JP2009279699A - Position-force reproducing method and position-force reproducing device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a position-force reproducing method and position-force reproducing device capable of storing, in the form of data, and reproducing an operation in which contact and non-contact are repeatedly performed. <P>SOLUTION: This position-force reproducing device includes an actuator 201 as a means for acquiring position information and force contact information in time series, a control unit 203 for analyzing and processing the acquired position information and force contact information, and a storage means 204 for storing the result of the analysis and processing. The position information and the force contact information are reproduced according to the result of the stored analysis and processing. Consequently, both the position and force in contact and in non-contact can be reproduced, such an operation that contact and non-contact are repeated can be stored and reproduced, both the position and force of an operator A can be reproduced, and the operation of the operator A can be reproduced irrespective of time and space. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a position / force reproduction method and a position / force reproduction apparatus that store and reproduce contact and non-contact operations as data.

  Among human senses, visual, auditory, and tactile sensations are important senses that connect machines and humans, or humans and humans that exist in remote places. Image and sound processing technologies that correspond to vision and hearing are rapidly spreading as interfaces with the development of information and communication engineering, and are indispensable as industrial basic technologies. For example, the invention of photography, television, telephones, phonographs, etc. enabled the transmission, storage, and playback of information related to vision and hearing, and the broadcasting of vision and hearing information was also possible using video cameras and televisions. . This is equivalent to making the human sense of vision and hearing transcend time and space. Currently, the development of digital signal processing technology that handles the analysis and processing of such visual and auditory information is progressing day by day.

  On the other hand, tactile information, which is one of sensory information, is extracted, stored, processed, transmitted, and artificially reproduced as new multimedia information after visual and auditory information. Development of such technology is required. On the other hand, remote operation between the robot system by the master system and the slave system, and the haptic display that displays the moving position and various information at the position through the tactile sense, the tactile information is displayed. A lot of research has been done on the technology to be handled.

  However, the visual and auditory information described above is passive and unidirectional sensory information, whereas tactile information is bidirectional sensory information that is bound by the “law of action and reaction” in the real world. Yes, the information can be obtained for the first time by actively touching the environment that is the object of contact, so the haptic information is reproduced using the model base and virtual reality, but the haptic information in the real world has been reproduced. Acquisition is difficult. Real-world haptics is an academic field that deals with the extraction and reproduction of tactile information in the real world. Bilateral force feedback and multilateral (multi-directional) tactile transmission technology from remote locations in the real world have been developed. ing.

  In order to realize such bilateral force feedback in the real world, for example, as disclosed in Non-Patent Documents 1 and 2, research has been conducted mainly for the purpose of improving “transparency”. There are two types of “transparency”: spatial transparency and temporal transparency, of which spatial transparency corresponds to the degree of freedom of tactile reproducibility, so efficient integration of multi-degree-of-freedom force feedback is desired. . On the other hand, since the time transparency corresponds to the frequency band of the tactile information to be reproduced, transmission without time delay from the direct current component to the high frequency component of the tactile information is required.

  In order to achieve force feedback from a real-world environment with multiple degrees of freedom and wide bandwidth, acceleration control is required to make the control stiffness zero without losing robustness, which is disclosed in Non-Patent Document 3. Such a realization method using a disturbance observer is widely used. In particular, Non-Patent Document 4 shows that wide-range force information can be acquired without using a force sensor by using a disturbance observer, and many studies have been made on further widening the disturbance observer.

  In this way, research on the extraction and reproduction of haptic information in the real world and real time has been conducted in the past, but many conventional systems using robots, etc., store and reproduce only the position or force. Therefore, it was only possible to reproduce the environment without contact or only with contact. In addition, although research and development have been extensively conducted on technologies for storing and reproducing audio information and image information, technologies for storing and reproducing tactile information have not been established yet.

