JP4968641B2 - Structure position / orientation control method, structure position / orientation control system, and program - Google Patents

Description

  The present invention relates to a structure / floor control method for maintaining the position and orientation of a structure that remains constant even when a suspended structure or the like floats in the sea or the ocean, for example, a stopped ship. The present invention relates to an azimuth control system and a program that causes a computer to execute the control method.

In some cases, it is desirable to keep the position and orientation of a ship constant even when a ship or the like navigating offshore stops and is subjected to disturbances such as waves, tidal currents, and winds. Also, in the case of an exploration ship that navigates to the exploration destination by self-propelled and searches for the seabed resources at the destination, the position of the exploration ship stopped at the destination and the direction at that time, such as waves, tidal currents, winds, etc. It is necessary to keep it constant without being affected by disturbance.
Such a ship is provided with a control function for keeping the position and orientation of the ship constant.

In the control for maintaining the position and direction, the ship is based on the ship position information (position information) obtained from the positioning sensor attached to the ship and the direction information obtained from the gyrocompass attached around the center of gravity of the ship. The ship's position and azimuth are set in the ship so that the position and azimuth of the ship are maintained at the set target position and azimuth even if they are affected by tidal currents, winds, and waves Integrated control of multiple thrust generators (actuators).
At that time, the integrated control of multiple thrust generators is based on the control force that should act on the center of gravity of the ship from the current position information and direction information of the ship and information on the ship's environmental conditions such as tidal current, wind, and waves. Is calculated and distributed to each of the plurality of thrust generators. The control force includes a force in the surge direction (force in the traveling direction of the ship), a force in the sway direction (force in the direction orthogonal to the traveling direction of the ship), and a moment around the center of gravity of the ship. That is, the thrust and the thrust direction applied to each thrust generator are calculated and controlled so that these control forces act on the position of the center of gravity of the ship.

Patent Document 1 below discloses a thrust generator control method for obtaining a thrust distribution to a thrust generator from a total thrust and a total moment to be applied to a ship by a thrust distribution calculation. Specifically, in the thrust distribution calculation, when calculating the sum of squares of the components of the thrust array vector, the weighted sum of squares obtained by multiplying the square value of each component by a weighting coefficient is used as an evaluation function, and each of the obtained thrusts generates thrust. While adjusting the weighting coefficient so as not to exceed the rated thrust of the generator, the thrust array vector is calculated so that the evaluation function is minimized in this condition.
However, in this thrust distribution calculation, the weighting coefficient is repeatedly adjusted so that the thrust to be obtained does not exceed the rated thrust of the thrust generator, so that it takes time to calculate the thrust array vector. Further, it is impossible to know in advance the number of repetitions of adjustment of the weighting coefficient that is performed so as not to exceed the rated thrust. For this reason, there exists a problem that control by quick thrust distribution cannot be performed. Furthermore, the thrust distribution calculation does not limit the operating range of the thrust generator in the thrust direction, so there is a shield in the thrust direction, or the thrust direction of the two thrust generators is the direction of the other thrust generator. In some cases, the thrusts facing each other interfere with each other. In this case, there also arises a problem that the thrust is reflected on the shield and interferes with its own thrust, and the desired control cannot be performed due to the mutual interference of the thrust.

  Patent Document 2 below discloses a control thrust distribution device that minimizes a change in the swing thruster angle, which is the thrust direction, when distributing the total thrust and the total moment. In this device, this evaluation function is minimized by using an evaluation function obtained by adding the sum of squares of the change amount of the thruster angle, the penalty function related to the thrust balance conditional expression, and the penalty function related to the moment balance conditional expression. As described above, the optimum solution of the thrust that minimizes the evaluation function is obtained while adjusting the value of the parameter used for the penalty function. That is, the optimum solution of thrust is obtained by repeatedly performing calculations while adjusting the parameter values. For this reason, similarly to Patent Document 1, it takes time to calculate the thrust array vector, and the time for finding the optimal solution is not constant, so that there is a problem that control by rapid thrust distribution cannot be performed.

JP 2001-219899 A JP-A-9-2393

  Therefore, in order to solve the above problems, the present invention performs position and orientation control of a structure that maintains the position and orientation of the structure constant even when the structure floating in or on the space is subjected to disturbance. When performing, the control force to be applied to the center of gravity of the structure is supplied, and at the same time, the thrust and the thrust direction can be distributed online, and the thrust direction is limited so that there is no interference between the thrusts. An object of the present invention is to provide a structure position / orientation control method for controlling thrust and thrust direction distribution, a program for executing the method by a computer, and a structure position / orientation control system for executing the method. .

In order to achieve the above object, the present invention provides a position / orientation control method for a structure that maintains the position and orientation of the structure constant even when the structure floating in or on the space is subjected to disturbance. The structure is provided with a plurality of actuators that control and output a desired thrust in a desired thrust direction with respect to a reference direction of the structure, and further, the operating range of the plurality of actuators in the thrust direction is limited. The vector of the component force when the thrust to be applied to each of the plurality of actuators to which the operating range in the thrust direction is restricted is decomposed in two directions orthogonal to each other is defined as an unknown, and the component force The result of applying a matrix defined by the information on the installation positions of the plurality of actuators to the vector is the center of gravity of the structure to maintain the position and orientation of the structure. A thrust distribution equation equal to the control force vector to be used is defined, and a first solution of the distribution equation having an unknown vector of the thrust component force of the actuator whose operating range in the thrust direction is limited is defined as The operating range in the thrust direction of the plurality of actuators is not limited, and the second solution of the distribution equation when the sum of squares of the thrust is minimized is obtained when singular value decomposition of the matrix is performed. When the obtained orthogonal matrix is expressed as a linear sum obtained by adding the processing vectors obtained by acting on the desired adjustment vector,
Obtaining the adjustment vector based on a limitation of each operating range of the thrust direction given to the plurality of actuators;
When controlling the position / orientation of the structure, the control force is received, the second solution is obtained based on the control force vector, and the orthogonal matrix is applied to the adjustment vector. Adding the obtained processing vector to the second solution to obtain the first solution, and setting the component of the first solution as a component of thrust to be applied to the plurality of actuators. The structure position / orientation control method is provided.

