CN108227502A - One kind is overdrived AUV executing agencies method for handover control - Google Patents

One kind is overdrived AUV executing agencies method for handover control Download PDF

Info

Publication number
CN108227502A
CN108227502A CN201810057813.2A CN201810057813A CN108227502A CN 108227502 A CN108227502 A CN 108227502A CN 201810057813 A CN201810057813 A CN 201810057813A CN 108227502 A CN108227502 A CN 108227502A
Authority
CN
China
Prior art keywords
auv
control
state
speed
space model
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201810057813.2A
Other languages
Chinese (zh)
Other versions
CN108227502B (en
Inventor
张铁栋
刘超
邢炜
张韧然
辛坚
刘自豪
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Harbin Engineering University
Original Assignee
Harbin Engineering University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Harbin Engineering University filed Critical Harbin Engineering University
Priority to CN201810057813.2A priority Critical patent/CN108227502B/en
Publication of CN108227502A publication Critical patent/CN108227502A/en
Application granted granted Critical
Publication of CN108227502B publication Critical patent/CN108227502B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
    • G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
    • G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
    • G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Feedback Control In General (AREA)
  • Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)

Abstract

The present invention provides one kind and overdrives AUV executing agencies method for handover control, including：Establish AUV vertical planes kinetics equation, motion state spatial model；During low speed executing agency using promote mainly with vertical channels propeller, using promoting mainly and the rudder wing during high speed；A transitional region is set at high low speed switching point, in transitional region, by defining a soft-constraint to executing agency, makes vertical channels propeller and the rudder wing smooth variation in this section, and keep Angle of Trim constant；Using model predictive control method, calculated according to control instruction and promote mainly thrust, thrust of hanging down and rudder angle；It brings control input into model and obtains AUV posture informations.The present invention can avoid the shake by a relatively large margin that the speed as caused by switching AUV executing agencies suddenly and trim generate, reduce the oscillation amplitude that control occurs, ensure that AUV keeps seamlessly transitting for motion state while executing agency is switched, and work capacities of the AUV in handoff procedure is unaffected to be ensured for the control of trim.

