A method for designing a controller of a quadrotor unmanned aerial vehicle based on an event-triggered mechanism when a moment of inertia matrix is unknown

Abstract

The present application relates to the field of unmanned aerial vehicle. The controller design method of quadrotor unmanned aerial vehicle based on event-triggered mechanism when inertia matrix is unknown, comprising the following steps: step 1, introducing rotation matrix to describe the attitude of quadrotor unmanned aerial vehicle; step 2, giving the attitude dynamics system of quadrotor unmanned aerial vehicle; step 3, defining angular velocity estimation value and inertia matrix estimation value dynamics equation; step 4, introducing event-triggered mechanism; step 5, designing system controller u. The present application introduces event-triggered mechanism without knowing the accurate inertia matrix of unmanned aerial vehicle in advance, and designs the controller which can save control cost effectively while ensuring the stability of the system.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicles (UAVs), and more specifically to a controller design method. Background Technology

[0002] Quadrotor drones, with their unique advantages such as simple structure, high maneuverability, and vertical takeoff and landing capabilities, have been increasingly widely used in military reconnaissance, agricultural plant protection, logistics distribution, aerial photography and surveying, and emergency rescue, demonstrating enormous development potential. As application scenarios continue to expand and become more complex, higher demands are being placed on the comprehensive flight capabilities and reliability of drones. Attitude control, as the foundation of drone flight control, directly determines the stability and mission execution capability of the drone.

[0003] However, attitude control research for quadcopter UAVs faces numerous challenges and difficulties. First, their dynamic models exhibit highly nonlinear and strongly coupled characteristics. More problematic is that, in practical applications, due to factors such as load variations, mechanical wear, or external disturbances, key system model parameters—such as the inertia matrix—are often difficult to obtain accurately. The unknown inertia matrix significantly reduces the performance of controllers designed based on accurate models, and can even lead to system instability. Therefore, designing a controller insensitive to the model's inertia matrix parameters when the inertia matrix is ​​unknown has become a critical technical challenge that urgently needs to be addressed.

[0004] On the other hand, at the engineering implementation level of the controller, the traditional time-triggered mechanism based on periodic sampling has inherent drawbacks. Under this mechanism, the controller needs to update at a fixed, sufficiently high frequency to ensure system stability. However, this "always-on" working mode means that even in a stable phase where the system state approaches equilibrium and frequent control is unnecessary, computation and communication continue. This results in the unnecessary consumption of significant computing resources, communication bandwidth, and the lifespan of sensors and actuators. For airborne embedded systems with extremely limited computing power, storage space, and battery energy, this waste of resources directly restricts the drone's endurance and ability to handle complex tasks.

[0005] Unlike time-triggered mechanisms, which operate on a periodic "time-driven" basis, event-triggered mechanisms employ a "state-driven" principle. They pre-define a trigger condition, and only when this condition is met are the controller driven to update and communicate. This "on-demand allocation" approach significantly reduces the number of controller updates while ensuring system stability, resulting in substantial savings in computing, communication, and energy resources.

[0006] In view of this, and addressing the practical engineering challenge of an unknown inertial matrix, and focusing on overcoming the resource efficiency bottleneck of traditional time-triggered control methods, this invention proposes a quadrotor UAV controller design method based on an event-triggered mechanism when the inertial matrix is ​​unknown. This method can achieve attitude stabilization control through an event-triggered controller without prior knowledge of the UAV's precise inertial matrix, realizing efficient and stable control of the UAV system in resource-constrained environments. This provides a feasible technical approach to broaden the application of quadrotor UAVs in complex real-world scenarios. Summary of the Invention

[0007] The purpose of this invention is to provide a design method for a quadrotor UAV controller based on an event-triggered mechanism when the inertial matrix is ​​unknown, so as to solve the practical engineering problem of unknown inertial matrix and the resource efficiency bottleneck caused by traditional time-triggered mechanism control methods.