When obtaining the adjustment vector, the limit of the operating range in the thrust direction is expressed by an inequality using the component of the adjustment vector, and this inequality is used as a constraint condition. Under this constraint condition, the sum of squares of the components of the adjustment vector is It is preferable to obtain the component of the adjustment vector so as to be minimized. At that time, the control force is provided with an upper limit value and a lower limit value, and the control force represented in the constraint condition is replaced with the upper limit value and the lower limit value of the control force to obtain a plurality of inequalities. It is preferable to obtain the component of the adjustment vector using an inequality.
Further, the adjustment vector is obtained in advance before the control of the position / orientation of the structure is received upon receiving the control force to be applied to the center of gravity of the structure, and the first solution is the structure When the position / orientation is controlled, it is preferable to receive the input of the control force to be applied to the center of gravity of the structure, and to calculate using the control force and the adjustment vector or the processing vector.

  The structure body is, for example, a floating body floating in the sea having the reference direction, and the control force preferably includes a component force in two directions on a plane and a moment about a center of gravity axis perpendicular to the plane of the floating body. .

  Furthermore, the present invention provides a position / orientation control system for a structure that maintains the position and orientation of the structure constant even when the structure floating in or on the space is subjected to disturbance, A plurality of actuators that are provided at a position and control a desired thrust in a desired thrust direction with respect to a reference direction of the structure, and are limited in an operating range in the thrust direction; and A control device for controlling the thrust and the thrust direction of the plurality of actuators, and the control device orthogonally crosses the thrusts to be applied to each of the plurality of actuators limited in the operating range in the thrust direction. The vector of the component force when decomposed in two directions is an unknown, and a matrix defined by the information on the installation positions of the plurality of actuators is applied to this component vector. This results in a thrust distribution equation that is equal to the control force vector to be applied to the center of gravity of the structure to maintain the position and orientation of the structure, and the operating range in the thrust direction is limited. The first solution of the distribution equation with the vector of the thrust component of the actuator as an unknown is not limited in the operating range in the thrust direction of the plurality of actuators, and the sum of squares of the thrust is minimized. When the second solution of the distribution equation is expressed as a linear sum obtained by adding a processing vector obtained by applying an orthogonal matrix obtained by performing singular value decomposition of the matrix to a desired adjustment vector, The adjustment vector is obtained based on each operating range in the thrust direction given the restriction, and this adjustment vector or a processing vector obtained by applying the orthogonal matrix to this adjustment vector When controlling the position / orientation of the processing unit to be stored and the structure, the control force is received, the second solution is obtained based on the control force vector, and the adjustment vector is The processing vector is obtained by operating the orthogonal matrix, or the processing vector stored and read is read out, and the obtained processing vector is added to the second solution to obtain a first solution, And a control unit for controlling the actuator by determining a thrust and a thrust direction for controlling the actuator from the component of the thrust as a component of the thrust applied to the plurality of actuators. A position / orientation control system for a structure is provided.

The processing unit represents the limitation of the operating range in the thrust direction as an inequality using the component of the adjustment vector, and the inequality is used as a constraint condition. Under this constraint condition, the sum of squares of the component of the adjustment vector is minimum. It is preferable to obtain the component of the adjustment vector. At that time, the control force is provided with an upper limit value and a lower limit value, and the control force represented in the constraint condition is replaced with the upper limit value and the lower limit value of the control force to obtain a plurality of inequalities. It is preferable to obtain the component of the adjustment vector using an inequality.
The processing unit receives an input of a control force to be applied to the center of gravity of the structure and calculates an adjustment vector in advance before controlling the position / orientation of the structure. When the position / orientation is controlled, an input of a control force to be applied to the center of gravity of the structure is received, and the first solution is calculated using the control force and the adjustment vector or the processing vector. It is preferable.

  The structure is, for example, a floating body floating in the sea having the reference direction, and the control force includes a component force in two directions on a plane and a moment about the center of gravity axis perpendicular to the plane of the floating body. It is preferable.

Furthermore, the present invention can be executed by a computer that controls the position and orientation of a structure so that the position and orientation of the structure are kept constant even when the structure floating in or on the space is subjected to disturbance. The structure is provided with a plurality of actuators for controlling and outputting a desired thrust in a desired thrust direction with respect to a reference direction of the structure, and the thrust direction of the plurality of actuators. The operation range is limited to a certain amount, and the vector of the component force when the thrust to be applied to each of the plurality of actuators to which the operation range in the thrust direction is limited is decomposed in two directions orthogonal to each other is defined as an unknown quantity. The result of applying a matrix defined by the information on the installation positions of the plurality of actuators to the component force vector is the position and orientation of the structure. A thrust distribution equation equal to the control force vector to be applied to the center of gravity of the structure to be held, and the vector of the thrust component force of the actuator whose operating range in the thrust direction is limited as an unknown The first solution of the distribution equation is the second solution of the distribution equation when the operating range in the thrust direction of the plurality of actuators is not limited and the sum of squares of the thrust is minimized. When expressed as a linear sum obtained by adding the processing vectors obtained by applying the orthogonal matrix obtained when performing singular value decomposition of the matrix to the desired adjustment vector,
Based on each operating range of the thrust direction given to the plurality of actuators, the calculation means of the computer is calculated by the computer, and the adjustment vector or the processing vector obtained by applying the orthogonal matrix to the adjustment vector is calculated by the computer. A procedure for storing in the storage means;
In response to the input of the control force, the calculation means of the computer calculates the second solution based on the vector of the control force, and reads the adjustment vector stored in the storage means. The processing vector obtained by operating the orthogonal matrix or the processing vector read from the storage means is obtained, and the first solution obtained by adding the processing vector to the second solution is used as the computing means of the computer. And a procedure for calculating the thrust and the thrust direction applied to the plurality of actuators from the component of the first solution and applying the thrust and the thrust direction to the plurality of actuators. To do.

When calculating the adjustment vector, the limitation of the operating range in the thrust direction is represented by an inequality using the component of the adjustment vector, and the calculation means of the computer creates the inequality as a constraint condition. It is preferable to cause the calculation unit of the computer to calculate the component of the adjustment vector so that the sum of squares of the component of the adjustment vector is minimized. At that time, the control force is provided with an upper limit value and a lower limit value, and the control force represented in the constraint condition is replaced with the upper limit value and the lower limit value of the control force to obtain a plurality of inequalities. It is preferable to cause the computing unit of the computer to calculate the component of the adjustment vector using an inequality.
The calculation of the adjustment vector is performed before the position / orientation of the structure is controlled by receiving the input of the control force to be applied to the center of gravity of the structure, and the calculation is performed on the calculated adjustment vector or the adjustment vector. The processing vector subjected to the processing is stored and held in the storage means, and when controlling the position / orientation of the structure, an input of a control force to be applied to the center of gravity of the structure is received. It is preferable to cause the calculation means to calculate the first solution using the adjustment vector or the processing vector stored and held.

