Unmanned rotorcraft anti-interference safety control method based on obstacle avoidance strategy

Abstract

A rotor unmanned aerial vehicle anti-interference safety control method based on an obstacle avoidance strategy comprises the following steps that a rotor unmanned aerial vehicle attitude and height nonlinear system model is constructed by considering constant attitude angle and height constraints and unknown external interference; designing an obstacle avoidance algorithm according to the defined upper and lower bound values of the constraint system and the output value of the constructed model, obtaining a safety expectation tracking signal, obtaining a zero-order derivative through the safety expectation tracking signal, and obtaining a converted switching error system with jump by combining an error with a rotor unmanned aerial vehicle attitude and height nonlinear system model; designing a robust switching tracking controller based on an obstacle avoidance algorithm by combining a disturbance observer, a filter and a backstepping method based on the constructed switching error system; and the anti-interference safety control of the unmanned rotorcraft can be completed based on the robust switching tracking controller. The problem of limitation in the flight process can be solved under the condition that the influence of constant attitude angle, height constraint and unknown external interference is considered.

Description

Technical Field

[0001] The invention relates to the technical field of unmanned aerial vehicles, and in particular to an anti-interference safety control method for a rotary-wing unmanned aerial vehicle based on an obstacle avoidance strategy. Background Art

[0002] In recent years, rotary-wing UAVs have achieved remarkable performance in high-altitude photography, waterway inspection, earthquake prevention and disaster relief, epidemic prevention and disinfection, and fixed-point material delivery, relying on their advantages such as fixed-point hovering, vertical take-off and landing, wide application environment, and flight in any direction. Rotary-wing UAV systems have the characteristics of high nonlinearity, strong coupling, time variation, and open-loop instability. It is very difficult to achieve the flight goals of rotary-wing UAV systems such as fixed-point hovering, forward and backward flight, and autonomous take-off and landing. However, since rotary-wing UAVs have extremely high application value in many fields in the real world, the development of autonomous flight control technology for rotary-wing UAVs is of far-reaching significance. There is no doubt that advanced autonomous flight control technology is the core of solving the above problems. During the flight of a rotorcraft UAV, factors such as engine jitter and changes in the external environment (such as gusts of wind) may cause problems with the stability of the system. Especially when the flight control system is running, if the impact of system uncertainty and external disturbances is not considered, the system performance may deteriorate. In addition, there must be various safety protection mechanisms during the flight of a rotorcraft UAV, such as attitude restrictions and altitude constraints. Therefore, the designed controller must ensure that the system state quantity always remains within the safety boundary in order to complete the preset tasks safely and effectively. If the restricted problems during the flight of the UAV cannot be solved in time, it may lead to irreversible losses such as system loss of control and crash. Summary of the invention

[0003] In order to solve the above technical problems, the present invention provides an anti-interference safety control method for a rotorcraft UAV based on an obstacle avoidance strategy, which has the advantages of enabling the rotorcraft UAV to complete tracking of preset expected tracking signals while ensuring system safety and stability, taking into account constant attitude angle and height constraints and the influence of unknown external interference, and on this basis, it can solve the problem of limitations during flight.

