Aircraft shaking target safe landing control method based on motion constraint

Abstract

The invention discloses an aircraft shaking target safe landing control method based on motion constraints, and belongs to the field of aircraft autonomous control. The implementation method comprises the following steps: respectively establishing a dynamics model and a kinematics model of the aircraft; a continuous reference height is established by considering a safe landing condition; establishing position error dynamics and attitude error dynamics between the aircraft and the landing platform, so that the aircraft tracks the reference height trajectory; establishing a position loop force controller, constructing a constraint curved surface based on relative height, establishing a control barrier function, and constraining the output of the force controller to obtain a position loop saturation force controller with safety constraint; a roll angle range and a pitch angle range are given, a control barrier function of an attitude ring is established, output of a torque controller is constrained through the control barrier function, and an attitude ring torque controller with safety constraints is obtained; solving to obtain aircraft landing control output, and realizing aircraft shaking target safe landing control according to the aircraft landing control output.

Description

Technical Field

[0001] The present invention relates to a method for controlling the safe landing of an aircraft swaying target based on motion constraints, and belongs to the technical field of autonomous aircraft control. Background Art

[0002] In recent years, multirotor aircraft, owing to their maneuverability and high environmental adaptability, have been widely used in jungle exploration, urban security monitoring, maritime search and rescue, and other fields. To improve the aircraft's endurance and meet the needs of material transfer, autonomous landing on dynamic platforms such as moving vehicles or surface ships is required. However, complex terrain interference, external environmental disturbances, and insufficient navigation measurement accuracy still pose significant challenges to the safe and precise landing of aircraft on dynamic targets.

[0003] To address these issues, existing research has proposed various algorithms, including robust control, sliding mode control, adaptive control, neural network control, and fuzzy control, and has also introduced iterative learning strategies to address the control of uncertain systems. However, robust control and sliding mode control rely on conservative upper bounds on dynamic uncertainty, which can lead to excessively large control variables. Adaptive and learning algorithms require complex online parameter tuning or offline training, limiting their practicality. Furthermore, while disturbance estimators can simplify controller design, their position controllers fail to consider attitude amplitude constraints, potentially causing pitch or roll angles to exceed safe limits (-π / 2, π / 2), posing a risk of vehicle rollover. In the field of dynamic platform landing, existing research focuses on methods based on visual servoing, optimal control, and formation control, but most ignore the impact of platform swaying motion. Some studies have adopted a split landing strategy, terminating platform trajectory tracking during landing. This results in large landing errors at high speeds and lacks proper landing timing planning, which can easily lead to collisions and slips. Furthermore, while UWB-vision fusion positioning in GPS-denied environments improves measurement accuracy, existing results rely heavily on numerical simulations and lack actual flight experiments, making it difficult to handle the high-frequency platform swaying motion caused by complex terrain. Therefore, there is an urgent need to develop a robust hierarchical control strategy that balances attitude safety constraints, robust dynamic platform adaptability, and ease of engineering implementation to enable safe and autonomous landing of a quadrotor on a dynamic swaying mobile platform. Summary of the Invention

[0004] In order to solve the problems of lack of attitude safety, insufficient adaptability of the dynamic platform, high complexity of parameter setting, inaccurate positioning accuracy, etc. during the autonomous landing of a quadrotor aircraft on a dynamic swaying mobile platform, the purpose of the present invention is to propose a safe landing control method for a swaying target aircraft based on motion constraints, plan a reference trajectory for guidance of aircraft descent, tracking and landing, and provide feasible landing conditions for targets with swaying motion; construct a robust hierarchical control strategy using a saturation force controller and an attitude constraint torque controller; ensure the safe landing of the aircraft by constructing a constraint surface based on relative height, constrain the control force input by controlling the obstacle function; constrain the control torque input by controlling the obstacle function, so that the roll angle and pitch angle are within a reasonable range, thereby ensuring the safety of the aircraft.