  The structure is, for example, a floating body floating in the sea having the reference direction, and the control force includes a component force in two directions on a plane and a moment about the center of gravity axis perpendicular to the plane of the floating body. It is preferable.

  In the present invention, when the optimum solution (first solution) relating to the thrust distributed to each actuator is obtained, the adjustment vector used for this solution is obtained based on the limitation of each operating range in the thrust direction given to the actuator. A solution satisfying the condition in which the thrust direction is limited can be obtained. In addition, the adjustment vector can be calculated in advance offline regardless of the control force before the control is performed upon receiving the input of the control force that changes depending on time. For this reason, the calculation for calculating the optimum solution, which is the first solution, is only a short-time calculation according to the following formula (5), which will be described later. Can be calculated in a short time. Accordingly, the thrust can be distributed online.

  Hereinafter, a structure position / orientation control method, structure position / orientation control system, and program according to the present invention will be described in detail with reference to the preferred embodiments shown in the accompanying drawings.

FIG. 1 is a schematic configuration diagram of an example of a structure position / orientation control system according to the present invention that implements the structure position / orientation control method according to the present invention.
In the structure position / orientation control system (hereinafter referred to as system) 10 shown in FIG. 1, a ship is applied as a structure. In addition to the ship, a bridge platform platform floating on the ocean, a submarine resource exploration ship, underwater navigation The present invention can also be applied to a running body, a flying object flying in the air, and an artificial satellite in outer space.

  The system 10 includes an actuator group 12 that is mounted on a ship and outputs a controllable thrust in a controllable direction, and a control device 14 that is mounted on the ship and controls the actuator group 12 and is configured by a computer. Configured. The system 10 optimally distributes the thrust and thrust direction of each actuator so that the ship floating on the ocean is kept constant in the ship's position (position) and direction even if it is subjected to wind, wave and tidal disturbances.

  The actuator group 12 includes a plurality of actuators 1 to n that are provided at a plurality of bottom positions of the ship and output a desired thrust in a desired direction (thrust direction) with respect to the reference direction of the ship. The reference direction of the ship is, for example, a surge direction (front-rear direction of the ship, x direction in FIG. 2). The actuators 1 to n are, for example, azimuth thrusters that are provided on the bottom of the ship and in which a rotation shaft of a propeller that generates thrust rotates in a desired direction. In addition to the azimuth thruster, the actuator group 14 may include a side thruster that generates thrust only in the lateral direction of the ship (sway direction). However, the actuator group 12 includes two or more azimuth thrusters that output at least controllable thrust in a controllable direction (thrust direction).

The control device 14 includes various memories 16 such as RAM and ROM, and a CPU 18, and by executing a program stored in the memory 16, a control force calculation unit 20, a preprocessing unit 22, and a control whose functions are modularized are executed. A unit 24 is formed.
The control force calculation unit 20 acquires information on the current position supplied using a GPS (Global Positioning System) (not shown) mounted on the ship, and is supplied using a gyrocompass provided on the ship. The direction information of the surge direction of the ship (ship direction) is acquired. Moreover, the environmental information obtained by measuring instruments, such as an anemometer provided in the ship, is acquired. These pieces of information are digitized at a predetermined sampling interval in the control force calculation unit 20 and used for calculation of the control force.

The control force calculation unit 20 uses the acquired information on the current position and direction of the ship for feedback control via the Kalman filter, and uses the acquired environment information for feed-forward control to affect the center of gravity of the ship. The power in the power surge direction, the force in the sway direction, and the moment about the center of gravity axis of the ship are calculated as control forces. The control force calculation unit 20 supplies the calculated control force to the control unit 24 at a predetermined cycle.
The method for calculating the control force performed in the control force calculation unit 20 is not particularly limited in the present invention, but the position and direction are kept constant based on at least the current ship position and direction information and environment information. As described above, the control force to be applied to the center of gravity is calculated according to the feedback control and feedforward control scheme.

  The preprocessing unit 22 offline adjusts the adjustment vector used when calculating the thrust and the thrust direction to be applied to each actuator of the actuator group 12 before controlling the position and direction of the ship in response to the input of the control force. This is a part to be calculated in advance. The calculated adjustment vector is stored and held in the memory 16. The operating range of each actuator in the actuator group 12 in the thrust direction is limited according to the input by the operator. The operating range in the thrust direction is limited by the two actuators facing each other so that the reflected force does not interfere with the thrust due to the instruments provided in the thrust direction acting as a barrier. This is because the change of the thrust direction is small and the movement time for the actuator to change the direction is short. Based on the limitation of the operating range in the thrust direction, an adjustment vector is calculated as described later. The adjustment vector and the calculation method thereof will be described later in detail based on a specific example shown in FIG.

The control unit 24 reads the adjustment vector calculated by the preprocessing unit 22 and stored in the memory 16, and uses the adjustment vector and the control force supplied from the control force calculation unit 20 at a predetermined cycle. This is a part for calculating the thrust applied to each actuator of the actuator group 12 and the thrust direction. Specifically, within the limited operating range of the thrust direction, the thrust component to be applied to each actuator, ie, the surge direction (reference direction) of the thrust to be applied, and the sway direction perpendicular thereto are decomposed. The solution of the vector related to the thrust component force is calculated by adding the processing vector to the solution of the force component vector that minimizes the sum of the squares of the thrusts of the actuator group 12. Here, the solution of the thrust component vector that minimizes the sum of squares of the thrusts of the actuator group 12 is a solution when there is no constraint including the constraint of the operating range in the thrust direction. The processing vector is a vector obtained by applying an orthogonal matrix obtained when performing singular value decomposition of an extended thrust distribution matrix described later to an adjustment vector described later. A more specific description will be described later based on the specific example shown in FIG.
The control unit 24 determines the thrust and the thrust direction from the calculated thrust component, and supplies the thrust and the thrust direction as a control signal to a drive unit (not shown) of each actuator.