[0004] To achieve the above technical purpose, the adopted technical solution is: a rotor UAV anti-interference safety control method based on obstacle avoidance strategy, comprising the following steps: Step 1: Considering the constant attitude angle and altitude constraints and unknown external disturbances, a nonlinear system model of the attitude and altitude of the rotorcraft is constructed. in: H is the attitude rotation matrix, x1=[Ω T ,Z] T ,x2=[ωT ,w] T ,Ω=[φ,θ,ψ] T , φ, θ, ψ are the roll angle, pitch angle and yaw angle of the drone under the body axis respectively, ω=[p,q,r] T , p, q, r are the roll angular velocity, pitch angular velocity and yaw angular velocity respectively, Z and w represent the vertical height and vertical velocity of the rotorcraft in the ground coordinate system respectively, u=[Σ T ,T mr ] T , J is the moment of inertia matrix, d Σ ,d T is the unknown disturbance vector under attitude angle and attitude angular velocity; Step 2: Ensure that the roll angle φ, pitch angle θ, yaw angle ψ and vertical height Z correspond to the expected tracking signal φ d ,θ d ,ψ d ,Z d and its first-order piecewise derivative is bounded. According to the upper and lower bounds of the defined constraint system and the output value y constructed in step 1, the obstacle avoidance algorithm is designed to obtain the safety expected tracking signal φ corresponding to the roll angle φ, pitch angle θ, yaw angle ψ and vertical height Z. s ,θ s ,ψ s ,Z s , through the safety expectation tracking signal φ s ,θ s ,ψ s ,Z s Get its zero-order derivative Defining vectors Calculate x1 and y σ,s The error e σ,1 , x2 and virtual control law α σ The error e σ,2 , error e σ,1 、e σ,2 Combining the attitude of the rotorcraft UAV with the highly nonlinear system model, the converted switching error system with jumps is obtained; Step 3: Based on the switching error system constructed in step 2, a robust switching tracking controller based on the obstacle avoidance algorithm is designed by combining the disturbance observer, filter and backstepping method. in, is the positive definite matrix to be designed for the kth switching subsystem, The filter input signal λ k The time derivative of is a positive definite matrix, and the disturbance observer is is an auxiliary variable, Γ is a positive definite matrix; Step 4: Based on the robust switching tracking controller, the anti-interference safety control of the rotorcraft UAV can be completed.

[0005] The obstacle avoidance algorithm has three situations. Situation 1 is when the expected tracking signal has crossed the boundary, a fixed slope is used to pull the signal back into the safety boundary to generate a safe expected tracking signal; Situation 3 is when the expected tracking signal predicts an out-of-bounds behavior, an imaginary obstacle circle is defined, and a safe expected tracking signal is constructed along the tangent direction; Situation 3 is when the expected tracking signal is tracked normally, then the expected tracking signal is used as a safe expected tracking signal.

[0006] The converted switching error system with jump is expressed as follows: in, is the vector y σ,s The time derivative of is the virtual control law α σ The time derivative of σ It is the control input under the switching system characteristics.

[0007] The beneficial effects of the present invention are: 1. The present invention designs an obstacle avoidance algorithm with the help of the real-time output of the rotor UAV system and the constant attitude angle and height constraints, and converts the tracking error system into a switching system with a jump state. At the same time, an interference observer is used for estimation of unknown external interference to construct an obstacle avoidance switching tracking controller to solve the problem of safe flight control of rotor UAVs with constant attitude angle and height constraints and external interference. It has been verified that the obstacle avoidance switching tracking controller designed by the present invention can complete the tracking of the preset expected tracking signal on the basis of ensuring the safety and stability of the system.

[0008] 2. The present invention simultaneously considers constant attitude angle and altitude constraints as well as unknown external interference, and constructs a nonlinear system model of attitude and altitude of a rotorcraft UAV that can be used for safe tracking control; the present invention designs a class of obstacle avoidance algorithms with the help of the real-time output of the rotorcraft UAV system and the constant attitude angle and altitude constraints, and converts the tracking error system into a switching system with a jump state, thereby realizing the construction of an unconstrained error system and solving the problem of piecewise differentiability of the safe expected tracking signal; the present invention adopts the interference observer method to approximate the unknown external interference, and constructs an obstacle avoidance switching tracking controller based on the backstepping method to realize the safe flight intention of the rotorcraft UAV system. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 is a system control flow chart of the present invention; Figure 2 This is a classification diagram of the out-of-bounds situation of the obstacle avoidance algorithm of the present invention. DETAILED DESCRIPTION

[0010] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0011] See also Figure 1 , Figure 2 , an anti-interference safety control method for a rotary-wing UAV based on an obstacle avoidance strategy, comprising the following steps: Step 1: Considering constant attitude angle and altitude constraints as well as unknown external disturbances, a nonlinear system model of the rotorcraft attitude and altitude is constructed.