[0005] The purpose of the present invention is achieved through the following technical solutions:

[0006] The present invention discloses a method for controlling the safe landing of an aircraft swaying target based on motion constraints, comprising the following steps:

[0007] Step 1: Establish the dynamic and kinematic models of the quadrotor respectively.

[0008] Step 2: During the tracking landing process of the aircraft, consider the safe landing conditions and establish a continuous reference altitude.

[0009] Step 3: Based on the dynamic and kinematic models of step 1, the position error dynamics and attitude error dynamics between the aircraft and the landing platform are established so that the aircraft can track the reference altitude trajectory proposed in step 2.

[0010] Step 4: Establish a position loop force controller and construct a constraint surface based on relative height. Establish a control obstacle function based on the constraint surface. Constrain the output of the force controller through the control obstacle function to obtain a position loop saturated force controller with safety constraints.

[0011] Step 5: Given a reasonable range of roll angle and pitch angle, establish a control obstacle function for the attitude loop, constrain the output of the torque controller through the control obstacle function, and obtain an attitude loop torque controller with safety constraints.

[0012] Step 6: Generate a safe landing control output for the aircraft by solving the position loop saturation force controller and attitude loop torque controller established based on the motion constraints in steps 4 and 5. Drive the aircraft based on this output to ultimately achieve safe and precise landing control relative to the moving and shaking target platform.

[0013] Furthermore, the steps for establishing the dynamic and kinematic models of the quadrotor aircraft in step 1 are as follows:

[0014] Establishing an inertial coordinate system and the body coordinate system of the quadrotor The coordinate center of B is the geometric center point of the quadrotor. The rotation matrix R is introduced to describe the rotational motion.

[0015] Where SO(3) represents the Lie group of third-order special orthogonal matrices, Represents m×n Euclidean space. I3 represents the 3rd-order identity matrix. Rotate to inertial coordinate system The corresponding rotation matrix is

[0016]

[0017] in, and Then, the position p and attitude γ of the aircraft are [φ,θ,ψ] T Expressed as:

[0018]

[0019] in, is the center of gravity of the quadrotor in the inertial coordinate system, is the linear velocity, g is the local acceleration of gravity, m is the mass of the quadrotor, is the thrust applied in the body coordinate system, is the unit vector in the Z direction in the inertial coordinate system, is the disturbance acceleration, is the inertia matrix, is the angular velocity, τ=[τ x ,τ y ,τ z ] T is the applied torque, is the disturbance torque, for x = [x1, x2, x3] T , uppercase × is defined as x × =[0,-x3,x2;x3,0,-x1;-x2,x1,0], W is defined as follows:

[0020]

[0021] Furthermore, the process of establishing a continuous reference height in step 2 is as follows:

[0022] According to the high-order polynomial function, the reference approach height Given by formula (7):

[0023]

[0024] where t a >0 is a pre-assigned time constant, h is a constant that satisfies h>sup(|p sz |) constant, a k is the constant coefficient determined by the following linear equation. t is time. p a (t) represents the reference approach altitude trajectory, which is expressed as:

[0025]

[0026] p z (t) is the vertical position. z (t) Velocity in the vertical direction.

[0027] When the aircraft reaches the height h, it starts to track the platform. When the platform rises and swings at a large angle, the aircraft introduces and To ensure the aircraft lands safely when the platform descends and swings at a small angle.

[0028]

[0029] Among them, φ s is the roll angle of the target platform. s is the pitch angle of the target platform. b1 and b2 are positive real numbers. By choosing appropriate b1 and b2, the platform will turn to a relatively stable situation in a short time.

[0030]

[0031] p sz is the height of the platform's vertical motion plane.

[0032] Satisfy at the same time and The aircraft performs the final landing operation. Switch to reference landing altitude

[0033]

[0034] in is the switching function to promote continuous switching, expressed as:

[0035]

[0036] where t s Represents the time to meet the safe landing conditions, n≥2 is a positive integer, t l >0 is the total time of the final landing operation.