Description

One kind is overdrived AUV executing agencies method for handover control
Technical field
It overdrives AUV executing agencies method for handover control the present invention relates to one kind, belongs to AUV motion control fields.
Background technology
Due to the needs of the detection exploitation of marine resources, both at home and abroad to AUV technology pay attention to day by day.Since AUV is being performed not During with task, the speed of robot in itself there are certain requirements, if necessary the conversion work state between different task, that The switching between high low speed will necessarily just occur.Because thruster is efficient in low speed, steerage rate is low, and in high speed, Thruster efficiency declines, and steerage rate increases, and many AUV especially high-speed motion AUV are using the rudder wing and propeller mixing control. AUV uses the control of the rudder wing instead in high speed and rises in low speed using vertical channels thruster control heave and trim under normal circumstances Heavy executing agency's configuration mode with trim.This there is the switching of a control mode, if in a certain speed of a ship or plane, turns off push away immediately Power device opens rudder, and due to the mutation of control instruction, control will appear the oscillation of a period of time, and speed and trim can all occur one Fixed shake, the variation of Angle of Trim so that task difficulty increases AUV in the process.
Document [1] will control output to be set as a linear function, and controller input is provided by the FLC of TS forms, to adopt With one fuzzy logic controller of the VSC design of control law of smooth function, the fuzzy controller with smooth function is formd Method.Compared to PID control, the method reduce overshoot, enhance the robustness of AUV controls.
Document [2] is had studied in AUV path tracking procedures, and drift angle is trembled when being switched to drive lacking by driving entirely for AUV The problem of amplitude of moving is larger, the smooth function for employing trigonometric function form constrain traversing control input.Emulation The result shows that this approach reduce the maximum value of drift angle shake, but the absolute value of minimum value increased.
Document [3] during actuator switching control during AUV rudder paddle united controls for be easy to causeing asking for output mutation Topic, it is proposed that a kind of switching function is taken over seamlessly to achieve the purpose that eliminate mutation.
Pertinent literature shows in the prior art, to have no that effect is preferable, can be during AUV switches executing agency both It can guarantee seamlessly transitting for switching, and AUV Angle of Trims can be made not do the method significantly alterred.
The data that the present invention refers to is as follows：
[1]Wang Y,Shen Y,Wang K,et al.Fuzzy controller used smoothing function for depth control of autonomous underwater vehicle[C]//Oceans.IEEE, 2016:1-5.
[2]Xiang X,Lapierre L,Jouvencel B.Smooth transition of AUV motion control:From fully-actuated to under-actuated configuration[J].Robotics& Autonomous Systems,2015,67:14-22.
[3] Zhang Lei, Pang Yongjie, Li Ye wait autonomous type underwater robot motion control [J] Beijing of rudder paddle united controls Polytechnical university's journal, 2011,37 (1):40-46.
Invention content
The purpose of the invention is to provide one kind to overdrive AUV executing agencies method for handover control, can make to overdrive Seamlessly transitting for motion state is kept when AUV executing agencies switch.
The object of the present invention is achieved like this：Step 1：According to the hydrodynamic force coefficient of AUV, AUV is established relative to satellite The advance and retreat of coordinate system, heave, pitching the equation of motion；
Step 2：Find main state variables, by retreating, heaving, the kinematical equation of pitching and kinetics equation derive into For state-space model, and state-space model discretization is obtained by augmented state spatial mode according to the sampling rate of controller Type；
Step 3：The cost function of quadratic term form is defined according to augmented state space model, cost function is that controller is set Error between definite value and system in future track inputs the sum of change in future amplitude with control；
Step 4：The speed interval of high speed and low speed is set separately, and transition speed section is set in boundary, it will be at a high speed When executing agency be set as promoting mainly with the rudder wing, executing agency during by low speed, which is set as promoting mainly and hang down, to push away, transitional region executing agency Be set as promoting mainly, hang down push away with the rudder wing, set a soft-constraint for rudder wing angle and thrust of hanging down in transitional region, make thrust and rudder angle It is seamlessly transitted in transition section；
Step 5：Cost function minimum is made to solve control to input and act in system, the speed of obtained AUV is indulged Incline, depth information；
Step 6：The posture information that step 5 obtains judges whether to meet control instruction as feedback：If satisfied, control knot Beam；If not satisfied, posture information is supplied to state-space model and cost function, prediction locus is obtained, performs step 5, directly Terminate to current control.
The invention also includes some such structure features：
1. step 1 is specially：
AUV is heaved, the kinematical equation of pitching is relative to the advance and retreat of satellite coordinate system：
Kinetics equation is：