[0012] The process of establishing the nonlinear system model of the attitude and height of the rotary-wing UAV in step 1 is as follows: a. Nonlinear system model of the attitude and height of the rotary-wing UAV Considering the constant attitude angle and height constraints and unknown external disturbances, combined with the aerodynamics and kinematics of the rotorcraft, the attitude and height nonlinear model is described as: Where Ω=[φ,θ,ψ] T and ω=[p,q,r] T are the attitude angle vector and attitude angular velocity vector of the rotorcraft in the body coordinate system, φ, θ, ψ are the roll angle, pitch angle and yaw angle of the drone under the body axis, p, q, r are the roll angular velocity, pitch angular velocity and yaw angular velocity, respectively; Z and w represent the vertical height and vertical velocity of the rotorcraft in the ground coordinate system, respectively; m0 and g are the mass and gravity acceleration of the rotorcraft, respectively; d Σ ,d T is the unknown disturbance vector under attitude angle and attitude angular velocity, T mr and Σ are the main rotor thrust and the moment under the fuselage axis, Σ=[Σ x ,Σ y ,Σ z ] T With T mr They represent the resultant moment of the helicopter in the three directions of the body axis and the main rotor thrust, J is the moment of inertia matrix and the attitude rotation matrix is Then system (1) can be restated as the following second-order system: in, x1=[Ω T ,Z] T ,x2=[ω T ,w] T ,d=[d Σ T ,d T ] T , u=[Σ T ,T mr ] T ,

[0013] Before controller design, the following assumptions and lemmas are required: Assumption 1: External interference vector The derivative of is bounded, that is, there exists a positive constant β such that the time derivative of the disturbance vector d is The inequality holds:

[0014] Assumption 2: The expected tracking signal y d =[φ d ,θ d ,ψ d ,Z d ] T and its r-order derivative are bounded (r = 1, 2), that is, there exists a positive constant δ j (j=1,...,4) so ​​that the following inequality always holds:

[0015] Assumption 3: The roll angle and pitch angle of the rotorcraft always remain within a reasonable range (-π / 2, π / 2).

[0016] Lemma 1: For any time t ≥ 0, consider the following nonlinear switching system with jumps Among them, t - is the instant before t, is the system state under mode σ(t), represents the switching function, Fk and H k,l (k,l∈Q) is a piecewise smooth function, Represents x σ(t) (t), Δx(t) represents the right continuous derivative of x σ(t) (t) At the switching time t j The increment of , assuming that there exists the following function χ1(·), χ2(·)∈κ∞ ,a0>1,d0>0,ρ>0,C>0 and positive definite function V k (x k ),V l (x k )∈C 1 Make χ1(||x k ||)≤V k (x k )≤χ2(||x k ||) (6) Where k,l∈Q,z σ(t) ≠0, then the switching system (5) satisfies the average residence time τ a >(lna0+d0) / ρ, for all switching signals σ(t) is eventually uniformly bounded, Indicates V k (x k ) time derivative. b. Attitude and altitude constraints and external disturbance description

[0017] Considering that rotorcraft drones are often subject to altitude constraints and attitude angle constraints, the upper and lower bounds of the system are defined as y up =[φ up ,θ up ,ψ up ,Z up ] T With y low =[φ low ,θ low ,ψ low ,Z low ] T , due to the unpredictability of the constraints, the preset expected tracking signal y d =[φ d ,θ d ,ψ d ,Z d ] T There is a possibility that the boundary value may be exceeded, at which point the rotorcraft control will give priority to its safety issues.

[0018] In addition, during the flight of the rotorcraft, the existence of factors such as engine jitter and changes in the external environment (such as gust interference) will cause problems to the stability of the system. The unknown external interference signal d(t) satisfies the norm bounded condition, that is, there is a positive constant β that makes the following inequality hold

[0019] Step 2: Using constraint information and real-time system output, design a class of obstacle avoidance algorithms to obtain piecewise differentiable safety expectation tracking signals; and combined with the system model constructed in step 1, convert the tracking error system into a switching system with jump dynamics; the process of constructing a new safety expectation tracking signal in step 2 is as follows: a. Without loss of generality, for any time t≥0, taking the roll angle φ(t) (i.e. φ) as an example, on the basis of assuming that φ(t) still changes at the current rate, its predicted value at time (t+Δt) can be obtained as: It can be obtained by a first-order differentiator of the following form Among them, Δt s is the system sampling time.

