A long-range heavy-load composite wing take-off unmanned aerial vehicle and a flight path optimization control method thereof

Abstract

The application discloses a long-haul heavy-load composite wing vertical take-off unmanned plane and a flight path optimization control method thereof, and relates to the field of unmanned planes. The unmanned plane comprises an unmanned plane body and a sling rope connected to the bottom of the unmanned plane body and used for hoisting and fixing goods. The method comprises the following steps: establishing a cargo-carrying unmanned plane dynamics model, respectively establishing an unmanned plane body model and a goods dynamics model; designing a flight path optimization index, taking energy as the index of flight path optimization, and combining a swing angle definition to establish an optimization objective function of the energy function of the cargo-carrying unmanned plane. The application realizes the safety of the flight path of the cargo-carrying unmanned plane, transportation efficiency and goods stability by means of a prediction time domain rolling optimization method and a multi-target collaborative constraint mechanism, i.e. dynamically solving the objective function of the speed stability, goods swing suppression and control efficiency in a time window, synchronously strictly constraining the dynamic characteristics, obstacle avoidance safety boundary and actuator physical limit value.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicles (UAVs), and more particularly to a long-range, high-payload, repetitive wing-jointing vertical takeoff and landing UAV and its trajectory optimization control method. Background Technology

[0002] Cargo drones, as core equipment for modern intelligent logistics and special operations, have been widely used in express delivery, agricultural plant protection, disaster relief, and power line inspection. Their core advantages lie in their ability to overcome terrain limitations, reach areas with inconvenient transportation, and significantly improve the efficiency of material delivery; and in their ability to perform high-risk tasks in complex environments with vertical take-off and landing and precise hovering capabilities, significantly reducing the risks associated with manual operations.

[0003] In the field of unmanned aerial vehicle (UAV) platform technology, compound-wing vertical takeoff and landing (VTOL) UAVs have become the mainstream configuration that balances VTOL capabilities with efficient cruising. Existing patents (e.g., CN110217391B) have disclosed compound-wing structures combining fixed-wing and multi-rotor configurations, achieving VTOL through symmetrically arranged lift propulsion systems, with the fixed wing providing the primary lift during level flight. Meanwhile, for heavy payload requirements, existing technologies (e.g., CN209258372U) have also disclosed fuselage load-bearing structures made of carbon fiber composite materials, optimizing the load-bearing frame to meet load-bearing requirements while controlling weight. Furthermore, the use of high-capacity battery systems and corresponding power system configurations for long-endurance flight are also conventional techniques in this field. Therefore, for those skilled in the art, the basic structure, materials, and power configurations used in compound-wing VTOL UAVs to achieve long-endurance and heavy payload capabilities are known prior art.

[0004] However, the inventors of this application have discovered that although UAV platform structure technology is relatively mature, existing trajectory optimization methods still have significant shortcomings when performing actual lifting tasks. Especially in dynamic and complex environments, existing methods fail to fully consider the strong coupling constraints between cargo swaying, obstacle avoidance, and energy consumption optimization. Traditional path planning algorithms (such as A and RRT) often use the shortest path or shortest time as the sole optimization objective, ignoring the nonlinear swaying characteristics of suspended cargo due to inertia. This is particularly problematic during high-maneuver turns or when encountering sudden wind disturbances, easily leading to cargo collisions with the aircraft or UAV instability and overturning. Meanwhile, while traditional sway suppression controllers such as PID and LQR can compensate for swaying at a local level, they are isolated from the global path planning layer and cannot coordinately optimize trajectory smoothness, obstacle avoidance safety, and flight energy efficiency at the system level. This makes it difficult to simultaneously ensure trajectory safety, transportation efficiency, and cargo stability in complex environments.