[0008] The design method of a quadcopter UAV controller based on an event-triggered mechanism when the inertial matrix is ​​unknown is characterized by the fact that, without prior knowledge of the UAV's precise inertial matrix, the designed controller only triggers the calculation and update of the control task when the system state deviates from the expected trajectory to a certain preset threshold (i.e., trigger condition). This can effectively reduce the number of unnecessary updates to the controller while ensuring system stability, thereby significantly saving control costs such as computing, communication, and energy resources.

[0009] Specifically, the steps include the following:

[0010] Step 1: Introduce a rotation matrix to describe the attitude of the quadcopter drone:

[0011] Let SO(3) be the set of rotation matrices, then...

[0012] Step 2: Provide the attitude dynamics system of the quadcopter UAV:

[0013]

[0014]

[0015] in, The inertia matrix represents the positive definite inertia matrix. Indicates the system's control input, It is the angular velocity defined in a fixed coordinate system. Indicates system interference;

[0016] Step 3: Define the angular velocity estimate and the inertia matrix estimate, and the dynamic equations:

[0017] Define the dynamic equation for the angular velocity estimate:

[0018]

[0019] in, These represent the estimated values ​​of angular velocity and attitude, respectively. The error between the true and estimated attitude values;

[0020] Define the dynamic equation for the inertia matrix estimate:

[0021]

[0022] in, This represents the estimated value of the inertia matrix. This represents the error between the estimated and true values ​​of the inertia matrix, i.e.

[0023] Step 4: Introduce an event triggering mechanism:

[0024] Assume the trigger times are t1, t2, ..., t i ..., the system control input u satisfies:

[0025]

[0026] The trigger condition is defined as follows:

[0027]

[0028] Where, constant Used to represent the state of the system, here e R Let e ​​be the attitude error vector. Ω This is the angular velocity error vector. t∈[t i ,t i+1 () indicates error;

[0029] Step 5: Design the system controller u:

[0030]

[0031] in

[0032] Preferably, when there is no system interference, the constant c in the system controller u satisfies the following equation:

[0033]

[0034] Preferably, when dynamic disturbances exist, the constant c in the system controller u satisfies the following equation:

[0035]

[0036]

[0037] Beneficial effects: 1. The controller designed in this invention does not require prior acquisition of the precise inertial matrix parameters of the quadcopter UAV, thus effectively overcoming the influence of model parameter uncertainty and solving the practical engineering problem of unknown inertial matrix. 2. This invention introduces an event-triggered mechanism into the controller design, realizing "on-demand execution" of control tasks. This effectively reduces redundant updates to the controller, significantly saves control costs, improves resource utilization efficiency, and solves the resource efficiency bottleneck caused by traditional time-triggered control methods. Attached Figure Description

[0038] Figure 1 This is a schematic diagram of the steps of the present invention;

[0039] Figure 2 When there is no system interference, the attitude error vector e R Simulation diagram;

[0040] Figure 3 When there is no system interference, the angular velocity error vector e Ω Simulation diagram;

[0041] Figure 4 For the absence of system interference, the inertial error matrix Simulation diagram;

[0042] Figure 5 For the existence of dynamic disturbance Δ=3e A At that time, the attitude error vector e R Simulation diagram;

[0043] Figure 6 For the existence of dynamic disturbance Δ=3e A At that time, the angular velocity error vector e Ω Simulation diagram;

[0044] Figure 7 For the existence of dynamic disturbance Δ=3e A At that time, the inertial error matrix Simulation diagram;

[0045] Figure 8 When there is no system interference and no event triggering mechanism is introduced, the attitude error vector e R Simulation diagram;

[0046] Figure 9 When there is no system interference and no event triggering mechanism is introduced, the angular velocity error vector e Ω Simulation diagram;

[0047] Figure 10 The inertial error matrix is ​​calculated when there is no system interference and no event triggering mechanism is introduced. Simulation diagram;