[0020] In order to ensure the boundedness of the designed desired tracking signal and its first and second order piecewise derivatives, the maximum prediction time Δt to be designed is defined. max and the minimum prediction time Δt min , and there is Δt max >Δt min >Δt s >0, from the above definition, when the following conditions are met, the rotor UAV system will predict the occurrence of the roll angle signal out-of-bounds behavior, that is, φ(t)∈(φ low ,φ up )and According to φ(t) and The value of, three different cases are given below; b. By defining the maximum prediction time Δt to be designed max and the minimum prediction time Δt min , according to φ(t) and The obstacle avoidance algorithm is divided into the following three different cases: Scenario 1: (The system has crossed the line); This situation indicates that the system attitude angle has exceeded the safety boundary. In order to pull the attitude angle signal back to the safety boundary, the algorithm will generate a safety expectation tracking signal with a fixed slope, φ s (t) is the safety expectation tracking signal, and its time derivative information is shown as follows: Among them, c1<0 and c2>0 are constants to be designed; Case 2: φ(t)∈(φ low ,φ up )and (anticipating out-of-bounds behavior); In this case, the system roll angle itself has not crossed the boundary, but has a tendency to cross the boundary. The goal of the algorithm is to avoid the occurrence of crossing the boundary. Therefore, the algorithm defines a circle with a center of (t+Δt φ ,φ(t+Δt φ )), with a radius of Δt φ The imaginary obstacle circle is constructed along the tangent direction to construct the safety expectation tracking signal. For the convenience of description, let Then the radius of the imaginary obstacle circle is Δt φ The definition is as follows: To find the tangent point of the imaginary obstacle circle (t+Δt φcom ,φ com ), can be divided into four cases according to geometric relationships, such as Figure 2 shown.

[0021] according to Figure 2 , according to geometric theorems and trigonometric relationships, we can obtain the coordinates of the tangent point in case 2 (t+Δt φcom ,φ com ), from which the safety expectation tracking signal φ can be obtained s The time derivative information expression of (t) is as follows Case 3: φ(t)∈(φ low ,φ up )and (Normal tracking) In this case, the system attitude angle itself has not crossed the boundary, and the predicted signal has not crossed the boundary either. It can be considered that the system attitude angle is far from the boundary. In this case, the algorithm will maintain the original expected tracking signal φ d (t) remains unchanged, is φ d The time derivative of (t) is But according to Figure 2 In the cases (b) and (d) of Figure 1, the safety expectation tracking signal will satisfy the triggering condition of case 3 under the premise of case 2. At this time, the algorithm will frequently switch between cases 2 and 3, resulting in poor constraint effect and increased computational burden. To avoid this phenomenon, the algorithm will add an item under the condition of case 3: when |φ d (t)|≥min{|φ low (t)|,|φ up (t)|}, take Then there is Summarizing cases 1-3, the first-order and second-order time-dependent piecewise derivative expressions of the safety expectation tracking signal are as follows: Then the safety expectation tracking signal φ s The expression of (t) is: Similarly, for the safety expectation tracking signal θ s The expression of (t) is: For the safety expectation tracking signal ψ s The expression of (t) is: For safety expectations tracking signal Z s The expression of (t) is: c. Switching system establishment with jump state Similarly, for the pitch angle θ, yaw angle ψ and vertical height Z of the rotor UAV system, repeating the above signal reconstruction process for the attitude angle φ, the safety expectation tracking signal and its r-order derivative in the same form as equations (16)-(18) can be obtained respectively: as well as On this basis, we define the vector is φ s The zero-order derivative of (t), for The time derivative of for The time derivative of Then construct the switching law σ(y(t),t) of the whole system, and the number of modes is M=6 4 ,Right now,

[0022] d. Define the attitude angle and vertical height tracking error with jump, as well as the attitude angular velocity and vertical velocity tracking error as follows: e σ,1 =x1-y σ,s (19) e σ,2 =x2-α σ (20) Among them, α σ is the virtual control law under the σth switching subsystem.