[0037] The overall reference altitude for a complete landing is given by

[0038]

[0039] By finding p zd For a time guide, the vertical reference velocity is:

[0040]

[0041] Furthermore, the process of establishing the position error dynamics in step 3 is as follows:

[0042] According to the overall reference height p in step 2 zd and vertical reference velocity Let the reference position and reference velocity be p d =[p sx ,p sy ,p zd ] T and The position and velocity errors are defined as and According to equations (2) and (3), the position error dynamics is expressed as:

[0043]

[0044] in is the force controller, the equivalent disturbance acceleration γ c =[φ c ,θ c ,ψ c ] T are the command Euler angles, given by:

[0045]

[0046] In addition, in order to ensure that there is no singularity in attitude control, the following conditions should also be met:

[0047]

[0048] Thrust T = m‖u‖ is given by get.

[0049] Furthermore, the establishment process of the attitude error dynamics in step 3 is as follows:

[0050] Define an auxiliary manifold:

[0051]

[0052] κ1>0 and is a constant.

[0053] The attitude error dynamics is expressed as

[0054]

[0055] definition

[0056] Furthermore, the method of the position loop force controller in step 4 is as follows:

[0057] In order to compensate for the unmeasurable dynamic characteristics, a dynamics estimator is introduced and the following auxiliary system is constructed:

[0058]

[0059] Given an initial value s υ (0) = 0 and s u (0)=0, where ∈ t >0 is a constant parameter.

[0060] According to the position error dynamics in step 3, the nominal dynamic system is obtained:

[0061]

[0062] disturbed term By R(γ)→R(γ c )and Sure.

[0063] Formula (22) is transformed into t / (s+∈ t ) is filtered, and combined with equations (19) and (20), we get:

[0064]

[0065] in is u a The estimated value of the system The output of . Considering equations (19) and (23), the estimator is designed as:

[0066]

[0067] Define a new variable:

[0068]

[0069] Where Γ1=diag(γ 1x ,γ 1y ,γ 1z ), γ 1x ,γ 1y ,γ 1z is a positive constant. The force controller is established as:

[0070]

[0071] where Γ2 = diag(γ 2x ,γ 2y ,γ 2z ) is a positive definite matrix.

[0072] Furthermore, the method for establishing a constraint surface based on relative height in step 4 is as follows:

[0073] Design the following continuous smooth curve:

[0074]

[0075] Among them, h1 and h2 are positive real numbers satisfying h1>h2, and the definition is Indicates that the function is continuously differentiable n times. It is expressed as follows:

[0076]

[0077] Wherein, δ1 and δ2 are positive real numbers satisfying δ1>δ2, and n≥2 is a positive integer. z about The partial derivatives of are as follows:

[0078]

[0079] in,

[0080]

[0081] The constraint surface based on relative height is:

[0082]

[0083] where k s is a positive real number.

[0084] Furthermore, the method for establishing the position loop saturation force controller with safety constraints in step 4 is as follows: According to formula (31), the control obstacle function is obtained as follows:

[0085]

[0086] Define the following function

[0087]

[0088] Among them, α1(ψ0(t))=k1ψ0(t), α2(ψ1(t))=k2ψ1(t), k1 and k2 are positive real numbers.

[0089] The position loop saturation force controller with safety constraints is established using equations (32) and (33):

[0090]

[0091] Among them, u iref The actual force control input for the safety of the aircraft. i is the nominal force controller output.

[0092]

[0093]

[0094] In formula (36), in,

[0095]

[0096] Furthermore, the process of establishing the attitude loop control obstacle function and the constrained attitude torque controller in step 5 is as follows:

[0097] will u τ The dynamics estimator of is designed as follows:

[0098]

[0099] where ∈ r >0 is a constant parameter, z z and z τ Generated by the following auxiliary systems:

[0100]

[0101] Given an initial state s z (0) = 0 and s τ (0) = 0, then the attitude loop torque controller is designed as:

[0102]

[0103] Where κ2>0 is a constant. Definition is the estimation error.