Wherein：X and z is longitudinal velocity and vertical velocity under earth coordinates respectively, and θ is the Angle of Trim of AUV, and m is AUV Quality, IyIt is the moment of inertia of the hull for y-axis；Xλ、Zλ、Xf、ZfIt is the fluid inertia force item as caused by u, w speed term and towing Power item, XrWith ZrIt is the drag that the rudder wing generates；Mq、MhIt is the turning moment and hull torque that AUV hydrodynamic forces generate, MrIt is rudder The torque that the wing generates；Xp、ZTf、ZTrItem is tail portion main thruster, head and the tail vertical channels propeller thrust；LTf、LTrIt is that head and the tail are vertical The arm of force of conduit propeller thrust；G, ▽ represents the gravity and displacement of volume of hull；G represents acceleration of gravity；Δ z represents them Center to Z axis distance.
2. in step 2, state-space model is：
Wherein： ρ is the density of water, with footmark items α and β be respectively measure with the relevant constant of hydrodynamic force, δ is rudder Angle；
State-space model is reduced to：
Wherein：
State-space model discretization is obtained into augmented state model：
Wherein, I is unit matrix.
3. in step 3：The cost function of quadratic term form is defined by augmented state space model, given pre- for one If value r (ki), in sample time kiInterior, in prediction space, the target of Predictive Control System is that the output terminal of prediction to the greatest extent may be used It can be close to preset value, it is assumed that preset value remains unchanged in window is optimized, NPIt is respectively prediction step and control step-length with Nc, then With NPFor length, the vector comprising preset value is：
Cost function J is defined to reflect control targe：
In formula：U is control output,It is a diagonal matrix,rw>=0, rwIt is penalty factor.
4. in step 4, it is contemplated that the limitation of dead zone and motor speed uses restraint to promoting mainly thrust：
Xmin≤Xp≤Xmax
ΔXmin≤ΔXp≤ΔXmax
Constraint to thrust of hanging down：
Tmin≤Tp≤Tmax
ΔTmin≤ΔTp≤ΔTmax
Constraint to rudder angle：
δmin≤δp≤δmax
Δδmin≤Δδp≤Δδmax
The method that prediction output obtains is to obtain predicted state variable using augmented state space model：
x(ki+2|ki)=Ax (ki+1|ki)+BΔu(ki+1)
=A2x(ki)+ABΔu(ki)+BΔu(ki+1)
Wherein,
Obtaining prediction output from predicted state variable is：
y(ki+1|ki)=CAx (ki)+CBΔu(ki)
y(ki+2|ki)=CA2x(ki)+CABΔu(ki)+CBΔu(ki+1)
y(ki+3|ki)=CA3x(ki)+CA2BΔu(ki)+CABΔu(ki+1)+CBΔu(ki+2)
Wherein, C=[0m I]。
Definition vector：
Y=[y (ki+1|ki) y(ki+2|ki) y(ki+3|ki) ... y(ki+Np|ki)]TY represents state output, and Following control domain the Δ U, length NC, it is expressed as：
Δ U=[Δ u (ki) Δu(ki+1) Δu(ki+2) ... Δu(ki+Nc-1)]TIt can obtain：
Wherein：
5. in step 5, in order to which the Δ U for finding best minimizes J, by state output：
Bringing cost function into, then J is:
The first derivative of function J is：
Become the condition that J is minimized：
Compared with prior art, the beneficial effects of the invention are as follows：1st, the present invention is using using model predictive control method, root It is calculated according to control instruction and promotes mainly thrust, hang down thrust and rudder angle can naturally handle multiple-input and multiple-output (MIMO) control pair well As some methods controlled can preferably control AUV than tradition.In design process constraint can be systematically included in, this with Traditional control method is different, therefore this method is not limited by latent device physical condition.And MPC has the function of prediction, can be right Following error is predicted.Therefore the present invention has the advantages of applicability is wide, and control accuracy is high.2nd, the present invention switches in high low speed A transitional region is set at point, in transitional region, by defining a soft-constraint to executing agency, ensures that trim is constant Meanwhile make vertical channels propeller and the rudder wing smooth variation in this section, it can avoid being drawn by switching AUV executing agencies suddenly The shake by a relatively large margin that the speed and trim risen generates reduces control and the risk that oscillation even results in control diverging occurs, AUV is made to keep seamlessly transitting for motion state.
Description of the drawings
The depthkeeping constant speed that Fig. 1 is the present invention moves longitudinal velocity simulation result figure；
The depthkeeping constant speed that Fig. 2 is the present invention moves depth simulation result figure；
The depthkeeping constant speed that Fig. 3 is the present invention moves Angle of Trim simulation result figure；
Thrust simulation result figure is promoted mainly in the depthkeeping constant speed movement that Fig. 4 is the present invention；
The depthkeeping constant speed that Fig. 5 is the present invention moves rudder angle simulation result figure；
Fig. 6 is the vertical thrust simulation result figure of depthkeeping constant speed movement of the present invention；
Fig. 7 is the flow chart of the present invention.
Specific embodiment
The present invention is described in further detail with specific embodiment below in conjunction with the accompanying drawings.
Embodiment one：As shown in the figure, one kind of the present invention is overdrived, AUV executing agencies method for handover control includes following step Suddenly：
Step 1：Establish the AUV equations of motion.According to the hydrodynamic force coefficient of AUV, establish AUV relative to satellite coordinate system into It moves back, heaves, the equation of motion of pitching.Because the AUV of most of movements belong to the weak motion of automobile, second and higher order hydrodynamic force item Influence for AUV movements is smaller than single order power item, therefore only considers single order hydrodynamic force, ignores more than second-order hydrodynamic , then AUV is heaved, the kinematical equation of pitching is relative to the advance and retreat of satellite coordinate system：
