrees of freedom motion of the mobile platform and the drift errors of its own navigation system, ensuring centimeter-level positioning accuracy during the final landing phase.

[0047] 4. The architecture combining planning and correction balances global optimization and local response: In the far-to-near phase, a hybrid A* planner generates a global waypoint sequence, possessing forward-looking global optimization capabilities; upon entering the near phase, the position vectors calculated by the dual vector synthesis channel are no longer output independently, but are superimposed on the final waypoints planned by S1 as corrections for the terminal waypoints, enabling the guidance system to simultaneously possess global path continuity and high-frequency correction capabilities at the terminal. The final landing point of the S1 waypoint sequence is corrected in real time through the position vectors calculated by S2, achieving a unified closed loop of global planning and precise terminal correction.

[0048] 5. The algorithm is computationally efficient and adaptable to airborne embedded computing platforms: the reverse tracing method uses a fixed heading change. The gradual construction of transition routes involves only geometric operations, resulting in minimal computational cost. The hybrid A* algorithm expands nodes on a rasterized map using discretized control inputs and performs heuristic search by maintaining open and closed lists, eliminating the need for complex nonlinear optimization solvers. The entire planning algorithm is suitable for real-time operation on airborne embedded computing platforms with limited computing power. Attached Figure Description

[0049] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments, and the advantages of the present invention in the above and / or other aspects will become clearer.

[0050] Figure 1 This is a flowchart of the method of the present invention.

[0051] Figure 2 This is a schematic diagram of the construction geometry of the "fixed turning route" during the far-to-near transition guidance phase. Detailed Implementation

[0052] This invention provides a combined landing guidance method for gasoline-powered unmanned helicopters operating on mobile platforms, which is executed in two stages based on the relative distance D between the unmanned helicopter and the mobile platform: when the relative distance D is greater than a distance threshold... At that time, execute the steps This refers to the integrated near-far guidance phase; when the relative distance D is less than or equal to the distance threshold. Then, switch to step S2, which is the close-range precision guidance stage. The implementation methods of the two stages are explained in detail below.

[0053] Step S1, the integrated remote-near guidance phase;

[0054] The goal of this phase is to guide the unmanned helicopter from a distance (e.g., 30 to 100 kilometers away) to the exit point behind the mobile platform. And ensure that the unmanned helicopter completes terminal alignment with the correct azimuth and stable attitude. Exit point This indicates the target point that the unmanned helicopter should reach after completing its transition flight; it is located directly behind the mobile platform, and is also the exit point. Expected course Current course of the mobile platform Consistent Indicates the current direction of movement of the mobile platform. Exit point. The distance to the mobile platform depends on the mobile platform's current speed and the unmanned helicopter's minimum turning radius. Dynamic adjustment.

[0055] The core of this stage is to adopt a closed-loop solution strategy of "hybrid A* pre-planning - backtracking - quadratic programming" to generate a path from the current position of the unmanned helicopter to the exit point. A three-dimensional waypoint sequence. This waypoint sequence satisfies the following motion constraints and safety constraints:

[0056] Minimum turning radius constraint: The turning radius at each turning point must be greater than the minimum turning radius of the unmanned helicopter. This is to ensure that turning maneuvers are feasible.

[0057] Climb / dive rate constraint: Altitude variation between adjacent waypoints The ratio to flight time (i.e., rate of climb or rate of dive) must be less than or equal to the maximum rate of climb. And greater than or equal to the maximum dive rate This is to ensure the feasibility of longitudinal movement. This represents the vertical height difference between two adjacent waypoints.

[0058] Minimum segment length constraint: the distance between the i-th waypoints of adjacent waypoints It must be greater than the preset minimum flight segment length This allows unmanned helicopters sufficient distance to complete turns and altitude changes. subscript Number the waypoints.

[0059] Airspace and no-fly zone constraints: Waypoints must be located within the pre-defined permitted airspace and avoid known no-fly zones. If the planned route crosses a no-fly zone, it must be automatically detoured or replanned.

[0060] Segment angle constraint: The angle between adjacent segments should be an acute angle, i.e. less than 90 degrees, to ensure the smoothness of the heading change and avoid the impact of sharp turns on the attitude of the unmanned helicopter.

[0061] (I) Hybrid A* Algorithm Principle: Based on the traditional A* algorithm, the hybrid A* algorithm directly embeds the kinematic constraints of the unmanned helicopter into the node expansion process, making the generated trajectory physically executable.

