(19)
(11)EP 3 228 427 B1

(12)EUROPEAN PATENT SPECIFICATION

(45)Mention of the grant of the patent:
22.07.2020 Bulletin 2020/30

(21)Application number: 15865660.3

(22)Date of filing:  02.12.2015
(51)International Patent Classification (IPC): 
B25J 9/16(2006.01)
G06N 3/00(2006.01)
B25J 9/22(2006.01)
G06N 3/08(2006.01)
(86)International application number:
PCT/JP2015/005981
(87)International publication number:
WO 2016/088367 (09.06.2016 Gazette  2016/23)

(54)

ACTION TRANSFER DEVICE, ACTION TRANSFER METHOD, AND NON-TRANSITORY COMPUTER-READABLE MEDIUM IN WHICH PROGRAM IS STORED

AKTIONSÜBERTRAGUNGSVORRICHTUNG, AKTIONSÜBERTRAGUNGSVERFAHREN UND NICHTFLÜCHTIGES COMPUTERLESBARES MEDIUM MIT DARAUF GESPEICHERTEM PROGRAMM

DISPOSITIF DE TRANSFERT D'ACTION, PROCÉDÉ DE TRANSFERT D'ACTION ET SUPPORT NON TRANSITOIRE LISIBLE PAR ORDINATEUR SUR LEQUEL UN PROGRAMME EST STOCKÉ


(84)Designated Contracting States:
AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

(30)Priority: 02.12.2014 JP 2014243740

(43)Date of publication of application:
11.10.2017 Bulletin 2017/41

(73)Proprietor: Soinn Inc.
Tokyo 1940004 (JP)

(72)Inventors:
  • KIMURA, Daiki
    Tokyo 152-8550 (JP)
  • HASEGAWA, Osamu
    Tokyo 152-8550 (JP)

(74)Representative: Guérin, Jean-Philippe 
Opilex 32 rue Victor Lagrange
69007 Lyon
69007 Lyon (FR)


(56)References cited: : 
JP-A- H0 869 314
JP-A- 2007 276 052
US-B1- 6 272 396
JP-A- 2001 051 713
JP-A- 2011 110 621
  
  • BALAJI LAKSHMANAN ET AL: "Transfer learning across heterogeneous robots with action sequence mapping", INTELLIGENT ROBOTS AND SYSTEMS (IROS), 2010 IEEE/RSJ INTERNATIONAL CONFERENCE ON, IEEE, PISCATAWAY, NJ, USA, 18 October 2010 (2010-10-18), pages 3251-3256, XP031920116, DOI: 10.1109/IROS.2010.5649422 ISBN: 978-1-4244-6674-0
  • OSAMU HASEGAWA ET AL: "Home Robots, Learn by Themselves", 21 July 2013 (2013-07-21), UNIVERSAL ACCESS IN HUMAN-COMPUTER INTERACTION. APPLICATIONS AND SERVICES FOR QUALITY OF LIFE, SPRINGER BERLIN HEIDELBERG, BERLIN, HEIDELBERG, PAGE(S) 48 - 53, XP047034370, ISBN: 978-3-642-39193-4 * the whole document *
  • D Kimura: "Knowledge Transfer on Robots", Tokyo (JP), 25 September 2015 (2015-09-25), pages 1-143, XP055500113, Retrieved from the Internet: URL:http://t2r2.star.titech.ac.jp/rrws/fil e/CTT100695875/ATD100000413/ [retrieved on 2018-08-15]
  • MAKONDO NDIVHUWO ET AL: "Knowledge transfer for learning robot models via Local Procrustes Analysis", 2015 IEEE-RAS 15TH INTERNATIONAL CONFERENCE ON HUMANOID ROBOTS (HUMANOIDS), IEEE, 3 November 2015 (2015-11-03), pages 1075-1082, XP032837519, DOI: 10.1109/HUMANOIDS.2015.7363502 [retrieved on 2015-12-22]
  
Note: Within nine months from the publication of the mention of the grant of the European patent, any person may give notice to the European Patent Office of opposition to the European patent granted. Notice of opposition shall be filed in a written reasoned statement. It shall not be deemed to have been filed until the opposition fee has been paid. (Art. 99(1) European Patent Convention).


Description

Technical Field



[0001] The present invention relates to an action transfer device, an action transfer method, and a program. For example, the present invention relates to a technique for transferring an action from a transfer source robot by using a small number of learning samples acquired by a transfer destination robot.

Background Art



[0002] Currently, the use of robots has been expanding in human society including in the home. At present, an intellectual development mechanism for robots is still being developed, and thus actions that can be executed by robots and things that can be understood by robots are limited. On the other hand, various types of robots having different body characteristics have been developed. Assuming that such robots having the above limitations are introduced into the standard home, it would be inefficient to cause each of these robots to learn actions and the like independently.

[0003] Accordingly, there is a need for a mechanism for sharing the knowledge learned by each robot, in particular, action information, among other robots. The technique for obtaining an action by sharing action information among robots is referred to as "action transfer". In other words, the action transfer is a technique in which a transfer destination (target domain) robot efficiently learns an action by using action information obtained by a transfer source (source domain) robot.

[0004] Information (action information) based on which a robot acts can be considered as, for example, being information obtained by accumulating correspondence relations between a joint angle (joint value) and coordinates (end effector) of a leading end of an arm of a robot having certain physical properties (e.g., the length of an arm, or the number of joints). The use of such action information enables the robot to act. Accordingly, the physical properties have an important effect on the robot action transfer. However, it is difficult to identify the physical properties of a robot among various types of robots. Therefore, a mechanism for adapting action information obtained from another robot to the physical properties of the robot is important in the action transfer.