Processing performed in the preprocessing unit 22 and the control unit 24 will be described in more detail.
FIG. 2 is an explanatory diagram of the ship 30 to be controlled by the system 10. Three azimuth thrusters (shown by ● in FIG. 2) 32, 34, and 36 are provided on the bottom of the ship 30. The azimuth thrusters 32 and 34 are provided on both the left and right sides of the rear side of the ship, and are separated from the center of gravity G of the ship by distances l _x1 and l _{x2 in the} surge direction (x direction), respectively, and l _{y1 in the} sway direction (y direction). , L _y2 away. Further, the azimuth thruster 36 is provided in front of the ship, and is separated from the center of gravity G of the ship by a distance l _x3 forward in the surge direction (x direction) and is separated by a distance l _{y3 in} the sway direction (y direction). The thrusts of the azimuth thrusters 32, 34, and 36 are represented by thrusts T ₁ , T ₂ , and T ₃ , respectively, and the thrust directions are represented by angles δ ₁ , δ ₂ , and δ ₃ with respect to the surge direction, respectively.
In the ship 30, the control force calculated by the control force calculation unit 20, that is, the force in the surge direction (x direction) to be applied to the center of gravity G is τ _x and the sway direction (y direction) to be applied to the center of gravity G. When the force is τ _y , the yaw (moment) about the center of gravity axis perpendicular to the plane including the surge direction and the sway direction is τ _z , the relationship is expressed by the following equation (1). The matrix A is a thrust distribution matrix that distributes thrust to each azimuth thruster.

The components of the matrix A, which is the thrust distribution matrix, include angles δ ₁ , δ ₂ , and δ ₃ representing the thrust direction of the azimuth thruster, so the characteristics of the matrix A are represented by angles δ ₁ , δ ₂ , and δ ₃ . Dependent. Therefore, a vector τ (control force vector) of forces τ _x , τ _y and moment τ _z to be applied to the center of gravity G is used, and the thrust to be applied to each of the azimuth thrusters is divided into the surge direction and the sway direction. A vector of component forces (T _1x , T _1y , T _2x , T _2y , T _3x , T _3y ) is represented by T, and the relationship between the vector τ and the vector T is represented. This relationship is defined by information on the installation positions of the azimuth thrusters (distances l _x1 , l _x2 , l _y1 , l _y2 , l _x3 , l _y3 ), and is a matrix A _ex that does not depend on the angles δ ₁ , δ ₂ , δ _3. Specifically, it is represented by the following formulas (2) and (3). The matrix A _ex in the equation (2) is a matrix obtained by extending the matrix A in the equation (1), and is referred to as an extended thrust distribution matrix. Since this extended thrust distribution matrix is defined by information on the installation positions of the azimuth thrusters (distances l _x1 , l _x2 , l _y1 , l _y2 , l _x3 , and l _y3 ), the position of the azimuth thruster of the ship 30 is determined. If so, it is a matrix to be determined. The following equation (3) is a thrust distribution equation in the present invention.

From this, the vector T of the thrust component, which is an unknown, is calculated using the vector τ of the control force supplied from the control force calculation unit 20 and the known matrix A _ex . That is, the distribution equation relating to the vector T expressed by Expression (3) is solved. Since the matrix A _ex has a matrix size of 3 × 6, the solution of the vector T cannot be uniquely obtained for the distribution equation of Expression (3). For this reason, the norm of the vector T is limited to the operating range in the thrust direction, that is, the operating range in which the angles δ ₁ , δ ₂ , and δ ₃ that are the swing angles of the azimuth thrusters 32, 34, and 36 are set. An optimum solution is calculated that is as small as possible (the sum of squares of the components of the vector T is as small as possible). Conventionally, when calculating the solution of the vector T, the optimal solution in which the norm of the vector T is as small as possible is obtained without limiting the operating range of the angles δ ₁ , δ ₂ , and δ _3. Different from the invention.

More specifically, as shown in the following equation (4), the matrix A _ex is subjected to a well-known singular value decomposition, and the diagonal matrix Σ ₁ having the matrix A _ex as a diagonal component and the orthogonal matrix U ₁ , Decompose using U ₂ , V ₁ , V ₂ . Using this singular value decomposition, an optimal solution T _ex ^* (first solution) is calculated as shown in the following equation (5). Here, C is an adjustment vector that acts on the orthogonal matrix V ₂ and is added to the expression of the first term. The optimal solution T _ex ^* satisfies Expression (3) because the orthogonal matrices V ₁ and V ₂ have orthogonality. The adjustment vector C is a vector that is known by a method described later. The optimal solution T _ex ^* is the first expressed by using the diagonal matrix Σ ₁ and the orthogonal matrices V ₁ and U ₁ obtained by singular value decomposition for the control force vector τ supplied from the control force calculation unit 20. The term V ₁ Σ ₁ ⁻¹ U ₁ ^T is represented by a linear sum of the second term V ₂ C which is a processing vector represented by using the adjustment vector C and the orthogonal matrix V ₂ . The first term V ₁ Σ ₁ ⁻¹ U ₁ ^T is Lagrange's undetermined constant solution (second solution) that minimizes the sum of squares of the thrust of the azimuth thruster without any constraint, and the optimal solution T _ex ^* Is obtained by adding the processing vector V ₂ C obtained by applying the orthogonal matrix V ₂ to the adjustment vector C to the Lagrange undetermined constant solution.

The square of the norm of the optimal solution T _ex ^* , that is, the square sum of the vector components of the optimal solution T _ex , that is, the square sum of the thrust of each azimuth thruster is determined by the above Lagrange undetermined as shown in Equation (6). The square sum of the constant solution is added to the square of the norm of the adjustment vector C. Since the supplied control force, diagonal matrix Σ ₁ and orthogonal matrix U ₁ have known values, the first term in equation (6) is known. For this reason, in order to make the sum of squares of the thrust of the azimuth thruster as small as possible, the adjustment vector C is adjusted under the condition that the operating range of the angles δ ₁ , δ ₂ , and δ ₃ that are the swing angles of the azimuth thruster is limited. It is necessary to calculate the value of the component of the adjustment vector C so that the sum of squares of the component is minimized.