[0104] In order for the aircraft to operate within a safe range of roll and pitch, it is necessary to ensure that the pitch or roll angle remains within the safe range (-π / 2,π / 2), so the constraint area is designed as Where v = φ, θ. The control barrier function is obtained as follows:

[0105]

[0106] Combined with formula (33), the attitude loop torque controller with safety constraints is established as follows:

[0107]

[0108] Among them, τ iref is the actual torque control input for safety of the aircraft. i is the nominal attitude torque controller output.

[0109]

[0110] In formulas (43)(44)(45)(46), k3 and k4 are positive real numbers.

[0111] [ξ1 ξ2 ξ3] T =W -1 J -1 (47)

[0112] [η1 η2 η3] T =W -1 J -1 (d τ -ω × Jω) (48)

[0113] Furthermore, the process of establishing the safe landing control method based on relative motion constraints in step 6 is as follows:

[0114] By solving the position loop saturation force controller and attitude loop torque controller established based on the motion constraints in steps 4 and 5, a safe aircraft landing control output is generated. The aircraft is driven based on this output, ultimately achieving safe and precise landing control relative to the moving and shaking target platform.

[0115] Beneficial effects:

[0116] 1. The present invention discloses a method for controlling the safe landing of an aircraft with a swaying target based on motion constraints. By establishing a continuously differentiable reference altitude trajectory, setting safe landing trigger conditions, and accurately modeling position and attitude errors, the aircraft can achieve high-precision tracking of the trajectory and attitude of a relatively stable moving swaying target platform, thereby achieving precise, smooth, and safe autonomous landing.

[0117] 2. The present invention discloses a method for controlling the safe landing of an aircraft with a shaking target based on motion constraints. By introducing a constraint surface based on relative height and a control obstacle function for horizontal position error and vertical height error, the relative position safety between the aircraft and the moving shaking target platform and the aircraft's own posture safety during landing are improved, collisions and risks are avoided, and the reliability of landing is ensured.

[0118] 3. The present invention discloses a method for controlling the safe landing of a swaying aircraft target based on motion constraints. By applying online quadratic programming in real time in the position loop and attitude loop, the safety constraints represented by the control obstacle function are seamlessly integrated into a high-performance nominal controller. This framework actively and continuously monitors and guides the aircraft to avoid dangerous states, significantly reducing the risk of collision or instability in complex swaying environments, realizing autonomous safety assurance in dynamic environments, and reaching a safety protection level that exceeds that of traditional passive fault-tolerant mechanisms. BRIEF DESCRIPTION OF THE DRAWINGS

[0119] Figure 1 Schematic diagram of the aircraft coordinate system of the present invention.

[0120] Figure 2 This is a flow chart of the algorithm of the present invention.

[0121] Figure 3 This is the control obstacle function and landing trajectory diagram of the present invention.

[0122] Figure 4 This is a trajectory diagram of a moving and shaking target tracked by the aircraft of the present invention. DETAILED DESCRIPTION

[0123] In order to better illustrate the purpose and advantages of the present invention, the invention is further described below with reference to the accompanying drawings and examples.

[0124] Example 1:

[0125] like Figure 2 As shown, this embodiment discloses a method for controlling the safe landing of an aircraft swaying target based on motion constraints, and the specific implementation steps are as follows:

[0126] Step 1: Establish the dynamic and kinematic model of the quadrotor.

[0127] Establishing an inertial coordinate system and the body coordinate system of the quadrotor like Figure 1 As shown; the coordinate center of B is the geometric center point of the quadrotor; the rotation matrix R is introduced to describe the rotational motion, Where SO(3) represents the Lie group of third-order special orthogonal matrices, Represents m×n Euclidean space; I3 represents the third-order unit matrix; the body coordinate system Rotate to inertial coordinate system The corresponding rotation matrix is

[0128]

[0129] in, and Then, the position p and attitude γ of the aircraft are [φ,θ,ψ] T Expressed as:

[0130]

