JP4757838B2 - Positioning control device for towed mechanism - Google Patents

Description

  The present invention relates to a positioning control device for a towed mechanism, and more particularly to a towed mechanism suitable for bending a distal end side of an operation target in the up / down and left / right directions with a towed mechanism formed to be bendable as an operation target. The present invention relates to a positioning control device.

  2. Description of the Related Art Conventionally, an endoscope is used for observing an organ in a body cavity or performing various treatments on an affected part in a body cavity. The endoscope is provided in a curved portion on the distal end side of the elongated insertion portion formed in a cylindrical shape, and is connected to a monitor. For this reason, by inserting the insertion portion into the body cavity, it is possible to observe the organ in the body cavity with the endoscope of the bending portion, and as necessary, the treatment tool while observing the inside of the body cavity with the endoscope And the like can be inserted into the insertion portion to perform various treatments and treatments on the affected portion. Industrial endoscopes are widely used not only in the medical field but also in the industrial field to observe and inspect internal scratches and corrosion in boilers, turbines, engines, chemical plants, and the like. Yes.

  When driving the bending portion provided with the endoscope, for example, as described in Patent Document 1, a wire is connected to the bending portion provided on the distal end side of the endoscope insertion portion, and this wire There is one in which the bending portion is bent up and down / left and right by pulling the portion by the driving force of the electric motor. When the electric motor is driven, a motor drive circuit that drives the electric motor in proportion to an operation amount from an operation lever provided in the operation unit is employed. In this case, the bending angle of the bending portion can be set by pulling the pulling wire with the electric motor and moving the bending portion in the vertical and horizontal directions.

  In Patent Document 2, a traction wire is attached to a bending portion provided on the distal end side of an endoscope insertion portion, and the traction wire is connected to a universal cord via a relay pulley provided in an operation portion. And the traction wire is pulled by the driving force of the electric motor. Furthermore, this publication detects the loose state of the pulling wire by comparing the rotation angle of the pulley provided in the operation unit with the rotation angle of the electric motor, and when the pulling wire is in the relaxed state, the electric motor is set to the maximum speed. The method of releasing the slack of the pulling wire instantaneously by rotating with the rotation is shown.

Japanese Patent Publication No. 63-59329 JP-A-6-22904

  Among the above prior arts, the former one is configured to detect an operation amount using a strain gauge attached to an operation lever, and to directly apply a voltage according to the detected operation amount to the motor. The electric motor can be driven in proportion to the amount of operation of the control lever, but the friction between the pulling wire and the inner surface of the coil sheath for guiding the pulling wire and the slackness of the pulling wire are not fully considered. The amount does not correspond to the amount of bending at the distal end of the bending portion, which is not sufficient for improving the operability of the observer (endoscope operator).

  In addition, the latter of the conventional techniques employs a technique for releasing the slack of the wire that occurs in the universal cord between the motor and the relay pulley in order to improve operability. It is not considered about the slack and friction that occurs from the relay pulley to the pulling wire of the endoscope insertion part, and it is not considered about the release of the slack when there is no relay pulley. Is not enough.

  The subject of this invention is providing the positioning control apparatus of the towed mechanism which can make the operation command value follow the instantaneous value of the bending position of operation object.

  In order to solve the above problems, the present invention provides a command value signal output means capable of outputting a command value signal corresponding to an operation, and a drive signal corresponding to the command value signal output from the command value signal output means. Drive signal generation means for outputting, drive means for generating a drive force corresponding to the drive signal output from the drive signal generation means, and traction means for performing a traction operation based on the drive force generated by the drive means A sensor that detects an operation amount of the traction means and outputs a detection signal corresponding to the operation amount, and the drive signal generation means changes a correction amount based on the detection signal output from the sensor. And a feedforward control means for outputting a feedforward control signal using a signal obtained by correcting the command value signal based on the filter as the drive signal. A structure positioning control device of the traction mechanism.

The present invention also provides a command value signal output means capable of outputting a command value signal according to an operation, and a drive signal generation means for outputting a drive signal according to the command value signal output from the command value signal output means. Driving means for generating a driving force corresponding to the driving signal output from the driving signal generating means, traction means for performing a traction operation based on the driving force generated by the driving means, and bending by the traction means A towed mechanism having a bendable portion, a sensor that outputs an internal tension signal of the traction means, the command signal output from the command value signal output means, and the internal tension signal output from the sensor. determine the specific, and outputs the state estimation result the estimates a frequency characteristic which the traction mechanism has according to a preset function showing a relationship between a frequency characteristic of the said and the ratio towed mechanism A state estimation means, wherein the includes a frequency characteristic of the end position of the internal tension signal and said traction means as previously plurality of models, and models to select the model corresponding to the state estimation result from the model of the plurality of, is said selected Based on the position estimation means for estimating the tip position of the bending portion of the towed mechanism from the internal tension signal, and the command value signal output means based on the estimation result of the tip position output from the position estimation means. Feedback control means for feedback-correcting the command value signal, and feed-forward correction of the command value signal input to the drive signal generation means based on an output signal output from the feedback control means to drive the drive signal The tow mechanism positioning control device having the feedforward control means for outputting as