[0062] State Space and Node Representation: In the hybrid A* algorithm, the state space not only contains position coordinates but is a higher-dimensional space including position, velocity, and orientation information. Each node corresponds to a state X, containing three-dimensional spatial position and heading angle information. The map is rasterized, but unlike the ordinary A* algorithm, the hybrid A* algorithm uses a continuous state space for path search and state transitions within each grid cell, rather than being limited to state transitions based on the grid center. This allows the algorithm to consider more subtle steering and position adjustments.

[0063] The hybrid A* algorithm performs node search by maintaining an Open table and a Close table. The Open table stores nodes to be checked, and the Close table stores nodes that have already been processed. The cost function of the algorithm is:

[0064] ,

[0065] in, Indicates starting from the initial state via the current state Reaching the target state The estimated total cost; Indicates starting from the initial state To the current state The actual costs already incurred include path length costs, climb or dive maneuver penalties, and course change penalties; Indicates starting from the current state To the target state The heuristic cost estimate is calculated using a weighted sum of Euclidean distance and heading difference. In the region near the target point, the heuristic cost... Adjustments are made based on the direction of entry into the target to guide the trajectory's final heading to meet requirements. (Subscript) Assign node index numbers, Indicates the initial state. Indicates the target state.

[0066] State transition equation: During node expansion, the parent node's state... To its child node state The transition must conform to the kinematic laws of the unmanned helicopter. Let the dynamic model of the unmanned helicopter at node state X be the state-space equation:

[0067] ,

[0068] in, Let X represent the derivative of state X with respect to time, A be the state transition matrix, and B be the control input matrix. The state transition matrix A and the control input matrix B are matrix parameters calculated based on the aerodynamic characteristics of the unmanned helicopter.

[0069] Discretizing the dynamic model, then during the sampling period Internally, from the parent node state To child node state The state transition process is described by the following discretized state equations:

[0070] ,

[0071] in, The state of the child nodes; The state of the parent node; The sampling period is denoted by 'e', ​​which represents the natural constant. For the integration variable, in the range of 0 to... Integral over the interval; For the i-th discretized control input, i.e., the continuous control input Discretize the value range into a set of optional values, each Each corresponds to a control method, including left turn, right turn, climb, dive, and combinations of different amplitudes.

[0072] In engineering implementation, a deterministic discretized control input is adopted. Then, the state of the corresponding child node is directly calculated using the state transition equation. The technical approach. Each parent node selects different... This generates a set of discretely distributed child nodes, located within a fan-shaped region with the parent node's position as the vertex and the parent node's velocity direction as the axis. The trajectory from the parent node to the child node is shown. The length of the search step is called the search step size, and the angle between the edge trajectory of the sector and the velocity direction of the parent node is called the search angle. The maximum value of the search angle is determined by the minimum turning radius. and sampling period A joint decision.

[0073] Altitude-direction planning: In the route planning process, the desired speed from the starting node to the target node is usually the speed of the starting node. The expected height is the height of the target node. During the search process, the node's height will move towards the desired height under the constraints of maximum climb rate and maximum dive rate. Gradually approaching.

[0074] (ii) Closed-loop solution strategy;

[0075] The closed-loop solution strategy consists of three steps: pre-planning, backtracking, and quadratic planning.

[0076] (1) Pre-planning: The pre-planning stage starts with the current position and attitude of the unmanned helicopter. exit point The desired position is the endpoint, and the terminal heading is not constrained for now. A hybrid A* algorithm is run to plan a coarse trajectory that satisfies both kinematic and spatial constraints. When the algorithm searches for a point close to the exit point... When selecting nodes within the preset transition zone, extract the heading at that node and record it as the transition entry heading. .

[0077] (2) Reverse tracing: transitioning into the course Entry angle, exit point The destination is the destination, and the desired heading at the destination is... Current course of the mobile platform Consistent. A path is constructed from the transition entry point using a reverse tracing method. To exit point A smooth transition route. Transition entry point. The position of the corresponding node when the pre-planned coarse trajectory enters the preset transition zone.