[0005] To address this issue, this invention proposes a long-range, high-payload, repetitive wing-mounted vertical takeoff and landing unmanned aerial vehicle (UAV) and its trajectory optimization control method to solve the comprehensive optimization problem under the aforementioned strong coupling constraints. Summary of the Invention

[0006] The purpose of this invention is to address the shortcomings of existing technologies by proposing a long-range, high-payload, repetitive folding-wing vertical takeoff and landing unmanned aerial vehicle (UAV) and its trajectory optimization control method.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] A method for trajectory optimization control of a long-range, high-payload, repeatable unidirectional vertical takeoff and landing (VTOL) unmanned aerial vehicle (UAV) includes the following steps:

[0009] S1: Establish a dynamic model of the cargo drone, including the drone body model and the cargo dynamic model;

[0010] S2: Design trajectory optimization indicators, using energy as the indicator for trajectory optimization, and establish an optimization objective function by combining the definition of swing angle and the energy function of cargo drones;

[0011] S3: Environmental perception and map building, which uses LiDAR or visual sensors to build point cloud data and construct flight route maps;

[0012] S4: Global path planning. A cost function is established based on the map and objective function. The cost function is then optimized to perform path planning. At the same time, a local trajectory optimization model is established to optimize the local trajectory based on the position of obstacles during flight.

[0013] In step S1, the cargo dynamics model is as follows: ,in:

[0014] For the location of the goods, ;

[0015] For the quality of goods;

[0016] Let the velocity vector of the cargo be... ;

[0017] For cargo aerodynamic drag;

[0018] The tension of the rope on the cargo;

[0019] e3 is the unit vector along the Z-axis;

[0020] In step S1, the constraint equation for the rope is: , Let be the length of the rope, and let the direction vector of the rope be . .

[0021] Preferably: In step S2, the swing angle is defined as follows in the body coordinate system: , ,in The swing angle is the vertical deviation. This is the azimuth angle, which is the horizontal orientation. Let r be the position vector of the cargo in the body coordinate system, and r be the displacement vector of the cargo in the inertial coordinate system.

[0022] In step S2, the oscillation energy function is: E k For the kinetic energy of the cargo, E p This refers to the potential energy of the cargo.

[0023] Preferably, in step S2, the objective function is: ,in For the drone's acceleration vector, , , These are the velocity term weight, oscillation suppression weight, and acceleration smoothing weight, respectively. t0 is the track start time, and tf is the track end time.

[0024] Preferably, in step S4, the cost function is: Let g(n) be the actual cost distance from the starting point to node n, and h(n) be the Euclidean distance from n to the ending point. The penalty for the oscillation energy at node n;

[0025] In step S4, the logic of the local trajectory optimization model is as follows:

[0026] S41: State sampling. At time k, the current state of the UAV is acquired through sensors. The current state includes position. ,speed Attitude angle Given the swing energy E(k), the UAV state expression is: x(k|k) = [ , , [E(k)] T ;

[0027] S42: Establish a predictive time-domain rolling optimization model within the time window [k, k+N-1]. ,in (j) is the number of steps to predict the speed of the drone. For j-step predictive control input, For j-step prediction of cargo swing energy, Q1, Q2, and R are weight matrices;

[0028] S43: Execute the first step control, first apply only the current step to solve the optimal solution of the model, and then discard the remaining control sequence;

[0029] S44: Move to the next time step k+1, and then repeat steps S41-S43;

[0030] In step S42, the predicted time-domain rolling optimization model satisfies dynamic constraints, obstacle avoidance constraints, and control quantity constraints, which are as follows:

[0031] Dynamic constraints: ;

[0032] Obstacle avoidance constraints: , Location of the obstacle. Minimum safe distance;

[0033] Control constraints: , These are the lower and upper limits of the control quantity, respectively.

[0034] Preferably: In step S1, the UAV body model is ,in:

[0035] Let be the position vector of the UAV. ;

[0036] For the drone's velocity vector, ;

[0037] For drone quality;

[0038] g is the acceleration due to gravity;

[0039] The rotation matrix from the body coordinate system to the inertial coordinate system;

[0040] For the body coordinate system thrust, ;

[0041] For the aerodynamic drag of the drone;

[0042] The tension of the rope on the drone;

[0043] For the attitude angle of the drone, , This is the roll angle. The pitch angle, Yaw angle;

[0044] The rotational inertia matrix of the UAV;

[0045] e3 is the unit vector along the Z-axis;

[0046] This is the angular velocity vector of the UAV;

[0047] This is the total torque vector acting on the UAV;

[0048] This is the control torque vector acting on the UAV.