Further, (1) traction means for pulling the operation target in accordance with a driving force with a towed mechanism formed to be bendable, and an operation command value corresponding to the target position of the operation target in response to the operation Operation command value signal output means for outputting a signal; feedforward control means for generating a feedforward control signal by compensating the operation command value signal according to a feedforward compensation value; and a drive signal based on the feedforward control signal. The driving signal generating means for generating and the driving means for driving the traction means with the driving force according to the driving signal can be employed.
(2) In the positioning control device for the towed mechanism according to (1), a drive detection means for outputting the drive state detection signal, and a feedforward control signal generated by the feedforward control means for the drive means. A deviation calculating means for calculating a deviation between the command signal and the detection signal of the drive detecting means as a command signal, and driving by performing an operation for suppressing the deviation to zero based on the deviation calculated by the deviation calculating means It can be set as the structure provided with the calculating means which produces | generates a signal.
(3) In the positioning control device for a towed mechanism according to (1) and (2), a plurality of traction means connected to each other as the traction means and sequentially transmitting a driving force by driving of the driving means are provided. A drive detection means for detecting a position associated with driving of the relay traction means that receives driving force from the drive means from other traction means among the plurality of traction means and outputting a relay state signal; and the feed By calculating a deviation between the command signal and a signal output from the drive detection means using a feedforward control signal generated by the forward control means as a command signal for the drive detection means, and by calculating the deviation calculation means An arithmetic means for generating a drive signal by performing an operation for suppressing the deviation to zero based on the deviation can be provided.
(4) In the towed mechanism positioning control device according to (1), (2) or (3), the feedforward control signal generated by the feedforward control means has a phase with respect to the operation command value signal. It can be advanced.
(5) In the positioning control device for a towed mechanism according to (1), (2), or (3), the drive state detection signal may be a drive detection unit that outputs a drive position.
(6) In the positioning control device for a towed mechanism according to (1), (2), or (3), the drive state detection signal is a drive detection unit that outputs a tension associated with the traction of the traction unit, and an operation command Command value conversion means for converting a value signal into a tension command value signal can be provided. .
(7) In the positioning control device for a towed mechanism according to (1), (2), or (3), a tow state amount detection that detects a tow state amount of the tow means and outputs a tow state amount detection signal. A command value / state quantity deviation calculating means for calculating a deviation between the operation command value signal output from the operation command value signal output means and the traction state quantity detection signal, and the command value / state quantity deviation calculating means. Feedback control signal generating means for generating a feedback control signal for suppressing the deviation to zero based on the calculated deviation, and correcting the feedforward control signal generated by the feedforward control means with the feedback control signal; Position command value signal generation means for generating a position command value signal for the drive means can be provided.
(8) In the positioning control device for a towed mechanism according to (7), the traction state amount detection unit detects a displacement associated with the traction of the traction unit and corresponds to a curved position of the operation target. Can be constituted by a bending position detecting means for outputting as a traction state quantity detection signal.
(9) In the positioning control device for a towed mechanism according to (7), the towed state amount detecting means is based on a tension detecting means for detecting a tension accompanying pulling of the towing means, and detection by the tension detecting means. The position estimation means can estimate the bending position of the operation target from the tension and the drive position detection signal output from the drive position detection means, and output the estimation result as a traction state quantity detection signal.
(10) In the positioning control device for a towed mechanism according to (7), the towed state amount detecting means includes a tension detecting means for detecting a tension accompanying pulling of the towing means, and the operation command value signal output means. The state estimation means for estimating the state of the operation target from the operation command value signal by the output of the force and the tension detected by the tension detection means, and the dynamics of the feedforward control means and the feedback control means by the state estimation means, respectively. Dynamics changing means for changing can be configured.
(11) In the positioning control device for a towed mechanism according to (3), the relay state signal may be a relay state amount detection unit that outputs a relay position.
(12) In the positioning control device for a towed mechanism according to (3), the relay state signal can be output from a relay state amount detecting unit that outputs a tension accompanying pulling of the relay pulling unit.
(13) In the positioning control device of the traction mechanism according to (10), can be made is composed of a dynamics changing means for changing the dynamics of the feed forward control means.
(14) In the positioning control device of the traction mechanism according to (10), can be made is composed of a dynamics changing means for changing the dynamics of the feedback control means.
According to the means described in (1) to (14), when the operation command value signal is output in response to the operation of the operator, the operation command value signal is compensated by the feedforward control means, and the feedforward control means Because the traction means is driven by the driving force according to the drive signal generated based on the feedforward control signal generated by generating the traction means, even if there is a delay associated with the driving of the traction means during traction, this delay is controlled by feedforward control. Compensated by the means, the bending position of the operation target can be made to follow the operation command value instantaneously, which can contribute to the improvement of operability.
Further, as the traction state quantity of the traction means, the displacement (wire position) associated with the traction of the traction means attenuated by friction or slack is detected, the traction state quantity is fed back, the traction state quantity detection signal, the operation command value signal, By generating a feedback control signal based on the deviation and correcting the feedforward control signal with the feedback signal, non-linearity due to friction and slack (wire slack) contained in the traction means is compensated can do.

  According to the present invention, since the operation command value is compensated by the feedforward control means, the bending position of the operation target can be instantaneously followed by the operation command value, which can contribute to improvement in operability.

A reference example and an embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is an overall configuration diagram of a bending positioning control device for an electric endoscope showing a first reference example of the present invention. In FIG. 1, the electric endoscope has a joystick 10 that is attached to an operation unit and can be operated up and down / left and right, a bending portion 12 configured to be bendable, and a distal end side of the bending portion 12 to bend in the vertical direction. The pulling wires 14 and 16, the pulling wires 14 and 16 are wound around and a pulley 18 that applies a pulling force to the pulling wires 14 and 16, and the pulley 18 is connected to the pulley 18 via a gear mechanism to drive the pulley 18 to rotate ( An electric motor) 20 is provided, and the pulling wires 14 and 16 are formed of a single continuous wire, and the pulling wires 14 and 16 are connected to the bending portion 12 through the coil sheaths 22 and 24, respectively. The bending portion 12 is constituted by a serpentine tube in which a plurality of node rings 26 are rotatably connected to each other as an operation target constituting the main element of the towed mechanism, and the bending portion 12 is formed in a substantially cylindrical shape. It is configured to be able to bend in the up / down / left / right directions.

In the present reference example , the bending portion 12 is bent in the vertical direction by pulling the pulling wires 14 and 16. Specifically, when the motor 20 rotates forward or backward and the pulley 18 rotates, one of the wires 14 and 16 wound around the pulley 18 is drawn, the other is drawn out, and the bending portion 12 is moved up and down. It is designed to bend in the direction. That is, the pulling wires 14 and 16 and the pulley 18 are configured as pulling means that pulls the bending portion 12 as an operation target by driving the motor 20.

In this reference example , for the sake of simplification of the drawing, only a pair of pulling wires 14 and 16 for bending the bending portion 12 up and down are shown, and the bending portion 12 is bent in the left-right direction. Necessary pulling wires, pulleys, motors, and control devices are omitted because they can be basically the same as those for bending the bending portion 12 up and down.