[0023] Combining errors (19), (20) with system (3), the converted switching error system with jump is expressed as follows: in, is the switching law; u σ is the control input under the switching system characteristics; through model conversion, the safety tracking control problem of the attitude and altitude system (3) of the rotorcraft UAV is converted into the stabilization control problem of the switching error system (21) with jumps.

[0024] According to the definition of the switching law σ(y(t),t) in the previous section, the switching interval always satisfies t j -t j-1 =n·Δt min ≥Δt min (j,n=1,2,...), then the average residence time τ of system (21) a Must satisfy τ a ≥t j -t j-1 ≥Δt min (twenty two)

[0025] Step 3: Based on the switching error system constructed in step 2, combined with the disturbance observer and the backstepping method, a robust switching tracking controller based on the obstacle avoidance algorithm is designed; the controller design process of step 3 is as follows: a. Assume that for the kth switching subsystem, a virtual control law α is designed k as follows: in, is a positive definite matrix; to compensate for unknown interference, the interference observer is designed as follows: in, is an auxiliary variable, Γ is a positive definite matrix, and the error variable is defined Then there is b. To avoid the virtual control law α k Directly derivate , and pass it through a filter of the following form: Among them, λ k is the filter output signal, is k Time derivative, τ k is a positive definite diagonal matrix, and defines the error variable Then there is in, is a continuous function vector, and under the given initial value, it satisfies ||B k (·)||B k .

[0026] Combining the switching system (21), the disturbance observer (24) and the filter (26), the switching controller u is designed. k as follows: in, is a positive definite matrix, is the positive definite matrix to be designed for the kth switching subsystem.

[0027] The present invention uses obstacle avoidance algorithm and switching system control strategy to study the safe tracking control problem of a rotorcraft UAV system with attitude and altitude constraints and unknown external interference. The control goal is to design an obstacle avoidance switching tracking controller to ensure that the closed-loop system (1) satisfies the following properties: a. Convergence: Under zero initial conditions, system (1) is uniformly bounded; b. Security: At any time, the system (1) satisfies the following conditions in,

[0028] After the above analysis and discussion, the following conclusions are drawn: Conclusion 1: For the kth switching subsystem, considering the switching error system with jump (21), the disturbance observer is designed as (24), and the control input is designed as (23) and (28) respectively. Under the condition that the initial value is bounded and the inequality ρ>0 holds, the closed-loop system error signal under the kth switching subsystem is bounded. Proof: For the kth subsystem, take the Lyapunov function as V k Taking the derivative with respect to time t, we get in, In order to ensure the stability of the closed-loop system under the kth switching subsystem, it is necessary to design the relevant parameters P k ,K1,K2,τ1,τ2 and L make the inequality ρ>0 hold, and the conclusion is proved.

[0029] Conclusion 2: Considering the attitude and altitude system of the rotorcraft under output constraints and external unknown interference (1), for the preset expected tracking signal y dBased on a class of obstacle avoidance algorithms, a safety expectation tracking signal (18) is generated. Based on a disturbance observer (24), a switching virtual control law (23) and a switching controller (28) are designed to track the signal. When the initial value is bounded and the parameters are appropriate, the error signals of all closed-loop systems are bounded when the switching law σ(y(t), t) satisfies the following conditions: The closed-loop system of the rotorcraft UAV can track the safety expectation tracking signal y s On this basis, it is guaranteed that all signals are eventually bounded and always satisfy the constraints.

[0030] Proof: To prove the stability of all error signals in the closed-loop system, multiple Lyapunov candidate functions V l (x l (t))(l∈Q) is still selected as the form of formula (30), then Must exist Satisfy the inequality

[0031] In other words, by selecting Then the following inequality holds: Combining equation (34) and equation (35), the following inequality holds: Combined with Lemma 1, under the conditions that the initial value is bounded, ρ>0 and inequality (33) holds, the error signal e of all closed-loop systems is σ,1 ,e σ,2 , and All are bounded, and the proof is complete.

[0032] According to the above steps, the rotorcraft UAV can complete the tracking of the preset expected tracking signal while considering the constant attitude angle and altitude constraints and the influence of unknown external interference and ensuring the safety and stability of the system.

[0033] A device comprises: a memory for storing a computer program; and a processor for implementing an anti-interference safety control method for a rotary-wing unmanned aerial vehicle based on an obstacle avoidance strategy when executing the computer program.

[0034] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it can perform anti-interference safety control on a rotary-wing unmanned aerial vehicle.

[0035] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An anti-interference safety control method for a rotary-wing UAV based on obstacle avoidance strategy. Features: The following steps are involved: Step 1: Considering the constant attitude angle and altitude constraints and unknown external disturbances, a nonlinear system model of the attitude and altitude of the rotorcraft is constructed. in: H is the attitude rotation matrix, Ω=[φ,θ,ψ] T , φ, θ, ψ are the roll angle, pitch angle and yaw angle of the drone under the body axis respectively, ω=[p,q,r] T , p, q, r are the roll angular velocity, pitch angular velocity and yaw angular velocity respectively, Z and w represent the vertical height and vertical velocity of the rotorcraft in the ground coordinate system respectively, u=[Σ T ,T mr ] T , J is the moment of inertia matrix, d Σ ,d T is the unknown disturbance vector under attitude angle and attitude angular velocity; Step 2: Ensure that the roll angle φ, pitch angle θ, yaw angle ψ and vertical height Z correspond to the expected tracking signal φ d ,θ d ,ψ d ,Z d and its first-order piecewise derivative is bounded. According to the upper and lower bounds of the defined constraint system and the output value y constructed in step 1, the obstacle avoidance algorithm is designed to obtain the safety expected tracking signal φ corresponding to the roll angle φ, pitch angle θ, yaw angle ψ and vertical height Z. s ,θ s ,ψ s ,Z s , through the safety expectation tracking signal φ s ,θ s ,ψ s ,Z s Get its zero-order derivative Defining vectors Calculate x 1 With y σ,s The error e σ,1 , x 2 With virtual control law α σ The error e σ,2 , error e σ,1 、e σ,2 Combining the attitude of the rotorcraft UAV with the highly nonlinear system model, the converted switching error system with jumps is obtained; Step 3: Based on the switching error system constructed in step 2, a robust switching tracking controller based on the obstacle avoidance algorithm is designed by combining the disturbance observer, filter and backstepping method. in, is the positive definite matrix to be designed for the kth switching subsystem, The filter input signal λ k The time derivative of is a positive definite matrix, and the disturbance observer is is an auxiliary variable, Γ is a positive definite matrix; Step 4: Based on the robust switching tracking controller, the anti-interference safety control of the rotorcraft UAV can be completed.

2. The anti-interference safety control method for a rotary-wing UAV based on an obstacle avoidance strategy as claimed in claim 1, Features: The obstacle avoidance algorithm has three situations. Situation 1 is when the expected tracking signal has crossed the boundary, a fixed slope is used to pull the signal back into the safety boundary to generate a safe expected tracking signal; Situation 3 is when the expected tracking signal predicts an out-of-bounds behavior, an imaginary obstacle circle is defined, and a safe expected tracking signal is constructed along the tangent direction; Situation 3 is when the expected tracking signal is tracked normally, then the expected tracking signal is used as a safe expected tracking signal.

3. The anti-interference safety control method for a rotary-wing UAV based on an obstacle avoidance strategy as claimed in claim 1, Features: The converted switching error system with jump is expressed as follows: in, is the vector y σ,s The time derivative of is the virtual control law α σ The time derivative of σ It is the control input under the switching system characteristics.

Images