[0049] A long-range, high-payload compound-wing vertical takeoff and landing (VTOL) unmanned aerial vehicle (UAV) includes: a UAV body, employing a compound wing configuration combining a fixed wing and a multi-rotor; the multi-rotor includes multiple independently driven rotors symmetrically arranged at predetermined positions on the fuselage and / or wings of the UAV body; the fixed wing is connected to the fuselage and employs a laminar flow airfoil; a horizontal propulsion system includes at least one thruster; an energy system includes a high-capacity onboard battery; a cargo lifting system includes a high-strength lifting rope, the upper end of which is connected to the bottom of the UAV body; the UAV body is made of carbon fiber composite material; the material and structure of the lifting rope are selected according to actual needs; the UAV is configured to execute the trajectory optimization control method as described in claims 1-5.

[0050] The beneficial effects of this invention are as follows:

[0051] This invention establishes a coupled dynamic model of the UAV body and the suspended cargo, and constructs a multi-objective optimization function that includes speed stability, sway suppression and control efficiency based on the definition of sway angle and energy analysis. By combining global path planning and local prediction time-domain rolling optimization, it achieves integrated and coordinated optimization of trajectory safety, transportation efficiency and cargo sway suppression under multiple constraints of dynamic characteristics, obstacle avoidance safety boundary and actuator physical limit. Attached Figure Description

[0052] Figure 1 This is a flowchart of a trajectory optimization control method for a long-range, high-payload, repetitive drooping-wing vertical takeoff and landing UAV proposed in this invention. Detailed Implementation

[0053] The technical solution of the present invention will be further described in detail below with reference to specific embodiments.

[0054] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "setting" should be interpreted broadly. For example, they can refer to a fixed connection or setting, a detachable connection or setting, or an integral connection or setting. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0055] Example 1:

[0056] A long-range, high-payload compound-wing vertical takeoff and landing (VTOL) unmanned aerial vehicle (UAV) includes a UAV body, employing a compound wing configuration combining a fixed wing and a multi-rotor. The multi-rotor comprises multiple independently driven rotors symmetrically arranged on both sides of the fuselage of the UAV body and on the fixed wing. The fixed wing is connected to the fuselage and employs a high lift-to-drag ratio laminar flow airfoil, specifically the NACA 6 series laminar flow airfoil, with a lift-to-drag ratio of not less than 18 during cruise. The horizontal propulsion system includes an electric propeller located at the rear of the fuselage. The energy system includes a high-capacity onboard battery. The cargo lifting system includes a high-strength lifting rope, the upper end of which is connected to the bottom of the UAV body. The UAV body is made of lightweight, high-strength carbon fiber composite material, and its load-bearing structure is optimized through a load-bearing frame and reinforcing structure to meet the load-bearing capacity and overall structural rigidity required for heavy-load transportation. The material and structure of the lifting rope are selected and designed according to the heavy-load requirements to ensure stability and operational safety during lifting.

[0057] Example 2:

[0058] A trajectory optimization control method for a long-range, high-payload, repetitive wing-jointed vertical takeoff and landing unmanned aerial vehicle (UAV) includes the following steps:

[0059] S1: Establish a dynamic model of the cargo drone, including the drone body model and the cargo dynamic model;

[0060] S2: Design trajectory optimization indicators, using energy as the indicator for trajectory optimization, and establish an optimization objective function by combining the definition of swing angle and the energy function of cargo drones;

[0061] S3: Environmental perception and map building, which uses LiDAR or visual sensors to build point cloud data and construct flight route maps;

[0062] S4: Global path planning. A cost function is established based on the map and objective function. The cost function is then optimized to perform path planning. At the same time, a local trajectory optimization model is established to optimize the local trajectory based on the position of obstacles during flight.