  Further, the bending portion 12 and the coil sheaths 22 and 24 are inserted and protected in an insertion portion (not shown) formed of an elongated elastic pipe, and the distal end of the bending portion 12 serving as the distal end of the insertion portion is provided with a CCD. A CCD camera constituted by the above is attached (not shown), and this CCD camera is connected to a monitor (not shown) via a cable wired in the insertion portion. Images of organs in the body cavity and industrial piping taken by the CCD camera are displayed on the screen of the monitor. In this case, the operator basically performs the bending operation of the joystick 10 up and down or left and right with the left hand while watching the image displayed on the monitor screen, and holds the middle of the insertion portion with the right hand to insert The part is moved back and forth, and sometimes the insertion part is rotated to push the insertion part into the body cavity or the like, and after the distal end side of the insertion part has reached the observation point, the distal end of the bending part 12 is bent, An observation point can be imaged in an appropriate state, or an appropriate treatment can be performed using a treatment tool inserted into the insertion portion.

  For this reason, since the operator can observe the movement seen from the distal end of the insertion portion in real time by looking at the monitor, if the movement of the distal end of the bending portion 12 can be freely controlled by appropriately controlling the motor 20, The burden on the operator is reduced and the operability is greatly improved.

  In an endoscope that is electrically driven by a wire, the bending portion 12 on the distal end side of the insertion portion is pulled by the pulling wires 14 and 16 in accordance with the driving force of the motor. What must be considered when towing is friction between the towing wires 14 and 16 and the coil sheaths 22 and 24, occurrence of slack in the pulling wire on the side not pulled by the pulley 18 (drawing tow wire), Alternatively, the amount of friction or slack between the pulling wires 14 and 16 and the coil sheaths 22 and 24 varies due to a change in the bent shape in the middle of the insertion portion. That is, the bending position of the bending portion 12 cannot be made to follow the operation command value instantaneously unless the problems of friction, slack, and characteristic fluctuation are overcome.

Therefore, in the present reference example , a feedforward control system that advances the phase of the operation command value is used to move the dynamics from the pulley position where the phase is delayed to the tip position of the pulling wire 14 or 16, and the feedforward control system However, a configuration is adopted in which a tracking error that cannot be compensated for is compensated by a feedback control system.

  Specifically, the control system including the feedforward control system and the feedback control system is configured as follows.

First, in configuring the feedback control system, in the present reference example , first, sensors 28 and 30 for observing the moving amounts of the pulling wires 14 and 16 are attached to the distal ends of the coil sheaths 22 and 24. . The sensors 28 and 30 detect movement amounts of the pulling wires 14 and 16 using, for example, an optical linear encoder and a resistance linear potentiometer as sensors for detecting the movement amounts of the pulling wires 14 and 16. . In this case, since the distance from each sensor 28, 30 to the tip of the bending portion 12 is shorter than the entire length of the insertion portion, the position of the pulling wires 14, 16 detected by each sensor 28, 30 is curved. It is equal to the amount of movement on the tip side of the portion 12, that is, the bending position. Wire position signals 32 and 34 detected by the sensors 28 and 30 are input to a wire position detector 36, respectively. In this case, the wire position signals 32 and 34 show positive values when the pulling wires 14 and 16 are moved in the directions of the arrows, respectively, and the wire position detector 36 receives the wire position signals 32 and 34. The wire tip position signal 38 is output with the average value as the position of the tip of the bending portion 12.

  That is, the sensors 28 and 30 and the wire position detector 36 detect the displacement accompanying the pulling of the pulling wires 14 and 16 together with the pulling state amount detecting means, and pull the wire tip position signal 38 corresponding to the bending position of the bending portion 12. It is configured as a bending position detecting means for outputting as a quantity detection signal.

  The wire tip position signal 38 is input to the adder 42 of the controller 40, and the command value signal 46 is input to the adder 42 from the command value detector 44. The command value detector 44 responds to the operation of the joystick 10 when the joystick 10 is operated in the up / down / left / right directions, and a command value signal (operation command value signal) 46 corresponding to the target bending position of the bending portion 12. Is configured as an operation command value signal output means. The adder 42 is configured as a command value / state quantity deviation calculating means for calculating a deviation between the command value signal 46 and the wire tip position signal 38 and outputting a signal related to the calculated deviation to the feedback controller 48. The feedback controller 48 performs only gain compensation on the deviation calculated by the adder 42 to generate a feedback control signal for suppressing the deviation to zero, and outputs the feedback control signal to the adder 50. It is configured as control signal generation means.

  A feedforward signal is input to the adder 50 from the feedforward controller 52. When the command value signal 46 is input from the command value detector 44, the feedforward controller 52 compensates the command value signal 46 according to the feedforward compensation value as a process for advancing the phase of the command value signal 46. It is configured as feedforward control means for performing a calculation to generate a feedforward control signal as a result of the calculation. Specifically, the feedforward controller 52 is composed of a secondary phase advance filter, as shown by the following equation (1).

ここで、ｓはラプラス演算子、ｗ１とｗ２は設定する周波数で、ｗ１＜ｗ２の関係にあり、ｚ１とｚ２は減衰係数で、ｚ１＞ｚ２の関係に設定されている。このフィルタの特性は、加算器５０の生成によるモータ位置指令値５４から牽引ワイヤ１４、１６の先端位置までの伝達特性を測定し、その逆モデルを近似して得られたものである。このため、基本的には、操作指令値が指令値信号４６としてフィードフォワード制御器５２を通過し、後述するモータフィードバック閉ルーループ系、牽引ワイヤ１４、１６の先端位置までの伝達関数はほぼ１となり、牽引ワイヤ１４、１６の先端位置である湾曲部１２の湾曲位置は操作指令値とほぼ等しい動きをすることとなる。

Here, s is a Laplace operator, w1 and w2 are frequencies to be set, and a relationship of w1 <w2, z1 and z2 are attenuation coefficients, and a relationship of z1> z2 is set. The characteristics of this filter are obtained by measuring the transmission characteristics from the motor position command value 54 generated by the adder 50 to the tip positions of the pulling wires 14 and 16, and approximating the inverse model. Therefore, basically, the operation command value passes through the feedforward controller 52 as the command value signal 46, and the transfer function to the tip positions of the motor feedback closed loop system and the pulling wires 14 and 16 described later becomes approximately 1. The bending position of the bending portion 12 which is the tip position of the pulling wires 14 and 16 moves substantially equal to the operation command value.

  The adder 50 to which the feedforward control signal is input serves as position command value signal generating means for correcting the feedforward control signal with the feedback control signal and generating a motor position command value signal 54 as a position command value signal for the motor 20. The motor position command value signal 54 is input to the adder 56. In addition to the motor position command value signal 54, the motor position signal 60 is input from the motor position detector 58 to the adder 56. A signal indicating the rotation angle is input to the motor position detector 58 from the rotary encoder 62 that detects the rotation angle of the motor 20, and the motor position detector 58 is in accordance with the rotation angle of the motor 20 detected by the rotary encoder 62. A motor position signal 60 is generated. That is, the rotary encoder 62 and the motor position detector 58 are configured as drive position detection means for detecting a position associated with the drive of the motor 20 and outputting a motor position signal 60 as a drive position detection signal to the adder 56.