[0131] in, is the center of gravity of the quadrotor in the inertial coordinate system, is the linear velocity, g is the local acceleration of gravity, which is 9.8N / s 2 ; m is the mass of the quadrotor, which is 1.1 kg; is the thrust applied in the body coordinate system, is the unit vector in the Z direction in the inertial coordinate system, is the disturbance acceleration, is the inertia matrix, diag(0.0064,0.0064,0.0128)kg m 2 ; is the angular velocity, τ=[τ x ,τ y ,τ z ] T is the applied torque, is the disturbance torque, for x = [x1, x2, x3] T , uppercase × is defined as x × =[0,-x3,x2;x3,0,-x1;-x2,x1,0], W is defined as follows:

[0132]

[0133] Step 2: Establish a continuous reference height

[0134] According to the high-order polynomial function, the reference approach height Given by formula (6):

[0135]

[0136] where t a >0 is the pre-assigned time constant, which is 5s; h is the value that satisfies h>sup(|p sz |) constant, take 2m; a k is the constant coefficient determined by the following linear equation; t is time; p a (t) represents the reference approach altitude trajectory, which is expressed as:

[0137]

[0138] p z (t) is the vertical position; v z (t) represents the velocity in the vertical direction.

[0139] When the aircraft reaches the height h, it starts to track the platform. When the platform rises and swings at a large angle, the aircraft introduces and To ensure the aircraft lands safely when the platform descends and swings at a small angle;

[0140]

[0141] Among them, φ s is the roll angle of the target platform; θ s is the pitch angle of the target platform; b1 and b2 are positive real numbers; by choosing appropriate b1 and b2, the platform will turn to a relatively stable situation in a short time;

[0142]

[0143] p sz is the height of the platform's heaving motion plane;

[0144] Satisfy at the same time and When the aircraft performs the final landing operation; the reference altitude is Switch to reference landing altitude

[0145]

[0146] in is the switching function to promote continuous switching, expressed as:

[0147]

[0148] where t s represents the time to meet the safe landing conditions, n≥2 is a positive integer, take 2; t l >0 is the total time of the final landing operation;

[0149] Therefore, the overall reference altitude for a complete landing is given by

[0150]

[0151] By finding p zd For a time guide, the vertical reference velocity is:

[0152]

[0153] (3) Establishing position error dynamics

[0154] According to the overall reference height p in step 2 zd and vertical reference velocity Let the reference position and reference velocity be p d =[p sx ,p sy ,p zd ] T and The position and velocity errors are defined as and According to equations (2) and (3), the position error dynamics is expressed as:

[0155]

[0156] in is the force controller, the equivalent disturbance acceleration γ c =[φ c ,θ c ,ψ c ] T are the command Euler angles, given by:

[0157]

[0158] In addition, in order to ensure that there is no singularity in attitude control, the following conditions should also be met:

[0159]

[0160] Thrust T = m‖u‖ is given by get.

[0161] (4) Establishing attitude error dynamics

[0162] Define an auxiliary manifold:

[0163]

[0164] κ1>0 and is a constant;

[0165] Combining equations (4) and (5), the attitude error dynamics is expressed as

[0166]

[0167] definition

[0168] (5) Establishing a position loop force controller

[0169] In order to compensate for the unmeasurable dynamic characteristics, a dynamics estimator is introduced and the following auxiliary system is constructed:

[0170]

[0171] Given an initial value s υ (0) = 0 and su (0)=0, where ∈ t >0 is a constant parameter, which is set to 0.1;

[0172] According to the position error dynamics in step 3, the nominal dynamic system is obtained:

[0173]

[0174] disturbed term It can be achieved by R(γ)→R(γ c )and Sure.

[0175] Formula (22) is transformed into t / (s+∈ t ) is filtered, and combined with equations (19) and (20), we get:

[0176]

[0177] in is u a The estimated value of the system The output of ; Considering (19) and (23), the estimator can be designed as:

[0178]

[0179] Define a new variable:

[0180]

[0181] Where Γ1=diag(γ 1x ,γ 1y ,γ 1z ), γ 1x ,γ 1y ,γ 1z is a positive constant, take diag(4,4,1.5); the force controller is established as:

[0182]

[0183] where Γ2 = diag(γ 2x ,γ 2y ,γ 2z ) is a positive definite matrix, take diag(0.14,0.14,0.23);