[0078] The specific steps of the reverse tracing method are as follows: starting from the exit point Starting point, exit point Expected course Current course of the mobile platform Consistent. From the exit point Initially, with a fixed change in heading. Add waypoints incrementally in reverse order. Each time a waypoint is added, ensure that the turning radius between adjacent waypoints is greater than [a certain value]. And the first Distance between waypoints More than twice the minimum turning radius For each waypoint added sequentially, calculate the altitude change between that waypoint and the previous waypoint. And verify whether the climb or dive rate corresponding to the flight time estimated from the current flight speed is within the range. and Within the range. Continue constructing until a waypoint. ( The expected course for tracing back waypoint numbers. Transition into course The difference is less than a preset threshold (e.g., 30 degrees), and the waypoints With transition entry point The distance between them must be greater than twice the minimum turning radius. At that time, and Connecting them completes the construction of the reverse tracing route. Among them... Waypoint numbers added during the reverse tracing process, such as Figure 2 As shown.

[0079] (3) Secondary planning: The reverse tracing route will be used at the transition entry point. The heading at the point is used as the terminal heading constraint, which requires the secondary planning trajectory to reach the transition entry point. The heading at time equals Exit point Desired position and desired course As the final terminal constraint, the hybrid A* algorithm is run again to plan the transition point from the current pose of the unmanned helicopter. The trajectory. Due to the transition into the course. The path has been verified as reachable during the pre-planning phase, and quadratic programming ensures the existence of a solution that satisfies all constraints. The trajectory obtained from quadratic programming is connected end-to-end with the reverse tracing route, forming a path from the starting position to the exit point. The complete flight waypoint sequence.

[0080] This closed-loop solution strategy ensures that the unmanned helicopter, whether entering from any direction, can ultimately reach the exit point with the correct azimuth and a stable attitude. This solves the problem that gasoline-powered unmanned helicopters cannot simultaneously meet the starting and ending boundary conditions due to their large turning radius.

[0081] Step S2: Close-range precision guidance phase;

[0082] When the relative distance between the unmanned helicopter and the mobile platform Less than or equal to the distance threshold At this point, the system switches from step S1 to step S2. This stage requires high-precision and high-reliability relative pose measurement to support the final accurate landing. Based on the waypoint sequence planned in S1, this stage uses the first and second vector synthesis channels to calculate the position vector of the unmanned helicopter's center of gravity relative to the landing target point of the mobile platform in real time. This is used as the offset correction amount for the final waypoint. Specifically, it involves adjusting the final waypoint in the S1 waypoint sequence. The data is replaced or offset to the calculated current landing target point position, forming a dynamically corrected sequence of target waypoints, thus providing continuous landing guidance commands to the flight control system. The two channels are completely independent in terms of physical medium and information source, constituting an inherently heterogeneous redundancy. The system defaults to using the first vector synthesis channel as the primary channel and the second vector synthesis channel as a hot backup. When the data quality of the primary channel deteriorates, the system automatically and seamlessly switches to the backup channel.

[0083] (a) First vector synthesis channel;

[0084] The first vector synthesis channel is based on the algebraic superposition of the differential global navigation satellite system (GNSS) and the fixed vector of the antenna. It is applicable when the GNSS signal is good and the data link is unobstructed.

[0085] Hardware infrastructure: The mobile platform is equipped with a high-precision differential global satellite navigation system reference station; the unmanned helicopter is equipped with a high-dynamic global satellite navigation system receiver, which can calculate its own position and velocity information and include a timestamp.

[0086] The calculation formula for the first vector synthesis channel is:

[0087] ,

[0088] in, This represents the position vector from the aircraft's center of gravity to the landing target point, calculated by the first vector synthesis channel. The three-dimensional position of the antenna on the unmanned helicopter is indicated by the receiver of the onboard high dynamic global satellite navigation system; The three-dimensional position of the mobile platform's ground antenna is provided by the platform's differential reference station and transmitted to the unmanned helicopter via a high-speed data link; and High-precision timestamp alignment eliminates measurement errors caused by data link transmission delays; This represents the relative position vector between the onboard antenna and the ground antenna, obtained through a timestamp-aligned differential global navigation satellite system. The known fixed vector from the ground antenna to the landing target point is obtained through pre-calibration on the mobile platform; The known fixed vector representing the distance from the onboard antenna to the aircraft's center of gravity is obtained through pre-calibration on the unmanned helicopter.

[0089] (ii) Second vector synthesis channel;

[0090] The second vector synthesis channel is based on the superposition of airborne pod laser ranging and cooperation identification vectors. It is applicable to scenarios involving close range, strong electromagnetic interference, GGN signal rejection, or data link interruption.

[0091] Hardware infrastructure: The unmanned helicopter is equipped with a high-precision inertial navigation system (INS) and a controllable two-axis electro-optical pod, which includes a built-in laser rangefinder and a high-resolution camera. Cooperation markers of known size and shape are placed near the landing area of ​​the mobile platform.