  The adder 56 is configured as a position deviation calculating means for calculating a deviation between the motor position command value signal 54 and the motor position signal 60, and a signal relating to the position deviation calculated by the adder 56 is input to the motor controller 64. ing. The motor controller 64 includes, for example, a PID (proportional / integral / differential) compensator, performs a calculation for making the deviation generated by the adder 56 zero, generates a drive signal, and converts the drive signal into a motor amplifier. 66 is configured as a calculation means for outputting to 66. The motor amplifier 66 amplifies the drive signal and outputs the amplified drive signal to the motor 20.

  In the above configuration, when the operator operates the joystick 10, the operation command value according to this operation is output as the command value signal 46, and the deviation between the command value signal 46 and the wire tip position signal 38 is calculated by the adder 42. A feedback control signal in accordance with this deviation is generated by the feedback controller 48, and a feedforward compensation calculation for advancing the phase of the position command value signal 46 is performed by the feedforward controller 52. Generated. The feedforward control signal is corrected with the feedback control signal to generate a motor position command value signal 54, and a position deviation corresponding to the deviation between the motor position command value signal 54 and the motor position signal 60 is obtained by the adder 56, A drive signal based on this positional deviation is generated by the motor controller 64. When the motor 20 is driven to rotate forward or backward by this drive signal, the pulling wire 14 or 16 is pulled, and the bending portion 12 The tip side curves upward or downward. In this case, the motor 20 cancels the reaction force of the tension of the pulling wires 14 and 16 applied to the pulley 18, and the rotational position of the motor 20 can follow the motor position command value 54 almost without delay. Therefore, if the rattling of the gear is ignored, the wire movement amount in the vicinity of the pulley 18 moved by the rotation of the pulley 18 is calculated by multiplying the motor position command value signal 54 by the gear ratio and the radius of the pulley 18. can do.

  However, the moving amount of the pulling wire 14 or 16 near the pulley 18 moved by the rotation of the pulley 18 and the moving amount of the pulling wire 14 or 16 on the distal end side of the bending portion 12 are not in a proportional relationship. This is because the tension between the pulling wires 14 and 16 and the coil sheaths 22 and 24 attenuates the pulling tension of the pulling wires 14 and 16 halfway, and when the pulley 18 rotates in the reverse direction, the slack is wound up. This is considered to be caused by the phase delay due to the time gap (dead zone) and the dynamics trying to prevent the bending portion 12 formed of the serpentine tube and the tube constituting the bending portion 12 from bending.

Thus, in this reference example , the feedforward controller 52 for advancing the phase of the command value signal 46 is used to move the dynamics from the pulley 18 position where the phase is delayed to the distal end position of the bending portion 12, and this cannot be compensated by this. Is compensated by the feedback controller 46, the tip position (curved position) of the bending portion 12 can be made to follow the operation command value without delay.

In this reference example , since the movement amounts of the pulling wires 14 and 16 are detected by the sensors 28 and 30, it is possible to detect the position of the pulling wires 14 and 16 attenuated by friction and slack, and this detection. By feeding back the signal, it is possible to compensate for non-linearity caused by friction or loosening of the wire included in the system constituting the traction means.

In this reference example , since the wire mechanism of the insertion portion is basically a stable system and the offset with respect to the operation command value may be adjusted by the operator looking at the monitor, the feedback controller 48 However, only gain compensation is performed without having integral characteristics or differential characteristics. However, since the gain may increase in a specific frequency band by applying feedback, the characteristics of the closed loop system are corrected as necessary by a notch filter or the like.

  Further, when the gain of the feedback controller 48 is increased, the target operation command value is followed. However, if the gain is increased too much, the reversing operation of the motor is repeated and becomes oscillating. It is set to double. In this case, the operation command value and the tip position of the bending portion 12 do not completely coincide with each other as the feedback gain is lowered, but the tip of the bending portion 12 is advanced by advancing the phase of the operation command value signal 46 by the feedforward controller 52. The position can be made to follow the operation command value instantaneously.

Next, the experimental results of the apparatus according to this reference example and the prior art will be described with reference to FIGS. FIG. 2 shows an example of an experimental result when the bending positioning operation of the endoscope is performed by the control device shown in FIG. FIG. 3 shows a response result according to the prior art in which a position command value obtained by operating a joystick is directly generated as a motor position command value. 2 and 3, the upper stage (a) shows the time waveform at the tip position of the upper and lower traction wires, and the lower stage (b) shows the tension acting on the upper and lower traction wires of the endoscope insertion portion. The time waveform measured on the side is shown. In each figure, the upper and lower figures both show a triangular waveform with a period of 7 seconds, which is applied as a command value 300 assuming a command value from a joystick. In the section A, the pulley 18 shown in FIG. 1 rotates in the direction of the arrow (clockwise), and in the section B, it rotates counterclockwise.

  In FIG. 3, it can be seen that the tip positions 307 and 308 of the upper and lower wires in the prior art are trapezoidal with the phase delayed by about 20 degrees with respect to the target command value 300, and the gain of the triangular shape is crushed. . Therefore, in the case of the prior art, if the operator wants to make the movement of the bending portion tip follow the triangular wave, the phase of the bending portion tip position is delayed only by inputting the triangular wave command value from the joystick. In addition, since it has a trapezoidal waveform, it means that the operator cannot perform accurate positioning. Therefore, in order to achieve the object, the operator is forced to correct the command value so that the position of the distal end of the bending portion follows the triangular wave while watching the monitor. For this reason, skill is required for operation, and the burden on the operator increases on the contrary by electrification of the endoscope.

On the other hand, in the reference example , as shown in FIG. 2, the tip positions 303 and 304 of the upper and lower traction wires and the target command value (operation command value) 300 substantially coincide. For this reason, the operator moves the tip positions of the traction wires 14 and 16 according to the command value, so that the operation burden is reduced by electrification, and the tip of the bending portion can be quickly positioned at a target location with high accuracy. become.