[0184] (6) Establish a constraint surface based on relative height

[0185] Design the following continuous smooth curve:

[0186]

[0187] Among them, h1 and h2 are positive real numbers satisfying h1>h2, and are 16 and 0.04 respectively. The constraint surface is as follows Figure 3 As shown, the definition Indicates that the function is continuously differentiable n times; It is expressed as follows:

[0188]

[0189] Among them, δ1 and δ2 are positive real numbers satisfying δ1>δ2, and are 8 and 1 respectively. n≥2 is a positive integer, and is 5. z about The partial derivatives of are as follows:

[0190]

[0191] in,

[0192]

[0193] Therefore, the constraint surface based on relative height is:

[0194]

[0195] where k s is a positive real number, take 1;

[0196] (7) Establish a position loop saturation force controller with safety constraints:

[0197] According to formula (31), the control barrier function is:

[0198]

[0199] Define the following function

[0200]

[0201] Where, α1(ψ0(t))=k1ψ0(t), α2(ψ1(t))=k2ψ1(t), k1 and k2 are positive real numbers, 3 and 3 respectively.

[0202] The position loop saturation force controller with safety constraints is established using equations (32) and (33):

[0203]

[0204]

[0205] Among them, u iref The actual force control input for the safety of the aircraft; u iis the nominal force controller output;

[0206]

[0207] In formula (36), in,

[0208]

[0209] (8) Establish the attitude loop control obstacle function and the constrained attitude loop torque controller

[0210] will u τ The dynamics estimator of is designed as follows:

[0211]

[0212] where ∈ r >0 is a constant parameter, taking 0.025, z z and z τ Generated by the following auxiliary systems:

[0213]

[0214] Given an initial state s z (0) = 0 and s τ (0) = 0, then the attitude loop torque controller is designed as:

[0215]

[0216] Where κ2>0 is a constant, which is set to 0.15; definition is the estimation error;

[0217] In order for the aircraft to operate within a safe range of roll and pitch, it is necessary to ensure that the pitch or roll angle remains within the safe range (-π / 2,π / 2), so the constraint area is designed as Where v = φ, θ; the control barrier function is obtained as follows:

[0218]

[0219] Then, combined with formula (33), the attitude loop torque controller with safety constraints is established as follows:

[0220]

[0221] Among them, τ iref is the actual torque control input for safety of the aircraft; τ i is the nominal attitude torque controller output;

[0222]

[0223] In formulas (43)(44)(45)(46), k3 and k4 are positive real numbers, 2, 2, respectively.

[0224] [ξ1 ξ2 ξ3] T =W -1 J -1 (47)

[0225] [η1η2η3] T =W -1 J -1 (d τ -ω × Jω) (48)

[0226] (9) Establish a safe landing control method based on relative motion constraints

[0227] By solving the position loop saturation force controller and attitude loop torque controller established based on the motion constraints in steps 4 and 5, a safe aircraft landing control output is generated. The aircraft is driven based on this output, ultimately achieving safe and precise landing control relative to the moving and shaking target platform.

[0228] Substitute the above parameters into the corresponding formula to solve and conduct flight verification; record the relative position relationship between the aircraft trajectory and the constraint surface during landing, such as Figure 3 As shown; record the trajectory of the aircraft and the trajectory of the moving shaking platform during the landing process, such as Figure 4 As shown;

[0229] The above specific description further illustrates the purpose, technical solutions and beneficial effects of the invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for controlling the safe landing of an aircraft swaying target based on motion constraints, characterized by: The following steps are included: Step 1: Establish the dynamic and kinematic models of the quadrotor aircraft respectively; Step 2: During the tracking landing process of the aircraft, consider the safe landing conditions and establish a continuous reference altitude; Step 3: Based on the dynamic and kinematic models of step 1, establish the position error dynamics and attitude error dynamics between the aircraft and the landing platform so that the aircraft tracks the reference altitude trajectory proposed in step 2; Step 4: Establish a position loop force controller and construct a constraint surface based on relative height. Use the constraint surface to establish a control obstacle function. Use the control obstacle function to constrain the output of the force controller to obtain a position loop saturation force controller with safety constraints. Step 5: Given a reasonable range of roll angle and pitch angle, establish a control obstacle function for the attitude loop, constrain the output of the torque controller through the control obstacle function, and obtain an attitude loop torque controller with safety constraints; Step 6: Generate a safe landing control output for the aircraft by solving the position loop saturation force controller and attitude loop torque controller established based on the motion constraints in steps 4 and 5. Drive the aircraft based on this output to ultimately achieve safe and precise landing control relative to the moving and shaking target platform.