On the other hand, the present inventors have proposed “Haptograph” as a technique for visualizing tactile information (see Non-Patent Document 5, Japanese Patent Application No. 2007-212585, etc.). A haptograph can express a tactile sensation with a color like a photograph by performing frequency analysis on the acquired tactile information. In addition, there has been a demand for the development of technology related to storage and reproduction of information that has never existed using such technology.
Dee. A. D. A. Lawrence “Stability and Transparency in Bilateral Teleoperation”, IEEE Transactions on Robotics and Automation, Volume 9 No. 5 (Vol.9, No.5), pp. 624-637, October 1993 (October, 1993) Yokoko Yasuyoshi, Yoshikawa Tsuneo (T. Yoshikawa), “Bilateral Control of Master-Slave Manipulators for Ideal Kinesthetic Coupling-Formulation and Experiment ”, IEEE Transactions on Robotics and Automation, Vol. 10, No. 5, Vol. 10, No. 5, pp. 605-620, October 1994 (October, 1993) Koji Ohnishi, Masaaki Shibata, Toshiyuki Murakami (K. Ohnishi, M. Shibata, T. Murakami), “Motion Control for Advanced Mechatronics”, The Institute of Electrical and Electronics Engineers / Mechanical Society of Mechanical Engineers (American Society of Mechanical Engineers) IEEE / ASME Transactions on Mechatronics), Volume 1 (Vol.1), Issue 1 (No.1), pp. 56-67, March 1996 (March, 1996) Position / force reproduction method Toshiyuki Murakami, Toshiyuki Tsuji, Tohru Onishi (T. Murakami, F. Yu, K. Ohnishi), “Torque Sensorless Control in Multidegree-of-freedom Manipulator”, American Institute of Electrical and Electronics Engineers IEEE Transactions on Industrial Electronics, Vol. 40, No. 2 (Vol.40, No.2), pp. 259-265, April 1993 (April, 1993) Seiichiro Katsura, Kohei Irie, Kiyoshi Oishi (S. Katsura, K. Irie, K. Ohishi), “Visualization of Environmental Information by Haptograph Based on Wideband Force Control, Institute of Electrical and Electronics Engineers 33rd Annual Conference of the Industrial Electronics Division (IEEE Industrial Electronics Society), IECON 2007-Taipei (IECON'07-TAIPEI), pp. 368-373, 2007 November (November, 2007)

  The present invention has been made in view of the above problems, and an object of the present invention is to provide a position / force reproduction method and a position / force reproduction apparatus that can store and reproduce the operation of repeating contact and non-contact as data. There is.

  In order to solve the above-described problem, the position / force reproduction method according to the present invention acquires time-series position information and force contact information, analyzes the acquired position information and force contact information, and performs the analysis process. The result is stored in a storage means, and the position information and force contact information are reproduced based on the stored analysis processing result.

  In this case, the position information and force contact information can be reproduced at a variable speed.

  Further, the position information and the contact information of the force can be reproduced in reverse time series.

  Further, a large number of pieces of individual time-series position information and force contact information are obtained, the obtained many pieces of position information and force contact information are analyzed, and the analysis processing results are stored in a storage means. According to an analysis processing result, the multiple pieces of position information and force contact information are reproduced simultaneously or individually.

  Corresponding to the above method, the position / force reproduction device according to the present invention includes means for acquiring time-series position information and force contact information, and processing means for analyzing the acquired position information and force contact information. And storage means for storing the analysis processing result, and the position information and the force contact information are reproduced based on the stored analysis processing result.

  In this case, it is preferable to include reproduction condition setting means for reproducing the position information and the contact information of the force at a variable speed.

  In addition, it is preferable that a reproduction condition setting unit that reproduces the position information and the contact information of the force in reverse time series is provided.

  Furthermore, means for acquiring a large number of individual time-series position information and force contact information, a processing means for analyzing the acquired numerous position information and force contact information, and a storage means for storing the analysis processing results And a plurality of position information and force contact information are reproduced simultaneously or individually according to the stored analysis processing result.

  According to the method of claim 1 and the apparatus of claim 5, reproduction of both the position and force at the time of contact and non-contact can be realized, and an operation that repeats contact-non-contact can be stored and reproduced. it can.

  According to the method of claim 2 and the apparatus of claim 6, the operation of repeating contact-noncontact can be reproduced at constant speed and variable speed based on the stored data.

  According to the method of claim 3 and the apparatus of claim 7, the operation of repeating contact-noncontact can be reversely reproduced.

  According to the method of claim 4 and the apparatus of claim 8, the operation of repeating contact-noncontact by a large number of persons can be stored and reproduced simultaneously or individually.

  Preferred embodiments of the present invention will be described with reference to the accompanying drawings. The embodiments described below do not limit the contents of the present invention described in the claims. In addition, all the configurations described below are not necessarily essential requirements of the present invention.

  FIG. 1 shows a flow of storage and reproduction of audiovisual information and tactile information showing a preferred embodiment 1 of the present invention. As shown in the figure, the audiovisual information is stored by a microphone or camera and reproduced using a speaker or a display. Here, we propose a motion copy system that realizes the preservation and reproduction of tactile information. The motion copy system is a system that stores and reproduces tactile information using an actuator. The motion copy system includes a motion storage system 200 and a motion reproduction system 300. The motion storage system 200 stores the operation of the operator A, and the motion reproduction system 300 can reproduce both the position and the force of the operator A based on the stored operation data. That is, the operation of the operator can be reproduced across time and space. By using this motion copy technology, it is possible to save the operation of a skilled engineer and to allow the robot to inherit the technology.