[0005] As the above mechanism, a technique in which physical properties of a transfer destination robot are obtained and then action information about a transfer destination robot is processed to be adapted to the physical properties of the transfer source robot is generally employed. However, in this technique, some advance preparation such as measurement of physical properties of the transfer destination robot is required. Further, if the physical properties of the transfer destination robot are changed, or if an error occurs in the measurement thereof, it is difficult to perform an action accurately. Furthermore, humans and animals can learn actions without obtaining information about their own physical properties in advance. Considering this, it seems to be a more realistic approach to achieve an action transfer based on experiences of a real robot, without obtaining information about the physical properties of the transfer destination robot in advance.

[0006] In this regard, Non Patent Literature 1 proposes a technique for transferring action samples of a transfer source robot to a transfer destination robot by using the respective numbers of action samples acquired from both the transfer source robot and the transfer destination robot that are the same as each other even when some of the physical properties of the transfer destination robot are unknown. Note that in this case, the transfer is achieved by fitting using a matrix calculation. Document "Transfer learning across heterogeneous robots with action sequence mapping" by Balaji Lakshmanan and published on October 18 2010 in Intelligent robots and systems (IROS), IEEE/RSJ International Conference ON, IEEE, PISCATAWAY, NJ, USA, discloses an action transfer method and an associated device. The action transfer device comprises a transfer source action information acquisition unit configured to acquire a plurality of pieces of action information about a transfer source robot, a transfer destination action information acquisition unit configured to acquire a plurality of pieces of first action information about a transfer destination robot, and a correction unit configured to generate a plurality of pieces of second action information about the transfer destination robot by correcting the action information by a prescribed update formula.

Citation List


Non Patent Literature



[0007] Non Patent Literature 1: Botond Bocsi, and two others, "Alignment-based Transfer Learning for Robot Models", [online], 2013, [searched on October 6, 2014], URL: http://www.cs.ubbcluj.ro/~bboti/pubs/ijcnn_2013.pdf

Summary of Invention


Technical Problem



[0008] However, Non Patent Literature 1 fails to describe details regarding the number of data samples required for the action transfer. It is considered that it is necessary to acquire the respective numbers of action samples from both the transfer source robot and the transfer destination robot that are the same as each other. Specifically, the technique needs to acquire a large number of learning samples not only in the transfer source but also in the transfer destination. This causes a problem that it takes a large amount of time for the technique to perform action transfer, it takes a large number of man-hours, cost, and the like, and it is difficult to achieve the action transfer rapidly and accurately.

[0009] In general, the use of a real robot on a simulator or in an experiment facility, or the like as a transfer source robot makes it relatively easy to acquire a large number of action samples. However, it is assumed that the transfer destination robot is a real robot provided for an actual operation at home or the like. Under such a situation, it is considered that it is difficult to acquire a number of learning samples the same as that of the transfer source robot. Accordingly, in the transfer destination robot, a mechanism for achieving the action transfer with high efficiency without the step of acquiring a large number of learning samples is required.

[0010] The present invention has been made to solve the above-mentioned problems, and an object of the present invention is to provide an action transfer device, an action transfer method, and a program which are capable of performing an action transfer from a transfer source robot by using a small number of learning samples acquired by a transfer destination robot.

Solution to Problem



[0011] An action transfer device according to the present invention is defined in appended claim 1.

[0012] An action transfer method according to the present invention is defined in the appended claim 4.

[0013] A program according to the present invention is a program for causing a computer to execute the action transfer method described above.

[0014] In the present invention, the action transfer device modifies a large number of data sets about the transfer source robot (action information about the transfer source robot) by using a small number of action samples (first action information about the transfer destination robot) acquired by the transfer destination robot, thereby generating a large number of data sets (second action information about the transfer destination robot) that are optimized by the transfer destination robot. Thus, the action of the transfer source robot can be transferred to the transfer destination robot without obtaining a large number of action samples in the transfer destination robot.

Advantageous Effects of Invention



[0015] According to the present invention, it is possible to provide an action transfer device, an action transfer method, and a program which are capable of performing an action transfer from a transfer source robot by using a small number of learning samples acquired by a transfer destination robot.

Brief Description of Drawings



[0016] 

Fig. 1 is a diagram showing a configuration of an action transfer device 100 according to a first example;

Fig. 2 is a flowchart showing an action of the action transfer device 100 according to the first example;

Fig. 3 is a diagram showing an example of an action algorithm of the action transfer device 100 according to the first example;

Fig. 4 is a flowchart showing an action of the action transfer device 100 according to an embodiment;

Fig. 5 is a diagram showing an example of an action algorithm of the action transfer device 100 according to the embodiment;

Fig. 6 is a diagram showing physical properties of a robot used for a comparative experiment;

Fig. 7 is a table showing physical properties of a robot used for a comparative experiment;

Fig. 8 is a table showing results of comparative experiments;

Fig. 9 is a diagram showing results of comparative experiments;

Fig. 10 is a diagram showing results of comparative experiments;

Fig. 11 is a diagram showing a robot used for a comparative experiment;

Fig. 12 is a diagram showing a robot used for a comparative experiment;

Fig. 13 is a table showing physical properties of the robot used for the comparative experiment;

Fig. 14 is a diagram showing results of comparative experiments;

Fig. 15 is a diagram showing results of comparative experiments;

Fig. 16 is a diagram showing results of a comparative experiment;

Fig. 17 is a diagram showing results of a comparative experiment;

Fig. 18 is a conceptual diagram showing a transfer of an action according to the present invention; and

Fig. 19 is a diagram showing an example of an algorithm for obtaining joint values for causing an arm to perform an action using a data set.


Description of Embodiments


<Introductory information>



[0017] Prior to the description of the invention, conditions under which an action transfer method according to the present invention is implemented will be described.

[0018] The present invention proposes a technique for implementing an action transfer between robots under the following conditions (Fig. 18). Embodiments of the present invention described below are also implemented under the following conditions. Note that the action of the robot refers to an operation for moving a specific part of the robot, such as an end effector of an arm (a tip end of an arm which corresponds to a hand of a human) to an arbitrary location (coordinate value).