2. The method for controlling the safe landing of an aircraft swaying target based on motion constraints according to claim 1, characterized in that: The steps for establishing the dynamic and kinematic model of the quadrotor aircraft in step 1 are as follows: Establishing an inertial coordinate system and the body coordinate system of the quadrotor The coordinate center of B is the geometric center point of the quadrotor; the rotation matrix R is introduced to describe the rotational motion. Where SO(3) represents the Lie group of third-order special orthogonal matrices, Represents m×n Euclidean space; I3 represents the third-order unit matrix; the body coordinate system Rotate to inertial coordinate system The corresponding rotation matrix is in, and Then, the position p and attitude γ of the aircraft are [φ,θ,ψ] T Expressed as: in, is the center of gravity of the quadrotor in the inertial coordinate system, is the linear velocity, g is the local acceleration of gravity, m is the mass of the quadrotor, is the thrust applied in the body coordinate system, is the unit vector in the Z direction in the inertial coordinate system, is the disturbance acceleration, is the inertia matrix, is the angular velocity, τ=[τ x ,τ y ,τ z ] T is the applied torque, is the disturbance torque, for x = [x1, x2, x3] T , uppercase × is defined as x × =[0,-x3,x2;x3,0,-x1;-x2,x1,0], W is defined as follows:

3. The method for controlling the safe landing of an aircraft swaying target based on motion constraints according to claim 2, characterized in that: The process of establishing continuous reference height in step 2 is as follows: According to the high-order polynomial function, the reference approach height Given by formula (7): where t a >0 is a pre-assigned time constant, h is a constant that satisfies h>sup(|p sz |) constant, a k is the constant coefficient determined by the following linear equation; t is time; p a (t) represents the reference approach altitude trajectory, which is expressed as: p z (t) is the vertical position; v z (t) velocity in the vertical direction; When the aircraft reaches the height h, it starts to track the platform. When the platform rises and swings at a large angle, the aircraft introduces and To ensure the aircraft lands safely when the platform descends and swings at a small angle; Among them, φ s is the roll angle of the target platform; θ s is the pitch angle of the target platform; b1 and b2 are positive real numbers; by choosing appropriate b1 and b2, the platform will turn to a relatively stable situation in a short time; p sz is the height of the platform's heaving motion plane; Satisfy at the same time and When the aircraft performs the final landing operation; the reference altitude is Switch to reference landing altitude in is the switching function to promote continuous switching, expressed as: where t s Represents the time to meet the safe landing conditions, n≥2 is a positive integer, t l >0 is the total time of the final landing operation; The overall reference altitude for a complete landing is given by By finding p zd For a time guide, the vertical reference velocity is:

4. The method for controlling the safe landing of an aircraft swaying target based on motion constraints according to claim 3, characterized in that: The process of establishing the position error dynamics in step 3 is as follows: According to the overall reference height p in step 2 zd and vertical reference velocity Let the reference position and reference velocity be p d =[p sx ,p sy ,p zd ] T and The position and velocity errors are defined as and According to equations (2) and (3), the position error dynamics is expressed as: in is the force controller, the equivalent disturbance acceleration γ c =[φ c ,θ c ,ψ c ] T are the command Euler angles, given by: In addition, in order to ensure that there is no singularity in attitude control, the following conditions should also be met: Thrust T = m‖u‖ is given by get.