  FIG. 2 is a diagram illustrating the concept of the motion storage system described above. As shown in the figure, the motion storage system 200 is operated by a bilateral control unit 203 by two actuators 201 and 202. The operator A on the master side can perceive the feel of the environment E on the slave side by the bilateral control unit 203. At this time, the tactile information on the master side is stored as operation data in the storage device 204 as storage means. The stored operation data is used when the motion reproduction system 300 reproduces the operation of the operator A.

  FIG. 3 is a diagram illustrating the concept of the motion reproduction system 300 described above. The motion reproduction system 300 uses one actuator 301, which is on the slave side. The actuator on the master side is virtually prepared, and this virtual master system 2K is configured by operation data stored in the storage device 204 of the motion storage system 200. Based on the stored operation data, tactile information is transmitted to the slave side, and the actuator 301 on the slave side operates accordingly. By performing the control in this way, the operation data of the master operator A stored in the past can be reproduced at any time.

  FIG. 4 shows an example of an experimental device of the motion storage system 200. The motion storage system 200 includes two actuators 201 and 202 each having a linear motor. As shown in the figure, the actuator 201 of the master system 2M is made of a metal having a rectangular plane as a fixing plate for installing the actuator 201. A base 12 made of a plate material is provided, and a round bar-shaped shaft body 13 made of metal is provided on the upper surface of the base 12 so as to be movable in the front-rear direction along the longitudinal direction of the upper surface of the base 12. In addition, a linear motor 14 fixed on the upper surface of the base 12 is mounted at a substantially central position in the longitudinal direction of the shaft body 13. The linear motor 14 allows the shaft body 13 to move in the longitudinal direction. Further, a position encoder 15 as a detecting means is attached to the other end of the shaft body 13 in the longitudinal direction. The position encoder 15 can measure the position and acceleration of the shaft body 13.

  Similarly, the actuator 202 of the slave system 3S is also provided with a base 22 made of a metal plate having a rectangular plane as an installation fixing plate for the actuator 202, and is made of a metal along the longitudinal direction on the upper surface of the base 22. A round rod-like shaft body 23 is provided on the upper surface of the base 22 so as to be movable in the front-rear direction. Further, a linear motor 24 fixed on the upper surface of the base 22 is mounted at a substantially central position in the longitudinal direction of the shaft body 23. The linear motor 24 allows the shaft body 23 to move in the longitudinal direction. Further, a position encoder 25 as a detecting means is attached to the other end of the shaft body 23 in the longitudinal direction. With the position encoder 25, the position and acceleration of the shaft body 23 can be measured.

  In the motion storage system 200, the operator A on the master side can touch the environment E on the slave side by the bilateral control unit 203. The slave-side environment E used in the experimental apparatus is an iron block. The motion storage system 200 stores the master position information and force information in the memory of the storage device 204 for each control period, and stores the information as an operation data file.

  FIG. 5 shows an example of an experimental apparatus of the motion reproduction system 300. As described above, the actuator 301 of the motion reproduction system 300 includes only one linear motor 24 on the slave side. The tip of the slave is set so that it can contact the iron block as environment E, like the motion storage system 200. Based on the operation data file stored in the storage device 204 of the motion storage system 200, the motion reproduction system 300 operates the actuator 301 of the slave system 3S. In addition, each system 200 and 300 comprises each part other than actuator system 201,202,301 by a control program, the control program is mounted by RT-Linux, and the sampling time is set to 100 microseconds. The control parameters are shown in Table 1 below.

FIG. 6 is a block diagram showing the concept of the motion storage system 200 described above. The motion storage system 200 includes a bilateral control unit 203, and is roughly divided into a joint space in the real world and a mode space in the virtual world. In the figure, the master motor is the linear motor 14 of the master system 2M, and the slave motor is the linear motor 24 of the slave system 3S. Each master is provided with disturbance observers 31 and 32, respectively, to compensate for disturbances and enable acceleration control. Parameters can be converted from joint space to mode space using a Quarry matrix. In the following description, except for figures and mathematical expressions, the first-order differential is represented as “•” and the second-order differential is represented as “••” for convenience, and written after the corresponding symbol. In addition, “^” representing an estimated value is additionally written before the corresponding symbol.

A quadratic Quarry matrix Q ₂ is shown in Equation 1 below.

The reaction force estimation observers 33 and 34 in the figure estimate the master reaction force F _M ^ext and the slave reaction force F _S ^ext . The estimated master and sub reaction forces F _M ^ ^ext and F _S ^ ^ext are summed mode response F _C ^res and difference mode response F _D ^{res according to the} Quarry matrix Q ₂ shown in Equation 2 below. Is converted to

Further, the master position response x _M ^res and the slave position response x _S ^res obtained from the position encoders 15 and 25 as position sensors are also converted into a mode space by the following equation (3).

