A Fault-Tolerant Control Method for Logistics of a Nonplanar Hexrotor UAV Based on a High-Order All-Drive System

Abstract

This invention relates to a fault-tolerant control method for non-planar hexacopter UAVs based on a high-order all-drive system. Specifically, it relates to the field of UAV control, specifically a fault-tolerant control method for non-planar hexacopter UAVs based on a high-order all-drive system. The purpose of this invention is to solve the problem of non-planar hexacopter UAVs failing to deliver packages due to actuator failures during delivery. The process is as follows: Step 1: Establish an actuator failure model considering incomplete failure; obtain a failure model for a non-planar hexacopter UAV carrying asymmetric loads; Step 2: Establish a second-order all-drive system; introduce nonlinear integral terms and variable parameters to form a new sliding mode surface; obtain the UAV input u when the UAV does not fail; obtain the ideal control input U of the actuator through inverse kinematics; calculate the deviation between u and U as compensation after the UAV fails; based on the compensation after the UAV fails, obtain the compensated control law, which is the controller.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) control, and more specifically to a fault-tolerant logistics control method for a non-planar hexacopter UAV based on a high-order all-drive system. Background Technology

[0002] With the rapid development of e-commerce, the express delivery industry faces ever-increasing transportation demands and fierce market competition. Traditional delivery methods can no longer meet the challenges of modern logistics. Problems such as traffic congestion, insufficient delivery to remote areas, and excessively long package processing times necessitate innovative solutions from the express delivery industry. Drones, as an emerging transportation tool, are gradually demonstrating enormous potential in express delivery due to their high efficiency and unique advantages. First, drones can avoid ground traffic congestion and quickly deliver packages to their destinations, meeting the e-commerce industry's demand for fast delivery and improving customer satisfaction. Second, the flexibility of drones allows them to cover remote or inaccessible areas, solving blind spots that traditional delivery methods cannot cover, thereby optimizing the balance of logistics services. Furthermore, drones can significantly reduce manual delivery costs and improve delivery efficiency, especially during peak periods, alleviating the workload of delivery personnel. With continuous technological advancements, the application prospects of drones in the express delivery industry are becoming increasingly broad.

[0003] Hexacopter drones offer several advantages in express delivery. First, they boast superior stability; the six-rotor design provides enhanced stability and control, enabling safe flight even in strong winds and complex environments. Second, they offer greater payload capacity; compared to quadcopters, the additional rotors allow them to carry heavier loads, making them suitable for transporting various types of packages. Furthermore, the six rotors provide redundancy, ensuring continued flight even if one or two rotors fail, thus enhancing safety. In terms of maneuverability, hexacopter drones offer more flexible control, making hovering, turning, and obstacle avoidance easier, adapting to the complex flight requirements of urban environments. Simultaneously, during horizontal movement, hexacopter drones require less tilting, thereby improving transport efficiency and shortening delivery times.

[0004] Compared to non-planar hexacopter all-wheel-drive designs, traditional hexacopter drones face numerous challenges in complex environments. Traditional hexacopter drones typically employ a planar layout, relying on six independent rotors to provide lift and thrust, but they have limitations in complex maneuvers, energy efficiency, and adaptability. For example, traditional hexacopter drones consume a significant amount of power, limiting their range; their larger structure and rotor layout also affect maneuverability, portability, and storage convenience. They exhibit poor adaptability in extreme weather and complex terrain, with the additional rotors increasing power consumption and complicating fault detection and maintenance. Non-planar hexacopter all-wheel-drive designs, through their unique construction philosophy, successfully overcome these limitations, demonstrating superior flight performance and adaptability. The most significant feature of this design is that the rotors are no longer confined to a planar layout but are distributed at multiple angles in three-dimensional space, allowing the drone to maneuver flexibly in a wider flight space. Through the rational configuration of rotors at different spatial angles, non-planar hexacopter drones enable each rotor to provide more balanced lift and thrust in different flight attitudes. This layout enables drones to achieve more precise control and faster response, significantly enhancing their flight capabilities, especially in complex environments. For example, in narrow city streets or mountainous terrain, the non-planar design allows drones to effectively avoid obstacles and achieve more flexible flight paths. This feature allows non-planar hexacopter drones to easily navigate around obstacles such as tall buildings and trees when performing delivery tasks, ensuring efficient and safe completion of delivery missions. Another significant advantage of the non-planar hexacopter design is its ability to achieve static hovering in any attitude. Traditional hexacopter drones may struggle to maintain stability in certain flight attitudes, especially when encountering changes in wind speed or complex airflow. However, the non-planar hexacopter, through its optimized rotor layout and control system, can maintain static hovering even at high altitudes or in complex terrain, thus providing a significant guarantee for accurate delivery. In terms of energy efficiency, the non-planar hexacopter is more efficient than the traditional hexacopter. Because its rotor layout allows for more precise control of the power output of each rotor, the thrust of each rotor can be more rationally distributed during flight, effectively reducing unnecessary energy waste. This feature is particularly important for long-duration flights or long-distance travel, especially in express delivery missions. Energy efficiency optimization can effectively improve the drone's working efficiency and reduce the risks associated with insufficient power. Compared to traditional hexacopter drones, the non-planar hexacopter's rapid response capability allows the drone to take off quickly and reach the target location rapidly. The flight control system can quickly adjust its flight attitude and path after receiving mission commands, achieving efficient and precise express delivery. Non-planar hexacopter drones not only improve stability but also optimize the design of power distribution and control systems, enabling them to exhibit better adaptability and maneuverability in complex environments. Compared to traditional hexacopter drones, their unique advantages greatly enhance the convenience of drone applications in express delivery.

[0005] This non-planar design, with its rotors rationally configured at different spatial angles, demonstrates a significant advantage in overcoming the underactuated characteristics of common multi-rotor configurations. By ensuring the full rank of the control assignment matrix, it enables six independent control variables, allowing the drone to move horizontally without additional tilting. The drone can achieve stable and continuous flight within a small angle range, reducing the need for additional rotation and stabilization devices in express delivery, thus increasing the possibility of modular design. The simplified design not only reduces production costs but also significantly lowers the failure rate, enhancing the reliability of the drone in express delivery. The advantages of this design are particularly pronounced when facing complex environmental factors such as strong winds or sudden weather changes. Rapid response capability is crucial for ensuring timely delivery when dealing with urgent delivery needs. However, non-planar hexacoach drones inevitably encounter mechanical failures such as sensor malfunctions, battery failures, shaft failures, and actuator failures when performing express delivery tasks. These failures can affect mission completion and lead to drone crashes and damage. Summary of the Invention

[0006] The purpose of this invention is to solve the problem of non-planar hexacopter drones failing to deliver packages due to actuator failures during delivery, and to propose a fault-tolerant control method for non-planar hexacopter drone logistics based on high-order all-drive system theory.