  Next, when comparing with the tension, when observing the waveform of the prior art in FIG. 3 (b), the tension 305 at the moment when the direction of rotation (operation direction) changes from the A section to the B section is the largest. In spite of the change of direction, the tension gradually decreases. Conversely, the tension 306 of the lower wire has no tension (0), that is, from the slack state to the B section, and the direction of rotation has changed. Nevertheless, it can be seen that the tension begins to increase after about 0.7 seconds due to the slackness, and the direction of the tip position changes with it.

On the other hand, as shown in FIG. 2B, it can be seen that the tension 302 starts to be applied almost instantaneously after the section B is reached in the reference example . This is an effect of the feedforward controller 52 that advances the phase of the target command value (operation command value) and the feedback controller 48 that feedback-controls the tip position of the wire. When the tension of the pulling wires 14 and 16 is increased, the tip position of the pulling wires 14 and 16 can be controlled by the pulley 18. In other conventional techniques that focus on slack and release the slack quickly, the tip position of the wire greatly deviates from the target command value during that time, so that the operator adjusts the tip position of the wire to the command value. Since it is necessary to try to correct the error by yourself, skill is required, and the electric load increases the burden on the operator.

Next, a second reference example of the present invention will be described with reference to FIG. This reference example is obtained by removing the feedback control system for feedback control of the tip position of the pulling wire from the control device shown in FIG. 1, and the other configurations are the same as those in FIG.

  In this embodiment, the phase of the command value signal 46 output from the command value detector 44 is advanced by the feedforward controller 52 to generate a feedforward control signal, and the deviation between the feedforward control signal and the motor position signal 60 is calculated. Since the motor controller 64 generates a drive signal for obtaining this deviation by the adder 56 and makes this deviation zero, the drive signal is amplified by the motor amplifier 66 and the motor 20 is driven to rotate. The tip position of the bending portion 12 due to the pulling of the pulling wires 14 and 16 can follow the operation command value without delay. That is, the phase is delayed from the movement of the pulley 18 to the tip position of the pulling wires 14 and 16 due to non-linear effects such as friction between the pulling wires 14 and 16 and the coil sheaths 22 and 24 and slack of the pulling wires 14 and 16. This phase delay can be compensated for by the feedforward controller 52. For this reason, it is possible to advance the movement of the pulley 18 with respect to the operation command value by advancing the phase of the operation command value by the feedforward controller 52.

In this reference example , when the relationship between the tip positions of the pulling wires 14 and 16 and the operation command value was measured, an experimental result as shown in FIG. 5 was obtained. 5B shows the characteristics of the triangular wave 300 indicating the operation command value and the waveform 311 of the motor position command value output from the feedforward controller 52. From FIG. 5, it can be seen that an appropriate offset is added to the command each time the operation direction is changed by operating the joystick 10, and the command value is corrected in the direction in which the phase advances. Thereby, compared with the tip positions 307 and 308 shown in the upper part of FIG. 3, the tip positions 307 and 308 of the pulling wires 14 and 16 are approaching the triangular wave 300 of the command value in (a) shown in the upper part of FIG. I understand.

Thus, in this reference example , unlike the reference example , there is no feedback system, so it cannot cope with characteristic fluctuations such as friction of the insertion portion, but by advancing the phase of the operation command value. It can be seen that the movement of the distal end of the bending portion 12 is closer to the operation command value.

In addition, as the feedforward controller 52 in this reference example, a configuration in which the sign changes as shown in the following equation (2) may be used.

ここで、Ｒは目標指令値、ｓｇｎはその符号を表す。またゲインｇの大きさは、一定値よりも指令値の速度が小さいときにはその速度に比例し、指令値の速度が一定値以上大きくなったときには一定値とするように構成した方が良い。これにより、速度の符号の変化に対し、モータ２０の速度を徐々に変化させることができ、牽引ワイヤ１４、１６などの振動を抑えることができる。

Here, R represents a target command value, and sgn represents its sign. The magnitude of the gain g is proportional to the speed when the speed of the command value is smaller than a certain value, and is preferably set to a constant value when the speed of the command value becomes larger than the certain value. Thereby, the speed of the motor 20 can be gradually changed with respect to the change of the sign of the speed, and vibrations of the pulling wires 14 and 16 can be suppressed.

Further, in this reference example , the endoscope insertion part does not have special sensors 28 and 30 and a feedback control system, so that the cost can be reduced as compared with the reference example .

Next, an embodiment of the present invention according to FIGS. 6 to 11. In this embodiment, instead of feeding back the tip position of the pulling wire, the tension of the pulling wires 14 and 16 is detected and the tension is fed back, and the pulling wires 14 and 16 are pulled toward the pulley 18 side. Tension sensors 68 and 70 for detecting the tension of the wires 14 and 16 are provided, and the detected values of the tension sensors 68 and 70 are input to the wire tension detector 72, respectively.

  For example, as shown in FIG. 7, each of the tension sensors 68 and 70 includes a strain gauge 74, a bridge circuit 76, and a differential amplifier circuit 78. Are installed. When a change in the amount of strain is detected by the strain gauge 74, a signal indicating the change in the amount of strain is amplified by the differential amplifier circuit 78 via the bridge circuit 76 whose resistance changes in proportion to the change in the amount of strain. It has become so. In this case, when no external force is applied to the pulling wires 14 and 16, the bridge circuit 76 is balanced by the resistance value of the strain gauge 74, and the output of the differential amplifier circuit 78 is zero. On the other hand, when the pulling wire 14 or 16 is pulled by the pulley 18, an external force is applied to the pulling wire 14 or 16, the resistance value of the strain gauge 74 changes, the balance of the bridge circuit 76 is lost, and the output of the differential amplification circuit 78 A voltage is generated at the end. This output voltage is output to the wire tension detector 72 as a signal indicating the tension acting on each pulling wire 14 or 16. The wire tension detector 72 adds the tensions detected by the tension sensors 68 and 70 and outputs an internal tension signal 80 indicating a substantial internal tension for pulling the tip of the bending portion 12. That is, the tension sensors 68 and 70 and the wire tension detector 72 are configured as tension detecting means, and the internal tension signal 80 is output to the endoscope state estimator 82 and the position estimator 84. Here, the positive and negative tensions detected by the tension sensors 68 and 70 have the same relationship as the characteristics shown in FIG.