Kinetics equation is：
Wherein, x and z is the longitudinal velocity and vertical velocity under earth coordinates respectively, and θ is the Angle of Trim of AUV, and m is AUV Quality, IyIt is the moment of inertia of the hull for y-axis, Xλ、Zλ、Xf、ZfIt is the fluid inertia force item as caused by u, w speed term and towing Power item, XrWith ZrIt is the drag that the rudder wing generates；Mq、MhIt is the turning moment and hull torque that AUV hydrodynamic forces generate, MrIt is rudder The torque that the wing generates；Xp、ZTf、ZTrItem is tail portion main thruster, head and the tail vertical channels propeller thrust；LTf、LTrIt is that head and the tail are vertical The arm of force of conduit propeller thrust；G、Represent the gravity and displacement of volume of hull；G represents acceleration of gravity；Δ z represents them Center to Z axis distance.
Step 2：Derive state-space model.Main state variables are found, by retreating, are heaved, the kinematical equation of pitching Being derived with kinetics equation becomes state-space model, and according to the sampling rate of controller further by state-space model Discretization obtains augmented state space model；
To formula (1) derivation, and wushu (3) is brought into obtain：
Wushu (4), (5) take into account again, and formula (1) becomes：
Wherein, XλCos θ, ZλSin θ represents the inertia force in E ξ axis directions, and the sum of they are：
Wherein, λξRepresent the additional mass in E ξ axis directions.
Similar, XfCos θ and ZfSin θ is expressed as：
Formula (6) can be written as form：
Similar, formula (2) can be expressed as with (3)：
For the AUV studied, Angle of Trim is limited in θ ∈ (- 15 °, 15 °), therefore：
Sin θ ≈ θ, cos θ ≈ 1
Definition：m-λξ=mx, m- λζ=mz
So the equation of motion for studying AUV in vertical plane is：
For equation (14), it is found that drag and rudder power are the functions of longitudinal velocity square, and resistance coefficient and rudder Force coefficient is the function of θ and δ respectively, so these will be approximate in state-space model.
Obviously, longitudinal velocity variation is affected to longitudinal velocity equation than θ and δ variations, so longitudinal flow velocity is as master The variable wanted.The drag item of linearisation is represented using newest longitudinal velocity, and pulls force coefficient and rudder force coefficient calculating then Using newest pitch angle and rudder angle, then equation (14) is rewritten as：
Wherein：
θ pairs is considered for formula (15)Influence, therefore withθ ignores surplus buoyancy influence as main state variables, withθ, δ are obtained as main state variables：
Wherein:
Q and θ is primary variables, is changed to：
Wherein：ρ is the close of water Degree, with footmark items α and β be respectively measure with the relevant constant of hydrodynamic force, δ is rudder angle.
In conclusion state-space expression is：
Obtain state-space model：
It can be expressed as：
Wherein：
Control exports：
Y (t)=Cmxm(t)
To formula (22), (23) discretization, can obtain：
xm(k+1)=Mmxm(k)+Kmum(k) (24)
ym(k+1)=Nmxm(k+1) (25)
The present invention predicts the track of system in future using MPC algorithm, and the output of system is that a function current state becomes The vector rather than input vector of change.
MPC algorithm used herein is embedded into model, allows the controller to handle inaccurate model.By the sum of U and Δ U As the input of system, Δ U is the difference between state currently and previously.
Differential state equation can be expressed as:
Δxm(k+1)=MmΔxm(k)+KmΔum(k) (26)
It is obtained by equation (25)
Y (k+1)-y (k)=Nm(xm(k+1)-xm(k))=NmΔxm(k+1)=NmMmΔxm(k)+NmKmΔu(k) (27)
Composite equation (24) and (25) can be expanded state equation:
Step 3：Define the cost function of quadratic term form.The cost function of quadratic term form is by augmented state space model Definition is that the error between controller setting value and system in future track inputs the sum of change in future value with control, for one Given preset value r (ki), in sample time kiInterior, in prediction space, the target of Predictive Control System is by the defeated of prediction Outlet is as close possible to preset value, herein we assume that preset value remains unchanged in window is optimized.NPIt is respectively pre- with Nc Step-length and control step-length are surveyed, then with NPFor length, the vector comprising preset value is：
We define cost function J to reflect control targe：
First item is related with the purpose of error between preset value to minimizing prediction output, and Section 2 reflection is to work as target Function J hours as much as possible, the size of Δ u.
It is a diagonal matrix,
Wherein, rwIt is penalty factor.Work as rwWhen=0, cost function J, situation as expression：We are not concerned with the big of Δ U Small, our unique objects are to make first item error small as far as possible.
And work as rw>When 0, cost function need to consider the size of Δ U and Section 2 error simultaneously.
Step 4：Setting constraint.It is set separately at a high speed and the speed interval of low speed, and executing agency during by high speed is set as Promote mainly with the rudder wing, executing agency during by low speed, which is set as promoting mainly and hang down, to push away, and transitional region executing agency be set as promoting mainly, hang down push away and The rudder wing, but a soft-constraint is set for rudder wing angle and thrust of hanging down in transitional region, make thrust and rudder angle in transition section Inside seamlessly transit；