[0092] The calculation formula for the second vector synthesis channel is:

[0093] ,

[0094] in, This represents the position vector from the aircraft's center of gravity to the landing target point, calculated by the second vector synthesis channel. The vector representing the distance from the airborne pod to the cooperation logo is obtained from the laser rangefinder inside the pod. azimuth angle of the pod relative to the unmanned helicopter body and pitch angle Calculations show that This indicates the straight-line distance from the pod to the cooperation logo. This indicates the angle of deflection of the pod's optical axis in the horizontal plane relative to the nose direction of the unmanned helicopter. This indicates the pitch angle of the pod's optical axis relative to the horizontal plane; The known fixed vector from the cooperation identifier to the landing target point is obtained through pre-calibration on the mobile platform; The known fixed vector representing the distance from the aircraft's center of gravity to the pod is obtained by pre-calibration on the unmanned helicopter.

[0095] calculate At that time, firstly, in the spherical coordinate system, distance is used Azimuth and pitch angle Determine the three-dimensional position of the cooperation identifier relative to the pod:

[0096] ,

[0097] ,

[0098] ,

[0099] Then, the relative position vector is transformed from the pod coordinate system to the unmanned helicopter body coordinate system using a rotation matrix, and then combined with the current attitude angle (roll angle) of the unmanned helicopter provided by the inertial navigation system (INS). Pitch angle Yaw angle The coordinates are transformed to the local navigation coordinate system using the direction cosine matrix.

[0100] (III) Primary / backup hot-switching mechanism;

[0101] The first and second vector synthesis channels run simultaneously and are solved independently. The system defaults to using the first vector synthesis channel for solving. The final waypoint is corrected and output. When a deterioration in the data quality of the first vector synthesis channel is detected, including situations such as loss of GPS satellite lock, differential data link interruption, or positioning accuracy factor (PDOP) exceeding a preset threshold, the system automatically switches to the second vector synthesis channel within milliseconds, adopting the same data quality. The final waypoint output is corrected, and the waypoint transition is smoothed during the switching process. The system can switch back once the first vector synthesis channel returns to normal operation.

[0102] Set distance threshold When the relative distance between the unmanned helicopter and the mobile platform Greater than At that time, the system executes step S1, tracking the waypoint sequence generated by the integrated planning. When the relative distance... Less than or equal to At this point, the system switches to step S2, where the position vector from the aircraft's center of gravity to the landing target point is calculated using two vector synthesis channels. This information is used to correct the final waypoint of the S1 waypoint sequence before being output to the flight control system. To prevent frequent switching near the threshold due to distance measurement noise or platform motion, a hysteresis interval is introduced. The condition for switching back to step S1 from step S2 is the relative distance. Greater than Plus .

[0103] The following section provides a detailed explanation of the implementation process of the method of the present invention in typical application scenarios, using specific parameters and calculation examples.

[0104] In this embodiment, the unmanned helicopter is a certain type of 800 kg class gasoline-powered unmanned helicopter, and its key motion constraint parameters are as follows: minimum turning radius. meters, maximum rate of ascent m / s, maximum dive rate meters per second, minimum flight segment length Meters. Distance threshold. The value is 1500 meters, and the hysteresis interval is... The value is set to 200 meters. The update frequency for long-distance, low-speed data links is 2 Hz, and the update frequency for short-distance, high-speed data links is 20 Hz. Exit point. Located directly behind the mobile platform, the exit point The distance to the mobile platform depends on the mobile platform's current speed and minimum turning radius. The value is dynamically adjusted and set to 800 meters in this embodiment. The sampling period in the hybrid A* algorithm... The time interval is 3 seconds, and the search step size is 300 meters. The fixed heading change is used in the reverse tracing method. The value is 30 degrees. The preset transition zone range is set to the exit point. A circular area with a center and a radius of 2000 meters.

[0105] Step S1: Integrated Far-Near Guidance Phase ( );

[0106] 1. The unmanned helicopter returns to a position approximately 80 kilometers from the mobile platform. The ground station then transmits the mobile platform's current location, including longitude, latitude, altitude, and heading, via a low-speed data link at a frequency of 2 Hz. Due to the long distance at this stage, the system is not sensitive to data link latency and can tolerate delays of up to a second.