  The position estimator 84 estimates the bending position of the bending portion 12 from the internal tension signal 80 and the estimation result 86 of the endoscope state estimator 82, and outputs the estimation result 88 to the adder 42 as a traction state amount detection signal. It is comprised as a position estimation means to do. This position estimator 84 is configured by modeling the dynamics from the internal tension signal 80 to the tip positions of the pulling wires 14 and 16, and a specified model from a plurality of models according to the state estimation result 86. The position of the distal end of the insertion portion can be estimated by selecting. As a model used for this position estimator 84, for example, a plurality of primary low-pass filters are used. When a model is selected from the state estimation result 86, it is necessary to select a designated model according to the state of the insertion portion of the endoscope. That is, since the insertion portion is inserted into a body cavity or the like, the insertion start is almost straight, but when inserted along the shape of the intestine or the like, the bending rate in the middle of the insertion portion increases depending on the location. There is. As the bending rate increases, the frictional force between the coil sheaths 22 and 24 and the pulling wires 14 and 16 increases, and the dynamics from the measured internal tension signal 80 to the leading end positions of the pulling wires 14 and 16 are in a straight state. The characteristics fluctuate compared to Furthermore, the characteristics of dynamics vary depending on the usage environment of the endoscope and changes over time. Therefore, in order to further improve the estimation accuracy of the tip position (curved position) of the pulling wire, the state is estimated, a model according to this estimation is selected, and the feedback controller 48 is controlled according to the selected model. There is a need to.

  Therefore, in the present embodiment, in the endoscope state estimator 82, an appropriate value is determined according to a predetermined function from a relational ratio between the command value signal 46 indicating the operation command value and the internal tension signal 80 indicating the magnitude of the internal tension. A state estimation result 86 for selecting a simple model is obtained, and this state estimation result 86 is output to the position estimator 84. This state estimation result 86 is obtained in consideration of the fact that the dynamic characteristic variation from the internal tension signal 80 to the tip position of the pulling wires 14 and 16 appears most in the tension. That is, even when the operation command value is the same, when the motor 20 is driven based on the operation command value, when the insertion portion is in a straight state, the tension applied to the pulling wires 14 and 16 is small, and conversely, the insertion portion is bent. When the rate is large and the middle of the insertion portion rotates around, the tension applied to the pulling wires 14 and 16 increases. For this reason, the endoscope state estimator 82 estimates the state of the bending portion 12 from the operation command value and the internal tension, and the endoscope state estimator 82 determines the operation command value signal 46 and the internal tension. It is configured as state estimation means for estimating the state of the bending portion 12 from the signal 80.

  As a model used for the position estimator 84, for example, four first-order low-pass filters can be used as shown in FIG. The four models are represented by characteristics 320, 321, 322, and 323, and the crossover frequencies of the characteristics are 1 Hz, 2 Hz, 4 Hz, and 8 Hz, respectively. When the insertion portion is in a straight state, a crossover frequency of 4 Hz and a characteristic 322 is used. Further, when the middle of the insertion portion is rotated and friction is large, for example, a crossover frequency of 1 Hz and a characteristic of 320 is used. Here, when the position of the tip of the wire was estimated using a low-pass filter having a crossover frequency of 4 Hz as a model, an experimental result as shown in FIG. 9 was obtained. In this case, the shape of the trapezoidal waveform indicating the position of the tip of the wire does not completely match the shape of the command value, but the phase characteristics match, and the tip position 307, 308 of the pulling wire and the position estimator 84 It was confirmed that the estimated result 312 was in agreement.

  Further, as shown in FIG. 10, the endoscope state estimator 82 includes low-pass filters 90 and 92, absolute value calculation circuits 94 and 96, integration circuits 98 and 100, an evaluation value calculation circuit 102, a function reference circuit 104, and a model determination. A circuit 106 is provided. The command value signal 46 based on the operation of the joystick 10 is input to the low-pass filter 90. When the command value signal 46 passes through the low-pass filter 90, a noise component is cut and only a signal in a necessary band passes. The signal that has passed through the low-pass filter 90 is converted into a positive-only signal by the absolute value calculation circuit 94, the signal is integrated by the integration circuit 98 for a fixed time T0, and the integrated value is an evaluation value as the magnitude RI of the operation command value. It is output to the calculation circuit 102. In this case, if the operation command value is R, RI is expressed by the following equation (3):

となる。

It becomes.

  On the other hand, the internal tension signal 80 is input to the low-pass filter 92, and the noise component is removed from the internal tension signal 80 by the low-pass filter 92, and only a necessary band signal passes through the low-pass filter 92. The signal that has passed through is converted into a positive signal by the absolute value calculation circuit 96. When this signal is integrated by the integration circuit 100 for a predetermined time T0, this integration value is output to the evaluation value calculation circuit 102 as a signal indicating the magnitude CI of the internal tension C. In this case, the magnitude CI of the internal tension C is expressed by the following equation (4):

となる。

It becomes.

  The evaluation value calculation circuit 102 calculates the evaluation value related to the state of the endoscope insertion portion by dividing the magnitude RI of the operation command value by the magnitude CI of the internal tension. When referring to a function in the function reference circuit 104 according to this evaluation value, the relationship between the evaluation value and the model breakpoint frequency (crossover frequency) as shown in FIG. 11 is set. This relationship is expressed by the following equation (5):

で表される。

It is represented by

  When the model breakpoint frequency (crossover frequency) corresponding to the evaluation value is determined, the model of the model breakpoint frequency corresponding to the evaluation value is determined by the model determination circuit 106, and this determination is output as the state estimation result 86. The

  For example, if the evaluation value is greater than 8, that is, if the magnitude of the tension is large relative to the operation command value, this means that the delay in the dynamics of the insertion portion increases due to friction or the like, and the model is broken. A low-pass filter having a point frequency of 1 Hz is selected. On the other hand, when the evaluation value is between 1 and 8, the model breakpoint frequency is calculated according to equation (5). When the evaluation value is smaller than 1, the model breakpoint frequency is set to 8 Hz. For example, when the evaluation value is 2 and the intersection with the function is A, the evaluation value is in a straight state, and a model with a break frequency of 4 Hz is selected as the model at this time.

  Instead of integrating the signals that have passed through the absolute value arithmetic circuits 94 and 96 for a predetermined time, it is also possible to use a signal that has passed through a low-pass filter of about 0.1 Hz for a predetermined time. The model may be updated at regular intervals.