[0048] Figure 11 For the existence of dynamic disturbance Δ=3e A Furthermore, without an event-triggered mechanism, the attitude error vector e R Simulation diagram;

[0049] Figure 12 For the existence of dynamic disturbance Δ=3e A Furthermore, without an event-triggered mechanism, the angular velocity error vector e Ω Simulation diagram;

[0050] Figure 13 For the existence of dynamic disturbance Δ=3e A Furthermore, without an event-triggered mechanism, the inertial error matrix The simulation diagram. Detailed Implementation

[0051] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below with reference to specific illustrations.

[0052] This invention relates to the following mathematical symbols:

[0053]

[0054]

[0055] Design method for a quadrotor UAV controller based on an event-triggered mechanism when the inertial matrix is ​​unknown, see [link to relevant documentation]. Figure 1 It includes the following steps:

[0056] Step 1: Introduce a rotation matrix to describe the attitude of the quadcopter UAV;

[0057] Step 2: Provide the attitude dynamics system of the quadcopter UAV;

[0058] Step 3: Define the angular velocity estimate and the inertia matrix estimate, and the dynamic equations.

[0059] Step 4: Introduce an event triggering mechanism;

[0060] Step 5: Design the system controller u.

[0061] The details are as follows:

[0062] Step 1. Introduce the rotation matrix R to describe the attitude of the quadcopter UAV.

[0063] This invention selects rotation matrices to describe the attitude information of a quadcopter UAV. Let SO(3) be the set of rotation matrices, then...

[0064] For any aircraft attitude, there can be a uniquely determined rotation matrix R∈SO(3) to represent it.

[0065] Step 1.1. Define two coordinate systems: a fixed coordinate system and an inertial coordinate system, both of which satisfy the right-hand rule.

[0066] A fixed coordinate system is usually established on the ground and serves as the unified reference coordinate system for the entire system. An inertial coordinate system is often used to describe the position and attitude of an object in a fixed coordinate system. By rotating one coordinate system by a certain angle, it can coincide with another coordinate system. Common rotation angles include roll angle φ, pitch angle θ, and yaw angle ψ. That is, starting from the fixed coordinate system, rotating ψ around the Z-axis, rotating θ around the Y-axis, and finally rotating φ around the X-axis will result in coinciding with the inertial coordinate system.

[0067] Step 1.2. Let the fixed coordinate system be represented as... Inertial coordinate system is represented as The rotation matrix from the fixed coordinate system to the inertial coordinate system is expressed as: satisfy:

[0068]

[0069] Where R φ R θ and R ψ The rotation matrices representing the roll angle, pitch angle, and yaw angle are shown below:

[0070]

[0071] Then the rotation matrix The expression can be simplified as follows:

[0072]

[0073] For ease of understanding, the rotation matrix will be discussed later. Abbreviated as R, it satisfies the following equation:

[0074] RR T =R T R = I 3×3 ,

[0075] det(R) = 1

[0076] Let SO(3) be the set of rotation matrices, i.e.

[0077] For any aircraft attitude, there can be a uniquely determined rotation matrix R∈SO(3) to represent it.

[0078] Step 2. Give the attitude dynamics system of the quadcopter UAV.

[0079] Using Newton's second law of motion and Euler's equations, the attitude dynamics system of the quadcopter UAV is obtained:

[0080]

[0081]

[0082] in, The inertia matrix represents the positive definite inertia matrix. Indicates the system's control input, It is the angular velocity defined in a fixed coordinate system. This indicates system interference.

[0083] Step 2.1. Make the following assumptions:

[0084] 1) Ignoring factors such as the Earth's curvature and rotation, assume that the quadcopter drone moves within a plane;

[0085] 2) Treat the quadcopter drone as a rigid body, that is, do not consider its shape, size, mass, and deformation during motion;

[0086] 3) The origin of the inertial coordinate system coincides with that of the fixed coordinate system and is located at the center of the rigid body;

[0087] 4) The principal axes of the inertial coordinate system and the fixed coordinate system coincide, and the direction of the coordinate axes is determined by the task currently being performed by the rigid body.