[0019] Condition 1: The number of joints of a transfer source robot and the number of joints of a transfer destination robot are known and equal to each other.

[0020] Condition 2: Physical properties (e.g., the length of an arm) other than the number of joints are unknown.

[0021] Condition 3: The transfer source robot can extract a large number of action samples. The transfer source robot is typically a robot that is present on a simulator or in an experiment facility.

[0022] Condition 4: The transfer destination robot can extract only a small number of action samples. The transfer destination robot is typically a real robot

[0023] In the technique according to the present invention, an action is transferred to the transfer destination robot from the transfer source robot that satisfies the above-mentioned restrictive conditions. The action transfer indicates that an experiment database

corresponding to a work space for the transfer destination robot is created from an action experiment database Dsrc created on the transfer source robot. At this time,

is created by first causing the transfer destination robot to perform a small number of actions to create an experiment data set Dtgt and then correcting Dsrc by Dtgt. In this case, the size of Dtgt is smaller than the size of Dsrc, and the size of

is the same as the size of Dsrc. Thus, according to this technique, learning can be achieved using a smaller number of data sets than when data sets are generated by causing the transfer destination robot to actually perform a large number of actions. Further, the accuracy is improved as compared with a case where Dsrc is copied to create

Note that the experiment database, such as Dsrc, may be a simple set of pairs of joint values and coordinate values. In addition to a simple database (a database recording information about a simple set of pairs of joint values and coordinate values), for example, a database obtained by compressing information using SOM (Self Organizing Maps) by competitive learning, SOINN (Self Organizing Incremental Neural Network) which is a technique for online unsupervised learning capable of additional learning by modifying the SOM, or the like.

[0024] Note that the SOM and the COINN are described in the following Reference 1 and Reference 2, and thus detailed descriptions thereof are omitted.
(Reference 1) Kohonen T., "Self-organized formation of topologically correct feature maps", Biological Cybernetics Vo. 43, pp. 59-69, 1982
(Reference 2) Shen F., Hasegawa O., "A Fast Nearest Neighbor Classifier Based on Self-organizing Incremental Neural Network" Neural Networks, 2008

[0025] A method for creating an experiment database D for actions on robots will now be described. This method can be applied to the creation of Dsrc and Dtgt. In general, when the physical properties of robots are known, the use of a well-known technique of Inverse Kinematics (IK) makes it possible to control the end effector of the arm to be moved to an arbitrary location. However, the physical properties are unknown in the current assumption. Accordingly, in the technique according to the present invention, IK based on experiences is carried out to thereby move the end effector. Data to be acquired by the robot so as to execute the IK based on experiences and a method for executing the IK using the data are disclosed below.

[0026] First, the following joint value set which is data indicating joint angles of all joints of the target robot J = {J1, J2, ..., Jn|Jieach joint value} is generated. For example, in the case of an m-axis robot, i.e., a robot having m joints, the following joint value

is represented by an m-dimensional vector.

[0027] Further, the robot is caused to actually act using the joint values included in J, and absolute coordinate values (hereinafter referred to simply as coordinate values) of the end effector of the arm are obtained by various sensor information (e.g., visual information) about the robot. When the physical properties of the robot on the simulator or the like are known, the use of Forward Kinetics makes it possible to reduce a time required to acquire the coordinate values. Further, the following data set indicating the calculated coordinate values is generated.

At this time, the coordinate values of the end effector are uniquely determined from the joint values of the robot. In other words, there is a functional relation between the joint values and the coordinate values.

[0028] The joint values and the coordinate values which are acquired from these experiences, or actual actions, are represented by a data set D (Formulas (1) and (2)).





[0029] In this case, on the basis of the acquired data set D, a method for obtaining the following joint value

for moving the end effector of the arm to the coordinate value

which is randomly designated is described. In this case, k pieces of coordinate value data in the vicinity of the designated coordinate value

are searched from the data set D, and
inverse distance weighted average of the joint values

respectively corresponding to the k coordinate values is calculated to thereby obtain



[0030] In this case, a plurality of joint values that allow the end effector to move to the designated coordinate value may be present. This event is more likely to occur as the number of joints increases. Accordingly, even when the coordinate values are similar values, the joint values for reaching the respective coordinate values may greatly differ from each other. In this case, if an average value is calculated using only the coordinate value space, the average of the joint values that greatly differ from each other is used, which results in an increase in positional error. Accordingly, in this technique,

is calculated referring to the distance between the joint values in the joint value space.

[0031] Specifically, first, in the coordinate value space,

which is located closest to the designated coordinate value, and a peripheral coordinate value sample Xknn thereof are searched. Then, joint values

which respectively correspond to the searched coordinate values are obtained. Further, data Jselected which is closer to

is searched among the joint values Jknn. Then, inverse distance weighted average for the values is represented by

This algorithm is shown in Fig. 19. Note that a parameter k is set to 3 in this technique.

[0032] Subsequently, the embodiment of the present invention discloses a type of method for correcting Dsrc, which is acquired by the transfer source robot, to thereby obtain

that is, a data set to be adapted to the work space for the transfer destination robot. First, Transfer by SOM-algorithm is disclosed as a first example. Next, Transfer by Error propagation is disclosed as an embodiment.

<First example provided in an illustrative purpose>



[0033] Transfer by SOM-algorithm according to the first example is a method for acquiring a small number of action samples Dtgt in the transfer destination robot and modifying a coordinate value space Xsrc for the transfer source robot by using data Xtgt of the coordinate value space in Dtgt and an update formula created with reference to the SOM.