  As described above, in the present embodiment, the estimation result 88 estimated by the position estimator 84 is fed back and feedback control is executed, so that a tracking error that cannot be compensated for by the feedforward controller 52 is compensated by the feedback control system. can do.

  Furthermore, in the present embodiment, the feedforward controller 52 and the feedback controller 48 can update the model according to the state of the endoscope insertion unit at regular intervals, thereby achieving more accurate positioning. it can.

  In the present embodiment, the parameters of the feedback controller 48 and the feedforward controller 52 can be adjusted along with the model update.

Next, a third reference example of the present invention will be described with reference to FIG. In this reference example , instead of the endoscope state setting unit 82 and the position estimator 84 shown in FIG. 6, the motor position signal 60 indicating the rotational movement amount of the pulley 18 and the internal tension signal 80 are used for the pulling wires 14 and 16. A position estimator 108 for estimating the tip position is provided, and other configurations are the same as those in FIG. The position estimator 108 receives a motor position signal 60 detected by the motor position detector 58 and an internal tension signal 80 detected by the wire tension detector 72, and based on these signals, the tip positions of the traction wires 14 and 16 are detected. The position estimation means is configured to estimate (the bending position of the bending portion 12) and output the estimation result 110 to the adder 42 as a traction state amount detection signal.

  When the position estimator 108 estimates the tip positions of the pulling wires 14 and 16 from the motor position signal 60 corresponding to the rotational movement amount of the pulley 18 and the internal tension signal 80 indicating the internal tension, Assuming that the rigidity is known, the tip position of the pulling wire is estimated. In this case, the internal tension Ten is considered to be generated by the following equation (6).

ここで、Ｋは牽引ワイヤ１４、１６の剛性、Ｐｏｓｐはプーリ１８が回転することによって移動した牽引ワイヤ１４、１６の移動量、Ｐｏｓｆは牽引ワイヤの先端位置である。この式を牽引ワイヤの先端位置について解くと、次の（７）式となる。

Here, K is the rigidity of the pulling wires 14 and 16, Posp is the amount of movement of the pulling wires 14 and 16 moved by the rotation of the pulley 18, and Posf is the tip position of the pulling wire. When this equation is solved for the tip position of the pulling wire, the following equation (7) is obtained.

上記（７）式において、牽引ワイヤ１４、１６の剛性を予め測定しておくことで、Ｐｏｓｐのワイヤ移動量は、ロータリエンコーダ６２の検出値からギア比とプーリ１８の半径とを積演算することにより算出することができる。これにより、牽引ワイヤの先端位置をリアルタイムに推定することが可能である。

In the above equation (7), by measuring the rigidity of the pulling wires 14 and 16 in advance, the wire movement amount of Posp is calculated by multiplying the detected value of the rotary encoder 62 by the gear ratio and the radius of the pulley 18. Can be calculated. Thereby, it is possible to estimate the tip position of the pulling wire in real time.

In this reference example , feedback control is executed in accordance with the deviation between the position estimation result 110 and the operation command value, so that the tip position of the pulling wire can be made to follow the operation command value.

Next, a fourth reference example of the present invention will be described with reference to FIG. This reference example considers that the operation command value by the operation of the joystick 10 is a function of the position, and converts the command value signal 46 into the tension command value 114 by the command value converter 112, and the tension command value 114 and the internal tension. The deviation from the signal 80 is obtained by the adder 42, and this deviation is executed by the feedback controller 48. The other configuration is the same as that of FIG.

  The command value converter 112 includes, for example, a primary high-pass filter, and is configured as command value conversion means for converting the command value signal 46 into a tension command value 114.

In this reference example , since feedback control is performed according to the deviation between the internal tension signal 80 and the tension command value (tension command value signal) 114, the tip positions of the traction wires 14 and 16 are made to follow the tension command value. Is out.

  In the case where a tension command can be directly input to input means such as the joystick 10, a configuration in which a deviation between the signal and the internal tension is taken and input to the feedback controller 48 can be employed.

Next, a fifth reference example of the present invention will be described with reference to FIG. In this reference example , the traction wires 14 and 16 shown in FIG. 1 are wound around the relay pulley 118 provided in the operation unit 116, and one end side of the traction wires 120 and 122 is wound around the relay pulley 118 and the inside of the coil sheaths 124 and 126 is wound. The other ends of the pulling wires 120 and 122 are wound around the pulley 118, and the pulling wires 14 and 16 and the pulling wires 120 and 122 are connected to each other via the relay pulley 118 to form a plurality of pulling means. ing. The relay pulley 118 is provided with a potentiometer 128 that detects the rotation angle of the relay pulley 118, and the rotation angle signal 130 detected by the potentiometer 128 is output to the relay position detector 132. That is, in this reference example , the length of the insertion part is shortened by providing the operation part 116 via the relay pulley 118 in the middle of the insertion part. Further, the traction wires 120 and 122 and the coil sheaths 124 and 126 from the motor 20 to the relay pulley 118 are disposed in a universal cord in which a CCD power transmission system and the like attached to the distal end bending portion 26 are disposed. The operator can operate an input means (not shown) such as the joystick 10 attached to the operation unit 116 covering the relay pulley 118 with a thumb or the like.

  The relay position detector 132 outputs a relay position signal 134 indicating the rotation angle of the relay pulley 118 to the adder 136 based on the rotation angle signal 130. That is, the potentiometer 128 and the relay position detector 132 are configured as relay position detection means for detecting the position of the traction means when the relay traction means is driven and outputting the relay position signal 134. The adder 136 receives the motor position command value signal 54 output from the adder 50 as a relay pulley position command value, obtains a deviation between the relay pulley position command value and the relay position signal 134, and uses this deviation as a relay position feedback controller. 138 is output. The relay position feedback controller 138 generates a relay position feedback control signal for suppressing the deviation calculated by the adder 136 to zero, and outputs the relay position feedback control signal to the adder 56 as a motor position command value. It has become. In the adder 56, a deviation between the motor position command value and the motor position signal 60 is obtained, and a drive signal corresponding to the deviation is generated by the motor controller 64, and this drive signal is amplified by the motor amplifier 66 to be supplied to the motor. 20 is driven. That is, the motor controller 64 performs control to cancel the reaction of the tension applied to the pulley 18 and cause the rotation angle of the motor 20 to follow the motor position command value.

  On the other hand, the relay position feedback controller 138 cancels the influence of the slack generated at the end of the pulley 18 connected to the motor 20 and the friction of the pulling wires 120 and 122 in the universal cord, and the rotation angle of the relay pulley 118. Is controlled to follow the relay pulley position command value 54.