[0088] Step 2.2. Introduce the attitude dynamics system of the quadcopter UAV.

[0089] When disturbances are present, according to Newton's second law of motion and Euler's equations, the attitude dynamics system of the quadcopter UAV is as follows:

[0090]

[0091]

[0092] in, The inertia matrix represents the positive definite inertia matrix. Indicates the system's control input, It is defined as the angular velocity in a fixed coordinate system. This represents system interference. Mapping (.)^: Transform the three-dimensional vector into a 3×3 skew-symmetric matrix for any vector x = (x1, x2, x3). Satisfy the following formula:

[0093]

[0094] At the same time, define the mapping (.). ∧ The inverse mapping of is (.). ∨ :

[0095] Step 3. Define the angular velocity estimate and the inertia matrix estimate, and the dynamic equation.

[0096] The dynamic equation for the estimated angular velocity is defined as follows:

[0097]

[0098] in, These represent the estimated values ​​of angular velocity and attitude, respectively. The error between the true and estimated attitude values;

[0099] The dynamic equation for the estimated inertia matrix is ​​defined as follows:

[0100]

[0101] in, This represents the estimated value of the inertia matrix. This represents the error between the estimated value and the true value of the inertia matrix, i.e.:

[0102] Step 3.1. Design an angular velocity observer.

[0103] Design an angular velocity observer to obtain estimates of angular velocity and attitude, let Let represent the estimated values ​​of angular velocity and attitude, respectively. The estimated values ​​of angular velocity and attitude also satisfy the rigid body dynamics equations, i.e.:

[0104]

[0105] Since the magnitude of the control input is usually finite, it can be assumed that the angular velocity and its first derivative are bounded, i.e., that there exist positive values. Make:

[0106]

[0107] Step 3.2. Construct the error model.

[0108] The inertia error matrix is ​​defined as the error between the estimated value and the true value of the inertia matrix, denoted as . Right now:

[0109]

[0110] Let the error between the true and estimated attitude values ​​be Q∈SO(3), that is... when When, Q = I 3×3 .

[0111] Define the attitude error function Ψ and the attitude error vector e. R Angular velocity error vector e Ω They are respectively:

[0112]

[0113] Where G = diag(g1, g2, g3), It is a positive real number.

[0114] Since angular velocity and its first derivative are bounded, it is easy to obtain

[0115] Attitude error function Ψ, attitude error vector e R and angular velocity error vector e Ω It has the following properties:

[0116] Property 1:

[0117] Property 2:

[0118] Property 3: in, satisfy:

[0119] Property 4: in

[0120] Property 5: Suppose there exist Ψ≤ψ≤n1 such that:

[0121] Where, n1 = min{g1 + g2, g2 + g3, g3 + g1}, n2 = max{(g1 - g2)} 2 (g2-g3) 2 (g3-g1) 2},

[0122] n3 = max{(g1 + g2)} 2 (g2+g3) 2 (g3+g1) 2},n4=max{g1+g2,g2+g3,g3+g1},

[0123] n5=min{(g1+g2) 2 (g2+g3) 2 (g3+g1) 2}

[0124] Property 6: ||e R ||≤B2, where,

[0125] Attitude stability control aims to make the attitude error function, angular velocity error function, and inertia error function approach zero, that is:

[0126]

[0127] Step 3.3. Define the angular velocity estimate and the inertia matrix estimate, and the dynamic equation.

[0128] The dynamic equation for the estimated angular velocity is defined as follows:

[0129]

[0130] in, These represent the estimated values ​​of angular velocity and attitude, respectively. The error between the true and estimated attitude values;

[0131] The dynamic equation for the estimated inertia matrix is ​​defined as follows:

[0132]

[0133] in, This represents the estimated value of the inertia matrix. This represents the error between the estimated value and the true value of the inertia matrix, i.e.:

[0134] Step 4. Introduce an event triggering mechanism.