In the present invention, for the reason described above, the value of the adjustment vector C is calculated so that the sum of squares of the components of the adjustment vector C is minimized under the condition that the operating ranges of the angles δ ₁ , δ ₂ , and δ ₃ are limited. Specifically, when the matrix components of the diagonal matrix Σ ₁ , the orthogonal matrices V ₁ , V ₂ , U ₁ are expressed by the following formulas (7-1) to (7-4), the i th of the optimal solution T _ex ^* These components are written down as shown in the following formula (8).
Since the vector T is constituted by the component force in the surge direction and the sway direction of the thrust as shown in the equation (3), the ratio of the component force in the surge direction and the sway direction should be taken for the optimum solution T _ex ^*. Thus, the thrust direction can be obtained. For this reason, when the operating range in the thrust direction is limited, θ _min ≦ δ _m ≦ θ _max and −π with respect to the angle δ _m (m = 1, 2, 3) in the thrust direction represented by Expression (8). When / 2 ≦ δ _m ≦ π / 2, the conditions of this operating range are expressed by the following formulas (9-1) to (9-3).

Note that τ _k (t) in the equations (9-1) to (9-3) is a control force that is continuously supplied from the control force calculation unit 20 at a constant period and varies with time. For this reason, the inequalities represented by the equations (9-1) to (9-3) change with time. Therefore, the optimal solution T _ex ^*, it is difficult to calculate the thrust and thrust direction short time seeking adjustment vector C under the conditions of inequalities vary with time, therefore, the optimal solution T _ex ^* Online It cannot be sought and controlled. Therefore, in the present invention, the upper limit value and the lower limit value of the component of the vector τ of the control force are set in advance as shown in the following formulas (10-1) and (10-2) by an operator input, By substituting 10-1) and (10-2) into the equations (9-1) to (9-2), it is possible to define a constraint condition represented by an inequality that does not depend on time. The upper limit value and lower limit value of the component of the control force τ _k (t) are set in advance by the operator because the range of the output value of the control force is limited in advance in the control force calculation unit 20. In the present invention, the component of the adjustment vector C that minimizes the sum of squares of the components of the adjustment vector C is calculated under the constraint condition represented by this time-independent inequality. The calculated adjustment vector C is stored and held in the memory 16.

On the other hand, the control unit 24 reads the adjustment vector C stored and held in the memory 16 and uses the adjustment vector C and the control force vector τ supplied from the control force calculation unit 20 to obtain the above equation (5). Therefore, the optimal solution T _ex ^* is calculated. Here, since the orthogonal matrices V ₁ , U ₁ , V ₂ , and the diagonal matrix Σ ₁ are uniquely determined by the known matrix A _ex , in the equation (5) in advance before the control, V ₁ Σ ₁ ⁻¹ U ₁ ^T and the processing vector V ₂ C can also be obtained. When controlling the position and direction of the ship online with respect to the supplied control force, the control force vector unit τ supplied from the control force calculation unit 20 at a constant cycle is used according to the equation (5). The optimal solution T _ex ^* can be quickly calculated by simply calculating. Accordingly, the thrust T _m and the thrust direction angle δ _m (m = 1, 2, 3) can be calculated from the optimum solution T _ex ^* according to the following equations (11-1) and (11-2). .
The control unit 24 converts the calculated thrust T ^* _m and the thrust direction angle δ ^* _m into control signals for the azimuth thrusters 32, 34, and 36, and supplies the control signals to an azimuth thruster drive unit (not shown).

In the above example, the calculated adjustment vector C is stored in the memory 16 and the optimum solution T _ex ^* is calculated according to the equation (5) using the adjustment vector C and the control force vector τ during online control. However, in the present invention, V ₁ Σ ₁ ⁻¹ U ₁ ^T and the processing vector V ₂ C which can be obtained in advance are stored and held in the memory 16, and V ₁ Σ ₁ ⁻¹ U ₁ ^T The processing vector V ₂ C may be read from the memory 16 and the optimal solution T _ex ^* may be calculated according to equation (5). In this case, since it is not necessary to perform the calculation process of V ₁ Σ ₁ ⁻¹ U ₁ ^{T and} the calculation process of the processing vector V ₂ C, the optimal solution T _ex ^* can be calculated in a shorter time.
The configuration of the system 10 has been described above.

FIG. 3 is a diagram for explaining the flow of the position / orientation control method for the ship 30 that is carried out by the system 10.
First, the operating range of the swing angle (thrust direction) of the azimuth thruster is input and set in the system 10 by the operator. Setting the operating range prevents, for example, the instruments provided on the bottom of the ship from blocking the thrust flow and exerting the desired thrust, and the two azimuth thrusters face each other and the mutual thrust interferes. This is to prevent this. Further, an upper limit value and a lower limit value of the control force are input and set by the operator (step S100). Since the system 10 is mounted on the ship 30, position information (distances l _x1 , l _x2 , l _y1 , l _y2 , l _x3 , l _y3 ) of the azimuth thruster (actuator) of the ship is also input simultaneously.

Next, in the preprocessing unit 22, a matrix A _ex which is an extended thrust distribution matrix shown in Expression (2) is set, and the orthogonal matrix V ₁ , U of singular value decomposition performed according to Expression (4) is performed. ₁ , V ₂ , diagonal matrix Σ ₁ , and further using equations (10-1) and (10-2) using upper limit value τ _max and lower limit value τ _min of each component of control force vector τ A constraint condition composed of the inequalities substituted in (9-1) to (9-3) is set (step S102).
Note that, when three azimuth thrusters are mounted, the constraint condition is that either the upper limit value τ _max or the lower limit value τ _min for each of τ ₁ (t), τ ₂ (t), and τ ₃ (t). On the other hand, substitution is made for each of the three inequalities (9-1) to (9-3), so that a constraint condition consisting of a total of 24 (= 3 × 2 ³ ) inequalities is set. The inequalities are _obtained by replacing the control force τ _k (t), which varies with time only in the equations (9-1) to (9-3), with the upper limit value or the lower limit value. It is a fixed inequality that does not depend on.

Next, the component of the adjustment vector C that minimizes the sum of squares of the components of the adjustment vector C is calculated under the constraint conditions that are obtained from the 24 inequalities (step S104). The reason for minimizing the sum of squares of the components of the adjustment vector C is that it is desired to minimize the sum of squares of the components of the optimal solution T _ex ^* , that is, the sum of squares of the thrust of the azimuth thruster. That is, since the driving energy is limited in a marine vessel, it is desired to efficiently control the azimuth thruster by using this driving energy to the minimum (minimizing the sum of squares of the components of the adjustment vector C). Because.
The calculated value of the component of the adjustment vector C is stored and held in the memory 16 (step S106). Further, V ₁ Σ ₁ ⁻¹ U ₁ ^T and the processing vector V ₂ C in the equation (5) are calculated and stored in the memory 16.