Using the sum mode force response F _C ^res transformed by the Quarry matrix and the difference mode position response x _D ^res , the controller is configured as follows.

Here C _f is the gain of the force controller 35, the K _p proportional gain of the position controller 36, the K _d is the derivative gain. The acceleration reference value x ·· _C ^{ref of the} sum mode obtained by the force controller 35 and the acceleration reference value x _D ·· ^ref of the difference mode obtained by the position controller 36 are the Quarry inverse shown in the following equation (6). The matrix Q ₂ ⁻¹ is inversely transformed into the joint space master system 2M and the joint space slave system 3S, respectively.

Each actuator is controlled by the master acceleration reference value x ·· _M ^ref and the slave acceleration reference value x ·· _S ^ref which are inversely transformed as described above.

  In this way, force control is performed in the sum mode, and position control is performed in the difference mode, and the bilateral control unit 203 is configured. When the control unit 203 is constructed, the master operator A can grasp the tactile information of the slave.

The motion storage system 200 stores the master position response x _M ^res and the master reaction force F _M ^ ^ext in the memory of the storage device 204 and stores them as operation data. This motion data is used when the motion reproduction system 300 reproduces the motion of the operator A.

FIG. 7 shows a block diagram of the motion reproduction system 300. The basic structure of the motion reproduction system 300 is the same as that of the motion storage system 200. The motion control of the motion reproduction system 300 is the same as that of the motion storage system 200. That is, in the mode space, force control is configured in the sum mode, and position control is configured in the difference mode. The only difference from the motion storage system 200 is that there is no master system in the real world. Therefore, only one actuator 301 is used. In the virtual master system 2K of the motion reproduction system 300, a virtual master motor 14K is virtually prepared and configured by operation data stored in the motion storage system 200. The virtual master motor 14K unilaterally outputs time series data of the stored master reaction force F _M ^ ^ext and position response x _M ^res . By configuring the control unit 303 in this way, the stored master position and force can be reproduced without difference. That is, it is possible to reproduce the operation of the operator A stored in the past at any time.

  Next, experimental results using the experimental apparatus described with reference to FIGS. 4 and 5 will be described. FIG. 8 illustrates a case where the operator A operates the shaft body 13 and performs a free motion in which the shaft body 13 does not touch the environment E. FIG. 9 illustrates only the operation in which the shaft body 13 presses the environment E. FIG. 10 shows experimental results when the shaft body 13 collides with the environment E. FIG. 8A to 10B show position information and force information stored by the motion storage system 200, respectively. On the other hand, (c) and (d) of FIGS. 8 to 10 show the position and force reproduced by the motion reproduction system 300. From the results of this experiment, the reproduced force and position are perfect for the stored force and position, regardless of whether it is a free movement that does not contact the environment E, an action that pushes the environment E, or an action that collides with the environment E. You can see that Therefore, the motion copy system of the present invention can store and reproduce the operation of the operator A.

  As described above, in this embodiment, time-series position information and force contact information are acquired, the acquired position information and force contact information are analyzed, and the analysis processing result is stored in the storage device 204 serving as storage means. And a method for reproducing the position information and the force contact information based on the stored analysis processing result, and the position / force reproduction device is a means for acquiring time-series position information and force contact information. A certain actuator 201, a control unit 203 of the motion storage system 200, which is a processing means for analyzing the acquired position information and force contact information, and a storage device 204 for storing the analysis processing result, and depending on the stored analysis processing result The position information and force contact information are reproduced.

  As a result, it is possible to reproduce both the position and force at the time of contact and at the time of non-contact, and it is possible to store and reproduce the operation of repeating contact-non-contact. Then, both the position and force of the operator can be reproduced, and the operation of the operator can be reproduced over time and space. By using this motion copy technology, it is possible to save the operation of a skilled engineer and to allow the robot to inherit the technology.

  FIG. 11 is a conceptual diagram of a reproduction speed variable motion copy system showing a second preferred embodiment of the present invention. In this example, a motion copy system capable of varying the reproduction speed will be described. The motion copy system described above is an example that only reproduces a stored operation as it is at 1 × speed. If a motion copy system taking into account the variable reproduction speed is constructed, the stored operation of the operator A can be reproduced not only at 1 × speed but also at an arbitrary speed. This reproduction speed can be slow or conversely fast and can be set freely.

  The motion storage system 200 is not different from the storage system of the first embodiment. Therefore, the operation data stored in the normal motion storage system 200 shown in the first embodiment can be used as it is in the variable reproduction speed motion reproduction system.