[0034]  Transfer by SOM-algorithm is an algorithm conceived from the update formula with reference to the conventionally known SOM (Self Organizing Maps). The SOM indicates self organizing maps devised by Kohonen et al. as a technique for changing a network structure (topology) according to an input. The SOM is one type of artificial neural networks and has a mechanism for modifying a topology of neurons according to an input. The SOM learning method is competitive learning, and neurons in the vicinity of the input are updated.

[0035] On the other hand, since setting of a target problem in the conventional SOM differs from setting of a target problem in the Transfer by SOM-algorithm, a new finding is added to the method of changing an update ratio in the Transfer by SOM-algorithm. In the conventional SOM, a node in one space is updated and the update ratio is determined using the distance between the input and a neighboring node of the input in the space. On the other hand, Transfer by SOM-algorithm is based on the premise that two spaces, i.e., a joint value space and an end effector coordinate space, are present. In this method, the coordinate value space is appropriate as the space in which updating is performed depending on input data. However, as described above, since constant restrictive conditions are present between the coordinate values and the joint values, it is important to consider the distance between the joint values in the joint value space. Therefore, in this method, a mechanism in which the distance between the joint values in the joint value space can be reflected in the update ratio of the coordinate value space is provided.

[0036] A detailed method for obtaining

using Transfer by SOM-algorithm is disclosed below.

[0037] Referring first to Fig. 1, a configuration of an action information transfer device 100 according to the first example will be described.

[0038] The action transfer device 100 is an information processing device which is typically a server computer, a personal computer (PC), or the like. The typical action transfer device 100 includes an arithmetic processing unit, a volatile or non-volatile storage device, and an input/output device. The arithmetic processing unit executes various controls based on programs stored in the storage device, thereby logically implementing each processing unit to be described later. The action transfer device 100 is not necessarily a physically single device, but instead may be implemented by dispersion processing using a plurality of information processing devices. Further, the action transfer device 100 may be incorporated in, for example, the transfer destination robot, or may be separated from the transfer destination robot.

[0039] A transfer source action information acquisition unit 101 acquires the data set Dsrc for causing the transfer source robot to act. Dsrc includes a joint value group Jsrc and a coordinate value group Xsrc of the end effector. Typically, Dsrc is generated by causing the transfer source robot to act several times on the simulator or in the experiment facility and acquiring a large number of pairs of the coordinate value of the end effector and the joint value corresponding to the coordinate value of the end effector. Assume in this example that Dsrc is generated in advance and the transfer source action information acquisition unit 101 acquires Dsrc from the input device or the storage device.

[0040] A transfer destination action information acquisition unit 102 acquires the data set Dtgt obtained when the transfer destination robot is caused to perform a test action. Dtgt includes a joint value group Jtgt and a coordinate value group Xtgt of the end effector. Typically, Dtgt can be generated by causing the transfer destination robot to act several times randomly and acquiring some pairs of coordinate value of the end effector and the joint value corresponding to the coordinate value. In this case, the size of Dtgt may be extremely smaller than the size of Dsrc. In other words, the number of action trials of the transfer destination robot may be smaller than the number of action trials of the transfer source robot.

[0041] Assume in this example that Dtgt is generated in advance and the transfer destination action information acquisition unit 102 acquires Dtgt from the input device or the storage device.

[0042] A correction unit 103 performs processing for modifying Dsrc into

by using Dtgt. In this example, the correction unit 103 performs modification processing using Transfer by SOM-algorithm.

[0043] An output unit 104 outputs

which is generated by the correction unit 103 to the outside. The transfer destination robot acquires

which is output from the output unit 104, and acts based on this value, thereby making it possible to acquire an action similar to that of the transfer source robot.

[0044] Subsequently, an action of the action transfer device 100 according to the first example will be described with reference to the flowchart of Fig. 2. Note that an algorithm shown in Fig. 3 is an implementation example of processing indicated by the flowchart.

S101: Initialization of a transfer destination data set



[0045] The transfer source action information acquisition unit 101 acquires the data set Dsrc for the transfer source.

[0046] The correction unit 103 generates

as the transfer destination data set. In this case, a value that is the same as Dsrc is set as an initial value (Fig. 3, second line).

S102: Repeat processing for all pairs of joint values and coordinate values of a transfer destination action sample



[0047] The transfer destination action information acquisition unit 102 acquires the action sample Dtgt for the transfer destination robot.

[0048] The correction unit 103 repeatedly executes the processing of S103 and subsequent steps for each pair of the joint value Jtgt and the coordinate value Xtgt which are included in Dtgt.

S103: Calculate a distance between the coordinate value of the transfer destination action sample and the nearest coordinate value of the transfer destination data set



[0049] The correction unit 103 searches the coordinate value

which is nearest the coordinate value

which is included in the action sample Dtgt of the transfer destination robot, from the data set

of the transfer destination, and calculates a distance dnearest between the both coordinate values (Fig. 3, third and sixth to eighth lines).

[0050] More specifically, the correction unit 103 first searches, from the joint value group

in the data set of the transfer destination, the joint value

which is nearest

which is included in the action sample Dtgt of the transfer destination robot (Fig. 3, sixth line). Then, the correction unit 103 searches

which is a pair of

from the data set

of the transfer destination (Fig. 3, seventh line). Further, the correction unit 103 calculates the distance dnearest between

and

(Fig. 3, eighth line). In this manner, the joint value space is first calculated. This is because a plurality of joint values corresponding to a certain coordinate value may be present. In this case, if the calculation is started from the coordinate value space, the joint value corresponding to the coordinate value cannot be searched.

S104: Parameter setting



[0051] In this case, the correction unit 103 may set a parameter relating to modification processing for

(Fig. 3, fourth and fifth lines). In this example, a parameter for setting characteristics of Sigmoid function (described later) used for the modification processing can be set. The speed of modification processing (learning processing) can be adjusted by changing the characteristics of the Sigmoid function. Note that this parameter may be arbitrarily set.