  Steps S <b> 100 to S <b> 106 are preprocessing performed before the control force is supplied and the position and direction of the ship 30 are controlled offshore, and are offline processing. In the calculation of the adjustment vector C, a plurality of operating ranges of the swing angle of each azimuth thruster are determined, and a plurality of sets of constraint conditions that variously limit the swing angle of each azimuth thruster are determined. The adjustment vector C under the conditions may be calculated and stored in the memory 16.

Next, even if the ship 30 is stationary on the ocean and the ship receives disturbance, control is performed to keep the position and orientation of the ship constant.
First, the position and orientation information of the ship is supplied from the GPS and gyrocompass provided in the ship 30, and the environment information is supplied from a measuring instrument such as an anemometer to the control force calculation unit 20. The control force calculation unit 20 calculates the control force that should act on the center of gravity G of the ship 30 according to the set algorithm. The calculated control force is supplied to the control unit 24. Since the position and direction of the ship 30 are controlled on-line, the control force that changes depending on time calculated by the control force calculation unit 20 is supplied to the control unit 24 at 2 Hz, for example.
Next, V ₁ Σ ₁ ⁻¹ U ₁ ^T read from the memory 16 and the processing vector V ₂ C are used together with the control force supplied to the control unit 24 to obtain the optimal solution T _ex according to the equation (5). ^* Is calculated (step S108).
In the present invention, the adjustment vector C may be read instead of the processing vector V ₂ C, and the calculation with the orthogonal matrix V ₂ may be performed when calculating the optimum solution T _ex ^* .

Finally, the calculated optimal solution T _ex ^* components ^{_{(T 1x *, T 1 *}} y, T 2x *, T 2y *, T 3x *, T 3y *) is taken out, the formula (11-1), In accordance with (11-2), the thrust T _m ^* , the angle δ _m ^* , (m = 1, 2, 3) are calculated (step S110). These are the thrust to be applied to the azimuth thrusters 32, 34, and 36 shown in FIG. 2 and the direction of the thrust. Information on the thrust and the thrust direction is converted into a control signal and supplied to the drive unit of each azimuth thruster. Is done.

As described above, the vector and matrix other than the control force vector τ in the optimal solution T _ex ^* calculated according to the equation (5) for control can be calculated in advance by offline processing. When the control force is supplied from the unit 20, the optimal solution T _ex ^* can be calculated in a short time by calculating according to the equation (5) using the vector τ of the control force. In each of the conventional methods, since the optimum solution is calculated by repeatedly performing calculations, the thrust and the thrust direction cannot be calculated quickly, and it is difficult to control the position and direction of the ship by online processing. Eliminates this problem. Further, unlike the conventional method, it is possible to limit the thrust direction, so that it is possible to prevent thrust interference and the like. In the conventional method, when the control force is small, the thrust direction of the azimuth thruster may change greatly. However, in the present invention, since the swing angle (thrust direction) of the azimuth thruster is limited, the azimuth thruster is There is no significant change in the swing angle over time to go to the desired orientation. For this reason, the time until the azimuth thruster faces the target azimuth direction is shortened, and efficient and stable control is possible.

When the above-described control is performed on the computer, it is preferable to store the following program in the computer and read and execute it.
That is, as a procedure, the program causes the CPU 18 of the computer to calculate the adjustment vector C based on the operating range in the thrust direction given to a plurality of azimuth thrusters, and the calculated adjustment vector C or the adjustment vector C is an orthogonal matrix V _2. The processing vector V ₂ C on which is applied is stored in the memory 16. As the next procedure, upon receiving the control force, the CPU 18 calculates the Lagrange's undetermined constant solution based on the control force vector τ, and reads the adjustment vector C stored in the memory 16. obtain treated vector V ₂ C an acting orthogonal matrices V ₂ to C, or obtained by reading out a processing vector V ₂ C where the orthogonal matrix V ₂ to the adjustment vector C stored in the memory 16 to act, resulting The CPU 18 calculates the optimal solution T _ex ^* of the linear sum obtained by adding the processing vector V ₂ C to the Lagrange undetermined constant solution. Thrust T _m and thrust direction of the angle [delta] _m providing from optimal solution T _ex ^* components of the linear sum in azimuth thrusters to (m = 1, 2, 3) by calculating the thrust T _m and thrust direction of the angle [delta] _m To the azimuth thruster.
When calculating the adjustment vector C, the components of the adjustment vector C are set so that the sum of squares of the components of the adjustment vector C is minimized under the constraint condition that the limitation of the operating range in the thrust direction is expressed by an inequality using the components of the adjustment vector. Is preferably calculated by the CPU 18. Each component τ _k (t) of the control force vector τ is provided with an upper limit value and a lower limit value, and each component τ _k (t) is replaced with an upper limit value and a lower limit value to obtain a plurality of inequalities, It is preferable to cause the CPU of the computer to calculate the component of the adjustment vector C using the plurality of inequalities. Further, the adjustment vector C is calculated before the position / orientation of the ship is controlled, and the calculated adjustment vector C or the processing vector V ₂ C obtained by applying the orthogonal matrix V ₂ to the adjustment vector C is stored in the memory 16. When the position / orientation of the ship is controlled, the stored adjustment vector C is read and the processing vector V ₂ C is calculated, or the processing vector V ₂ C is read and obtained. The optimal solution T _ex ^* is preferably calculated by the CPU 18 using the control force that should act on the center of gravity of the ship.

  FIG. 4 shows an example of the control force supplied to the control calculation unit 24 of the ship 30. The control force has an upper limit value of 1 (N) or 1 (Nm) and a lower limit value of -1 (N) or -1 (Nm). Further, the installation position information of the azimuth thrusters provided in the ship 30 is shown in Table 1 below.

At this time, the orthogonal matrix U ₁ , the diagonal matrix Σ ₁ and the orthogonal matrices V ₁ and V ₂ are as follows.