[0135] Unlike time-triggered mechanisms, which operate on a periodic "time-driven" basis, event-triggered mechanisms employ a "state-driven" principle. They pre-define a trigger condition, and only when this condition is met are the controller driven to update and communicate. This "on-demand allocation" approach effectively reduces unnecessary controller updates while ensuring system stability, resulting in significant savings in computing, communication, and energy resources.

[0136] Assume the trigger times are t1, t2, ..., t i ..., the system control input u satisfies:

[0137]

[0138] The trigger condition is defined as follows:

[0139]

[0140] Here, constant s = e Ω +βe R , Used to describe the system state; t∈[t i ,t i+1 ), used to represent error.

[0141] Step 5. Design the system controller u.

[0142] For systems (1) and (2), under the angular velocity estimate and the inertia matrix estimate, the dynamic equations (3) and (4) are used to design the system controller u as follows:

[0143]

[0144] in k,k R ,k Ω ,δ,

[0145] Case 1. In the absence of system interference, set the conditions that the system controller u must satisfy so that the attitude error vector e R Angular velocity error vector e Ω Inertia matrix and error matrix All of them are exponentially stable, which in turn ensures the stability of the quadcopter UAV attitude dynamics system (1)-(2), thereby completing the stable control of the quadcopter UAV attitude.

[0146] Assume that the attitude control system of the quadcopter UAV is in an ideal state with no external interference, i.e., Δ = 0.

[0147] Preferably, when there is no system interference, the constant c in the system controller u satisfies the following equation (*).

[0148]

[0149] It can be proven that the attitude error vector e R Angular velocity error vector e Ω Inertia error matrix It is exponentially stable, thereby ensuring the stability of the quadcopter UAV attitude dynamics system (1)-(2) and completing the stable control of the quadcopter UAV attitude.

[0150] The proof is as follows:

[0151] Consider Lyapunov functions

[0152] From matrix theory, we can obtain

[0153]

[0154] cλ m ||e Ω ||≤ce Ω T Je R ≤cλ M ||e Ω ||

[0155] From property 5, we can obtain:

[0156]

[0157] Therefore

[0158]

[0159] That is, satisfying:

[0160] ξ T W1ξ≤V≤ξ T W2ξ·

[0161] in, matrix They are respectively:

[0162]

[0163] From Δ=0, we get:

[0164]

[0165] Taking the first derivative of the Lyapunov function V, we get:

[0166]

[0167] Proof: Similarly, we can conclude that:

[0168] because Depend on The definitions are as follows:

[0169]

[0170] but:

[0171]

[0172] because We can obtain:

[0173]

[0174] Therefore: From the definitions of s and e, we get:

[0175] s(t i ) = e + s = e + e Ω +βe R

[0176] but:

[0177]

[0178] And because The above expression can then be simplified to:

[0179]

[0180] Substituting the above result into the first derivative of the Lyapunov function V, we get:

[0181]

[0182] Among them, matrix W3 satisfies the following equation:

[0183]

[0184] In this step, given the constant c, matrices W1, W2, and W3 are positive definite. This can be proven using Lyapunov stability theory, which demonstrates that the attitude error vector e... R Angular velocity error vector e Ω Inertia error matrix It is exponentially stable, indicating that the controller u effectively ensures the stability of the quadcopter UAV attitude dynamics system (1)-(2), thereby completing the stable control of the quadcopter UAV attitude.

[0185] Case 2. When dynamic disturbances exist, set conditions that the system controller u must satisfy to ensure that the attitude error vector e R Angular velocity error vector e Ω Inertia matrix and error matrix All of them are exponentially stable, thus ensuring the stability of the quadcopter UAV attitude dynamics system (1)-(2) and completing the stable control of the quadcopter UAV attitude.