[0107] 2. After receiving the location information of the mobile platform, the onboard computer obtains the current location and status of the unmanned helicopter: approximately 65 kilometers away from the mobile platform, current heading 120 degrees, current altitude 2000 meters, and current speed 40 meters per second.

[0108] 3. The system first determines the exit point. Obtain the current real-time location of the mobile platform (based on the origin of the local navigation coordinate system, with coordinates [0,0,0]) and the current heading. It is 210 degrees. Exit point. Located 800 meters directly behind the mobile platform, the coordinates are calculated as follows:

[0109] Longitude deviation = 800 × sin(210°) = -400 meters;

[0110] Latitude offset = -800 × cos(210°) = approximately 693 meters;

[0111] Exit point Expected course Set to 210 degrees, relative to the current heading of the mobile platform. Consistent.

[0112] 4. Implement a closed-loop solution strategy:

[0113] (1) Pre-planning: The current position of the unmanned helicopter is taken as the starting state. (Position [65000,0,2000], heading 120 degrees, speed 40 m / s), from the exit point The expected position ([-400, 693, The endpoint is [ ], and the terminal heading is not constrained at this time. A coarse trajectory is planned using a hybrid A* algorithm.

[0114] During the execution of the hybrid A* algorithm, the parent node state To child node state The transition is calculated using the discretized state equations:

[0115] ,

[0116] Sampling period The value is 3 seconds, and the control input is discretized. Includes the following typical options: 30-degree left turn, 30-degree right turn, 5-degree climb, 5-degree dive, straight flight, and combinations of different angles. State transition matrix. and control input matrix The results were obtained based on pre-calculated aerodynamic characteristics of this type of 800 kg-class unmanned helicopter, among which... The matrix is ​​a 4×4 matrix. The matrix is ​​a 4×2 matrix, and the specific values ​​are obtained from the flight dynamics model.

[0117] For each extended child node, the following constraint verification must be satisfied: Turning radius verification: if the heading change of the child node is 60 degrees and the current segment speed is 40 meters per second, then the required turning radius is... Velocity, where g is the acceleration due to gravity and is taken as 9.8 m / s². For the roll angle, the calculation must ensure that... Greater than 150 meters; climb / dive rate check, if the altitude change between adjacent waypoints With a depth of -200 meters and an estimated flight time of 100 seconds, the dive rate is -2 meters per second, meeting the requirement of being within the range of -5 meters per second to 5 meters per second; minimum segment length verification, straight-line distance between adjacent waypoints. The distance must be greater than 300 meters; airspace constraint verification: all waypoints must be located within the preset available airspace, such as longitude range from 110°E to 115°E, latitude range from 18°N to 22°N, and altitude range from 500 meters to 3000 meters. If there is a no-fly zone (such as a circular area with a radius of 20 kilometers centered at 112°E and 20°N), the flight path must automatically detour; segment angle constraint verification: the angle between adjacent segments must be less than 90 degrees. For example, if the heading of the previous segment is 120 degrees, then the heading of the next segment should fall within the range of 30 degrees to 210 degrees.

[0118] When the algorithm search enters or exits the point When the preset transition zone is centered at a point with a radius of 2000 meters, the current node heading is extracted as the transition entry heading. In this embodiment, it is assumed that the extracted transition heading... 120 degrees, corresponding to the transition entry point It is approximately 5.5 kilometers from the mobile platform.

[0119] (2) Reverse tracing: transitioning into the course Use 120 degrees as the entry angle and the exit point. ([-400,693, [The course is] 0.210 degrees (with a fixed course change) as the endpoint. A transitional route is constructed in the opposite direction at 30 degrees. The specific construction process is as follows:

[0120] Exit point Starting point, desired course It is 210 degrees. Add the first waypoint in reverse. , requiring from arrive The heading change is 30 degrees, that is The desired heading is 240 degrees. Calculate from... arrive The required turning radius is determined by the calculated value; if it is greater than 150 meters, the requirement is met, and adjustments are needed. The location ensures that the turning radius at this waypoint is greater than .calculate and Spacing ,make sure Greater than 300 meters.

[0121] Continue adding waypoints in reverse. The heading needs to be changed from 240 degrees to 270 degrees. Repeat the turning radius and spacing verification steps.

[0122] Continue building until a waypoint Expected course and transition course The difference (120 degrees) is less than 30 degrees, and With transition entry point The distance between them must be greater than twice the minimum turning radius. (i.e., a constraint of 300 meters). and Connect and complete the construction of the reverse tracing route.