[0063] In step S1, the UAV body model is: ,in:

[0064] Let be the position vector of the UAV. ;

[0065] For the drone's velocity vector, ;

[0066] For drone quality;

[0067] g is the acceleration due to gravity;

[0068] The rotation matrix from the body coordinate system to the inertial coordinate system;

[0069] For the body coordinate system thrust, ;

[0070] For the aerodynamic drag of the drone;

[0071] The tension of the rope on the drone;

[0072] For the attitude angle of the drone, , This is the roll angle. The pitch angle, Yaw angle;

[0073] The rotational inertia matrix of the UAV;

[0074] In step S1, the cargo dynamics model is as follows: ,in:

[0075] For the location of the goods, ;

[0076] For the quality of goods;

[0077] Let the velocity vector of the cargo be... ;

[0078] For cargo aerodynamic drag;

[0079] This refers to the tension of the rope on the cargo.

[0080] In step S1, the constraint equation for the rope is: , Let be the length of the rope, and let the direction vector of the rope be . .

[0081] This method can effectively suppress cargo swaying in complex environments, while planning a safe, smooth and energy-efficient flight trajectory, thus improving the overall safety and efficiency of hoisting operations.

[0082] Example 3:

[0083] A trajectory optimization control method for a long-range, high-payload, repetitive wing-jointed vertical takeoff and landing unmanned aerial vehicle (UAV) includes the following steps:

[0084] S1: Establish a dynamic model of the cargo drone, including the drone body model and the cargo dynamic model;

[0085] S2: Design trajectory optimization indicators, using energy as the indicator for trajectory optimization, and establish an optimization objective function by combining the definition of swing angle and the energy function of cargo drones;

[0086] S3: Environmental perception and map building, which uses LiDAR or visual sensors to build point cloud data and construct flight route maps;

[0087] S4: Global path planning. A cost function is established based on the map and objective function. The cost function is then optimized to perform path planning. At the same time, a local trajectory optimization model is established to optimize the local trajectory based on the position of obstacles during flight.

[0088] In step S1, the UAV body model is: ,in:

[0089] Let be the position vector of the UAV. ;

[0090] For the drone's velocity vector, ;

[0091] For drone quality;

[0092] g is the acceleration due to gravity;

[0093] The rotation matrix from the body coordinate system to the inertial coordinate system;

[0094] For the body coordinate system thrust, ;

[0095] For the aerodynamic drag of the drone;

[0096] The tension of the rope on the drone;

[0097] For the attitude angle of the drone, , This is the roll angle. The pitch angle, Yaw angle;

[0098] The rotational inertia matrix of the UAV.

[0099] In step S1, the cargo dynamics model is as follows: ,in:

[0100] For the location of the goods, ;

[0101] For the quality of goods;

[0102] Let the velocity vector of the cargo be... ;

[0103] For cargo aerodynamic drag;

[0104] This refers to the tension of the rope on the cargo.

[0105] In step S1, the constraint equation for the rope is: , Let be the length of the rope, and let the direction vector of the rope be . .

[0106] In step S2, the swing angle is defined in the body coordinate system as: , ,in The swing angle is the vertical deviation. This is the azimuth angle, which is the horizontal orientation. Let r be the position vector of the cargo in the body coordinate system, and let r be the displacement vector of the cargo in the inertial coordinate system.

[0107] In step S2, the oscillation energy function is: E k For the kinetic energy of the cargo, E p This refers to the potential energy of the cargo.

[0108] In step S2, the objective function is: ,in For the drone's acceleration vector, , , These are the velocity term weight, oscillation suppression weight, and acceleration smoothing weight, respectively. t0 is the track start time, and tf is the track end time.

[0109] Example 4:

[0110] A trajectory optimization control method for a long-range, high-payload, repetitive wing-jointed vertical takeoff and landing unmanned aerial vehicle (UAV) includes the following steps:

[0111] S1: Establish a dynamic model of the cargo drone, including the drone body model and the cargo dynamic model;

[0112] S2: Design trajectory optimization indicators, using energy as the indicator for trajectory optimization, and establish an optimization objective function by combining the definition of swing angle and the energy function of cargo drones;

[0113] S3: Environmental perception and map building, which uses LiDAR or visual sensors to build point cloud data and construct flight route maps;

[0114] S4: Global path planning. A cost function is established based on the map and objective function. The cost function is then optimized to perform path planning. At the same time, a local trajectory optimization model is established to optimize the local trajectory based on the position of obstacles during flight.