  Further, the feedback controller 48 performs control to cancel the influence of slack and friction existing in the endoscope insertion portion at the distal end from the relay pulley 118 and to make the pulling wire distal end position follow the operation command value.

  Further, in the feedforward controller 52, control for advancing the phase of the operation command value is performed, and control for moving the dynamics from the pulley position where the phase is delayed to the wire tip position is performed.

In this reference example , when the potentiometer 128 or the like is attached to the relay pulley 118, the operation unit 116 becomes large and heavy. Therefore, when the rotation angle of the relay pulley 118 may not be detected, the motor position feedback system and the wire position feedback The system may be implemented. In addition, when the sensors 28 and 30 cannot be attached, the motor position command value may be created only by the feedforward controller 52 configured by the phase advance filter. Furthermore, a configuration may be adopted in which a tension sensor is attached to the insertion portion, and the tip position of the wire is estimated from the tension detected by the tension sensor. Further, a configuration may be adopted in which a tension sensor is attached to the pulling wire on the motor side and the position of the pulling wire is estimated and fed back.

This reference example is basically because of the combination of two of the endoscope insertion portion, to implement the providing a plurality of traction means also in the embodiment of the fourth reference example and the present invention from the first reference example be able to. Further, a configuration in which the relay pulley 118 is further increased by one stage can be employed.

In each of the above reference examples and the embodiments of the present invention, the bending portion 12 of the electric endoscope is described as an operation target, but other devices for pulling with a pulling wire or the like to operate the tip portion, For example, the present invention can be applied to a device that uses a wire or the like to open and close the mouth of a wire-driven robot arm (hand) or forceps.

Moreover, although the reference example and the embodiment of the present invention using the feedback controller and the feedforward controller have been described, the effect of providing each controller is achieved by providing only one of them. be able to.

Further, in each of the reference examples and the embodiments of the present invention , the control device is configured as an analog circuit. However, digital control using a microcomputer or the like is also possible.

Further, according to each of the above reference examples and the embodiment of the present invention , the operation command value accompanying the operation of the joystick 10 and the responsiveness of the bending operation of the bending portion 12 that is bent when the pulling wire is pulled by the motor 20 are improved. Therefore, the operability of positioning the electric endoscope can be improved.

1 is an overall configuration diagram of a positioning control device for an electric endoscope showing a first reference example of the present invention. FIG. It is a response waveform figure of the wire tip position and wire tension by the 1st reference example of the present invention. It is a response waveform diagram of the wire tip position and wire tension according to the prior art. It is a whole block diagram of the positioning control apparatus of the electric endoscope which shows the 2nd reference example of this invention. It is a response waveform figure of the wire tip position and feedforward controller by the 2nd reference example of the present invention. 1 is an overall configuration diagram of a positioning control apparatus for an electric endoscope showing an embodiment of the present invention. It is a circuit block diagram of a wire tension detector. It is a characteristic view of the model used with a position estimator. It is a response waveform diagram which shows the estimation result of the wire tip position in one Embodiment of this invention. It is a block block diagram of an endoscope state estimator. It is a diagram which shows the relationship between the evaluation value used for an endoscope state estimator, and a model. It is a whole block diagram of the positioning control apparatus of the electric endoscope which shows the 3rd reference example of this invention. It is a whole block diagram of the positioning control apparatus of the electric endoscope which shows the 4th reference example of this invention. It is a whole block diagram of the positioning control apparatus of the electric endoscope which shows the 5th reference example of this invention.

DESCRIPTION OF SYMBOLS 10 Joystick 12 Bending part 14, 16 Pulling wire 18 Pulley 20 Motor 22, 24 Coil sheath 28, 30 Sensor 36 Wire position detector 40 Controller 42 Adder 44 Command value detector 46 Command value signal 48 Feedback controller 50 Adder 52 Feed Forward controller 54 Motor position command value 56 Adder 58 Motor position detector 60 Motor position signal 62 Rotary encoder 64 Motor controller 66 Motor amplifier 82 Endoscope state estimator 84 Position estimator 108 Position estimator 112 Command value conversion 118 Relay pulley 128 Potentiometer 132 Relay position detector 136 Adder 138 Relay position feedback controller

Claims (3)

Command value signal output means capable of outputting a command value signal according to the operation;
Drive signal generating means for outputting a drive signal corresponding to the command value signal output from the command value signal output means;
Driving means for generating a driving force according to the driving signal output from the driving signal generating means;
Traction means for performing a traction operation based on the driving force generated by the driving means;
A towed mechanism having a bending portion that can be bent by the pulling means;
A sensor that outputs an internal tension signal of the traction means;
A ratio between the command signal output from the command value signal output means and the internal tension signal output from the sensor is obtained, and a ratio indicating the relationship between the ratio and the frequency characteristics of the towed mechanism is set in advance. State estimation means for estimating a frequency characteristic of the towed mechanism according to a function and outputting a state estimation result;
A frequency characteristic of the internal tension signal and the tip position of the traction means is provided in advance as a plurality of models, a model corresponding to the state estimation result is selected from the plurality of models, and the selected model and the internal tension signal are selected. Position estimating means for estimating the tip position of the curved portion of the towed mechanism;
Feedback control means for feedback correcting the command value signal output from the command value signal output means based on the estimation result of the tip position output from the position estimation means;
A feed-forward control unit that feed-forward corrects the command value signal input to the drive signal generation unit based on an output signal output from the feedback control unit and outputs the command value signal as the drive signal. Positioning control device. The positioning control device for a towed mechanism according to claim 1,
The state estimation means includes an evaluation value calculation circuit that obtains an evaluation value based on a ratio between the command value signal and the internal equal force signal, and a frequency characteristic of the towed mechanism that refers to the function according to the evaluation value. A function reference circuit for determining, a model determining circuit for determining the model to be selected by the position estimating means corresponding to the determined frequency characteristic of the towed mechanism, and outputting the model as the state estimation result; A positioning control device for a towed mechanism, comprising: In the control apparatus of the towed mechanism according to claim 1 or 2,
The position estimating means includes four low-pass filters having different characteristics as the model, and the function is a ratio between the command value signal and the internal tension signal or the evaluation value, and a cross of each low-pass filter. A positioning control device for a towed mechanism, characterized by being set in association with an over frequency.

Images