[0007] The specific process of a nonplanar hexacopter UAV logistics fault-tolerant control method based on high-order all-drive system theory is as follows:

[0008] Step 1: Establish an actuator failure model that considers incomplete failure; obtain a failure model of a non-planar hexacopter UAV carrying asymmetric loads that considers failure.

[0009] Step 2: Based on the fault model of the non-planar hexacopter UAV carrying asymmetric loads established in Step 1, design the controller; the specific process is as follows:

[0010] Step 21: Establish a second-order fully driven system;

[0011] Step 22: Introduce a nonlinear integral term H(e) and a variable parameter k. D This forms a new sliding surface s;

[0012] Steps 2 and 3: Based on the second-order fully driven system and the new sliding surface s, the control law of integral sliding mode is obtained;

[0013] Step 24: Proof of Lyapunov stability;

[0014] Step 25: Based on the fault model considering incomplete failure and the control law of integral sliding mode, obtain the ideal control input U of the actuator; based on the ideal control input U of the actuator, obtain the input u of the UAV when the UAV does not fail;

[0015] Step 26: Based on the second-order full-drive system and actuator fault model, obtain the control input U after actuator efficiency loss through inverse solution. r Based on the actuator fault model, the ideal control input U of the actuator is obtained;

[0016] Step 27: Calculate the deviation κ = uU between the drone input u obtained in Step 25 when the drone does not malfunction and the ideal control input U of the actuator obtained in Step 26. κ is used as compensation after the drone malfunctions. Based on the compensation κ after the drone malfunctions, the compensated control law is obtained. The compensated control law is the controller.

[0017] The beneficial effects of this invention are as follows:

[0018] This paper proposes a controller design based on a high-order all-drive system to address the problem of delivery failures caused by actuator malfunctions during drone delivery. The controller, designed using a high-order all-drive system, effectively improves the robustness of the drone system in the face of faults or anomalies, ensuring its adaptability to external disturbances and system uncertainties during execution. Simultaneously, this method enhances the system's rapid response capability, ensuring fast delivery of goods in emergency situations. Furthermore, the designed controller possesses strong fault tolerance; when some actuators fail, the system can compensate through control algorithms, ensuring stable drone flight and continued delivery. This high-order all-drive system control method not only improves the reliability of drones but also provides an effective solution for drone applications in complex environments. Attached Figure Description

[0019] Figure 1 This is a flowchart of the present invention;

[0020] Figure 2 The image shows the position tracking curves and error curves for the UAV. 'a' represents the position tracking curve along the x-axis, 'b' represents the position tracking curve along the y-axis, 'c' represents the position tracking curve along the z-axis, and 'd' represents the position tracking error.

[0021] Figure 3 The following are the attitude tracking curves and error curves of the UAV: ​​a is the roll angle tracking curve, b is the yaw angle tracking curve, c is the pitch angle tracking curve, and d is the attitude tracking error.

[0022] Figure 4The figures show the UAV position tracking curve and error curve after actuator failure. a is the position tracking curve after actuator failure, and b is the position tracking error after actuator failure.

[0023] Figure 5 The figures show the attitude tracking curves and error curves of the UAV after the actuator failure. a is the attitude tracking curve after the actuator failure, and b is the attitude tracking error after the actuator failure.

[0024] Figure 6 A schematic diagram of a fully-driven hexacopter UAV carrying a payload. Detailed Implementation

[0025] Specific Implementation Method 1: The specific process of this implementation method for a non-planar hexagonal UAV logistics fault-tolerant control method based on a high-order all-drive system is as follows:

[0026] Step 1: Establish an actuator failure model that considers incomplete failure; obtain a failure model of a non-planar hexacopter UAV carrying asymmetric loads that considers failure.

[0027] This paper mainly studies the design of fault-tolerant control strategies when actuators simultaneously experience multiplicative and additive faults.

[0028] Step 2: Based on the fault model of the non-planar hexacopter UAV carrying asymmetric loads established in Step 1, design the controller; the specific process is as follows:

[0029] Step 21: Establish a second-order fully driven system;

[0030] Step 22: Introduce a nonlinear integral term H(e) and a variable parameter k. D This forms a new sliding surface s;

[0031] Steps 2 and 3: Based on the second-order fully driven system and the new sliding surface s, the control law of integral sliding mode is obtained;

[0032] Step 24: Proof of Lyapunov stability;

[0033] Step 25: Based on the fault model formula (3) considering incomplete failure and the control law formula (29) of integral sliding mode, obtain the ideal control input U of the actuator (actual control law of the control system); based on the ideal control input U of the actuator, obtain the input u of the UAV when the UAV does not fail;

[0034] Step 26: Based on the second-order full-drive system (12) and actuator fault model (3), the control input U after actuator efficiency loss is obtained by inverse solution. r Based on the actuator fault model, the ideal control input U of the actuator (the actual control law of the control system) is obtained;

[0035] Step 27: Calculate the deviation κ = uU between the drone input u obtained in Step 25 when the drone does not malfunction and the ideal control input U of the actuator obtained in Step 26. κ is used as compensation after the drone malfunctions. Based on the compensation κ after the drone malfunctions, the compensated control law is obtained. The compensated control law is the controller (Equation 37).

[0036] The advanced all-wheel drive system is a stage 2 all-wheel drive system.

[0037] Specific Implementation Method Two: This implementation method differs from Specific Implementation Method One in that step one establishes an actuator failure model considering incomplete failure; a failure model of a non-planar hexacopter UAV carrying asymmetric loads considering failure is obtained; the specific process is as follows:

[0038] Establish a mathematical model for multiplicative actuator faults with effective factors; establish a mathematical model for additive actuator faults.

[0039] Based on the mathematical models of multiplicative actuator failures with effective factors and additive actuator failures, a fault model considering incomplete failures is established.

[0040] The dynamic model of the fully driven hexacopter aircraft is derived using the Newton-Euler method; based on the dynamic model of the fully driven hexacopter aircraft, the control input vector u of the UAV is obtained:

[0041] Obtain the inertial matrix of a multi-rotor UAV carrying asymmetric loads;

[0042] Based on the dynamic model of the fully driven hexarotor aircraft and the inertial matrix of the multirotor UAV carrying asymmetric loads, the dynamic equations of the aircraft rotating around the center of mass when carrying loads are obtained.

[0043] Based on the fault model considering incomplete failure, the dynamic model of the fully driven hexacopter aircraft, and the dynamic equation of the aircraft rotating around the center of mass when carrying a load, a fault model of a non-planar hexacopter UAV carrying asymmetric load considering faults is obtained.

[0044] In the control system of a hexacopter drone, the actuator is a crucial component. As the actuator of the drone, the healthy operation of its six motors is fundamental to ensuring normal flight. Common actuator failures include multiplicative and additive failures.

[0045] This paper mainly studies the design of fault-tolerant control strategies when actuators simultaneously experience multiplicative and additive faults.

[0046] The other steps and parameters are the same as in Specific Implementation Method 1.

[0047] Specific Implementation Method 3: This implementation method differs from Specific Implementation Method 1 or 2 in that it establishes a mathematical model for multiplicative actuator faults with effective factors; and establishes a mathematical model for additive actuator faults.

[0048] Based on the mathematical models of multiplicative actuator failures with effective factors and additive actuator failures, a fault model considering incomplete failures is established.

[0049] The dynamic model of the fully driven hexacopter aircraft is derived using the Newton-Euler method; based on the dynamic model of the fully driven hexacopter aircraft, the control input vector u of the UAV is obtained:

[0050] Obtain the inertial matrix of a multi-rotor UAV carrying asymmetric loads;

[0051] Based on the dynamic model of the fully driven hexarotor aircraft and the inertial matrix of the multirotor UAV carrying asymmetric loads, the dynamic equations of the aircraft rotating around the center of mass when carrying loads are obtained.

[0052] Based on the fault model considering incomplete failure, the dynamic model of the fully driven hexacopter aircraft, and the dynamic equation of the aircraft rotating around the center of mass when carrying a load, a fault model of a non-planar hexacopter UAV carrying asymmetric load considering faults is obtained.

[0053] In the control system of a hexacopter drone, the actuator is a crucial component. As the actuator of the drone, the healthy operation of its six motors is fundamental to ensuring normal flight. Common actuator failures include multiplicative and additive failures.

[0054] The specific process is as follows:

[0055] Step 11: Establish a mathematical model for multiplicative actuator faults with an effective factor; expressed as:

[0056] U r =τU (1)

[0057] in,

[0058] U r U represents the control input after efficiency loss of the actuator (six actuators in a six-rotor aircraft). r =[U r1 U r2 U r3 U r4 U r5 U r6 ] T U r1 U represents the control input after the efficiency loss of actuator 1. r2U represents the control input after the efficiency loss of actuator 2. r3 U represents the control input after the efficiency loss of actuator 3. r4 U represents the control input after the efficiency loss of actuator 4. r5 U represents the control input after the efficiency loss of actuator 5. r6 This represents the control input after the efficiency loss of actuator 6, and the superscript table T indicates the transpose;

[0059] U represents the ideal control input of the actuator, U = [U1 U2 U3 U4 U5 U6] T U1 represents the ideal control input of actuator 1, U2 represents the ideal control input of actuator 2, U3 represents the ideal control input of actuator 3, U4 represents the ideal control input of actuator 4, U5 represents the ideal control input of actuator 5, and U6 represents the ideal control input of actuator 6.

[0060] τ represents the effective factor of actuator failure, τ∈[0,1]. When τ=1, it means that the actuator has no failure; when τ=0, it means that the actuator has a complete efficiency loss; when 0<τ<1, it means that the actuator has a partial efficiency loss.

[0061] Steps 1 and 2: Establish a mathematical model for additive faults in the actuator; represented as:

[0062] U r =U+σc (2)

[0063] Where 'c' represents an additive fault in the actuator;

[0064] σ takes the value 0 or 1. When σ is 0, it indicates that the actuator has no additive fault; when σ is 1, it indicates that the actuator has an additive fault.

[0065] Step 13: Based on the mathematical models of multiplicative actuator failures with effective factors and additive actuator failures, establish a fault model considering incomplete failure; expressed as:

[0066] U r =τU+σc(0<τ<1) (3)

[0067] Step 14: Derive the dynamic model of the fully driven six-rotor aircraft using the Newton-Euler method; represented as:

[0068]

[0069] In the formula, F represents the total force on the fully driven hexacopter aircraft, and M represents the total torque on the fully driven hexacopter aircraft.

[0070] P = [x, y, z] TThe vector represents the position vector, where x represents the position along the x-axis of the Earth coordinate system, y represents the position along the y-axis of the Earth coordinate system, and z represents the position along the z-axis of the Earth coordinate system.

[0071] Represents the velocity vector. Represents the acceleration vector;

[0072] Ω=[p,q,r] T denoted by angular velocity, p represents the angular velocity of the roll angle φ, q represents the angular velocity of the pitch angle θ, and r represents the angular velocity of the yaw angle ψ;

[0073] Represents angular acceleration. Let Ω represent the second derivative;

[0074] m represents the total mass of the fully driven hexacopter aircraft;

[0075] J represents the airframe inertia matrix of a fully driven hexacopter aircraft;

[0076] g represents the acceleration due to gravity, taken as 9.81 m / s². -2 e3 represents [0 0 1] T ;

[0077] n = [n1 n2 n3 n4 n5 n6] T This represents the control input, i.e., the magnitude of the thrust generated by the propeller. n1 represents the magnitude of the thrust generated by propeller 1, n2 represents the magnitude of the thrust generated by propeller 2, n3 represents the magnitude of the thrust generated by propeller 3, n4 represents the magnitude of the thrust generated by propeller 4, n5 represents the magnitude of the thrust generated by propeller 5, and n6 represents the magnitude of the thrust generated by propeller 6.

[0078] k represents air resistance. x k y k z Indicates the air drag coefficient. This represents the velocity along the x-axis of the Earth's coordinate system. This represents the velocity along the y-axis of the Earth's coordinate system. This represents the velocity along the z-axis of the Earth coordinate system, and diag() represents a diagonal matrix.

[0079] F(α) represents the control efficiency matrix, α is the angle of rotation of the rotor around the arm, and sα and cα represent sinα and cosα, respectively;

[0080] The control efficiency matrix F(α) is expressed as follows:

[0081]

[0082] In the formula,

[0083] T′ represents an intermediate variable.

[0084] W represents an intermediate variable.

[0085] L is the rotor radius of the UAV. These are the constant parameters for the UAV;

[0086] Step 15: Based on the dynamic model of the fully driven six-rotor aircraft, derive the control input vector u of the UAV:

[0087] Based on formulas (4) and (5), the control input vector u of the UAV is derived as follows:

[0088] Represented as:

[0089]

[0090] In the formula, u represents the control input vector of the UAV, u1 represents input quantity one of u, u2 represents input quantity two of u, u3 represents input quantity three of u, u4 represents input quantity four of u, u5 represents input quantity five of u, and u6 represents input quantity six of u;

[0091] Step 16: Obtain the inertial matrix of the multi-rotor UAV carrying asymmetric loads;

[0092] The specific process is as follows:

[0093] When a drone carries an asymmetric payload, assuming in the aircraft's body coordinate system O b X b Y b Z b Below, the aircraft carries a rigid body load located in plane O of the aircraft's body coordinate system. b X b Y b Below, the load is considered as a point mass with mass m. load The load distance from the aircraft body coordinate system plane O b X b Z b The distance is d y The load distance from the aircraft body coordinate system plane O b Y b Z b The distance is d x The load distance from the aircraft body coordinate system plane O b X b Y b The distance is d z ;

[0094] The direction of the load's gravity is opposite to the Z direction.b Axis, X b Axis and Z b The axis is perpendicular, Y b Axis and X b Z b Plane perpendicular;

[0095] The aircraft with asymmetric loads is divided into a bare section and a loaded section. The moments of inertia of the bare section and the loaded section about each axis of the aircraft's body coordinate system are calculated separately. The specific process is as follows:

[0096] For X b Axis: Bare machine part relative to X b The moment of inertia of the shaft is I x The load part is related to X b The moment of inertia of the axis is

[0097] For Y b Axis: Bare machine part relative to Y b The moment of inertia of the shaft is I y The load part is related to Y b The moment of inertia of the axis is

[0098] For Z b Axis: Bare machine part relative to Z b The moment of inertia of the shaft is I z The load part is related to Z b The moment of inertia of the axis is

[0099] When calculating the product of inertia, it is also divided into the bare body and the load portion. Given the aircraft's centrally symmetrical shape, in the aircraft's body coordinate system, the product of inertia for the bare body is zero. Therefore, the product of inertia for the load portion can be obtained:

[0100]

[0101] In the formula, I xy I represents the cross-correlation product of inertia of the load rotating about the x-axis and y-axis. yx I represents the cross-correlation product of inertia of the load rotating about the y-axis and x-axis. xz I represents the cross-correlation product of inertia of the load rotating about the x-axis and z-axis. zx I represents the cross-correlation product of inertia of the load rotating about the z-axis and x-axis. yz I represents the cross-correlation product of inertia of the load rotating about the y-axis and z-axis. zy This represents the cross-correlation inertial product of the load rotating about the z-axis and y-axis;

[0102] In summary, the inertial matrix of the multi-rotor UAV carrying asymmetric loads can be obtained as follows:

[0103]

[0104]

[0105] In the formula, J1 represents the inertial matrix of the multi-rotor UAV;

[0106] Step 17: Based on the dynamic model of the fully driven six-rotor aircraft and the inertial matrix of the multi-rotor UAV carrying asymmetric loads, obtain the dynamic equation of the aircraft rotating around the center of mass when carrying loads.

[0107] The dynamic equations of the aircraft rotating around its center of mass while carrying a load are obtained from formulas (4) and (8);

[0108] Represented as:

[0109]

[0110] In the formula, The angular acceleration representing the roll angle φ The angular acceleration representing the pitch angle θ Angular acceleration representing the yaw angle ψ;

[0111] make

[0112]

[0113] In the formula, D1 represents an intermediate variable, D2 represents an intermediate variable, and D3 represents an intermediate variable;

[0114] Step 18: Based on the fault model considering incomplete failure, the dynamic model of the fully driven hexacopter aircraft, and the dynamic equation of the aircraft rotating around the center of mass while carrying a load, a fault model of a non-planar hexacopter UAV carrying asymmetric load considering fault is obtained.

[0115] Based on (3), (4), (9), and (10), a fault model for a non-planar hexagonal UAV carrying asymmetric loads considering faults is obtained.

[0116] Represented as:

[0117]

[0118] In the formula, This represents the acceleration along the x-axis of the Earth's coordinate system. This represents the acceleration along the y-axis of the Earth's coordinate system. This represents the acceleration along the z-axis of the Earth coordinate system.

[0119] Other steps and parameters are the same as in specific implementation method one or two.

[0120] Specific Implementation Method Four: This implementation method differs from Specific Implementation Methods One to Three in that a second-order fully driven system is established in step two-one; the expression is:

[0121]

[0122] In the formula, E represents a constant matrix. Let t represent a continuous vector function, and t represent time. Let u represent the n′×n′ dimensional real matrix space, v represent the external signal, and n′ represent a positive integer.

[0123] Represents state variables, express The first derivative, express The second derivative;

[0124] in The following full-drive conditions must be met:

[0125]

[0126] In the formula, det represents the determinant of the matrix, and R n′ Represents an n′-dimensional real vector space;

[0127] When the second-order fully driven system (12) satisfies the fully driven condition (13), for any given matrix A1, A0 ∈ R n′×n′ The linear time-invariant closed-loop system is obtained through control law (14);

[0128]

[0129] Where u = [u1 u2 u3 u4 u5 u6] T Control input U after actuator efficiency loss r Represented as

[0130] U r =[U r1 U r2 U r3 U r4 U r5 U r6 ] T (15)

[0131] In the formula, A1 represents the n′×n′ dimensional real matrix space, A0 represents the n′×n′ dimensional real matrix space, and R n′×n′ Let u represent the n′×n′ dimensional real matrix space; u represents the control input vector of the UAV. Let v represent an n′×n′ dimensional real matrix space, and v represent an external signal.

[0132]

[0133] v = [0 0 g 0 0 0] T (18)

[0134] In the formula, diag(B1,B2,B3,B4,B5,B6) represents a diagonal matrix, B1, B2, B3, B4, B5, and B6 represent intermediate variables; d1, d2, and d3 represent intermediate variables.

[0135] in

[0136]

[0137] The tracking error e1 and its first derivative e2 are defined as follows:

[0138]

[0139] in, Represents state variables, Let e2 be the expected input to the state variable, and let e1 be the first derivative of e1. express The first derivative, express The first derivative.

[0140] The other steps and parameters are the same as those in one of the specific implementation methods one to three.

[0141] Specific Implementation Method Five: This implementation method differs from Specific Implementation Methods One to Four in that a nonlinear integral term H(e) and a variable parameter k are introduced in step two. D This forms a new sliding surface s, as shown in the following equation:

[0142]

[0143] Where s represents the sliding surface, k D k represents a constant. P k represents a constant. I Let H(e) represent a constant, H(e) represent the nonlinear integral term H(e), e2(0) represent the initial condition of e2, and e1(0) represent the initial condition of e1. Let h(e) represent the first derivative of H(e); e represents the error.

[0144] in

[0145]

[0146] β eIt is a positive constant;

[0147] β represents a constant greater than 0, β = diag(β) x ,β y ,β z ,β φ ,β θ ,β ψ ), β x β represents a constant. y β represents a constant. z β represents a constant. φ β represents a constant. θ β represents a constant. ψ β represents a constant. x β y β z β φ β θ β ψ For setting;

[0148] Taking the derivative of H(e), we obtain a nonlinear function of the following form:

[0149]

[0150] The other steps and parameters are the same as those in one of the specific implementation methods one to four.

[0151] Specific Implementation Method Six: This implementation method differs from Specific Implementation Methods One through Five in that, in steps two and three, the control law for integral sliding mode is obtained based on the second-order fully driven system and the new sliding surface s; the specific process is as follows:

[0152] Differentiating formula (22), we get:

[0153]

[0154] in, This represents the rate of change of state over time. This represents the first derivative of e². This represents the first derivative of H(e);

[0155] in, Combining (12) and (25), we get:

[0156]

[0157] in, express The second derivative, express The second derivative;

[0158] k D k represents a constant.P k represents a constant. I Represents a constant;

[0159] k D =diag(k) Dx ,k Dy ,k Dz ,k Dφ ,k Dθ ,k Dψ )

[0160] k I =diag(k) Ix ,k Iy ,k Iz ,k Iφ ,k Iθ ,k Iψ )

[0161] k P =diag(k) Px ,k Py ,k Pz ,k Pφ ,k Pθ ,k Pψ )

[0162] Where diag() represents a diagonal matrix;

[0163] k Dx k represents a constant. Dy k represents a constant. Dz k represents a constant. Dφ k represents a constant. Dθ k represents a constant. Dψ k represents a constant. Dx k Dy k Dz k Dφ k Dθ k Dψ For setting;

[0164] k Ix k represents a constant. Iy k represents a constant. Iz k represents a constant. Iφ k represents a constant. Iθ k represents a constant. Iψ k represents a constant. Ix k Iy k Iz k Iφ k Iθ k Iψ For setting;

[0165] k Px k represents a constant.Py k represents a constant. Pz k represents a constant. Pφ k represents a constant. Pθ k represents a constant. Pψ k represents a constant. Px k Py k Pz k Pφ k Pθ k Pψ For setting;

[0166] To enable the system state to approach the sliding surface at a faster rate, thereby achieving a fast and effective system response, an exponential convergence law is adopted here:

[0167]

[0168] In the exponential reaching law, εsigm(s) is the constant-rate reaching term;

[0169] ε represents a constant, ε = diag(ε) x ,ε y ,ε z ,ε φ ,ε θ ,ε ψ ), ε x ε represents a constant. y ε represents a constant. z ε represents a constant. φ ε represents a constant. θ ε represents a constant. ψ Represents a constant; ε, ε x ε y ε z ε φ ε θ ε ψ For setting;

[0170] in b represents a constant (set);

[0171] ks is the exponential approaching term, k represents a constant, and k = diag(k x ,k y ,k z ,k φ ,k θ ,k ψ ), k x k represents a constant. y k represents a constant. z k represents a constant. φ k represents a constant. θ k represents a constant. ψ Represents a constant; k, kx k y k z k φ k θ k ψ For setting;

[0172] Combining (26) and (27), we get:

[0173]

[0174] in, Represents a continuous vector function. Represents an n′×n′ dimensional real matrix space;

[0175] Based on (28), the control input U after obtaining the efficiency loss of the actuator (six actuators of a six-rotor) is obtained. r U r The control law for integral sliding mode is expressed as:

[0176]

[0177] U r This indicates the control input after the efficiency loss of the actuator (six actuators in a six-rotor aircraft).

[0178] The other steps and parameters are the same as those in one of the specific implementation methods one to five.

[0179] Specific Implementation Method Seven: This implementation method differs from Specific Implementation Methods One through Six in that the Lyapunov stability proof in step two or four is as follows:

[0180]

[0181] Where V represents the Lyapunov function, The derivative of the Lyapunov function is represented;

[0182] Substituting expression (29) into expression (31), we get:

[0183]

[0184] It is negatively definite, thereby ensuring the stability of the control system (ensuring the overall stability of the invention).

[0185] The other steps and parameters are the same as those in one of the specific implementation methods one to six.

[0186] Specific Implementation Method Eight: This implementation method differs from Specific Implementation Methods One to Seven in that, in step two-five, the ideal control input U of the actuator (actual control law of the control system) is obtained based on the fault model formula (3) considering incomplete failure and the control law formula (29) of integral sliding mode; and the input u of the UAV when the UAV does not fail is obtained based on the ideal control input U of the actuator.

[0187] Represented as:

[0188]

[0189] Where τ=diag(τ x ,τ y ,τ z ,τ φ ,τ θ ,τ ψ ),c = diag(c x ,c y ,c z ,c φ ,c θ ,c ψ );

[0190] τ represents a constant, τ x τ represents a constant. y τ represents a constant. z τ represents a constant. φ τ represents a constant. θ τ represents a constant. ψ Represents a constant; τ, τ x τ y τ z τ φ τ θ τ ψ For setting;

[0191] c represents a constant, c x c represents a constant. y c represents a constant. z c represents a constant. φ c represents a constant. θ c represents a constant. ψ Represents a constant; c, c x c y c z c φ c θ c ψ For setting;

[0192] When τ=I 6×6 When σ is a zero matrix, the input u of the UAV when the UAV does not malfunction is:

[0193]

[0194] The other steps and parameters are the same as those in any of the specific implementation methods one to seven.

[0195] Specific Implementation Method Nine: This implementation method differs from Specific Implementation Methods One through Eight in that, in step two-six, the control input U after actuator efficiency loss is obtained by inverse solution based on the second-order full-drive system (12) and actuator fault model (3). r Based on the actuator fault model, the ideal control input U of the actuator (the actual control law of the control system) is obtained;

[0196] Represented as:

[0197]

[0198] Substituting the fault model (3) into formula (35), we obtain the ideal control input U of the actuator (the actual control law of the control system) as follows:

[0199]

[0200] The other steps and parameters are the same as those in one of the specific implementation methods one to eight.

[0201] Specific Implementation Method 10: This implementation method differs from one of the specific implementation methods 1 to 9 in that, in step 27, the deviation κ = uU between the input u of the UAV obtained in step 25 when the UAV does not malfunction and the ideal control input U of the actuator obtained in step 26 is calculated. κ is used as compensation after the UAV malfunctions. Based on the compensation κ after the UAV malfunctions, the compensated control law is obtained. The compensated control law is the controller.

[0202] The specific process is as follows:

[0203] The deviation κ = uU between the drone's input u (obtained in step 25 when the drone is not malfunctioning) and the actuator's ideal control input U (obtained in step 26) is calculated. κ serves as compensation after a drone malfunction. Based on the compensation κ after a drone malfunction, the compensated control law is obtained. The compensated control law is as follows:

[0204]

[0205] Among them, U κ This represents the control law after compensation.

[0206] The other steps and parameters are the same as those in any of the specific implementation methods one to nine.

[0207] Example:

[0208] To verify and demonstrate the effectiveness of the controller designed in this invention for the tracking control method of a non-planar hexacopter UAV based on the theory of high-order all-drive systems, the following numerical simulation experiments were conducted.

[0209] The parameters of the drone are: mass m = 12 kg, m load =8kg, when t≥1s, m load =1kg, gravitational acceleration g = 9.81N / kg, I x =0.2kg·m 2 ,I y =0.2kg·m 2 ,I z =0.37kg·m 2 ,k x =0.001N / (s·m) 2 ),k y =0.001N / (s·m) 2 ),k z =0.001N / (s·m) 2 ), Rotor radius L = 0.5m

[0210] , Table 1 and Table 2 show the controller parameters and fault effectiveness factor parameters, respectively.

[0211] Table 1 Controller Parameters

[0212]

[0213] Table 2 Fault Effective Factor Parameters

[0214]

[0215] Select position reference trajectory x d =sin(πt)m,y d =sin(0.5πt)m,z d =0.25tm; Attitude reference trajectory θ d =0.4cos(πt)rad, ψ d = 0.2sin(0.5πt)rad. The numerical experimental results are as follows: Figure 2 , 3 As shown;

[0216] Figure 2 and Figure 3The system displays the position and attitude tracking curves and errors of the hexacopter UAV. It achieves good tracking within 0.5 seconds with a tracking error of less than 0.005m, demonstrating high controller tracking accuracy and excellent tracking performance. This allows the UAV to accurately and quickly reach the designated location to perform its mission. A load change during the UAV's 1-second operation has minimal impact on the tracking position and attitude errors, indicating good system robustness and suitability for delivery tasks.

[0217] Figure 4 and Figure 5 The position and attitude tracking curves and errors of the hexacopter UAV when the actuator fails are shown respectively.

[0218] As shown in the figure, the UAV experienced a malfunction after 2 seconds. The maximum position tracking error was less than 0.006m, and the attitude tracking error was less than 0.01rad. It successfully recovered to normal operation after 0.5 seconds. The minimal impact of the malfunction on position and attitude tracking errors indicates excellent fault tolerance. This performance enables the UAV to effectively avoid delays caused by malfunctions when conducting express delivery in complex environments, ensuring timely delivery to the mission objective and demonstrating its high reliability and mission completion capability in practical applications.

[0219] This invention may have other embodiments. Without departing from the spirit and essence of this invention, those skilled in the art can make various corresponding changes and modifications according to this invention, but these corresponding changes and modifications should all fall within the protection scope of the appended claims.

Claims

1. A fault-tolerant logistics control method for a nonplanar hexacopter UAV based on a high-order all-drive system, characterized in that: The specific process of the method is as follows: Step 1: Establish an actuator failure model that considers incomplete failure; obtain a failure model of a non-planar hexacopter UAV carrying asymmetric loads that considers failure. Step 2: Based on the fault model of the non-planar hexacopter UAV carrying asymmetric loads established in Step 1, design the controller; the specific process is as follows: Step 21: Establish a second-order fully driven system; Step 22: Introduce a nonlinear integral term H(e) and a variable parameter k. D This forms a new sliding surface s; Steps 2 and 3: Based on the second-order fully driven system and the new sliding surface s, the control law of integral sliding mode is obtained; Step 24: Proof of Lyapunov stability; Step 25: Based on the fault model considering incomplete failure and the control law of integral sliding mode, obtain the ideal control input U of the actuator; based on the ideal control input U of the actuator, obtain the input u of the UAV when the UAV does not fail; Step 26: Based on the second-order full-drive system and actuator fault model, obtain the control input U after actuator efficiency loss through inverse solution. r Based on the actuator fault model, the ideal control input U of the actuator is obtained; Step 27: Calculate the deviation κ = uU between the drone input u obtained in Step 25 when the drone does not malfunction and the ideal control input U of the actuator obtained in Step 26. κ is used as compensation after the drone malfunctions. Based on the compensation κ after the drone malfunctions, the compensated control law is obtained. The compensated control law is the controller. In step two, a second-order fully driven system is established; the expression is: In the formula, E represents a constant matrix. Let t represent a continuous vector function, and t represent time. Let u represent the n′×n′ dimensional real matrix space, v represent the external signal, and n′ represent a positive integer. Represents state variables, express The first derivative, express The second derivative; in The following full-drive conditions must be met: In the formula, det represents the determinant of the matrix, and R n′ Represents an n′-dimensional real vector space; When the second-order fully driven system (12) satisfies the fully driven condition (13), for any given matrix A1, A0 ∈ R n′×n′ The linear time-invariant closed-loop system is obtained through control law (14); Where u = [u1 u2 u3 u4 u5 u6] T u represents the control input vector of the UAV, u1 represents input quantity one of u, u2 represents input quantity two of u, u3 represents input quantity three of u, u4 represents input quantity four of u, u5 represents input quantity five of u, and u6 represents input quantity six of u; Control input U after actuator efficiency loss r Represented as U r =[U r1 U r2 U r3 U r4 U r5 U r6 ] T (15) In the formula, A1 represents the n′×n′ dimensional real matrix space, A0 represents the n′×n′ dimensional real matrix space, and R n′×n′ Let u represent the n′×n′ dimensional real matrix space; u represents the control input vector of the UAV. Let v represent an n′×n′ dimensional real matrix space, and v represent an external signal. U r1 U represents the control input after the efficiency loss of actuator 1. r2 U represents the control input after the efficiency loss of actuator 2. r3 U represents the control input after the efficiency loss of actuator 3. r4 U represents the control input after the efficiency loss of actuator 4. r5 U represents the control input after the efficiency loss of actuator 5. r6 This represents the control input after the efficiency loss of actuator 6, and the superscript table T indicates the transpose; v = [0 0 g 0 0 0] T (18) In the formula, diag(B1,B2,B3,B4,B5,B6) represents a diagonal matrix, and B1, B2, B3, B4, B5, B6 represent intermediate variables; k x k y k z Indicates the air drag coefficient; α is the angle of rotation of the rotor around the arm, and sα and cα represent sinα and cosα, respectively; m load Let m represent the total mass of the fully driven hexacopter aircraft, where m is the payload mass. I x For the bare metal part to X b moment of inertia of the shaft, I y For the bare metal part to Y b moment of inertia of the shaft, I z For the bare metal part to Z b Moment of inertia of the shaft; T′ represents an intermediate variable. W represents an intermediate variable. L is the rotor radius of the UAV. These are the constant parameters for the UAV; This represents the velocity along the x-axis of the Earth's coordinate system. This represents the velocity along the y-axis of the Earth coordinate system. This represents the velocity along the z-axis of the Earth coordinate system. The angular acceleration representing the roll angle φ The angular acceleration representing the pitch angle θ; d y The load distance from the aircraft body coordinate system plane O b X b Z b The distance, d x The load distance from the aircraft body coordinate system plane O b Y b Z b The distance, d z The load distance from the aircraft body coordinate system plane O b X b Y b The distance; D3 represents an intermediate variable; D3 = WU r6 +pq[I x -I y +m load (d y 2 -d x 2 )]+prm load d y d z -qrm load d x d z +(q 2 -p 2 )m load d x d y d1, d2, and d3 represent intermediate variables; in p represents the angular velocity of the roll angle φ, q represents the angular velocity of the pitch angle θ, and r represents the angular velocity of the yaw angle ψ; The tracking error e1 and its first derivative e2 are defined as follows: in, Represents state variables, Let e2 be the expected input to the state variable, and let e1 be the first derivative of e1. express The first derivative, express The first derivative.

2. The nonplanar hexacopter UAV logistics fault-tolerant control method based on a high-order all-drive system according to claim 1, characterized in that: In step one, an actuator failure model considering incomplete failure is established; a failure model of a non-planar hexacopter UAV carrying asymmetric loads considering failure is obtained; the specific process is as follows: Establish a mathematical model for multiplicative actuator faults with effective factors; establish a mathematical model for additive actuator faults. Based on the mathematical models of multiplicative actuator failures with effective factors and additive actuator failures, a fault model considering incomplete failures is established. The dynamic model of the fully driven hexacopter aircraft is derived using the Newton-Euler method; based on the dynamic model of the fully driven hexacopter aircraft, the control input vector u of the UAV is obtained: Obtain the inertial matrix of a multi-rotor UAV carrying asymmetric loads; Based on the dynamic model of the fully driven hexarotor aircraft and the inertial matrix of the multirotor UAV carrying asymmetric loads, the dynamic equations of the aircraft rotating around the center of mass when carrying loads are obtained. Based on the fault model considering incomplete failure, the dynamic model of the fully driven hexacopter aircraft, and the dynamic equation of the aircraft rotating around its center of mass while carrying a load, a fault model for a non-planar hexacopter UAV carrying an asymmetric load considering faults is obtained.

3. The nonplanar hexacopter UAV logistics fault-tolerant control method based on a high-order all-drive system according to claim 2, characterized in that: The mathematical model for actuator multiplicative faults with effective factors is established; the mathematical model for actuator additive faults is established. Based on the mathematical models of multiplicative actuator failures with effective factors and additive actuator failures, a fault model considering incomplete failures is established. The dynamic model of the fully driven hexacopter aircraft is derived using the Newton-Euler method; based on the dynamic model of the fully driven hexacopter aircraft, the control input vector u of the UAV is obtained: Obtain the inertial matrix of a multi-rotor UAV carrying asymmetric loads; Based on the dynamic model of the fully driven hexarotor aircraft and the inertial matrix of the multirotor UAV carrying asymmetric loads, the dynamic equations of the aircraft rotating around the center of mass when carrying loads are obtained. Based on the fault model considering incomplete failure, the dynamic model of the fully driven hexacopter aircraft, and the dynamic equation of the aircraft rotating around the center of mass when carrying a load, a fault model of a non-planar hexacopter UAV carrying asymmetric load considering faults is obtained. The specific process is as follows: Step 11: Establish a mathematical model for multiplicative actuator faults with an effective factor; expressed as: U r =τU (1) in, U r U represents the control input after actuator efficiency loss. r =[U r1 U r2 U r3 U r4 U r5 U r6 ] T ; U represents the ideal control input of the actuator, U = [U1 U2 U3 U4 U5 U6] T U1 represents the ideal control input of actuator 1, U2 represents the ideal control input of actuator 2, U3 represents the ideal control input of actuator 3, U4 represents the ideal control input of actuator 4, U5 represents the ideal control input of actuator 5, and U6 represents the ideal control input of actuator 6. τ represents the effective factor of actuator failure, τ∈[0,1]. When τ=1, it means that the actuator has no failure; when τ=0, it means that the actuator has a complete efficiency loss; when 0<τ<1, it means that the actuator has a partial efficiency loss. Steps 1 and 2: Establish a mathematical model for additive faults in the actuator; represented as: And r =U+σc (2) Where 'c' represents an additive fault in the actuator; σ takes the value 0 or 1. When σ is 0, it indicates that the actuator has no additive fault; when σ is 1, it indicates that the actuator has an additive fault. Step 13: Based on the mathematical models of multiplicative actuator failures with effective factors and additive actuator failures, establish a fault model considering incomplete failure; expressed as: U r =τU+σc(0<τ<1) (3) Step 14: Derive the dynamic model of the fully driven six-rotor aircraft using the Newton-Euler method; represented as: In the formula, F represents the total force on the fully driven hexacopter aircraft, and M represents the total torque on the fully driven hexacopter aircraft. P = [x, y, z] T The vector represents the position vector, where x represents the position along the x-axis of the Earth coordinate system, y represents the position along the y-axis of the Earth coordinate system, and z represents the position along the z-axis of the Earth coordinate system. Represents the velocity vector. Represents the acceleration vector; Ω=[p,q,r] T Indicates angular velocity; Represents angular acceleration. Let Ω represent the second derivative; J represents the airframe inertia matrix of a fully driven hexacopter aircraft; g represents the acceleration due to gravity, taken as 9.81 m / s². -2 e3 represents [0 0 1] T ; n = [n1 n2 n3 n4 n5 n6] T This represents the control input, i.e., the magnitude of the thrust generated by the propeller. n1 represents the magnitude of the thrust generated by propeller 1, n2 represents the magnitude of the thrust generated by propeller 2, n3 represents the magnitude of the thrust generated by propeller 3, n4 represents the magnitude of the thrust generated by propeller 4, n5 represents the magnitude of the thrust generated by propeller 5, and n6 represents the magnitude of the thrust generated by propeller 6. Represents air resistance; diag() represents a diagonal matrix. F(α) represents the control efficiency matrix; The control efficiency matrix F(α) is expressed as follows: Step 15: Based on the dynamic model of the fully driven six-rotor aircraft, derive the control input vector u of the UAV: Represented as: Step 16: Obtain the inertial matrix of the multi-rotor UAV carrying asymmetric loads; the specific process is as follows: When a drone carries an asymmetric payload, assuming in the aircraft's body coordinate system O b X b Y b Z b Below, the aircraft carries a rigid body load located in plane O of the aircraft's body coordinate system. b X b Y b Below, the load is considered as a point mass with mass m. load The load distance from the aircraft body coordinate system plane O b X b Z b The distance is d y The load distance from the aircraft body coordinate system plane O b Y b Z b The distance is d x The load distance from the aircraft body coordinate system plane O b X b Y b The distance is d z ; The aircraft with asymmetric loads is divided into a bare section and a loaded section. The moments of inertia of the bare section and the loaded section about each axis of the aircraft's body coordinate system are calculated separately. The specific process is as follows: For X b Axis: Bare machine part relative to X b The moment of inertia of the shaft is I x The load part for X b The moment of inertia of the axis is For Y b Axis: Bare machine part relative to Y b The moment of inertia of the shaft is I y The load part is related to Y b The moment of inertia of the axis is For Z b Axis: Bare machine part relative to Z b The moment of inertia of the shaft is I z The load part is related to Z b The moment of inertia of the axis is When calculating the product of inertia, it is also divided into the bare body and the load portion. Given the aircraft's centrally symmetrical shape, in the aircraft's body coordinate system, the product of inertia for the bare body is zero. Therefore, the product of inertia for the load portion can be obtained: In the formula, I xy I represents the cross-correlation product of inertia of the load rotating about the x-axis and y-axis. yx I represents the cross-correlation product of inertia of the load rotating about the y-axis and x-axis. xz I represents the cross-correlation product of inertia of the load rotating about the x-axis and z-axis. zx I represents the cross-correlation product of inertia of the load rotating about the z-axis and x-axis. yz I represents the cross-correlation product of inertia of the load rotating about the y-axis and z-axis. zy This represents the cross-correlation inertial product of the load rotating about the z-axis and y-axis; In summary, the inertial matrix of the multi-rotor UAV carrying asymmetric loads can be obtained as follows: In the formula, J1 represents the inertial matrix of the multi-rotor UAV; Step 17: Based on the dynamic model of the all-driven six-rotor aircraft and the inertial matrix of the multi-rotor UAV carrying an asymmetric load, the dynamic equation of the aircraft's rotation around its center of mass while carrying the load is obtained; expressed as: In the formula, Angular acceleration representing the yaw angle ψ; make In the formula, D1 represents an intermediate variable, D2 represents an intermediate variable, and D3 represents an intermediate variable; Step 18: Based on the fault model considering incomplete failure, the dynamic model of the fully driven hexacopter aircraft, and the dynamic equations of the aircraft rotating around its center of mass while carrying a load, a fault model for a non-planar hexacopter UAV carrying an asymmetric load, considering faults, is obtained; expressed as: In the formula, This represents the acceleration along the x-axis of the Earth's coordinate system. This represents the acceleration along the y-axis of the Earth's coordinate system. This represents the acceleration along the z-axis of the Earth coordinate system.

4. The nonplanar hexacopter UAV logistics fault-tolerant control method based on a high-order all-drive system according to claim 3, characterized in that: In step two, a nonlinear integral term H(e) and a variable parameter k are introduced. D This forms a new sliding surface s, as shown in the following equation: Where s represents the sliding surface, k D k represents a constant. P k represents a constant. I Let H(e) represent a constant, H(e) represent a nonlinear integral term, e2(0) represent the initial condition of e2, and e1(0) represent the initial condition of e1. Let h(e) represent the first derivative of H(e); e represents the error. in β e It is a positive constant; β represents a constant greater than 0, β = diag(β) x ,β y ,β z ,β φ ,β θ ,β ψ ), β x β represents a constant. y β represents a constant. z β represents a constant. φ β represents a constant. θ β represents a constant. ψ Represents a constant; Taking the derivative of H(e), we obtain a nonlinear function of the following form:

5. The nonplanar hexacopter UAV logistics fault-tolerant control method based on a high-order all-drive system according to claim 4, characterized in that: In steps two and three, the control law for integral sliding mode is obtained based on the second-order fully driven system and the new sliding surface s; the specific process is as follows: Differentiating formula (22), we get: in, This represents the rate of change of the state over time. This represents the first derivative of e². This represents the first derivative of H(e); in, Combining (12) and (25), we get: in, express The second derivative, express The second derivative; k D k represents a constant. P k represents a constant. I Represents a constant; k D =diag(k Dx ,k Dy ,k Dz ,k D φ,k D θ,k D ψ) k I =diag(k Ix ,k Iy ,k Iz ,k I φ,k I θ,k I ψ) k P =diag(k Px ,k Py ,k Pz ,k P φ,k P θ,k P ψ) Where diag() represents a diagonal matrix; k Dx k represents a constant. Dy k represents a constant. Dz k represents a constant. Dφ k represents a constant. Dθ k represents a constant. Dψ Represents a constant; k Ix k represents a constant. Iy k represents a constant. Iz k represents a constant. Iφ k represents a constant. Iθ k represents a constant. Iψ Represents a constant; k Px k represents a constant. Py k represents a constant. Pz k represents a constant. Pφ k represents a constant. Pθ k represents a constant. Pψ Represents a constant; Using the exponential convergence law: εsigm(s) is the isodynamic approach term; ε represents a constant, ε = diag(ε) x ,ε y ,ε z ,ε φ ,ε θ ,ε ψ ), ε x ε represents a constant. y ε represents a constant. z ε represents a constant. φ ε represents a constant. θ ε represents a constant. ψ Represents a constant; in b represents a constant; ks is the exponential approaching term, k represents a constant, and k = diag(k x ,k y ,k z ,k φ ,k θ ,k ψ ), k x k represents a constant. y k represents a constant. z k represents a constant. φ k represents a constant. θ k represents a constant. ψ Represents a constant; Combining (26) and (27), we get: in, Represents a continuous vector function. Represents an n×n′ dimensional real matrix space; Based on (28), the control input U after the actuator efficiency loss is obtained. r U r The control law for integral sliding mode is expressed as: U r This represents the control input after the actuator has lost efficiency.

6. The nonplanar hexacopter UAV logistics fault-tolerant control method based on a high-order all-drive system according to claim 5, characterized in that: The Lyapunov stability proof in step two of the above steps is as follows: Where V represents the Lyapunov function, The derivative of the Lyapunov function is represented; Substituting expression (29) into expression (31), we get: It is negative definite, thus ensuring the stability of the control system.

7. The nonplanar hexacopter UAV logistics fault-tolerant control method based on a high-order all-drive system according to claim 6, characterized in that: In step two, the ideal control input U of the actuator is obtained based on the fault model considering incomplete failure and the control law of integral sliding mode; the input u of the UAV when the UAV does not fail is obtained based on the ideal control input U of the actuator. Represented as: Where τ=diag(τ x ,t y ,t z ,t φ ,t θ ,t ψ ),c=diag(c x ,c y ,c z ,c φ ,c θ ,c ψ ); τ represents a constant, τ x τ represents a constant. y τ represents a constant. z τ represents a constant. φ τ represents a constant. θ τ represents a constant. ψ Represents a constant; c represents a constant, c x c represents a constant. y c represents a constant. z c represents a constant. φ c represents a constant. θ c represents a constant. ψ Represents a constant; When τ=I 6×6 When σ is a zero matrix, the input u of the UAV when the UAV does not malfunction is:

8. The nonplanar hexacopter UAV logistics fault-tolerant control method based on a high-order all-drive system according to claim 7, characterized in that: In step 26, based on the second-order full-drive system and actuator fault model, the control input U after actuator efficiency loss is obtained by inverse solution. r Based on the actuator fault model, the ideal control input U of the actuator is obtained; Represented as: Substituting the fault model (3) into formula (35), the ideal control input U of the actuator is obtained as follows:

9. A fault-tolerant logistics control method for a nonplanar hexacopter UAV based on a high-order all-drive system according to claim 8, characterized in that: In step 27, the deviation κ = uU between the drone's input u (obtained in step 25 when the drone is not malfunctioning) and the ideal control input U of the actuator (obtained in step 26) is calculated. κ serves as compensation after a drone malfunction. Based on the compensation κ, the compensated control law is obtained, which is the controller. The specific process is as follows: The deviation κ = uU between the drone's input u (obtained in step 25 when the drone is not malfunctioning) and the actuator's ideal control input U (obtained in step 26) is calculated. κ serves as compensation after a drone malfunction. Based on the compensation κ after a drone malfunction, the compensated control law is obtained. The compensated control law is as follows: Among them, U κ This represents the control law after compensation.

Images