In view of the limitation of dead zone and motor speed, use restraint to promoting mainly thrust：
Xmin≤Xp≤Xmax
ΔXmin≤ΔXp≤ΔXmax (32)
Constraint to thrust of hanging down：
Tmin≤Tp≤Tmax
ΔTmin≤ΔTp≤ΔTmax (33)
Constraint to rudder angle：
δmin≤δp≤δmax
Δδmin≤Δδp≤Δδmax (34)
Step 5：Set soft-constraint.For thrust and the rudder wing, because thruster is efficient in low speed, steerage rate of hanging down It is low, and in high speed, thruster efficiency declines, and steerage rate increases, and the AUV that overdrives uses vertical channels thruster control in low speed System heave and trim use rudder wing control heave and executing agency's configuration mode of trim instead in high speed, this there is a control The switching of mode processed.From low speed to when switching at high speed, this method does not turn off thruster in a certain speed of a ship or plane and opens rudder immediately, but Boundary setting transition speed section, is smoothly switched to rudder from thruster by control instruction, makes the control of each actuator Instruction is that total security maintains an equal level slow change.
Here v is the longitudinal velocity of AUV, [a1,a2] be soft-constraint effect speed transition section, speed a1、a2It will basis The efficiency of actual thrust device and rudder determines that T represents thrust of hanging down with speed of a ship or plane relationship, and δ represents rudder angle, and thrust is promoted mainly in P expressions, then Constraint is as follows：
Step 6：The prediction of state variable and output variable.In each sampling instant ki, state variable xi(k) can pass through It measures and obtains, state value xi(k) current information of object is provided, following increment is denoted as Δ u (ki),Δu(ki+1),…,Δu(ki +NC- 1), following state variable is：x(ki+1|ki,x(ki+2|ki),…,x(ki+Np|ki), future state vector can represent For：
Wherein,
Obtaining prediction output from predicted state variable is：
y(ki+1|ki)=CAx (ki)+CBΔu(ki) (36)
y(ki+2|ki)=CA2x(ki)+CABΔu(ki)+CBΔu(ki+1)
y(ki+3|ki)=CA3x(ki)+CA2BΔu(ki)+CABΔu(ki+1)+CBΔu(ki+2)
...
Wherein, C=[0m I]
Definition vector：
Y=[y (ki+1|ki) y(ki+2|ki) y(ki+3|ki) ... y(ki+Np|ki)]T (38)
Y represents state output, and following control length of field is NC, Δ U can be expressed as：
Δ U=[Δ u (ki) Δu(ki+1) Δu(ki+2) ... Δu(ki+Nc-1)]T (39)
It can obtain：
Wherein,
Step 7：Resolve control input.Cost function minimum is made to solve control input, acts in system, obtains The speed of AUV, trim, depth information；In order to which the Δ U for finding best minimizes J, by state output：
Cost function is brought into be expressed as J：
The first derivative of function J is：
Become the condition that J is minimized：
It is possible to it obtains：
Step 8：The posture information that step 6 obtains judges whether to meet control instruction as feedback：If satisfied, control knot Beam；If not satisfied, posture information is supplied to state-space model and cost function, prediction locus is obtained, performs step 5 to 7, Until current control terminates.
Embodiment two：The step of the present invention is as follows：
Step 1：According to the hydrodynamic force coefficient of AUV, advance and retreat of the AUV relative to satellite coordinate system are established, are heaved, the fortune of pitching Dynamic equation.Because the AUV of most of movements belongs to the weak motion of automobile, the shadow that second and higher order hydrodynamic force item move AUV Sound is smaller than single order power item, therefore only considers single order hydrodynamic force, ignores more than second-order hydrodynamic item；
Step 2：Find main state variables, by retreating, heave, the kinematical equation and kinetics equation of pitching derive into For state-space model, and state-space model discretization is further obtained by augmented state according to the sampling rate of controller Spatial model；
Step 3：The cost function of quadratic term form is defined according to augmented state space model, cost function is that controller is set Error between definite value and system in future track inputs the sum of change in future amplitude with control；
Step 4：The speed interval of high speed and low speed is set separately, and transition speed section is set in boundary, it will be at a high speed When executing agency be set as promoting mainly with the rudder wing, executing agency during by low speed, which is set as promoting mainly and hang down, to push away, and transitional region performs machine Structure is set as promoting mainly, hang down push away with the rudder wing, but set a soft-constraint for rudder wing angle and thrust of hanging down in transitional region, make thrust It is seamlessly transitted in transition section with rudder angle；
Step 5：Cost function minimum is made to solve control input, is acted in system, the speed of obtained AUV, trim, Depth information；
Step 6：The posture information that step 5 obtains judges whether to meet control instruction, if not satisfied, by position as feedback Appearance information is supplied to state-space model and cost function, obtains prediction locus, performs step 5, until current control terminates.
In step 1, after ignoring 2 rank more than hydrodynamic forces, AUV is relative to the advance and retreat of satellite coordinate system, heave, the movement of pitching Learning equation is：
Kinetics equation：