  On the other hand, FIG. 12 shows a block diagram of a motion reproduction system 300A in which variable reproduction speed and the like are considered. The basic structure of the control unit 303 is the same as that of the motion reproduction system 300 of the first embodiment. However, the virtual master system 2K is different from the virtual master system 2K in that the reproduction condition setting means 41 is provided and a reproduction speed command is given. This reproduction speed command can be set freely. The output speed of the stored operation data is changed by a reproduction speed command given from the outside by the reproduction condition setting means 41.

  In addition, the reproduction condition setting means 41 sets the data stored in the storage device 204 of the motion storage system 200 to be reversely reproduced with the time reversed, or the reproduction speed (output speed) can be freely set in this reverse reproduction. These reproduction conditions can be variably input and set in the reproduction condition setting means 41 by the condition setting input means 42. Examples of the reproduction condition input means 42 include a keyboard and a switching operation knob.

  Next, experimental examples will be described. The experimental apparatus is the same as in the case of the motion copy system of Example 1, and the experimental apparatus of the motion storage system 200 shown in FIG. 4 and the experimental apparatus of the motion reproduction system of FIG. 5 were used. The experiment was conducted only when the shaft body 13 was made to collide with the environment E.

  FIG. 13 shows the operation data stored by the motion storage system. In contrast, FIGS. 14 to 17 show experimental results of the reproduction speed variable motion reproduction system 300A. 14A to 17A show the position response, and FIGS. 14B to 17B show the force response. FIGS. 14 to 17 show the reproduction speed of 1 time, 2 times, 3 times, and 1/2 times, respectively.

  It can be understood from these experimental results that even if the reproduction speed command by the reproduction condition setting means 41 is set to an arbitrary value, the reproduced position response and force response are the position information and force stored by the motion storage system. To match the information. Therefore, it can be confirmed that the stored operation of the operator A can be reproduced at an arbitrary speed. It is possible to obtain object position information and force information by slow motion or reverse playback, instead of slow motion of audio and images that have been performed so far. In the experiment, the reproduction speed was 1, 2, 3 and 1/2 times. However, the reproduction speed may be set in an analog manner, or the reverse reproduction with the time reversed. Even in this case, the reproduction speed can be made variable.

  As described above, in this embodiment, the position information and the contact information of the force can be reproduced at a variable speed by the reproduction condition setting means 41, and the operation of repeating contact-noncontact based on the stored data is performed. Can be reproduced at high speed and variable speed.

  Further, since the reproduction condition setting means 41 that reproduces the position information and the contact information of the force in reverse time series is provided, the operation of repeating contact-noncontact can be reproduced in reverse, and further, reverse reproduction is performed at a variable speed. You can also.

  FIG. 18 is a block diagram of a motion copy system showing a third preferred embodiment of the present invention. In the motion copy system of this example, multilateral control is performed in consideration of the identity ratio. This multilateral control is an extension of the bilateral control, and exchanges force and position between a plurality of motors to enable sharing of tactile information. In conventional multilateral control, a plurality of motors perform the same operation. Therefore, when a plurality of masters try to operate one slave, the plurality of operators A, A... Influence each other and crosstalk occurs. That is, each operator A, A ... perceives not only the tactile information of the slave but also the operations of other operators A, A .... This is a problem in constructing future tactile broadcasting technology.

Therefore, we propose multilateral control using identity ratio. By providing the multilateral control unit 205 in the motion copy system, information separation between the operators A, A. The purpose of this control system is to operate one slave with N-1 masters. In the figure, the equivalent acceleration x ·· _M1 ^{Fres to} x ·· _MN-1 ^Fres of the reaction force of the masters 1 to N-1 and the equivalent acceleration x ·· _S ^Fres of the reaction force of the slave are ^represented by the Nth-order Quarry matrix Q _N. The control unit 205 includes mode decomposition means for mode decomposition and mode decomposition. Control is performed by a force servo (force controller) 35 in the mode space by using the equivalent acceleration x ·· _C ^Fres of the reaction force of the sum mode that is decomposed. On the other hand, it is controlled by the position regulator (position controller) 36 using the actual acceleration x ·· _D1 ^{Pres to} x ·· _DN-1 ^Pres similarly decomposed into the difference modes. The acceleration reference value x ·· _C ^{ref in the} sum mode and the acceleration reference value x ·· _D1 ^{ref to} x ·· _DN ^ref in the difference mode are the accelerations of the masters 1 to N-1 by the Quarry inverse matrix Q ₂ ^-1 . The reference value x ·· _M1 ^{ref to} x ·· _MN-1 ^ref and the slave acceleration reference value x ·· _S ^ref can be converted, and the control unit 205 includes arithmetic processing means 51 for this purpose. Note that a calculation algorithm of a computer can be used for the arithmetic processing means 51.