[0123] (3) Secondary planning: The reverse tracing route will be used at the transition entry point. The heading (120 degrees) at the exit point serves as the terminal heading constraint. Desired position and desired course (210 degrees) is used as the final terminal constraint. The hybrid A* algorithm is run again to plan the transition from the current pose of the unmanned helicopter to the transition entry point. The trajectory obtained from the secondary programming connects with the reverse tracing route, forming a path from the starting position to the exit point. The complete flight waypoint sequence.

[0124] The unmanned helicopter flies along the generated waypoint sequence, gradually approaching the mobile platform, and adjusting its altitude and speed during the tracking process.

[0125] Step S2: Close-range precision guidance phase ( );

[0126] When the relative distance between the unmanned helicopter and the mobile platform When the distance is less than or equal to 1500 meters, the system switches from step S1 to step S2. At this point, the unmanned helicopter has reached the exit point. Nearby, entering the final landing phase. The system simultaneously runs the first vector synthesis channel and the second vector synthesis channel, with the first vector synthesis channel being the primary channel by default and the second vector synthesis channel serving as a hot backup.

[0127] (a) First vector synthesis channel (primary):

[0128] The first vector synthesis channel is based on the algebraic superposition of the differential global navigation satellite system (GNSS) and the fixed vector of the antenna. The calculation formula is as follows:

[0129] ,

[0130] The specific calculation process is as follows:

[0131] On the mobile platform, the differential global navigation satellite system reference station generates differential positioning data at a frequency of 20 Hz and adds a precise Coordinated Universal Time (UTC) timestamp. The data includes the location of the mobile platform's ground antenna. And speed information.

[0132] The unmanned helicopter received the data packet via a high-speed data link; the data packet contained a Coordinated Universal Time (UTC) timestamp. .

[0133] Onboard computer reads timestamps from data packets And obtain the data from the data recorded by the local global satellite navigation system receiver at the same time through interpolation. ( Approximately equal to The location of the onboard antenna And speed information. Timestamp alignment eliminates measurement errors caused by data link transmission delays.

[0134] Calculate the relative position vector: Assume the aligned onboard antenna positions relative to ground antenna position The difference is =[-200,50,30] meters, that is, the eastward deviation is -200 meters, the northward deviation is 50 meters, and the altitude deviation is 30 meters.

[0135] The pre-calibrated fixed vector values ​​on the mobile platform are: =[2,-1,-3] meters, representing the three-dimensional offset from the ground antenna to the landing target point; the fixed vector values ​​pre-calibrated on the unmanned helicopter are: =[0.5,0,-1.5] meters, representing the three-dimensional offset of the onboard antenna from the aircraft's center of gravity.

[0136] Substitute the values ​​into the formula to calculate the position vector from the aircraft's center of gravity to the landing target point:

[0137] rice;

[0138] The position of the unmanned helicopter's center of gravity relative to the landing target point of the mobile platform is: -198.5 meters to the east, 49 meters to the north, and 28.5 meters in altitude.

[0139] The relative speed is obtained by subtracting the aligned speed vectors: Assuming the speed vector of the unmanned helicopter is [40,0,-0.5] m / s and the speed vector of the mobile platform is [5,0,0] m / s, then the relative speed vector is [35,0,-0.5] m / s.

[0140] The system is based on the above position vector The final waypoint in the waypoint sequence of step S1 The target is shifted to the position corresponding to the current landing target point of the mobile platform, and the updated target waypoint and relative velocity vector are output to the flight control system.

[0141] (ii) Second vector synthesis channel (hot backup):

[0142] The second vector synthesis channel is based on the superposition of airborne pod laser ranging and cooperative identification vectors, and the calculation formula is as follows:

[0143] ,

[0144] The specific calculation process is as follows:

[0145] A circular infrared cooperation marker with a diameter of 1 meter is deployed near the landing area of ​​the mobile platform. The center of the cooperation marker is relative to a known fixed vector of the landing target point. Pre-calibrated to [2, -1, 0] meters.

[0146] The laser rangefinder built into the two-axis electro-optical pod of the unmanned helicopter measures the cooperation marker and obtains the distance between the pod and the marker. The distance is 250 meters. Simultaneously, the pod servo system records the azimuth angle of the pod relative to the unmanned helicopter body. 5 degrees, pitch angle It is -10 degrees Celsius.