[0186] Assume the attitude control system of a quadcopter UAV is under dynamic disturbance. Define vectors. in For system disturbances, we assume that they are related to vector e. A Linear dependence, that is: there exists a positive number k such that Δ = keA .

[0187] When the constant c in the system controller u satisfies the following equation (**),

[0188]

[0189] It can be proven that the attitude error vector e R Angular velocity error vector e Ω Inertia error matrix It is exponentially stable, thereby ensuring the stability of the quadcopter UAV attitude dynamics system (1)-(2) and completing the stable control of the quadcopter UAV attitude.

[0190] The proof is as follows:

[0191] Consider Lyapunov functions

[0192] The proof method is the same as for case 1, and we have:

[0193] ξ T W1ξ≤V≤ξ T W2ξ·

[0194] in, matrix They are respectively:

[0195]

[0196] From property 4, we get:

[0197]

[0198] Taking the first derivative of the Lyapunov function, we get:

[0199]

[0200]

[0201] The proof method is the same as for case 1, and we have:

[0202]

[0203]

[0204] And Δ=ke A ,have:

[0205]

[0206] Substitute the above results into have to:

[0207]

[0208] in,

[0209]

[0210] In this step, given the constant c, matrices W1, W2, and W4 are positive definite. This can be proven using Lyapunov stability theory, which demonstrates that the attitude error vector e... R Angular velocity error vector e Ω Inertia error matrix It is exponentially stable, indicating that the controller u effectively ensures the stability of the quadcopter UAV attitude dynamics system (1)-(2), thereby completing the stable control of the quadcopter UAV attitude.

[0211] Simulation Experiment

[0212] A simulation experiment was conducted to test the stability of the attitude system of a quadrotor UAV under an event-triggered mechanism when the inertial matrix is ​​unknown. The performance of the proposed controller was intuitively demonstrated through numerical simulation.

[0213] The parameter values ​​are: R(0) = I 3×3 , J(0)=diag(1,2,3), G=diag(1,1,1), Ω=(1,3,2),k R =k Ω =5,k J =0.01, ε=3, δ=0.3, σ=0.05, c=2. In the simulation diagram below, the red, blue, and green lines represent the first, second, and third quantities of the error vector, respectively, for example: e Ω The i-th quantity.

[0214] Scenario 1. Simulation experiment on the attitude stability of a quadcopter UAV under no system interference. k=3, β=1, simulation graph as follows Figure 2 , Figure 3 , Figure 4 As shown.

[0215] From the figure, we can see the attitude error vector e R Angular velocity error vector e Ω Inertia error matrix Both exponents approach 0, demonstrating that our controller u can still effectively ensure system stability even without knowing the precise inertial matrix of the UAV, thus enabling the control of the quadcopter UAV's attitude.

[0216] Case 2. Simulation experiment on the attitude stability of a quadcopter UAV under dynamic disturbance. Take Δ = 3e A , k=3, β=1, simulation graph as follows Figure 5 , Figure 6 , Figure 7 As shown.

[0217] From the figure, we can see the attitude error vector e R Angular velocity error vector e Ω Inertia error matrix The fact that the exponents are all close to 0 demonstrates that our controller UU can still effectively ensure the stability of the system even without knowing the precise inertial matrix of the UAV, thus enabling the control of the attitude of the quadcopter UAV.

[0218] Analyze the impact of event triggering mechanisms on system convergence speed:

[0219] First, the stability of the attitude of the quadcopter UAV without an event triggering mechanism is simulated. To intuitively demonstrate the impact of the event triggering mechanism on the system convergence speed, the parameters are selected as above.

[0220] ① The attitude stability of a quadcopter UAV without system interference and without an event-triggered mechanism is shown in the simulation diagram. Figure 8 , Figure 9 , Figure 10 As shown.