Wherein, x and z is the longitudinal velocity and vertical velocity under earth coordinates respectively, and θ is the Angle of Trim of AUV, and m is AUV Quality, IyIt is the moment of inertia of the hull for y-axis, Xλ、Zλ、Xf、ZfIt is the fluid inertia force item as caused by u, w speed term and towing Power item, XrWith ZrIt is the drag that the rudder wing generates.Mq、MhIt is the turning moment and hull torque that AUV hydrodynamic forces generate, MrIt is rudder The torque that the wing generates.Xp、ZTf、ZTrItem is tail portion main thruster, head and the tail vertical channels propeller thrust.LTf、LTrIt is that head and the tail are vertical The arm of force of conduit propeller thrust.G、Represent the gravity and displacement of volume of hull；G represents acceleration of gravity；Δ z represents them Center to Z axis distance.
The parameter and the component of the moving axes of power that table 1 moves
In step 2, state-space model is：
Wherein, ρ is the density of water, with footmark items α and β be respectively measure with the relevant constant of hydrodynamic force, δ is rudder Angle.
State-space model can be abbreviated as following form：
Wherein:
In step 2, state-space model discretization is obtained into augmented state model:
I is unit matrix.
In step 3, the cost function of quadratic term form is defined by augmented state space model, given default for one Value r (ki), in sample time kiIt is interior, prediction space in, the target of Predictive Control System, be by the output terminal of prediction as far as possible Close to preset value, herein we assume that preset value remains unchanged in window is optimized.NPIt is respectively prediction step and control with Nc Step-length processed, then with NPFor length, the vector comprising preset value is：
We define cost function J to reflect control targe：
First item is related with the purpose of error between preset value to minimizing prediction output, and Section 2 reflection is to work as target Function J hours as much as possible, the size of Δ u.
U is control output,It is a diagonal matrix,
Wherein, rwIt is penalty factor.Work as rwWhen=0, cost function J, situation as expression：We are not concerned with the big of Δ U Small, our unique objects are to make first item error small as far as possible.
And work as rw>When 0, cost function need to consider the size of Δ U and Section 2 error simultaneously.
The method that prediction output obtains is to obtain predicted state variable using augmented state space model：
x(ki+2|ki)=Ax (ki+1|ki)+BΔu(ki+1)
=A2x(ki)+ABΔu(ki)+BΔu(ki+1)
Wherein,
Obtaining prediction output from predicted state variable is：
y(ki+1|ki)=CAx (ki)+CBΔu(ki)
y(ki+2|ki)=CA2x(ki)+CABΔu(ki)+CBΔu(ki+1)
y(ki+3|ki)=CA3x(ki)+CA2BΔu(ki)+CABΔu(ki+1)+CBΔu(ki+2)
Wherein, C=[0m I]
Definition vector：
Y=[y (ki+1|ki) y(ki+2|ki) y(ki+3|ki) ... y(ki+Np|ki)]T
Y represents state output, and following control domain Δ U, length NC, can be expressed as：
Δ U=[Δ u (ki) Δu(ki+1) Δu(ki+2) ... Δu(ki+Nc-1)]T
It can obtain:
Wherein,
In step 4, it is contemplated that the limitation of dead zone and motor speed uses restraint to promoting mainly thrust：
Xmin≤Xp≤Xmax
ΔXmin≤ΔXp≤ΔXmax
Constraint to thrust of hanging down：
Tmin≤Tp≤Tmax
ΔTmin≤ΔTp≤ΔTmax
Constraint to rudder angle：
δmin≤δp≤δmax
Δδmin≤Δδp≤Δδmax
In step 4, for thrust and the rudder wing of hanging down, because thruster is efficient in low speed, steerage rate is low, and at a high speed When, thruster efficiency declines, the increase of steerage rate, and the AUV that overdrives is heaved and indulged using the control of vertical channels thruster in low speed Incline, use rudder wing control heave and executing agency's configuration mode of trim instead in high speed, this there is cutting for control mode It changes.From low speed to when switching at high speed, this method does not turn off thruster in a certain speed of a ship or plane and opens rudder immediately, but sets one Control instruction is slowly switched to rudder by the intermediate stage from thruster, from high speed to low speed switch when similarly, that is to say, that it is each The control instruction of a actuator is that total security maintains an equal level slow change.
Here v is the longitudinal velocity of AUV, [a1,a2] be soft-constraint effect speed transition section, speed a1、a2It will basis The efficiency of actual thrust device and rudder determines that T represents thrust of hanging down with speed of a ship or plane relationship, and δ represents rudder angle, and thrust is promoted mainly in P expressions, then Constraint is as follows：
In step 5, in order to which the Δ U for finding best minimizes J, by state output：
Cost function is brought into be expressed as J:
The first derivative of function J is：
Become the condition that J is minimized：
To sum up, it overdrives Intelligent Underwater Robot (Autonomous Underwater the present invention is to provide one kind Vehicle, AUV) switching executing agency when control method.The present invention includes：Establish AUV vertical plane kinetics equations；It establishes AUV diving plane state-space models；During low speed executing agency using promote mainly with vertical channels propeller, using master during high speed It pushes away and the rudder wing；A transitional region is set at high low speed switching point, in transitional region, by defining one to executing agency Soft-constraint makes vertical channels propeller and the rudder wing smooth variation in this section, and keeps Angle of Trim constant；Using model prediction Control method calculates according to control instruction and promotes mainly thrust, thrust of hanging down and rudder angle；It brings control input into model and obtains AUV Posture information.The present invention can avoid the speed as caused by switching AUV executing agencies suddenly and trim from generating by a relatively large margin Shake, reduce control occur oscillation amplitude, ensure AUV the smoothed of motion state is kept while executing agency is switched It crosses, and work capacities of the AUV in handoff procedure is unaffected to be ensured for the control of trim.