[0147] Calculate the vector from the pod to the cooperation logo. In spherical coordinates, distance 250 meters, azimuth angle 5 degrees, pitch angle Given a value of -10 degrees, calculate the three-dimensional position of the cooperation logo in the pod coordinate system:

[0148] x direction ,

[0149] y direction ,

[0150] z direction ,

[0151] Will The process involves transforming the coordinate system from the pod to the body coordinate system, and then combining this with the current attitude angle (roll angle) of the unmanned helicopter provided by the inertial navigation system. 2 degrees, pitch angle 3 degrees, yaw angle The coordinates are transformed from 215 degrees to the local navigation coordinate system. The coordinate transformation process uses a conventional three-axis rotation matrix multiplication method, resulting in the coordinates in the navigation coordinate system. The vector is assumed to be [240, 15, -42] meters.

[0152] Known fixed vector from the aircraft's center of gravity to the pod The distance was pre-calibrated on the unmanned helicopter as [0.3, 0, -2.0] meters.

[0153] Substitute the values ​​into the formula to calculate the position vector from the aircraft's center of gravity to the landing target point:

[0154] rice,

[0155] Solving the second vector synthesis channel As a hot backup correction, when the first vector synthesis channel fails, the system uses... Update the final waypoint and output it; the correction method is the same as the first vector synthesis channel.

[0156] Phase transition:

[0157] When the relative distance between the unmanned helicopter and the mobile platform The distance gradually increases from below 1500 meters (e.g., due to the movement of the mobile platform causing an increase in relative distance). When the distance exceeds the 1500-meter plus 200-meter hysteresis interval, i.e., 1700 meters, the system switches back from step S2 to step S1 and re-executes the integrated trajectory planning. During the phase switch, the terminal waypoint correction amount calculated by the channel in step S2 gradually transitions from zero to full, and the weight of the original final waypoint in step S1 is reduced accordingly. The transition time is set to 2 seconds to ensure smooth changes in the target waypoints received by the flight control system.

[0158] Through the above steps, the unmanned helicopter completed the entire process of autonomous guidance from long-distance return to precise landing on the mobile platform.

[0159] This invention provides a combined landing guidance method for gasoline-powered unmanned helicopters on mobile platforms. Many methods and approaches exist for implementing this technical solution; the above description is merely a preferred embodiment. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of this invention, and these improvements and modifications should also be considered within the scope of protection of this invention. All components not explicitly stated in this embodiment can be implemented using existing technologies.

Claims

1. A combined landing guidance method for gasoline-powered unmanned helicopters on mobile platforms, characterized in that, Includes the following steps: Step 1, Integrated Long-Range Guidance Phase: Obtain the current position of the unmanned helicopter and the position of the mobile platform, and use a hybrid A* algorithm and reverse tracing method to plan a closed-loop path from the current position of the unmanned helicopter to the exit point behind the mobile platform. The three-dimensional waypoint sequence; the closed-loop planning process is as follows: first, using the exit point... Pre-planning is performed for the destination but not for the terminal heading, and the transition heading is extracted. ; and then For entry angle, for exit point To construct a smooth transition route, trace back to the endpoint; finally, use the reverse-tracing route at the transition entry point. The heading at the point is used as the terminal heading constraint for secondary planning to obtain a complete flight waypoint sequence; Step 2, Close-range Precision Guidance Phase: When the distance D between the unmanned helicopter and the mobile platform is less than the distance threshold... At the same time, the position vector of the unmanned helicopter's center of gravity relative to the landing target point of the mobile platform is calculated in parallel using the first vector synthesis channel and the second vector synthesis channel. The final waypoint in the waypoint sequence generated during the far-near integrated guidance phase is corrected using the position vector, and the corrected target waypoint is used as the guidance command for the flight control system. The position vector output is provided in a hot-swap mode; the first vector synthesis channel obtains relative vectors based on the Differential Global Navigation Satellite System (GNSS) and performs algebraic superposition, and the second vector synthesis channel obtains relative vectors based on airborne pod laser ranging and performs algebraic superposition.

2. The method according to claim 1, characterized in that, In step 1, the reverse tracing method in the far-near integrated guidance phase specifically includes: Exit point Starting point, exit point Expected course Current course of the mobile platform Consistent; From the exit point Initially, with a fixed change in heading. Add waypoints incrementally in reverse order, ensuring that the turning radius between adjacent waypoints is greater than the minimum turning radius each time a waypoint is added. And the first Distance between waypoints Greater than twice the minimum turning radius subscript Number the waypoints; For each waypoint added, calculate the change in altitude between the added waypoint and the previous waypoint. And verify whether the rate of climb or rate of dive is within the maximum rate of climb. and maximum dive rate Within the range; Continue building until waypoint Expected course Transition into course The difference is less than the preset heading difference threshold, and the waypoints With transition entry point The distance between them must be greater than twice the minimum turning radius. At that time, the transition entry point will be... With waypoints Connect them to complete the reverse tracing route; k is the waypoint number added during the reverse tracing process.