S105: Correct joint values and coordinate values in a transfer destination data set



[0052] The correction unit 103 performs processing for modifying the data set

of the transfer destination. Specifically, all coordinate values

which are included in

are corrected to thereby optimize mapping of the joint value

and the coordinate values for the transfer destination robot (Fig. 3, ninth to twelfth lines).

[0053] More specifically, the correction unit 103 first calculates a distance dj between the joint value

which is included in

and

(Fig. 3, tenth line). Then, the correction unit 103 calculates the difference

between

and

(Fig. 3, eleventh line). Further, the correction unit 103 corrects

by the following update formula using

and the distance dnearest calculated in S103 (Fig. 3, twelfth line).



[0054] It should be noted herein that in the above-mentioned update formula, information about the joint value space is added when the coordinate value space is updated on the basis of the SOM update formula of the related art. This is because, unlike the SOM of the related art, the present example needs to handle a plurality of spaces such as the joint value space and the coordinate value space.

[0055] In this example, the action transfer device 100 modifies the data set for the transfer source robot using Transfer by SOM-algorithm, thereby generating a data set optimized for the transfer destination robot. In this modification processing, action samples that are acquired in the transfer destination robot and are less than the action samples included in the data set are used. Thus, the transfer destination robot can acquire the action of the transfer source robot even if a large number of action samples are not obtained in the transfer destination robot. In other words, an equivalent action can be implemented and the action is transferred.

[0056] Further, in this example, the update formula obtained by modifying the SOM of the related art and two spaces with different properties are taken into consideration at the same time is introduced. Consequently, the example provides a method appropriate for problem setting or a problem to be solved, i.e., action transfer between robots.

[0057] Further, according to this example, the action transfer can be achieved with a smaller amount of calculation than that of the related art. In particular, since the number of learning samples in the transfer destination robot is smaller than that in the related art, the amount of calculation can be remarkably reduced.

< Embodiment>



[0058] Transfer by error propagation according to the embodiment is a method for calculating an error between Dsrc and Dtgt and modifying the coordinate value space Xsrc for the transfer source robot by using the error value. In other words, the error value calculated using a small number of action samples is propagated to peripheral data in Dsrc, thereby performing modification. Specifically, for each sample in the action sample data Dtgt obtained in the transfer destination robot, similar actions are searched in the data set Dsrc for the transfer source robot. Further, an error in the coordinate values of the end effector corresponding to the both actions is calculated. Furthermore, the error is propagated to the data in Dsrc having a similar joint value, thereby modifying the coordinate value space.

[0059] A detailed method for obtaining

using Transfer by Error propagation is disclosed below.

[0060] The configuration of the action transfer device 100 according to the embodiment is similar to that of the first example, and thus the description thereof is omitted.

[0061] Subsequently, the action of the action transfer device 100 according to the embodiment of the present invention will be described with reference to the flowchart of Fig. 4. Note that an algorithm shown in Fig. 5 is an implementation example of processing shown in the flowchart.

S201: Initialization of a transfer destination data set



[0062] Like in the first example, the transfer source action information acquisition unit 101 acquires the data set data set Dsrc for the transfer source. Further, the correction unit 103 generates

as the transfer destination data set. In this case, a value that is the same as Dsrc is set as an initial value (Fig. 5, second line).

S202: Repeat processing for all pairs of joint values and coordinate values of a transfer destination action sample



[0063] Like in the first example, the transfer destination action information acquisition unit 102 acquires the action sample Dtgt for the transfer destination robot. The correction unit 103 repeatedly executes the processing of S203 and subsequent steps on each pair of the joint value Jtgt and the coordinate value Xtgt included in Dtgt.

S203: Calculate a difference between the coordinate value of the transfer destination action sample and the nearest coordinate value of the transfer destination data set



[0064] The correction unit 103 searches, from the data set

of the transfer destination, the nearest coordinate value

of the coordinate value

which is included in the action sample Dtgt of the transfer destination robot, and calculates the error

between the both values (Fig. 5, third and fifth to seventh lines).

[0065] More specifically, the correction unit 103 first searches, from the joint value group

in the data set of the transfer destination,
the nearest joint value

in the joint value space of the joint value

which is included in the action sample Dtgt for the transfer destination robot (Fig. 5, fifth line). Then, the correction unit 103 searches a pair

of

from the data set

of the transfer destination (Fig. 5, sixth line). Further, the correction unit 103 calculates an error

between the current input

and

(Fig. 5, seventh line). Also in this case, like in the first example, the calculation for the joint value space is carried out first.

S204: Parameter setting



[0066] In this case, the correction unit 103 may set a parameter relating to modification processing for

(Fig. 5, fourth line). Also in this case, like in the first example, the speed of the modification processing (learning processing) can be adjusted by changing the characteristics of the Sigmoid function.

S205: Correct joint values and coordinate values in a transfer destination data set



[0067] The correction unit 103 performs processing for modifying the data set

of the transfer destination. Specifically, all coordinate values

which are included in

are corrected to thereby optimize mapping of the joint value

and the coordinate values for the transfer destination robot (Fig. 3, ninth to twelfth lines).

[0068]  More specifically, the correction unit 103 first calculates the distance dj between the joint value

which is included in

and

(Fig. 3, ninth line). Then, the correction unit 103 corrects

by the following update formula using dj and the displacement

which is calculated in S103 (Fig. 5, tenth line). Thus, the above-mentioned displacement is also propagated to other coordinates.



[0069] In this case, the above-mentioned update formula is obtained by modifying the update formula of the first example. As will be described in detail later, in Transfer by SOM-algorithm of the first example, modification processing is performed in such a manner coordinate values other than

approach the input coordinate values (the coordinate value group Xtgt of the action sample Dtgt of the transfer destination). On the other hand, in Transfer by Error propagation of the embodiment, an error between the input coordinate value and the corresponding coordinate value in

is also propagated to other coordinate values, and modification is performed so as to move based on the error. In view of the problem setting, i.e., the action transfer of robots, in the present invention, modification of the data set by the update formula of the embodiment is more efficient.