On the other hand, as shown in Table 2 below, conditions for the operating range of the swing angle were determined as in Examples 1 to 7. Table 3 shows the adjustment vector C calculated to minimize the norm square and the value of the norm square / 2 at this time.
As can be seen from Examples 1, 2, and 3 in Table 3, the square value of the norm of the adjustment vector C increases as the condition of the operating range of the swing angle is narrowed, that is, as the constraint is tightened. Recognize.

FIGS. 5A to 5D are diagrams showing the thrust and the thrust direction of the azimuth thruster calculated by Lagrange's undetermined constant solution with no constraints on the control force shown in FIG.
6 (a) to 6 (d) give the constraint condition of Example 1 (see Table 2) to the control force shown in FIG. 4 and determine the thrust and thrust direction of the azimuth thruster based on Equation (5). This is the result obtained.
7 (a) to 7 (d) give the constraint condition of Example 3 (see Table 2) to the control force shown in FIG. 4 and determine the thrust and thrust direction of the azimuth thruster based on equation (5). This is the result obtained.
8 (a) to 8 (d) give the constraint conditions of Example 6 (see Table 2) to the control force shown in FIG. 4 and determine the thrust and thrust direction of the azimuth thruster based on equation (5). This is the result obtained.

FIG. 6A to FIG. 7D, FIG. 7A to FIG. 7D, and FIG. 7A to FIG. It can be seen that the variation in the thrust direction calculated by the optimum solution T _ex ^* shown in 8 (a) to (d) is small and varies within the operating range of the given thrust direction. On the other hand, the thrust increases with respect to FIGS. This indicates that the variation in thrust and the variation in thrust direction are in a trade-off relationship, and the variation in thrust is increased by keeping the variation in the thrust direction small. As described above, in the present invention, by restricting the operating range in the thrust direction of the actuator, it is possible to suppress fluctuations in the thrust direction within a predetermined range while suppressing the thrust level within the allowable range. Further, as in the example shown in FIGS. 8A to 8D, the operating range in the thrust direction can be set separately for each azimuth thruster, so that the variation in thrust and the variation in thrust direction are more optimal ranges. Can be suppressed.

  As described above, the structure position / orientation control method, structure position / orientation control system, and program according to the present invention have been described in detail. However, the present invention is not limited to the above-described embodiment, and does not depart from the spirit of the present invention. Of course, various improvements and changes may be made.

It is a schematic block diagram of an example of the structure position / orientation control system of the present invention for implementing the structure position / orientation control method of the present invention. It is explanatory drawing of the ship made into the control object of the system shown in FIG. It is a figure explaining the flow of the position and direction control method of the ship implemented with the system shown in FIG. It is a figure which shows an example of the control force supplied to the control calculation unit of the ship shown in FIG. (A)-(d) is a figure which shows the thrust and thrust direction of an azimuth thruster calculated by Lagrange's undetermined constant solution without a constraint condition with respect to the control force shown in FIG. (A)-(d) is a figure which shows the result of having obtained the conditions of Example 1 with respect to the control force shown in FIG. 4, and having calculated | required the thrust and thrust direction of the azimuth thruster according to the method of this invention. . (A)-(d) is a figure which shows the result of having obtained the conditions of Example 3 with respect to the control force shown in FIG. 4, and having calculated | required the thrust and thrust direction of the azimuth thruster according to the method of this invention. . (A)-(d) is a figure which shows the result of having obtained the conditions of Example 6 with respect to the control force shown in FIG. 4, and having calculated | required the thrust and thrust direction of the azimuth thruster according to the method of this invention. .

Explanation of symbols

10 Position / Orientation Control System 12 Actuator Group 14 Controller 16 Memory 18 CPU
20 control force calculation unit 22 preprocessing unit 24 control unit 30 ship 32, 34, 36 azimuth thruster

Claims (15)