[0115] In step S1, the UAV body model is: ,in:

[0116] Let be the position vector of the UAV. ;

[0117] For the drone's velocity vector, ;

[0118] For drone quality;

[0119] g is the acceleration due to gravity;

[0120] The rotation matrix from the body coordinate system to the inertial coordinate system;

[0121] For the body coordinate system thrust, ;

[0122] For the aerodynamic drag of the drone;

[0123] The tension of the rope on the drone;

[0124] For the attitude angle of the drone, ;

[0125] The rotational inertia matrix of the UAV.

[0126] In step S1, the cargo dynamics model is as follows: ,in:

[0127] For the location of the goods, ;

[0128] For the quality of goods;

[0129] Let the velocity vector of the cargo be... ;

[0130] For cargo aerodynamic drag;

[0131] This refers to the tension of the rope on the cargo.

[0132] In step S1, the constraint equation for the rope is: l is the length of the rope, and the direction vector of the rope is... .

[0133] In step S2, the swing angle is defined in the body coordinate system as: , ,in The swing angle is the vertical deviation. This is the azimuth angle, which is the horizontal orientation. Let r be the position vector of the cargo in the body coordinate system, and let r be the displacement vector of the cargo in the inertial coordinate system.

[0134] In step S2, the oscillation energy function is: E k For the kinetic energy of the cargo, E p This refers to the potential energy of the cargo.

[0135] In step S2, the objective function is: ,in For the drone's acceleration vector, , , These are the velocity term weight, oscillation suppression weight, and acceleration smoothing weight, respectively. t0 is the track start time, and tf is the track end time.

[0136] In step S4, the cost function is: Let g(n) be the actual cost distance from the starting point to node n, and h(n) be the Euclidean distance from n to the ending point. The penalty for the oscillation energy at node n.

[0137] In step S4, the logic of the local trajectory optimization model is as follows:

[0138] S41: State sampling. At time k, the current state of the UAV is acquired through sensors. The current state includes position. ,speed Attitude angle Given the swing energy E(k), the UAV state expression is: x(k|k) = [ , , [E(k)] T ;

[0139] S42: Establish a predictive time-domain rolling optimization model within the time window [k, k+N-1]. ,in (j) is the number of steps to predict the speed of the drone. For j-step predictive control input, For j-step prediction of cargo swing energy, Q1, Q2, and R are weight matrices;

[0140] S43: Execute the first step control, first apply only the current step to solve the optimal solution of the model, and then discard the remaining control sequence;

[0141] S44: Move to the next time step k+1, and then repeat steps S41-S43.

[0142] In step S42, the predicted time-domain rolling optimization model satisfies dynamic constraints, obstacle avoidance constraints, and control quantity constraints, which are as follows:

[0143] Dynamic constraints: ;

[0144] Obstacle avoidance constraints: , Location of the obstacle. Minimum safe distance;

[0145] Control constraints: , These are the lower and upper limits of the control quantity, respectively.