[0070] According to this embodiment, the action transfer device 100 can achieve the action transfer more appropriately and more efficiently as compared with the first example.

<Experiments>



[0071] In order to verify the advantageous effects of the action transfer methods according to the first example and the embodiment, the inventor conducted two types of experiments. One of the experiments uses a two-axis robot with an end effector acting on a two-dimensional plane, and the other one of the experiments uses a six-axis real robot with an end effect acting in a three-dimensional space. A quantitative evaluation is mainly conducted using the former robot, i.e., the two-axis robot, and an effectiveness evaluation is conducted using the latter robot, i.e., the six-axis robot.

<Experiment using a two-axis robot>



[0072] Experiment environment: A two-axis one-arm robot is created on a simulator and an action is transferred on the simulator. After that a quantitative evaluation of a positional error of the end effector is conducted. Accordingly, four types of robots with different arm lengths were prepared. Fig. 7 shows the arm length of each robot. Fig. 6 shows a space in which the robots having different arm length can act. Experiments are conducted in two patterns. One of the patterns is that the range of movement of the joint value is limited to a range from 0 to 180 degrees (this case is hereinafter referred to as "the joint value has a limit", which corresponds to Fig. 6). The other one of the patterns is that the range of movement of the joint value is limited to a range from 0 to 360 degrees (this case is hereinafter referred to as "the joint value has no limit"). When the joint value has no limit, a plurality of candidates for the joint value for bringing the end effector to an arbitrary location may be present, and thus the generation of actions for the robots is not uniform.

[0073] In this experiment, the action transfer to robots A, B, and C is carried out using a robot Src as the transfer source robot. Further, the absolute value

of the positional error at that time is measured. In this case,

is a coordinate value designated as a target, and

is a coordinate value of the end effector where the robot has actually reached.

[0074] The data set Dsrc for the transfer source robot includes data of the joint values on a grid that are obtained by gradually moving the value of each joint of the robot Src by 0.0314≈π/100[rad], and the absolute coordinate value (XY-plane) of the end effector obtained from each of the joint values. As for the number of pieces of data, 10000 samples were used when the joint value has a limit, and 40000 samples were used when the joint value has no limit.

[0075] The action sample Dtgt for the transfer destination robot includes the joint values which are randomly set in the robots A, B, and C, and the coordinate values obtained from the respective joint values. The number of pieces of data is set to 50 samples, regardless of whether or not the joint value has a limit.

[0076] The evaluation in this experiment was carried out by conducting the experiment for moving the arm 200 times toward random targets 100 times and recording a positional error between the target and the end effector at that time. Parameters of as=1 and ae=0.1 were used. These values are a variety of values optimum for the respective parameters that are obtained by preliminarily carrying out the transfer, which is likely to be most difficult, at a small scale. In this experiment, the following six pieces of experimental data were compared.

STgt: when a small amount of sample data Dtgt is simply used as a data set for the transfer destination

Lsrc: when data Dsrc is simply copied and used as a data set for the transfer destination robot

TSom: when transfer is performed using Transfer by SOM-algorithm

TErr: when transfer is performed using Transfer by Error propagation

LTgt: when a large number of action samples are obtained in the transfer destination (theoretical minimum error)



[0077] Experimental result: Fig. 9 shows a box plot of positional errors when the joint value has a limit. Fig. 10 shows a box plot when the joint value has no limit. Fig. 8 shows average values and standard deviations when the joint value has a limit (L) and when the joint value has no limit (N).

[0078] As seen from Figs. 8 to 10, the use of Transfer by Error propagation (TErr) among several transfer methods makes it possible to suppress positional errors close to the theoretical minimum error (LTgt). This indicates that even when robots having different physical properties, such as an arm length, are used, if the transfer source robot has learned a sufficient number of samples (in this case, several tens of samples), the robots can achieve the action with about the same accuracy as that obtained when the robot has learned a large number of samples (several tens of samples), by collecting a small number of samples (50 samples) on the transfer destination robot. That is, time and effort for learning actions for each robot can be reduced and the action transfer can be performed efficiently.

[0079] Incidentally, even when Transfer by Error propagation is used, a certain number of positional errors are still present. This experiment is not sufficient to prove that the technique according to the present invention is actually effective as a robot action transfer method. Therefore, the effectiveness of the present invention in a real robot is verified as follows.

<Experiment using a six-axis real robot>



[0080] Experiment environment: In this experiment, a real robot shown in Fig. 12 is used as the transfer destination. This transfer source robot includes a six-axis left arm. As a transfer source robot for this robot, a robot with specs similar to the arm of the transfer destination is created on the simulator. Fig. 11 shows the appearance of the transfer source robot. Fig. 13 shows the lengths of the arms and wrists of this robot. In this case, two types of transfer source robots, i.e., a robot α and a robot β, were created on the simulator. The action transfer was carried out for the real robot by using the left arm of each of the transfer source robots α and β on the simulator and collecting a large number of databases.

[0081] Details regarding the data used during experiments will be described. The data set Dsrc acquired by the transfer source robot on the simulator is obtained by moving each joint by 0.0157≈π/200[rad], and the number of pieces of data is about 700,000 samples. A small number of action samples Dtgt acquired by the real transfer destination robot were obtained by randomly moving the real robot, and 100 samples were obtained.

[0082] In this case, two types of experiments, i.e., a quantitative evaluation using a six-axis robot and a demonstration experiment using a real robot, were conducted. The former experimental method is similar to the experiment using the two-axis robot described above. In the latter experiment, the real robot is caused to carry the end effector to an arbitrary target position and to grip an object that is placed at the location in advance. At this time, the absolute coordinates of the object position can be specified using a camera and a depth sensor which are mounted on the robot.