Claims (6)

  1. The AUV executing agencies method for handover control 1. one kind is overdrived, it is characterised in that：Step is as follows：
    Step 1：According to the hydrodynamic force coefficient of AUV, establish AUV relative to the advance and retreat of satellite coordinate system, heave, pitching movement side Journey；
    Step 2：Main state variables are found, by retreating, heaving, the kinematical equation of pitching and kinetics equation derive becomes shape State space model, and state-space model discretization is obtained by augmented state space model according to the sampling rate of controller；
    Step 3：The cost function of quadratic term form is defined according to augmented state space model, cost function is controller setting value Error between system in future track inputs the sum of change in future amplitude with control；
    Step 4：It is set separately at a high speed and the speed interval of low speed, and in boundary setting transition speed section, during by high speed Executing agency be set as promoting mainly with the rudder wing, executing agency during by low speed, which is set as promoting mainly and hang down, to push away, and transitional region executing agency is set as Promote mainly, hang down push away with the rudder wing, transitional region for rudder wing angle and hang down thrust set a soft-constraint, make thrust and rudder angle in mistake It crosses in section and seamlessly transits；
    Step 5：Cost function minimum is made to solve control to input and act in system, speed, trim, the depth of obtained AUV Spend information；
    Step 6：The posture information that step 5 obtains judges whether to meet control instruction as feedback：If satisfied, control terminates；If It is unsatisfactory for, posture information is supplied to state-space model and cost function, obtain prediction locus, perform step 5, until meeting Control instruction.
  2. The AUV executing agencies method for handover control 2. one kind according to claim 1 is overdrived, it is characterised in that：Step 1 has Body is：
    AUV is heaved, the kinematical equation of pitching is relative to the advance and retreat of satellite coordinate system：
    Kinetics equation is：
    Wherein：X and z is longitudinal velocity and vertical velocity under earth coordinates respectively, and θ is the Angle of Trim of AUV, and m is the matter of AUV Amount, IyIt is the moment of inertia of the hull for y-axis；Xλ、Zλ、Xf、ZfIt is fluid inertia force item and drag as caused by u, w speed term , XrWith ZrIt is the drag that the rudder wing generates；Mq、MhIt is the turning moment and hull torque that AUV hydrodynamic forces generate, MrIt is the rudder wing The torque of generation；Xp、ZTf、ZTrItem is tail portion main thruster, head and the tail vertical channels propeller thrust；LTf、LTrIt is head and the tail vertical channel The arm of force of road propeller thrust；G, ▽ represents the gravity and displacement of volume of hull；G represents acceleration of gravity；Δ z represents theirs Center is to the distance of Z axis.
  3. The AUV executing agencies method for handover control 3. one kind according to claim 2 is overdrived, it is characterised in that：Step 2 In, state-space model is：
    Wherein： ρ is the density of water, with footmark items α and β be respectively measure with the relevant constant of hydrodynamic force, δ is rudder Angle；
    State-space model is reduced to：
    Wherein：
    State-space model discretization is obtained into augmented state model：
    Wherein, I is unit matrix.
  4. The AUV executing agencies method for handover control 4. one kind according to claim 3 is overdrived, it is characterised in that：Step 3 In：The cost function of quadratic term form is defined by augmented state space model, for a given preset value r (ki), in sample This time kiIt is interior, in prediction space, the target of Predictive Control System, be by the output terminal of prediction as close possible to preset value, Assuming that preset value remains unchanged in window is optimized, NPIt is respectively prediction step and control step-length with Nc, then with NPFor length, Vector comprising preset value is：
    Cost function J is defined to reflect control targe：
    In formula：U is control output,It is a diagonal matrix,rw>=0, rwIt is penalty factor.
  5. The AUV executing agencies method for handover control 5. one kind according to claim 4 is overdrived, it is characterised in that：
    In step 4, it is contemplated that the limitation of dead zone and motor speed uses restraint to promoting mainly thrust：
    Xmin≤Xp≤Xmax
    ΔXmin≤ΔXp≤ΔXmax
    Constraint to thrust of hanging down：
    Tmin≤Tp≤Tmax
    ΔTmin≤ΔTp≤ΔTmax
    Constraint to rudder angle：
    δmin≤δp≤δmax
    Δδmin≤Δδp≤Δδmax
    The method that prediction output obtains is to obtain predicted state variable using augmented state space model：
    x(ki+2|ki)=Ax (ki+1|ki)+BΔu(ki+1)
    =A2x(ki)+ABΔu(ki)+BΔu(ki+1)
    Wherein,
    Obtaining prediction output from predicted state variable is：
    y(ki+1|ki)=CAx (ki)+CBΔu(ki)
    y(ki+2|ki)=CA2x(ki)+CABΔu(ki)+CBΔu(ki+1)
    y(ki+3|ki)=CA3x(ki)+CA2BΔu(ki)+CABΔu(ki+1)+CBΔu(ki+2)
    Wherein, C=[0m I]
    Definition vector：
    Y=[y (ki+1|ki) y(ki+2|ki) y(ki+3|ki) ... y(ki+Np|ki)]T
    Y represents state output, and following control domain Δ U, length NC, it is expressed as：
    Δ U=[Δ u (ki) Δu(ki+1) Δu(ki+2) ... Δu(ki+Nc-1)]T
    It can obtain：
    Wherein：
  6. The AUV executing agencies method for handover control 6. one kind according to claim 5 is overdrived, it is characterised in that：Step 5 In, in order to which the Δ U for finding best minimizes J, by state output：
    Bringing cost function into, then J is:
    The first derivative of function J is：
    Become the condition that J is minimized：
CN201810057813.2A 2018-01-22 2018-01-22 Over-driven AUV (autonomous Underwater vehicle) actuating mechanism switching control method Active CN108227502B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810057813.2A CN108227502B (en) 2018-01-22 2018-01-22 Over-driven AUV (autonomous Underwater vehicle) actuating mechanism switching control method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810057813.2A CN108227502B (en) 2018-01-22 2018-01-22 Over-driven AUV (autonomous Underwater vehicle) actuating mechanism switching control method