3. The method according to claim 2, characterized in that, In step 1, the exit point Located directly behind the mobile platform, the exit point The position depends on the speed of the mobile platform and the minimum turning radius. Dynamic adjustment.

4. The method according to claim 3, characterized in that, In step 1, the hybrid A* algorithm performs node search by maintaining an Open table and a Close table. The Open table stores nodes to be checked, and the Close table stores nodes that have already been processed. The cost function of the algorithm is: ， in, Indicates starting from the initial state via the current state Reaching the target state The estimated total cost; Indicates starting from the initial state To the current state They have already paid the price. Indicates starting from the current state To the target state Heuristic cost estimation; Suppose the dynamic model of the unmanned helicopter at node state X is a state-space equation: ， in, Representing state The derivative with respect to time, Here is the state transition matrix. To control the input matrix, To control the input vector; Discretizing the dynamic model, then during the sampling period Internally, from the parent node state To child node state The state transition process is described by the following discretized state equations: ， in, The sampling period is d, where d is the differential symbol. Represents the natural constant; For the integration variable, in the range of 0 to... Integral over the interval; For the first A discrete control input, that is, a continuous control input Discretize the value range into a set of optional values, each This corresponds to one type of operation.

5. The method according to claim 4, characterized in that, In step 1, during the closed-loop planning phase, terminal heading constraints are not applied in the pre-planning stage to verify the transition heading. The reachability is assessed, and in the reverse tracing phase, geometric constraints are used to construct the terminal transition route. In the secondary planning phase, the verified reachable transition routes are incorporated into the heading. It is applied as a hard constraint.

6. The method according to claim 5, characterized in that, In step 2, the first vector synthesis channel relies on radio data link to transmit differential GNSS information, while the second vector synthesis channel relies on laser ranging and photoelectric imaging to obtain relative positioning information. The first and second vector synthesis channels are completely independent in terms of physical medium and information source, forming heterogeneous redundancy. The first vector synthesis channel is selected as the primary channel by default. When a decrease in the data quality of the first vector synthesis channel is detected, the system switches to the second vector synthesis channel within milliseconds.

7. The method according to claim 6, characterized in that, In step 2, the calculation formula for the first vector synthesis channel is: ， in, This represents the position vector from the aircraft's center of gravity to the landing target point, calculated by the first vector synthesis channel. This indicates the three-dimensional position of the antenna on the unmanned helicopter. This indicates the three-dimensional position of the mobile platform's ground antenna. This represents a known, fixed vector from the ground antenna to the landing target point; Represents a known fixed vector from the onboard antenna to the aircraft's center of gravity; The calculation formula for the second vector synthesis channel is: ， in, This represents the position vector from the aircraft's center of gravity to the landing target point, calculated by the second vector synthesis channel. This represents the vector from the airborne pod to the cooperation logo. This represents a known fixed vector from the cooperation marker to the landing target point; This represents a known fixed vector from the aircraft's center of gravity to the pod.

8. The method according to claim 7, characterized in that, In step 2, calculate At that time, the straight-line distance from the pod to the cooperation logo is first used in the spherical coordinate system. The angle of deflection of the pod's optical axis in the horizontal plane relative to the nose direction of the unmanned helicopter. The pitch angle of the pod's optical axis relative to the horizontal plane Determine the three-dimensional position of the cooperation identifier relative to the pod: ， ， ， in , , This indicates the three-dimensional position coordinates of the cooperation identifier relative to the pod; Then , , The coordinates of the pod are transformed from the coordinate system to the coordinate system of the unmanned helicopter body through a rotation matrix, and then the current attitude angle of the unmanned helicopter provided by the inertial navigation system (INS) is transformed into the local navigation coordinate system through a direction cosine matrix.

9. An electronic device, characterized in that, It includes a processor and a memory, the memory storing program code that, when executed by the processor, causes the processor to perform the steps of the method as described in any one of claims 1 to 8.

10. A storage medium, characterized in that, It stores a computer program or instructions that, when run on a computer, perform the steps of the method as described in any one of claims 1 to 8.

Images