[0083] Also in this experiment, six pieces of experimental data of STgt, Lsrc, TSom, TErr, and LTgt were compared.

[0084] Experimental result: Fig. 14 shows a box plot of positional errors of the end effector when the robots α and β are used as transfer source robots and six methods are used. Fig. 15 shows average values and standard deviations. As seen from Figs. 14 and 15, like in the case of the two-axis robot, also in the six-axis robot, positional errors can be reduced with high efficiency by using Transfer by Error propagation. A time required for one transfer using Transfer by Error propagation is 56.8±3 [sec], which can be evaluated as a practically applicable time.

[0085] Next, an experiment for causing the real six-axis robot to grip an object placed at an arbitrary location by using information transferred from the robot β was conducted. As a result, it was confirmed that the robot can actually grip the object. Therefore, the technique according to the present invention can be evaluated as effective for the action transfer in the real environments.

<Comparison between examples>



[0086] According to the above experimental results, when the action transfer is performed by Transfer by SOM-algorithm according to the first example, a positional error is relatively large. On the other hand, according to Transfer by Error propagation of the embodiment, a positional error is extremely small.

[0087] Figs. 16 and 17 show the absolute coordinate space of the end effector in each of two modification methods when the action transfer from the robot A is performed using a two-axis robot as the transfer destination robot. Fig. 16 shows a case where Transfer by SOM-algorithm is used, and Fig. 17 shows a case where Transfer by Error propagation is used.

[0088] Referring to Figs. 16 and 17, each point A (represented by a circle) indicates the coordinate value Xtgt of a small number of action samples input on the transfer destination robot, and each point B (represented by a large black circle) indicates the coordinate value

which is obtained after the transfer. Each line segment indicates a locus of movement of coordinate values (correction modification for a data sets) by the action transfer. Each point C (represented by a small black circle) indicates a coordinate value assuming that a large number of samples can be acquired by the transfer destination robot. In this case, a state where the point B and the point C overlap each other indicates that the transfer method can sufficiently suppress errors in the transfer source robot, or can suppress errors in the same manner as in the case where the robot has learned about a number of samples the same as that of samples learned by the transfer source robot.

[0089] As is seen from Fig. 16, in the Transfer by SOM-algorithm, the point B, which indicates the coordinate values obtained after the transfer, is moved near the point A which indicates a small number of action samples input on the transfer destination robot. This is because the SOM on which the Transfer by SOM-algorithm is based was originally proposed as a technique for modifying a network for an unknown target, and is an algorithm for modifying the network in accordance with input information.

[0090] On the other hand, as seen from Fig. 17, in the transfer-by-error propagation, an error between the point A, which indicates a small number of action samples input on the transfer destination robot, and the coordinate point on the transfer source robot, i.e., the amount of correction for the coordinate point indicated by a line segment, is propagated to peripheral coordinate points, and the coordinate point of the data set is corrected. This is because the problem setting in Transfer by Error propagation is different from that in Transfer by SOM-algorithm, and Transfer by Error propagation is intended to "transfer" information to the transfer destination from a state where the information about the transfer source having an environment similar to that of the transfer destination is obtained in advance. As means for solving this problem, in Transfer by Error propagation, data is adapted to the transfer destination by using an error between a new environment (transfer destination) and a learned environment (transfer source).

[0091] That is, to solve the problem of transferring learned data from a certain environment to an environment similar to the environment, Transfer by Error propagation is suitably used and the use of Transfer by Error propagation makes it possible to achieve the transfer with less errors.

<Other Embodiments>



[0092] Note that the present invention is not limited to the embodiment described above and can be modified as appropriate without departing from the scope of the invention. For example, in the above embodiment, the Sigmoid function is employed for mapping of coordinate values of the transfer source and coordinate values of the transfer destination. However, instead of the Sigmoid function, any function, such as an arc tangent (arctan), may be employed.

[0093] Further, the above embodiment discloses methods for transferring action information on the premise that two spaces, i.e., the joint value space and the coordinate value space, are present. However, an arbitrary number N of spaces may be provided. In this case, the above-mentioned update formula can be arbitrarily extended depending on the number of spaces. For example, coordinate values may be defined as the functions of joint values and coordinate values by adding a sensor value space. Sensor values may be, for example, data indicating an arm state (e.g., a broken or bent arm). This enables the robot to act in consideration of the sensor values. Thus, it is considered that actions can be selected and generated depending on the robot itself and the peripheral state of the robot.

[0094] While the above embodiment has been described assuming that the present invention is mainly composed of hardware, the present invention is not limited to this. Any processing can be implemented by causing a CPU (Central Processing Unit) to execute a computer program. In this case, the computer program can be stored and provided to a computer using any type of non-transitory computer readable media. Non-transitory computer readable media include any type of tangible storage media. Examples of non-transitory computer readable media include magnetic storage media (such as floppy disks, magnetic tapes, hard disk drives, etc.), optical magnetic storage media (e.g. magneto-optical disks), CD-ROM (Read Only Memory), CD-R, CD-R/W, and semiconductor memories (such as mask ROM, PROM (Programmable ROM), EPROM (Erasable PROM), flash ROM, RAM (random access memory), etc.). The program may be provided to a computer using any type of transitory computer readable media. Examples of transitory computer readable media include electric signals, optical signals, and electromagnetic waves. Transitory computer readable media can provide the program to a computer via a wired communication line, such as an electric wire and an optical fiber, or a wireless communication line.

[0095]  This application is based upon and claims the benefit of priority from Japanese patent application No. 2014-243740, filed on December 2, 2014.