In the conventional multilateral control, each acceleration reference value is input to each master as it is, but here, it is input to each master after being multiplied by the identity ratio. When there are N-1 masters, the identity ratio IR _i of the master i can be obtained from Equation 7 below. Further, the control unit 205 also multiplies each acceleration reference value by the identity ratio.

Here, F _i ^ext is an estimated reaction force of the master i. The identity ratio is the ratio of its own input to the total input applied to the slave, and reflects this to allow each master to operate independently. Therefore, after information is separated between the masters, a plurality of masters can operate one slave and share tactile information.

FIG. 19 shows a block diagram of each of the master systems 2M and 2M and the slave system 3S. The disturbance observer 31 estimates the disturbance F ^dis applied to the actuator and compensates for this to realize acceleration control. The equivalent acceleration x ·· ^Fres of the reaction force is calculated by the reaction force F ^ext estimated by the reaction force observer and the inertia M _n . The multilateral control system shown in FIG. 18 is configured using the acceleration response value x ·· ^Pres and the equivalent acceleration x ·· ^Fres of the reaction force shown in FIG.

  The motion copy system in this example is composed of a motion storage system 200B and a motion reproduction system 300B. Here, one of them, the motion storage system 200B will be described.

  The basic structure is equivalent to multilateral control considering identity ratio. The parameters used in the motion copy system are shown below. The motion copy system of this example can cope with the case where there are any number of master systems 2M. Here, an example constituted by two master systems 2M, 2M and one slave system 3S will be described.

  FIG. 20 shows an example of an experimental apparatus of the motion storage system 200B. In the experiment of the motion storage system 200B of this example, three actuators 201, 201A, 202 having linear motors are used, of which two actuators 201, 201A are master 1, master 2, and the remaining one actuator 202 is a slave. . The actuators 201 and 201A of the master 1 and the master 2 are operated by the respective operators A and A to operate the slaves. The tip of the shaft 13 of the slave-side linear motor 24 contacts the environment E, and this feeling is transmitted to each master by multilateral control. In this experiment, a sponge was used for environment E on the slave side. The hardness of the sponge is not modeled. That is, the parameter is unknown to the control system. The actuators 201, 201A, and 202 are controlled by the control unit 303, and are program-controlled by RT-Linux. The operation of each operator A is saved in the memory of the storage device 204, and the operation data is output to a file. The stored operation data file is used in the motion reproduction system 300B.

  FIG. 21 shows an example of an experimental apparatus of the motion reproduction system 300B. Since the slave is virtually prepared, unlike the motion storage system 200B, the actuator 301 uses two linear motors 24 and 24A. The two linear motors 24 and 24A serve as a master 1 and a master 2, respectively. The tips of the shafts 23 and 23 of each master are set so that they can contact the sponge used in the motion storage system 200B. The motion reproduction system 300B is also controlled by the control unit 303, controlled by RT-Linux, and is reproduced based on the operation data file created by the motion storage system 200B.

FIG. 22 shows a block diagram of the motion storage system 200B. The motion storage system 200B has two spaces, a virtual mode space and a real-world joint space. Any physical parameter in the real-world joint space can be transformed into a virtual mode space by the Quarry matrix. The cubic Quarry matrix Q ₃ is given by the following equation (8).

Multi-lateral control can be realized by the Quarry matrix Q ₃ . The equivalent accelerations x ·· _M1 ^Fres , x ·· _M2 ^Fres , x ·· _S ^Fres of the reaction forces of the master systems 2M and 2M and the slave system 3S are decomposed into the equivalent accelerations of the respective modes by the following ^equation (9). The unit 205 includes a decomposition means for the decomposition.

Similarly, in the case of the actual acceleration x ·· _M1 ^Pres , x ·· _M2 ^Pres , x ·· _S ^Pres of the master and the slave, the joint space is converted to the mode space using the following equation 10, and the control unit 205 A change means for conversion is provided.

Of the accelerations converted to the mode space, the actual acceleration used in the common mode is x ·· _C ^Fres , the actual acceleration x ·· _D1 ^{Pres in the} differential mode ₁ and the differential mode. There are only 3 actual accelerations x _{2 and D2} ^Pres . In the common mode, the ^reactive acceleration equivalent acceleration x ·· _C ^Fres is used and is controlled by the force servo 35 as shown in ^Equation 11 below.

In Equation 11, C _f is the gain of the force controller 35.

In one differential mode, the position regulator 36 is configured by the following equations 12 and 13 using the actual acceleration x ·· _D1 ^Pres and x ·· _D2 ^Pres .