Publications (2)

Publication Number Publication Date
CN108227502A true CN108227502A (en) 2018-06-29
CN108227502B CN108227502B (en) 2020-09-25

Family

ID=62668133

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810057813.2A Active CN108227502B (en) 2018-01-22 2018-01-22 Over-driven AUV (autonomous Underwater vehicle) actuating mechanism switching control method

Country Status (1)

Country Link
CN (1) CN108227502B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108873925A (en) * 2018-08-02 2018-11-23 深圳市吉影科技有限公司 A kind of underwater unmanned plane determines pitch movement control method and its device
CN112666832A (en) * 2020-12-23 2021-04-16 大连海事大学 Non-periodic communication underwater glider cooperative controller structure and design method

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102385316A (en) * 2011-09-16 2012-03-21 哈尔滨工程大学 Deepening controlling method of underactuated automatic underwater vehicle based on neural network back stepping method
US8205570B1 (en) * 2010-02-01 2012-06-26 Vehicle Control Technologies, Inc. Autonomous unmanned underwater vehicle with buoyancy engine
CN104317197A (en) * 2014-10-16 2015-01-28 哈尔滨工程大学 Multi-model switching control method aiming at model change after task load arrangement of AUV (Autonomous Underwater Vehicle)
CN105511480A (en) * 2016-02-25 2016-04-20 哈尔滨工程大学 Under-actuated UUV (unmanned underwater vehicle) depth control method based on artificial bee colony optimized model predication
CN107024863A (en) * 2017-03-24 2017-08-08 哈尔滨工程大学 A kind of UUV Trajectory Tracking Control methods for avoiding differential from exploding

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8205570B1 (en) * 2010-02-01 2012-06-26 Vehicle Control Technologies, Inc. Autonomous unmanned underwater vehicle with buoyancy engine
CN102385316A (en) * 2011-09-16 2012-03-21 哈尔滨工程大学 Deepening controlling method of underactuated automatic underwater vehicle based on neural network back stepping method
CN104317197A (en) * 2014-10-16 2015-01-28 哈尔滨工程大学 Multi-model switching control method aiming at model change after task load arrangement of AUV (Autonomous Underwater Vehicle)
CN105511480A (en) * 2016-02-25 2016-04-20 哈尔滨工程大学 Under-actuated UUV (unmanned underwater vehicle) depth control method based on artificial bee colony optimized model predication
CN107024863A (en) * 2017-03-24 2017-08-08 哈尔滨工程大学 A kind of UUV Trajectory Tracking Control methods for avoiding differential from exploding

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
XIANBO XIANG: "Smooth transition of AUV motion control: From fully-actuated to under-actuated configuration", 《ROBOTICS AND AUTONOMOUS SYSTEMS》 *
万磊: "复杂环境下的欠驱动智能水下机器人定深跟踪控制", 《上海交通大学学报》 *
周焕银: "基于AUV 垂直面运动控制的状态增减多模型切换", 《哈尔滨工程大学学报》 *
周焕银: "基于权值范围设置的多模型稳定切换控制研究", 《控制与决策》 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108873925A (en) * 2018-08-02 2018-11-23 深圳市吉影科技有限公司 A kind of underwater unmanned plane determines pitch movement control method and its device
CN108873925B (en) * 2018-08-02 2021-08-24 深圳市吉影科技有限公司 Fixed pitch angle motion control method and device for underwater unmanned aerial vehicle
CN112666832A (en) * 2020-12-23 2021-04-16 大连海事大学 Non-periodic communication underwater glider cooperative controller structure and design method
CN112666832B (en) * 2020-12-23 2022-08-30 大连海事大学 Non-periodic communication underwater glider cooperative controller system and design method

Also Published As

Publication number Publication date
CN108227502B (en) 2020-09-25

Similar Documents

Publication Publication Date Title
Zhou et al. Adaptive robust sliding mode control of autonomous underwater glider with input constraints for persistent virtual mooring
Huang et al. Double-loop sliding mode controller with a novel switching term for the trajectory tracking of work-class ROVs
Skjetne et al. A nonlinear ship manoeuvering model: Identification and adaptive control with experiments for a model ship
US7366593B2 (en) Method and system for maneuvering movable object
Ihle et al. Formation control of marine surface craft: A Lagrangian approach
Lee et al. Discrete-time quasi-sliding mode control of an autonomous underwater vehicle
CN102768539B (en) AUV (autonomous underwater vehicle) three-dimension curve path tracking control method based on iteration
Caharija et al. Relative velocity control and integral LOS for path following of underactuated surface vessels
Kong et al. Dynamics modeling and motion control of an new unmanned underwater vehicle
CN108563130A (en) A kind of automatic berthing control method of underactuated surface vessel adaptive neural network, equipment and medium
CN112147899A (en) Underwater robot autonomous obstacle avoidance control method based on fuzzy sliding mode algorithm
CN110244556A (en) Based on the modified underactuated surface vessel course heading control method of desired course
CN108227502A (en) One kind is overdrived AUV executing agencies method for handover control
Liu et al. A hierarchical disturbance rejection depth tracking control of underactuated AUV with experimental verification
Zhou et al. Dynamic modeling and motion control of a novel conceptual multimodal underwater vehicle for autonomous sampling
Liu et al. Modified adaptive complementary sliding mode control for the longitudinal motion stabilization of the fully-submerged hydrofoil craft
CN109814583B (en) Dynamic state feedback control method based on autonomous underwater robot course motion
CN114967714A (en) Anti-interference motion control method and system for autonomous underwater robot
Li et al. Adaptive multi-mode switching strategy for the spherical underwater robot with hybrid thrusters
Dai et al. Dual closed loop AUV trajectory tracking control based on finite time and state observer
KR101690154B1 (en) Waypoint tracking apparatus of autonomous underwater system and method thereof
Astrov et al. Wind force model and adaptive control of catamaran model sailboat
Blintsov et al. Mathematical modeling of autonomous underwater vehicle propulsion and steering complex operation in oblique (beveled) water flow
CN106767778A (en) A kind of aircushion vehicle dead reckoning control method
Andersson Automatic Tuning of Motion Control System for an Autonomous Underwater Vehicle

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant