UUV trajectory tracking control method based on fast nonsingular integral terminal sliding mode
Through the UUV trajectory tracking control method based on fast non-singular integral terminal sliding mode, the problems caused by system uncertainty and external interference in UUV trajectory tracking control in the prior art are solved, and fast and robust trajectory tracking and high precision are achieved Control avoids the singularity problem and provides faster convergence speed and higher tracking accuracy.
Patent Information
- Application Number
- CN202411032823.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-30
- Publication Date
- 2024-11-08
AI Technical Summary
The existing UUV trajectory tracking control methods are difficult to achieve rapid convergence and robustness when facing system uncertainty and external environment interference, and the existence of singularity problems leads to system instability and cannot accurately track the expected trajectory.
Using the UUV trajectory tracking control method based on fast non-singular integral terminal sliding mode (FANITSM), the virtual speed control quantity and sliding mode surface are defined by establishing a six-degree of freedom kinematic model and dynamic model, and the controller is designed in combination with an adaptive algorithm. Using seven thrusters to distribute force and torque, trajectory tracking is achieved, and unknown parameters are estimated through the Lyapunov function and adaptive algorithm to avoid the singularity problem.
In the presence of system uncertainty and external interference, the fast and accurate tracking of UUV trajectory is achieved, avoiding system instability caused by singularity, providing faster convergence speed and higher tracking accuracy, and no system is required. The upper limit of uncertainty and perturbation.
Smart Images

Figure CN118915759A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the field of underwater robot motion control, and in particular relates to a UUV trajectory tracking control method based on fast non-singular integral terminal sliding mode. Background Art
[0002] In recent years, countries have attached increasing importance to the development of marine resources. Autonomous underwater vehicles (UUVs) play an important role in ocean exploration and military missions, such as ocean exploration, ocean measurement, underwater structure inspection, ocean forecasting, and mine detection. Therefore, the three-dimensional trajectory tracking algorithm of UUV has become a research hotspot in the field of control. However, due to the complex problems of system uncertainty and unknown external environmental interference in the trajectory tracking control of UUV, the design of an effective UUV trajectory tracking controller has always been challenging. In addition, the need to track time-varying trajectories imposes strong constraints on the convergence time, which increases the difficulty of the tracking controller. Therefore, the design of the tracking controller should meet the requirements of strong robustness and fast convergence at the same time.
[0003] Terminal sliding mode control (TSMC) has fast response, high precision and strong robustness, and has been widely used and studied in recent years. Compared with traditional SMC, TSMC's finite time convergence and high tracking accuracy make it widely used in trajectory tracking control of UUV. The existing integral fast terminal sliding mode control method (IFTSMC) can solve the finite time stable control problem of unmanned underwater vehicles (UUVs) with disturbances and parameter uncertainties. Moreover, the control method is non-singular, has strong robustness to model uncertainty and external environmental interference, and can ensure that the entire system remains stable in a finite time. However, the position tracking error and velocity tracking error can only be guaranteed to converge to a bounded region in a local time, but not to zero, and require an upper limit on system uncertainty and disturbance, which is difficult to achieve in practice. Qiao L (“Adaptive non-singular integral terminal sliding mode tracking control for autonomous underwater vehicles,” IET Control Theory & Applications, 11 (8): 1293-1306, 2017.) et al. proposed an adaptive non-singular integral terminal sliding mode control method for trajectory tracking control of autonomous underwater vehicles (UUV) with model uncertainty and external environmental interference. This control method does not require an upper limit on uncertainty and disturbance, and can ensure that the position tracking error and velocity tracking error converge to zero locally exponentially within a finite time. And this control method is non-singular. However, the convergence speed of this control method is much smaller than the exponential convergence speed in the area far from the equilibrium point. When the control system converges slowly, the UUV may not be able to accurately track the desired trajectory. Summary of the invention
[0004] The object of the present invention is to provide a UUV trajectory tracking control method based on fast adaptive non-singular integral terminal sliding mode (FANITSM).
[0005] The technical solution to achieve the purpose of the present invention is: a UUV trajectory tracking control method based on a fast non-singular integral terminal sliding mode, comprising the following steps:
[0006] Step (1): Establish the UUV six-degree-of-freedom kinematic model and dynamic model;
[0007] Step (2): define a virtual speed control variable to transform the trajectory tracking control of the UUV into a virtual speed control;
[0008] Step (3): Define the sliding surface FNITSM:
[0009]
[0010] Where e(t) represents the tracking error, α>0 is a constant that needs to be designed, P1, P2>0 are odd numbers that satisfy 1<P1 / P2<2, β>1 is a constant that needs to be designed, g1, g2>0 are odd numbers that satisfy g1 / g2>p1 / p2. m1, m2>0 are odd numbers that satisfy m1 / m2≤1;
[0011] Step (4): Based on the sliding surface FNITSM in step (3), the trajectory tracking controller of the UUV is obtained in combination with the adaptive algorithm;
[0012] Step (5): By measuring the real-time posture of the UUV and calculating the error with the desired trajectory, the error is used as the input of the controller to obtain the output force and torque, and finally the force and torque are distributed to the 7 thrusters of the UUV, and the thrusters drive the UUV to track the desired trajectory.
[0013] Furthermore, step (1) establishes the UUV six-degree-of-freedom kinematic model as follows:
[0014] The body coordinate system B-frame and the inertial coordinate system E-frame are used to describe the motion of the UUV body. The position and posture of the UUV are defined in the inertial coordinate system and expressed as η = [x, y, z, φ, θ, ψ] T , the velocity is defined in the body coordinate system and is expressed as v = [u, v0, w, p, q, r] T ;
[0015] Kinematic relationship between body coordinate system and inertial coordinate system:
[0016]
[0017] where J(η)∈R 6×6 is the Jacobian matrix, which has the form:
[0018]
[0019] Where J1: R 6 →R 3×3 and J2:R 6 →R 3×3 Defined as:
[0020]
[0021]
[0022] where s·, c, and t· represent sin(·), cos(·), and tan(·), respectively;
[0023] To avoid singularities, the following assumptions are made:
[0024] Assumption 1: The pitch angle θ is limited to |θ|≤π / 2;
[0025] Assumption 2: J(η) is reversible and bounded; there is a known constant Make
[0026] Further, step (1): establishing the UUV dynamics model is as follows:
[0027] The UUV dynamic model is as follows:
[0028]
[0029] where M∈R 6×6 is the inertia matrix, which includes the additional mass, M∈R 6×6 is the Coriolis force and centripetal force matrix, D(v)∈R 6×6 is the damping matrix, g(η)∈R 6 is the hydrostatic restoring force vector, g(η)∈R 6 is the unknown external disturbance vector, T v ∈R 6 are the input force and torque vectors;
[0030] The parameters in the UUV dynamics model are divided into a certain part and an uncertain part, namely
[0031]
[0032] in and Indicates uncertainty. and Indicates a definite item;
[0033] Based on this, the UUV dynamic model is rewritten as:
[0034]
[0035] where τ dη represents the total uncertainty of the UUV system, and
[0036]
[0037] Assumption 3: Unknown external disturbance ||τ d ||<l d is upper bounded, so that ||τ d ||<l d , where l d >0 is an unknown constant;
[0038] Based on hypothesis 3, hypothesis 4 is made:
[0039] Assumption 4: τ dη satisfy:
[0040] ||τ dη ||<λ0+λ1||υ||+λ2||υ|| 2
[0041] where λ i (i=0, 1, 2) are all unknown normal numbers.
[0042] Furthermore, step (3) also includes a sliding surface convergence speed verification step, which is as follows:
[0043] On the FNITSM sliding surface s(t)=0, we have
[0044]
[0045] Assume e I (t) = ∫0 t e(τ)dτ, that is
[0046]
[0047] Since g1 / g2>p1 / p2, and m1 / m2≤1, when far away from the equilibrium point, that is, |e I (t)|>>1, ignore e I The low-order terms of (t) are obtained,
[0048]
[0049] When approaching the equilibrium point, that is, |e I (t)|<<1, including e less than 1 I (t) exponential, ignoring e I (t), we get:
[0050]
[0051] Furthermore, step (4) is specifically as follows:
[0052] FANITSMC is divided into equivalent control law and switching terms:
[0053] τv=τ0+τ1+τ2
[0054] Where τ0 and τ1 are equivalent control laws, and τ2 is the switching term;
[0055] In the absence of system parameter uncertainty and external disturbances, the equivalent control law can achieve the required power alone, so and τ dη= 0 to obtain the equivalent control law:
[0056]
[0057] The switching term makes the system state slide on the sliding surface. At this time, the system dynamics behavior is independent of the control law and is robust. The adaptive algorithm is used to estimate the unknown parameter λ i (i=0, 1, 2), and use the estimated parameters as the control gain in the switching term;
[0058] Select the Lyapunov function as follows:
[0059]
[0060] in is a constant, satisfying Taking the derivative of V and simplifying it, we get:
[0061]
[0062] Therefore, the switch item is designed as:
[0063]
[0064] in
[0065]
[0066] in is λ i The estimated value of is given by the following adaptive law:
[0067]
[0068] where χ i >0 (i=0, 1, 2) is a constant adaptive gain, and Furthermore, the propeller configuration of the UUV in step (5) is given by
[0069] τ v =Eu v
[0070] where E∈R 6×7 is the thruster configuration matrix, u v ∈R 7 is the force produced by each of the seven thrusters, where the individual thrusts are calculated using the minimum norm solution:
[0071] u v =E T (EE T ) -1 τ v
[0072] Among them (EE T ) -1 is the pseudo-inverse matrix of E.
[0073] Compared with the prior art, the present invention has the following significant advantages:
[0074] (1) The sliding surface in the present invention does not involve negative fractional powers and will not cause singularity problems, thereby avoiding problems such as system instability, failure to converge, and difficulty in controller design caused by singularity problems.
[0075] (2) The present invention uses an adaptive algorithm to estimate unknown boundary parameters, which does not require prior knowledge of system uncertainty and unknown environmental interference.
[0076] (3) Compared with the existing ANITSMC scheme, FANITSMC provides faster convergence speed. BRIEF DESCRIPTION OF THE DRAWINGS
[0077] Figure 1 This is the controller structure diagram.
[0078] Figure 2 Propulsion arrangement for UUV
[0079] Figure 3 Tracking response of the spatial spiral trajectory under FANITSMC and ANITSMC schemes.
[0080] Figure 4 is the speed tracking error.
[0081] Figure 5 is the position tracking error.
[0082] Figure 6 The thrust of seven thrusters. DETAILED DESCRIPTION
[0083] The present invention is further described in detail below in conjunction with the accompanying drawings.
[0084] A UUV trajectory tracking control method based on fast adaptive non-singular integral terminal sliding mode (FANITSM) includes the following steps: firstly, a six-degree-of-freedom kinematic and dynamic model of the UUV is established:
[0085] Kinematic model:
[0086] Two coordinate systems are usually used to describe the motion of the UUV body, namely the body coordinate system (B-frame) and the inertial coordinate system (E-frame). The position and attitude of the UUV are defined in the inertial coordinate system and can be expressed as η = [x, y, z, φ, θ, ψ] T, the velocity is defined in the body coordinate system and is expressed as v = [u, v0, w, p, q, r] T .
[0087] Kinematic relationship between the UUV body coordinate system and the inertial coordinate system:
[0088]
[0089] where J(η)∈R 6×6 is the Jacobian matrix, which is of the form
[0090]
[0091] Where J1: R 6 →R 3×3 and J2:R 6 →R 3×3 Defined as
[0092]
[0093] where s·, c· and t· represent sin(·), cos(·) and tan(·), respectively.
[0094] To avoid singularity problems, the following assumptions can be made:
[0095] Assumption 1: The pitch angle θ is constrained to be |θ| ≤ π / 2.
[0096] Assumption 2: J(η) is reversible and bounded. There exists a known constant Make .
[0097] Normally, the upper limit of J(η) needs to be known, but since J(η) is a trigonometric combination of Euler angles and there is a measurement system to measure the corresponding angles, there is no need to assume that the upper limit of J(η) is unknown.
[0098] Kinetic model:
[0099] UUV dynamics are as follows:
[0100]
[0101] where M∈R 6×6 is the inertia matrix, which includes the additional mass, M∈R 6×6 is the Coriolis force and centripetal force matrix, D(v)∈R 6×6 is the damping matrix, g(η)∈R 6 is the hydrostatic restoring force vector, g(η)∈R 6 is the unknown external disturbance vector, T v ∈R 6are the input force and torque vectors.
[0102] Under actual conditions, the parameters of the UUV dynamic model are generally unable to obtain accurate values. Therefore, the parameters in the UUV dynamic model can be divided into a certain part and an uncertain part, namely
[0103]
[0104] in and Indicates uncertainty. and Indicates a confirmed item.
[0105] Therefore, the UUV dynamic model can be written as
[0106]
[0107] where τ dη represents the total uncertainty of the UUV system, and
[0108]
[0109] Assumption 3: Unknown external disturbance ||τ d ||<l d is upper bounded, so that ||τ d ||<l d , where l d >0 is an unknown constant.
[0110] According to Assumption 3, we can make the following assumptions:
[0111] Assumption 4: If the parameter uncertainty and external disturbance of the UUV model are within a certain range, and the UUV model tracks a relatively simple desired trajectory so that the thrust of the propeller will not be in a saturated state for a long time, then τ dη satisfy:
[0112] ||τ dη ||<λ0+λ1||υ||+λ2||υ|| 2 (7)
[0113] where λ i (i=0, 1, 2) are all unknown normal numbers.
[0114] (2) Design of virtual control speed
[0115] First, the reference trajectory of the UUV is defined as η d ∈R 6 , the position tracking error is defined as Combining formula (1), we can get
[0116]
[0117] Take speed v as the virtual control input of the above equation, and design v c Virtual control command as reference velocity vector:
[0118]
[0119] Where K1 = diag(k 11 , ..., k 16 ) and K2=diag(k 21 , ..., k 26 ) are all constant positive definite matrices that need to be designed, k 1i , j 2i (i=1, ..., 6) satisfies
[0120] Then, the velocity tracking error is defined as:
[0121]
[0122] When v = v c Sometimes, there are From (10), we can conclude that
[0123] v=v c
[0124] Substituting (9) into (8) we obtain
[0125]
[0126] Right now
[0127]
[0128] because can converge to zero exponentially. Therefore, it can be seen that in a finite time t sum Post-position tracking error The exponential converges to zero.
[0129] In addition, according to Assumption 1, the pitch angle is bounded, so the position tracking error is The local exponential converges to zero in finite time.
[0130] (3) Design of sliding surface
[0131] A sliding surface with faster convergence is defined:
[0132]
[0133] Where e(t) represents the tracking error, α>0 is a constant that needs to be designed, P1, P2>0 are odd numbers that satisfy 1<P1 / P2<2, β>1 is a constant that needs to be designed, g1, g2>0 are odd numbers that satisfy g1 / g2>p1 / p2, and m1, m2>0 are odd numbers that satisfy m1 / m2≤1.
[0134] On the FNITSM sliding surface s(t)=0, we have
[0135]
[0136] Assume e I (t) = ∫0 t e(τ)dτ, that is
[0137]
[0138] Since g1 / g2>p1 / p2, and m1 / m2≤1, when far away from the equilibrium point, that is, |e I (t)|>>1, e I The low-order terms of (t) can be ignored, that is, we can get
[0139]
[0140] Compared with LSM, it is e I (t) provides a faster convergence speed.
[0141] And close to the equilibrium point, that is, |e I (t)|<<1, including e less than 1 I (t) Index. e I The high-order terms of (t) can be ignored, that is, we can get:
[0142]
[0143] Compared to LSM, it is e I (t) provides a faster convergence speed.
[0144] (4) Controller design
[0145] Following the design principle of SMC, the design of FANITSMC method is divided into equivalent control law and switching term:
[0146] τ v =τ0+τ1+τ2 (18)
[0147] Where τ0 and τ1 are equivalent control laws, and τ2 is the switching term.
[0148] In the absence of system parameter uncertainty and external disturbances, the equivalent control law can achieve the required dynamics alone. and τ dη = 0 to obtain the equivalent control law:
[0149]
[0150]
[0151] The switching term can make the system state slide on the sliding surface. At this time, the system dynamics behavior is independent of the control law and is completely insensitive to the parameter uncertainty within the system and the external environment interference, that is, it has robustness. dη The parameter λ i (i=0,1,2), so an adaptive algorithm can be used to estimate the unknown parameter λ i (i=0, 1, 2), and the estimated parameters are used as control gains in the switching term to eliminate the influence of uncertainty and interference.
[0152] And select the Lyapunov function as follows:
[0153]
[0154] in is a constant, satisfying
[0155] Taking the derivative of V and simplifying it according to the above formula, we can get:
[0156]
[0157] Therefore, the switch item is designed as:
[0158]
[0159] in
[0160]
[0161] in is λ i The estimated value of is given by the following adaptive law:
[0162]
[0163]
[0164]
[0165] where χ i >0 (i=0, 1, 2) is a constant adaptive gain, and Substituting the designed switching term into equation (21), we can simplify it into:
[0166]
[0167]
[0168] make
[0169]
[0170] From ((26) and (27) we can see
[0171]
[0172] and So we can get
[0173]
[0174] Simulation verification:
[0175] The following simulation test verifies the effectiveness of the UUV trajectory tracking controller FANITSMC and compares it with the ANITSMC solution to show the superiority of the FANITSMC solution.
[0176] The simulation test was conducted using OCTO UUV. The specific structure of the UUV is shown in the figure. The figure also shows the arrangement of the seven thrusters on the UUV. It can be seen that the UUV is driven by distributing the thrust generated by the seven thrusters in six degrees of freedom. The seven thrusters are identical, and the saturation of each thruster is ±10N. The thruster configuration of the UUV is given by the following formula:
[0177] τ v =Eu v (29)
[0178] where E∈R 6×7 is the thruster configuration matrix, u v ∈R 7 is the force produced by each of the seven thrusters, where the individual thrusts can be calculated using the minimum norm solution:
[0179] u v =E r (EE T ) -1 τ v (30)
[0180] Among them (EE T ) -1 is the pseudo-inverse matrix of E.
[0181] The real-time position and posture of the UUV are measured by sensors, and the error with the expected trajectory is calculated. The error is used as the input of the controller to obtain the output force and torque, and finally the total force and total torque are distributed to the 7 thrusters of the UUV.
[0182] The spatial helix reference trajectory is described as follows: d (t) = sin(0.2t)my d (t) = cos(0.2t)m, z d (t) = 0.1tm
[0183] The ocean environment disturbances acting on UUV dynamics are selected as follows:
[0184] τ d1 =3+2sin(0.3t)N,τ d2 =4sin(0.2t)N
[0185] τ d3 =2+3sin(0.1t)N,τ d4 =1+2cos(0.1t)N·m
[0186] τ d5 =3cos(0.2t)N·m, τ a6 =2+cos(0.3t)N·m
[0187] The initial conditions of the UUV are as follows:
[0188] UUV initial position: x(0)=1.2m, y(0)=1.3m, z(0)=-0.6m, φ(0)=0.6rad, θ(0)=0.45rad, ψ(0)=0.4rad.
[0189] Initial velocity: u(0)=1.8m / s, v(0)=0.9m / s, w(0)=-1.4m / s, p(0)=0.8rad / s, q(0)=-0.7rad / s, r(0)=1rad / s.
[0190] The pitch angle θ is limited to |θ|≤π / 4rad.
[0191] The parameter designs used in the simulation are shown in the following table:
[0192]
[0193] Figure 3 , Figure 4 , Figure 5The simulation results of UUV tracking the spatial spiral trajectory under two schemes are given. It can be seen from the figure that despite the uncertainty of system parameters and unknown marine environment interference of UUV, the FANITSMC scheme proposed in this paper can still enable UUV to track the given desired trajectory well. Compared with the ANITSMC scheme, the UUV under the FANITSMC scheme has a faster convergence speed and higher tracking accuracy. Figure 6 It shows the thrust response curves of seven thrusters. It can be seen from the figure that the thrust changes of the thrusters are relatively stable.
Claims
1. A UUV trajectory tracking control method based on fast non-singular integral terminal sliding mode, characterized in that: The steps include: Step (1): Establish the UUV six-degree-of-freedom kinematic model and dynamic model; Step (2): define a virtual speed control variable to transform the trajectory tracking control of the UUV into a virtual speed control; Step (3): Define the sliding surface FNITSM: Wherein, e(t) represents the tracking error, α>0 is a constant to be designed, P1, P2>0 are odd numbers and satisfy 1<P1 / P2<2, β>1 is a constant to be designed, g1, g2>0 are odd numbers satisfying g1 / g2>p1 / p2, m1, m2>0 are odd numbers satisfying m1 / m2≤1; Step (4): Based on the sliding surface FNITSM in step (3), the trajectory tracking controller of the UUV is obtained in combination with the adaptive algorithm; Step (5): By measuring the real-time posture of the UUV and calculating the error with the desired trajectory, the error is used as the input of the controller to obtain the output force and torque, and finally the force and torque are distributed to the thrusters of the UUV, and the thrusters drive the UUV to track the desired trajectory.
2. The method according to claim 1, characterized in that Step (1) establishes the UUV six-degree-of-freedom kinematic model as follows: The body coordinate system B-frame and the inertial coordinate system E-frame are used to describe the motion of the UUV body. The position and posture of the UUV are defined in the inertial coordinate system and expressed as η = [x, y, z, φ, θ, ψ] T , the velocity is defined in the body coordinate system and is expressed as v = [u, v0, w, p, q, r] T ; Kinematic relationship between body coordinate system and inertial coordinate system: where J(η)∈R 6×6 is the Jacobian matrix, which has the form: Where J1: R 6 →R 3×3 and J2:R 6 →R 3×3 Defined as: where s·, c· and t· represent sin(·), cos(·) and tan(·), respectively; In order to avoid singularities, the following assumptions are made: Assumption 1: The pitch angle θ is limited to |θ|≤π / 2; Assumption 2: J(η) is reversible and bounded; there is a known constant Make 3. The method according to claim 2, characterized in that Step (1): Establish the UUV dynamic model as follows: The UUV dynamic model is as follows: where M∈R 6×6 is the inertia matrix, which includes the additional mass, M∈R 6×6 is the Coriolis force and centripetal force matrix, D(v)∈R 6 ×6 is the damping matrix, g(η)∈R 6 is the hydrostatic restoring force vector, g(η)∈R 6 is the unknown external disturbance vector, T v ∈R 6 are the input force and torque vectors; The parameters in the UUV dynamics model are divided into a certain part and an uncertain part, namely in and Indicates uncertainty. and Indicates a definite item; Based on this, the UUV dynamic model is rewritten as: where τ dη represents the total uncertainty of the UUV system, and Assumption 3: Unknown external disturbance ||τ d ||<l d is upper bounded, so that ||τ d ||<l d , where l d >0 is an unknown constant; Based on hypothesis 3, hypothesis 4 is made: Assumption 4: τ dη satisfy: ||t dη ||<λ0+λ1||υ||+λ2||υ|| 2 where λ i (i=0, 1, 2) are all unknown normal numbers.
4. The method according to claim 3, characterized in that Step (3) also includes a sliding surface convergence speed verification step, which is as follows: On the FNITSM sliding surface s(t)=0, we have set up Right now Since g1 / g2>p1 / p2, and m1 / m2≤1, when far away from the equilibrium point, that is, |e I (t)|>>1, ignore e I The low-order terms of (t) are obtained, When approaching the equilibrium point, that is, |e I (t)|<<1, including e less than 1 I (t) exponential, ignoring e I (t), we get:
5. The method according to claim 4, characterized in that Step (4) is as follows: FANITSMC is divided into equivalent control law and switching terms: t v =τ0+τ1+τ2 Where τ0 and τ1 are equivalent control laws, and τ2 is the switching term; In the absence of system parameter uncertainty and external disturbances, the equivalent control law can achieve the required power alone, so and τ dη = 0 to obtain the equivalent control law: The switching term makes the system state slide on the sliding surface. At this time, the system dynamics behavior is independent of the control law and is robust. The adaptive algorithm is used to estimate the unknown parameter λ i (i=0, 1, 2), and use the estimated parameters as the control gain in the switching term; Select the Lyapunov function as follows: in is a constant, satisfying Taking the derivative of V and simplifying it, we get: Therefore, the switch item is designed as: in in is λ i The estimated value of is given by the following adaptive law: where χ i >0 (i=0, 1, 2) is a constant adaptive gain, and 6. The method according to claim 5, characterized in that The thruster configuration of the UUV in step (5) is given by τ v =I v where E∈R 6×7 is the thruster configuration matrix, u v ∈R 7 is the force produced by each of the seven thrusters, where the individual thrusts are calculated using the minimum norm solution: u v =And T (AND AND T ) -1 t v Among them (EE T ) -1 is the pseudo-inverse matrix of E.
Citation Information
Cited By
MEMS galvanometer scanning method and laser radar system
CN119986606A
Unmanned autonomous submersible vehicle track tracking control method
CN120370718A