A structure position / orientation control method that maintains a constant position and orientation of a structure even when the structure floating in or on the space is subjected to disturbance,
The structure is provided with a plurality of actuators that control and output a desired thrust in a desired thrust direction with respect to a reference direction of the structure, and further, the operating range of the plurality of actuators in the thrust direction is limited. Given,
A vector of component force when the thrust to be applied to each of the plurality of actuators to which the operating range in the thrust direction is limited is decomposed in two directions orthogonal to each other is defined as an unknown, and A thrust distribution equation in which the result of applying a matrix defined by the information on the installation position of the actuator is equal to the vector of the control force to be applied to the center of gravity of the structure in order to maintain the position and orientation of the structure And the first solution of the distribution equation in which the vector of the thrust component force of the actuator whose operating range in the thrust direction is limited is unknown is limited to the operating range in the thrust direction of the plurality of actuators. There is no orthogonal row obtained when singular value decomposition of the matrix is performed as the second solution of the distribution equation when the sum of squares of the thrust is minimum When the expressed in linear sum obtained by adding a processing vector obtained by acting on the desired adjustment vector,
Obtaining the adjustment vector based on a limitation of each operating range of the thrust direction given to the plurality of actuators;
When controlling the position / orientation of the structure, the control force is received, the second solution is obtained based on the control force vector, and the orthogonal matrix is applied to the adjustment vector. Adding the obtained processing vector to the second solution to obtain the first solution, and setting the component of the first solution as a component of thrust to be applied to the plurality of actuators. The position / orientation control method of the structure.   When obtaining the adjustment vector, the limit of the operating range in the thrust direction is expressed by an inequality using the component of the adjustment vector. Using this inequality as a constraint, the square sum of the components of the adjustment vector is The structure position / orientation control method according to claim 1, wherein the component of the adjustment vector is calculated so as to be minimized. The control force is provided with an upper limit value and a lower limit value,
The control force represented in the constraint condition is replaced with an upper limit value and a lower limit value of the control force to obtain a plurality of inequalities, and the components of the adjustment vector are obtained using the plurality of inequalities. A method for controlling the position and orientation of a structure.   The adjustment vector is obtained in advance before performing control of the position / orientation of the structure upon receiving an input of a control force to be applied to the center of gravity of the structure, and the first solution is a position of the structure. When performing control of the azimuth, receiving an input of a control force that should act on the center of gravity of the structure, the control force and the adjustment vector or a processing vector obtained by applying the orthogonal matrix to the adjustment vector The position / orientation control method for a structure according to any one of claims 1 to 3, calculated by using the method.   The structure is a floating body floating in the sea having the reference direction, and the control force includes a component force in two directions on a plane and a moment about a center of gravity axis perpendicular to the plane of the floating body. The structure position / orientation control method according to any one of the above. A structure position / orientation control system that keeps the position and orientation of a structure constant even when the structure floating in or on space is subjected to disturbance,
The actuator is provided at a plurality of positions of the structure and outputs a desired thrust in a desired thrust direction with respect to a reference direction of the structure, and the operating range in the thrust direction is limited. Multiple actuators;
A control device that controls thrust and thrust direction of the plurality of actuators,
The controller is
A vector of component force when the thrust to be applied to each of the plurality of actuators to which the operating range in the thrust direction is limited is decomposed in two directions orthogonal to each other is defined as an unknown, and A thrust distribution equation in which the result of applying a matrix defined by the information on the installation position of the actuator is equal to the vector of the control force to be applied to the center of gravity of the structure in order to maintain the position and orientation of the structure And the first solution of the distribution equation in which the vector of the thrust component force of the actuator whose operating range in the thrust direction is limited is unknown is limited to the operating range in the thrust direction of the plurality of actuators. There is no orthogonal row obtained when singular value decomposition of the matrix is performed as the second solution of the distribution equation when the sum of squares of the thrust is minimum Is expressed as a linear sum obtained by adding the processing vectors obtained by acting on the desired adjustment vector, the adjustment vector is obtained based on each operating range in the thrust direction given the restriction, and this adjustment vector or this adjustment A processing unit for storing and holding a processing vector obtained by applying the orthogonal matrix to a vector;
When controlling the position / orientation of the structure, the control force is received, the second solution is obtained based on the control force vector, and the orthogonal matrix is applied to the adjustment vector. Obtaining the obtained processing vector or reading the stored processing vector, obtaining the processing vector, adding the obtained processing vector to the second solution to obtain a first solution, and components of the first solution And a control unit for controlling the actuator by determining a thrust and a thrust direction for controlling the actuator from the component force of the thrust, and a position of the structure.・ Direction control system.   The processing unit represents the limitation of the operating range in the thrust direction as an inequality using the component of the adjustment vector, and using the inequality as a constraint, the sum of squares of the component of the adjustment vector is minimum under the constraint. The structure position / orientation control system according to claim 6, wherein the component of the adjustment vector is calculated to be The control force is provided with an upper limit value and a lower limit value,
The control force represented in the constraint condition is replaced with an upper limit value and a lower limit value of the control force to obtain a plurality of inequalities, and the components of the adjustment vector are obtained using the plurality of inequalities. Structure position / orientation control system. The processing unit calculates an adjustment vector in advance before performing control of the position / orientation of the structure in response to an input of a control force to be applied to the center of gravity of the structure,
The control unit receives an input of a control force to be applied to the center of gravity of the structure when controlling the position / orientation of the structure, and the orthogonal matrix is applied to the control force and the adjustment vector or the adjustment vector. The structure position / orientation control system according to any one of claims 6 to 8, wherein the first solution is calculated using a processing vector obtained by applying the function.   The structure is a floating body floating in the sea having the reference direction, and the control force includes a component force in two directions on a plane and a moment about a center of gravity axis perpendicular to the plane of the floating body. 10. The structure position / orientation control system according to any one of items 9 to 9. A computer-executable program for controlling the position and orientation of a structure so that the position and orientation of the structure are maintained constant even when the structure floating in or on the space is subjected to disturbance,
The structure is provided with a plurality of actuators that control and output a desired thrust in a desired thrust direction with respect to a reference direction of the structure, and further, the operating range of the plurality of actuators in the thrust direction is limited. Given,
A vector of component force when the thrust to be applied to each of the plurality of actuators to which the operating range in the thrust direction is limited is decomposed in two directions orthogonal to each other is defined as an unknown, and A thrust distribution equation in which the result of applying a matrix defined by the information on the installation position of the actuator is equal to the vector of the control force to be applied to the center of gravity of the structure in order to maintain the position and orientation of the structure And the first solution of the distribution equation in which the vector of the thrust component force of the actuator whose operating range in the thrust direction is limited is unknown is limited to the operating range in the thrust direction of the plurality of actuators. There is no orthogonal row obtained when singular value decomposition of the matrix is performed as the second solution of the distribution equation when the sum of squares of the thrust is minimum When the expressed in linear sum obtained by adding a processing vector obtained by acting on the desired adjustment vector,
Based on each operating range of the thrust direction given to the plurality of actuators, the calculation means of the computer is calculated by the computer, and the adjustment vector or the processing vector obtained by applying the orthogonal matrix to the adjustment vector is calculated by the computer. A procedure for storing in the storage means;
In response to the input of the control force, the calculation means of the computer calculates the second solution based on the vector of the control force, and reads the adjustment vector stored in the storage means. The processing vector obtained by operating the orthogonal matrix or the processing vector read from the storage means is obtained, and the first solution obtained by adding the processing vector to the second solution is used as the computing means of the computer. And a program for calculating the thrust and the thrust direction to be applied to the plurality of actuators from the component of the first solution and applying the thrust and the thrust direction to the plurality of actuators.   When calculating the adjustment vector, the limitation of the operating range in the thrust direction is represented by an inequality using the component of the adjustment vector, and the calculation means of the computer creates the inequality as a constraint condition. The program according to claim 11, wherein the calculation unit of the computer calculates the component of the adjustment vector so that the sum of squares of the component of the adjustment vector is minimized. The control force is provided with an upper limit value and a lower limit value,
The control force represented by the constraint condition is replaced with an upper limit value and a lower limit value of the control force to obtain a plurality of inequalities, and the plurality of inequalities are used to calculate the components of the adjustment vector in the computing means of the computer. The program according to claim 12. The calculation of the adjustment vector is performed before the position / orientation of the structure is controlled by receiving the input of the control force to be applied to the center of gravity of the structure, and the calculation is performed on the calculated adjustment vector or the adjustment vector. Storing the processed processing vector in the storage means;
When controlling the position / orientation of the structure, it receives an input of a control force that should act on the center of gravity of the structure, and performs an arithmetic process on the control force and the adjustment vector stored or held. The program according to any one of claims 11 to 13, which causes the calculation means to calculate the first solution using the processing vector obtained in the above.   The structure is a floating body floating in the sea having the reference direction, and the control force includes a component force in two directions on a plane and a moment about a center of gravity axis perpendicular to the plane of the floating body. 14. The program according to any one of 14.

Images