5. The method for controlling the safe landing of an aircraft swaying target based on motion constraints according to claim 4, characterized in that: The establishment process of attitude error dynamics in step 3 is as follows: Define an auxiliary manifold: κ1>0 and is a constant; The attitude error dynamics is expressed as definition 6. The method for controlling the safe landing of an aircraft swaying target based on motion constraints according to claim 5, characterized in that: The method of position loop force controller in step 4 is as follows: In order to compensate for the unmeasurable dynamic characteristics, a dynamics estimator is introduced and the following auxiliary system is constructed: Given an initial value s υ (0) = 0 and s u (0)=0, where ∈ t >0 is a constant parameter; According to the position error dynamics in step 3, the nominal dynamic system is obtained: disturbed term By R(γ)→R(γ c )and Sure; Formula (22) is transformed into t / (s+∈ t ) is filtered, and combined with equations (19) and (20), we get: in is u a The estimated value of the system The output of ; Considering equations (19) and (23), the estimator is designed as: Define a new variable: Where Γ1=diag(γ 1x ,γ 1y ,γ 1z ), γ 1x ,γ 1y ,γ 1z is a positive constant; the force controller is established as: where Γ2 = diag(γ 2x ,γ 2y ,γ 2z ) is a positive definite matrix.

7. The method for controlling the safe landing of an aircraft swaying target based on motion constraints according to claim 6, characterized in that: The method for establishing a constraint surface based on relative height in step 4 is as follows: Design the following continuous smooth curve: h z =h2+s(h1-h2) (27) Among them, h1 and h2 are positive real numbers satisfying h1>h2, and the definition is Indicates that the function is continuously differentiable n times; It is expressed as follows: Among them, δ1 and δ2 are positive real numbers satisfying δ1>δ2, n≥2 is a positive integer; h z about The partial derivatives of are as follows: in, Therefore, the constraint surface based on relative height is: where k s is a positive real number.

8. The method for controlling the safe landing of an aircraft swaying target based on motion constraints according to claim 7, characterized in that: The method for establishing a position loop saturation force controller with safety constraints in step 4 is as follows: According to formula (31), the control barrier function is: Define the following function Where, α1(ψ0(t))=k1ψ0(t), α2(ψ1(t))=k2ψ1(t), k1 and k2 are positive real numbers; The position loop saturation force controller with safety constraints is established using equations (32) and (33): Among them, u iref The actual force control input for the safety of the aircraft; u i is the nominal force controller output; In formula (36), in, 9. The method for controlling the safe landing of an aircraft swaying target based on motion constraints according to claim 8, characterized in that: The process of establishing the attitude loop control obstacle function and the constrained attitude loop torque controller in step 5 is as follows: will u τ The dynamics estimator of is designed as follows: where ∈ r >0 is a constant parameter, z z and z τ Generated by the following auxiliary systems: Given an initial state s z (0) = 0 and s τ (0) = 0, then the attitude loop torque controller is designed as: Where κ2>0 is a constant; definition is the estimation error; In order for the aircraft to operate within a safe range of roll and pitch, it is necessary to ensure that the pitch or roll angle remains within the safe range (-π / 2,π / 2), so the constraint area is designed as Where v = φ, θ; the control barrier function is obtained as follows: Combined with formula (33), the attitude loop torque controller with safety constraints is established as follows: Among them, τ iref is the actual torque control input for safety of the aircraft; τ i is the nominal attitude torque controller output; In formulas (43)(44)(45)(46), k3 and k4 are positive real numbers. [ξ1 ξ2 ξ3] T =W -1 J -1 (47) [n1 n2 n3] T =W -1 J -1 (d τ -oh × (48) 10. The method for controlling the safe landing of an aircraft swaying target based on motion constraints according to claim 9, characterized in that: The process of establishing the safe landing control method based on relative motion constraints in step 6 is as follows: By solving the position loop saturation force controller and attitude loop torque controller established based on the motion constraints in steps 4 and 5, a safe aircraft landing control output is generated. The aircraft is driven based on this output, ultimately achieving safe and precise landing control relative to the moving and shaking target platform.

Images