Here, K _P and K _v are gains of the position regulator 36. The acceleration reference values of the respective modes calculated in the above equations 11 to 13 can be inversely transformed from the virtual mode space to the real world joint space by the Quarry inverse matrix Q _S ⁻¹ , and the control unit 303 performs the inverse transformation. Inverse conversion means is provided.

The acceleration reference value x ·· _M1 ^{ref to} x ·· _M2 ^ref of each master and the acceleration reference value x ·· _S ^{ref of the} slave are calculated by the above equation (14). In the case of performing multilateral control using an identity ratio, the acceleration reference value is actually input to the master and slave after being multiplied by each identity ratio. The identity ratio IR ₁ of the master 1 is calculated using the following formula 15, and the identity ratio IR ₂ of the master ₂ is calculated using the following formula 16.

This identity ratio is calculated online and changes from moment to moment with each master's input. Here, the identity ratio is valid when the input of each master is equal to or greater than the threshold value of 0.3N. The identity ratio forcibly changes the position threshold of the master in real space. Accordingly, it is necessary to virtually align position responses with other masters and slaves. Therefore, the response value of each master is multiplied by the reciprocal of the identity ratio. These controls are performed by the control unit 205.

In the motion preserving system 200B, the slave ^cycle reaction force equivalent acceleration x ·· _S ^Fres and actual acceleration x ·· _S ^Pres , master 1 identity ratio IR ₁ , master 2 identity ratio IR ₂ time-series data are controlled. Save every time. The stored data is used when the motion reproduction system 300B reproduces the operations of the operators A and A.

FIG. 23 shows a block diagram of the motion reproduction system 300B. The basic structure of the motion reproduction system 300B is the same as that of the motion storage system 200B. However, the difference from the motion storage system 200B is that a virtual slave system 3K is virtually prepared by regarding the stored data as a virtual slave. That is, the reproduction system 300B does not use a slave. This virtual slave unilaterally outputs time-series data of the stored slave reaction force equivalent acceleration x ·· _S ^Fres and actual acceleration x ·· _S ^Pres . Further, according to the identity ratio IR _{1 of} the master ₁ and the identity ratio IR ₂ of the master 2 stored in the motion storage system 200B, the respective identity ratios are also operated. In this way, the master 1 and the master 2 are controlled according to the slave prepared virtually using the operation data stored in the motion storage system 200B. With the proposed motion reproduction system 300B, the operations of the operator A of the master 1 and the operator A of the master 2 can be reproduced independently.

  Next, experimental results using the experimental apparatus shown in FIGS. 20 and 21 will be described. 24 shows the experimental results of the motion storage system 200B, FIG. 24 (a) shows the position response, FIG. 24 (b) shows the force response, and FIG. 24 (c) shows the change in the identity ratio. In conventional multilateral control, the position responses of all systems are the same. However, FIG. 24 (a) shows that the position response of each master and slave is different according to the identity ratio. From the identity ratio, it can be confirmed that the position response of the slave can be distributed to each master according to the input. In the experiment of the motion storage system 200B, the slave operation data and the identity ratio of each master are stored.

  FIG. 25 shows experimental results of the motion reproduction system 300B. The slave data and the identity ratio of each master in FIG. 25 are operation data stored by the motion reproduction system 300B. It can be confirmed that the position responses of the master 1 and the master 2 in FIG. 25A match the position responses of the master 1 and the master 2 shown in FIG. Also, the force response of each master shown in FIG. 25 (b) also matches the force response of each master shown in FIG. 24 (b). That is, the position and force of each master follows the position and force stored in the past. It can also be confirmed that the positions and forces of the operator A of the master 1 and the operator A of the master 2 can be reproduced. Therefore, it is shown that the master 1 and the master 2 can store and reproduce the motion independently.

  As described above, the motion data of each master can be stored by the motion storage system 200B and can be reproduced by the motion reproduction system 300B. As a result, it was confirmed that the position and force of the operator A of each master can be reproduced by a series of motion copy systems.

  As described above, in this embodiment, a large number of pieces of individual time-series position information and force contact information are acquired, the acquired many pieces of position information and force contact information are analyzed, and the analysis processing results are stored in the storage means. The actuator 201 serving as means for storing and reproducing a large number of position information and force contact information simultaneously or individually according to the stored analysis processing result, and acquiring a large number of individual time-series position information and force contact information, 201A, a control unit 205 serving as a processing unit that performs analysis processing on a large number of acquired position information and force contact information, and a storage device 204 serving as a storage unit that stores analysis processing results. Because it is configured to reproduce position information and force contact information simultaneously or individually, it is possible to store and reproduce the operation of repeating contact-non-contact by a large number of people simultaneously or individually. .

  Further, by performing control using the identity ratio that is the ratio of its own input to the total input applied to the slave, information separation between the operators A, A... Is possible, and the motion reproduction system 300B is based on the stored data. Thus, the operations of the operators A and A can be reproduced independently.