[0146] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for trajectory optimization control of a long-range, high-payload, repeatable uniplane vertical takeoff and landing (VTOL) unmanned aerial vehicle (UAV), characterized in that, Includes the following steps: S1: Establish a dynamic model of the cargo drone, including the drone body model and the cargo dynamic model; S2: Design trajectory optimization indicators, using energy as the indicator for trajectory optimization, and establish an optimization objective function by combining the definition of swing angle and the energy function of cargo drones; S3: Environmental perception and map building, which uses LiDAR or visual sensors to build point cloud data and construct flight route maps; S4: Global path planning. A cost function is established based on the map and objective function. The cost function is then optimized to perform path planning. At the same time, a local trajectory optimization model is established to optimize the local trajectory based on the position of obstacles during flight. In step S1, the cargo dynamics model is as follows: ,in: For the location of the goods, ; For the quality of goods; Let the velocity vector of the cargo be... ; For the aerodynamic drag of the cargo; The tension in the rope on the cargo; e3 is the unit vector along the Z-axis; In step S1, the constraint equation for the rope is: , Let be the length of the rope, and let the direction vector of the rope be . ; In step S4, the cost function is: Let g(n) be the actual cost distance from the starting point to node n, and h(n) be the Euclidean distance from n to the ending point. The penalty for the oscillation energy at node n; In step S4, the logic of the local trajectory optimization model is as follows: S41: State sampling. At time k, the current state of the UAV is acquired through sensors. The current state includes position. ,speed Attitude angle Given the swing energy E(k), the UAV state expression is: x(k|k) = [ , , [E(k)] T ; S42: Establish a predictive time-domain rolling optimization model within the time window [k, k+N-1]. ,in To predict the speed of the drone in step j, For j-step predictive control input, For j-step prediction of cargo swing energy, Q1, Q2, and R are weight matrices; S43: Execute the first step control, first apply only the current step to solve the optimal solution of the model, and then discard the remaining control sequence; S44: Move to the next time step k+1, and then repeat steps S41-S43.

2. The trajectory optimization control method for a long-range, high-payload, repetitive-wing vertical takeoff and landing UAV according to claim 1, characterized in that, In step S2, the swing angle is defined in the body coordinate system as: , ,in The swing angle is the vertical deviation. This is the azimuth angle, which is the horizontal orientation. Let r be the position vector of the cargo in the body coordinate system, and r be the displacement vector of the cargo in the inertial coordinate system. In step S2, the oscillation energy function is: E k For the kinetic energy of the cargo, E p This refers to the potential energy of the cargo.

3. The trajectory optimization control method for a long-range, high-payload, repetitive-wing vertical takeoff and landing UAV according to claim 1, characterized in that, In step S2, the objective function is: ,in For the drone's acceleration vector, , , These are the velocity term weight, oscillation suppression weight, and acceleration smoothing weight, respectively. t0 is the track start time, and tf is the track end time.

4. The trajectory optimization control method for a long-range, high-payload, repetitive-wing vertical takeoff and landing UAV according to claim 1, characterized in that, In step S42, the predicted time-domain rolling optimization model satisfies dynamic constraints, obstacle avoidance constraints, and control quantity constraints, which are as follows: Dynamic constraints: ; Obstacle avoidance constraints: , Location of the obstacle. Minimum safe distance; Control constraints: , These are the lower and upper limits of the control quantity, respectively.

5. The trajectory optimization control method for a long-range, high-payload, repetitive-wing vertical takeoff and landing UAV according to claim 1, characterized in that, In step S1, the UAV body model is: ,in: Let be the position vector of the UAV. ; For the drone's velocity vector, ; For drone quality; g is the acceleration due to gravity; The rotation matrix from the body coordinate system to the inertial coordinate system; For the body coordinate system thrust, ; For the aerodynamic drag of the drone; The tension of the rope on the drone; For the attitude angle of the drone, , This is the roll angle. The pitch angle, Yaw angle; The rotational inertia matrix of the UAV; e3 is the unit vector along the Z-axis; This is the angular velocity vector of the UAV; This is the total torque vector acting on the UAV; This is the control torque vector acting on the UAV.

6. A long-range, high-payload, reusable, vertical takeoff and landing unmanned aerial vehicle (UAV), characterized in that: include: The UAV body adopts a compound wing configuration combining fixed wings and multi-rotors; the multi-rotor system includes multiple independently driven rotors symmetrically arranged at predetermined positions on the fuselage and / or wings of the UAV body; the fixed wings are connected to the fuselage and adopt a laminar flow airfoil; the horizontal propulsion system includes at least one thruster; the energy system includes a high-capacity onboard battery; the cargo lifting system includes a high-strength lifting rope, the upper end of which is connected to the bottom of the UAV body; the UAV body is made of carbon fiber composite material; the material and structure of the lifting rope are selected according to actual needs; the UAV is configured to execute the trajectory optimization control method as described in claims 1-5.

Images