[0221] ② There is dynamic interference Δ=3e A Simulation results show the attitude stability of a quadcopter drone without an event-triggered mechanism, as illustrated in the figure. Figure 11 , Figure 12 , Figure 13 As shown.

[0222] Secondly, the impact of the event triggering mechanism on the system convergence speed is analyzed through simulation diagrams.

[0223] Case 1. In the absence of system interference, compare the simulation diagram without introducing an event triggering mechanism. Figure 8 , Figure 9 , Figure 10 ) and simulation diagrams introducing event triggering mechanisms ( Figure 2 , Figure 3 , Figure 4 Six simulation images show that the angular velocity error vector e Ω Attitude error vector e R and inertia matrix error matrix Ultimately, all values ​​tend to zero, thus ensuring system stability. However, comparing these six graphs reveals that the system under the controller without an event-triggered mechanism stabilizes quickly (around 200 seconds), while the system under the controller with an event-triggered mechanism stabilizes around 450 seconds.

[0224] Case 2. When dynamic disturbances exist, with Δ = 3e A For example, compare the simulation diagram without the introduction of an event triggering mechanism ( Figure 11 , Figure 12 , Figure 13 ) and simulation diagrams introducing event triggering mechanisms ( Figure 2 , Figure 3 , Figure 4 Six simulation images show that the angular velocity error vector e Ω Attitude error vector e R and inertia matrix error matrix Ultimately, all values ​​tend to zero, thus ensuring system stability. However, comparing these six graphs reveals that the system under a controller without an event-triggered mechanism stabilizes quickly (around 150 seconds), while the system under a controller with an event-triggered mechanism stabilizes around 500 seconds.

[0225] This demonstrates that introducing an event-triggered mechanism into the attitude control system of a quadcopter drone can effectively reduce system power consumption, but it also prolongs the time it takes for the system to reach stability. Therefore, in practical applications, it is necessary to determine, based on different mission requirements, whether to continuously update system signals, use an event-triggered mechanism to control signal updates, or adjust the event triggering conditions within the event-triggered mechanism.

[0226] The foregoing has shown and described the basic principles and main features of the present invention, as well as its advantages. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention.

Claims

1. A design method for a quadrotor UAV controller based on an event-triggered mechanism when the inertial matrix is ​​unknown, characterized in that: Without needing prior knowledge of the UAV's precise inertial matrix, the designed controller only triggers control task calculations and updates when certain conditions are met. This effectively reduces unnecessary controller updates while ensuring system stability, significantly saving control costs such as computation, communication, and energy resources. The process includes the following steps: Step 1: Introduce a rotation matrix to describe the attitude of the quadcopter drone: Let SO(3) be the set of rotation matrices, then... Step 2: Provide the attitude dynamics system of the quadcopter UAV: in, The inertia matrix represents the positive definite inertia matrix. Indicates the system's control input, It is the angular velocity defined in a fixed coordinate system. Indicates system interference; Step 3. Define the angular velocity estimate and the inertia matrix estimate, and the dynamic equations: The dynamic equation for the estimated angular velocity is defined as follows: in, These represent the estimated values ​​of angular velocity and attitude, respectively. The error between the true and estimated attitude values; The dynamic equation for the estimated inertia matrix is ​​defined as follows: in, This represents the estimated value of the inertia matrix. This represents the error between the estimated and true values ​​of the inertia matrix, i.e. Step 4: Introduce an event triggering mechanism: Assume the trigger times are t1, t2, ..., t i ..., the system control input u satisfies: The trigger condition is defined as follows: Here, constant s = e Ω +βe R , Used to describe the system state; Used to represent error. Step 5: Design the system controller u: in 2. According to the controller design method of claim 1, when there is no system interference, the constant c in the system controller u satisfies the following formula:

3. According to the controller design method of claim 1, when dynamic disturbances exist, the constant c in the system controller u satisfies the following formula:

Images