  In addition, this invention is not limited to the said Example, A various deformation | transformation implementation is possible in the range of the summary of this invention.

  With the position / force reproduction method and position / force reproduction device presented in the above embodiment, both the operator's position and force can be reproduced, and the operation of the operator can be reproduced over time and space. . By using this motion copy technology, it is possible to save the operation of a skilled engineer and to allow the robot to inherit the technology.

It is the schematic which shows preservation | save and reproduction of audiovisual information and tactile information in Example 1 of this invention. It is explanatory drawing which showed the concept of the motion preservation | save system same as the above. It is explanatory drawing which showed the concept of the motion reproduction system same as the above. It is explanatory drawing which shows the experiment apparatus of a motion preservation | save system same as the above. It is explanatory drawing which shows the experimental apparatus of a motion reproduction system same as the above. It is a block diagram of a motion preservation system same as the above. It is a block diagram of a motion preservation system same as the above. It is a graph which shows the result of an experiment by an experimental device, and shows the case of free motion. It is an experiment result by an experimental apparatus as above, Comprising: It is a graph which shows the case of pushing operation | movement. It is a graph which shows the case of a collision operation | movement as an experimental result by an experimental apparatus same as the above. It is explanatory drawing which showed the concept of the reproduction speed variable motion copy system in Example 2 of this invention. FIG. 2 is a block diagram of a motion reproduction system in which reproduction speed is considered. It is a graph figure which shows the operation data preserve | saved by the motion preservation | save system same as the above. FIG. 5 is a graph showing experimental results of a reproduction speed variable motion reproduction system, and showing a reproduction speed of 1 times. FIG. 5 is a graph showing experimental results of a reproduction speed variable motion reproduction system, and showing a case where the reproduction speed is doubled. FIG. 5 is a graph showing experimental results of a reproduction speed variable motion reproduction system, and showing a reproduction speed of 3 times. FIG. 6 is a graph showing experimental results of a reproduction speed variable motion reproduction system, and a reproduction speed of 1/2 times. It is a block diagram of the motion preservation | save system in Example 3 of this invention. FIG. 2 is a block diagram of each master system and slave system. It is explanatory drawing which shows the experiment apparatus of a motion preservation | save system same as the above. It is explanatory drawing which shows the experimental apparatus of a motion reproduction system same as the above. It is a block diagram of a motion preservation system based on multilateral control and identity ratio. FIG. 2 is a block diagram of a motion reproduction system using multilateral control. It is an experiment result by an experiment apparatus, Comprising: It is a graph which shows the result of a motion preservation | save system. It is a graph which shows the result of an experiment by an experimental apparatus, and shows the result of a motion reproduction system.

Explanation of symbols

2M Master system 3S Slave system 41 Reproduction condition setting means 200 Motion storage system 200A Motion storage system 200B Motion storage system 201 Actuator (means for acquiring)
201A Actuator (Means to acquire)
203 Bilateral control unit 204 Storage device (storage means)
205 Multi-Lateral Control Unit 300 Motion Reproduction System 300A Motion Reproduction System 300B Motion Reproduction System 301 Actuator 303 Control Unit

Claims (8)

Obtain time-series position information and force contact information,
Analyzing the acquired position information and force contact information;
A position / force reproduction method characterized in that the analysis processing result is stored in a storage means, and the position information and force contact information are reproduced based on the stored analysis processing result.   2. The position / force reproduction method according to claim 1, wherein the position information and force contact information are reproduced at a variable speed.   3. The position / force reproduction method according to claim 1, wherein the position information and force contact information are reproduced in reverse time series. Acquire a lot of individual time-series position information and force contact information,
Analyzing the acquired position information and contact information of force,
A position / force reproduction method characterized in that the analysis processing result is stored in a storage means, and the position information and force contact information are reproduced simultaneously or individually by the stored analysis processing result. Means for acquiring time-series position information and force contact information;
Processing means for analyzing the acquired position information and force contact information;
Storage means for storing the analysis processing result,
A position / force reproduction apparatus configured to reproduce the position information and force contact information based on the stored analysis processing result.   6. The position / force reproduction method according to claim 5, further comprising reproduction condition setting means for reproducing the position information and force contact information at a variable speed.   7. The position / force reproduction method according to claim 5, further comprising reproduction condition setting means for reproducing the position information and force contact information in reverse time series. Means for acquiring a large number of individual time-series position information and force contact information;
Processing means for analyzing and processing the acquired multiple position information and force contact information;
Storage means for storing the analysis processing result,
A position / force reproduction method, wherein the plurality of position information and force contact information are reproduced simultaneously or individually based on the stored analysis processing result.