Reference Signs List



[0096] 
100
ACTION TRANSFER DEVICE
101
TRANSFER SOURCE ACTION INFORMATION ACQUISITION UNIT
102
TRANSFER DESTINATION ACTION INFORMATION ACQUISITION UNIT
103
CORRECTION UNIT
104
OUTPUT UNIT



Claims

1. An action transfer device (100) comprising:

a transfer source action information acquisition unit (101) configured to acquire a plurality of pieces of action information about a transfer source robot, a robot action being an operation for moving a specific part of a robot to an arbitrary location defined by its coordinate value;

a transfer destination action information acquisition unit (102) configured to acquire a plurality of pieces of first action information about a transfer destination robot that has different physical characteristics from the transfer source robot, the number of joints of the transfer source robot and the number of joints of the transfer destination robot being known and equal to each other; and

a correction unit (103) configured to generate a plurality of pieces of second action information about the transfer destination robot by correcting the action information about the transfer source robot by a prescribed update formula using the first action information about the transfer destination robot, the correction being performed by the update formula for each piece of first action information about the transfer destination robot by extracting, from the plurality of pieces of action information about the transfer source robot, a piece of action information comprising a second joint value located near a first joint value included in the piece of first action information about the transfer destination robot, calculating an error between a first coordinate value included in the piece of first action information about the transfer destination robot and corresponding to the first joint value and a second coordinate value included in the extracted piece of action information about the transfer source robot and corresponding to the second joint value, and propagating the error to the coordinate value of another piece of action information about the transfer source robot, wherein

the number of the pieces of the first action information about the transfer destination robot is smaller than the number of the pieces of the action information about the transfer source robot, and

the number of the pieces of the generated second action information about the transfer destination robot is larger than the number of the pieces of the first action information about the transfer destination robot;

an output unit (104) configured to output the generated pieces of second action information to the outside.


 
2. The action transfer device (100) according to Claim 1, wherein
the action information about the transfer source robot, the first action information about the transfer destination robot, and the second action information about the transfer destination robot each include a plurality of pieces of data defined by a plurality of different spaces, and
the update formula satisfies all restrictive conditions for the different spaces.
 
3. The action transfer device (100) according to Claims 1 or 2, wherein the action information about the transfer source robot, the first action information about the transfer destination robot, and the second action information about the transfer destination robot include two sets of data on a robot joint value and a coordinate value corresponding to a robot specific part.
 
4. An action transfer method comprising:

a transfer source action information acquisition step of acquiring a plurality of pieces of action information about a transfer source robot, a robot action being an operation for moving a specific part of a robot to an arbitrary location defined by its coordinate value;

a transfer destination action information acquisition step of acquiring a plurality of pieces of first action information about a transfer destination robot, the number of joints of the transfer source robot and the number of joints of the transfer destination robot being known and equal to each other; and

a correction step of generating a plurality of pieces of second action information about the transfer destination robot by correcting the action information about the transfer source robot by a prescribed update formula using the first action information about the transfer destination robot, the correction being performed by the update formula for each piece of first action information about the transfer destination robot by extracting, from the plurality of pieces of action information about the transfer source robot, a piece of action information comprising a second joint value located near a first joint value included in the piece of first action information about the transfer destination robot, calculating an error between a first coordinate value included in the piece of first action information about the transfer destination robot and corresponding to the first joint value and a second coordinate value included in the extracted piece of action information about the transfer source robot and corresponding to the second joint value, and propagating the error to the coordinate value of another piece of action information about the transfer source robot, wherein

the number of the pieces of the first action information about the transfer destination robot is smaller than the number of the pieces of the action information about the transfer source robot, and

the number of the pieces of the generated second action information about the transfer destination robot is larger than the number of the pieces of the first action information about the transfer destination robot;

a step of outputting the generated pieces of second action information to the outside.
 
5. A non-transitory computer-readable medium storing a program for causing a computer to execute the method according to Claim 4.
 


Ansprüche

1. Aktionsübertragungsvorrichtung (100), umfassend:

eine Übertragungsquellen-Aktionsinformationserfassungseinheit (101), die so gestaltet ist, dass sie eine Mehrzahl von Elementen von Aktionsinformationen über einen Übertragungsquellenroboter erfasst, wobei eine Roboteraktion eine Operation zur Bewegung eines bestimmten Teils eines Roboters an eine wahlfreie durch dessen Koordinatenwert definierte Position ist;

eine Übertragungsziel- Aktionsinformationserfassungseinheit (102), die so gestaltet ist, dass sie eine Mehrzahl von Elementen erster Aktionsinformationen über einen Übertragungszielroboter erfasst, der sich von dem Übertragungsquellenroboter unterscheidende physische Eigenschaften aufweist, wobei die Anzahl der Gelenke des Übertragungsquellenroboters und die Anzahl der Gelenke des Übertragungszielroboters bekannt sind und jeweils gleich sind; und

eine Korrektureinheit (103), die so gestaltet ist, dass sie eine Mehrzahl von Elementen zweiter Aktionsinformationen über den Übertragungszielroboter erzeugt, indem die Aktionsinformationen über den Übertragungsquellenroboter um eine vorgeschriebene Aktualisierungsformel unter Verwendung der ersten Aktionsinformationen über den Übertragungszielroboter korrigiert werden, wobei die Korrektur durch die Aktualisierungsformel für jedes Element der ersten Aktionsinformationen über den Übertragungszielroboter durchgeführt wird, indem aus der Mehrzahl von Elementen von Aktionsinformationen über den Übertragungsquellenroboter ein Element von Aktionsinformationen extrahiert wird, das einen zweiten Gelenkwert umfasst, der sich nahe einem ersten Gelenkwert befindet, der in dem Element erster Aktionsinformationen über den Übertragungszielroboter enthalten ist, Berechnen eines Fehlers zwischen einem ersten Koordinatenwert, der in dem Element erster Aktionsinformationen über den Übertragungszielroboter enthalten ist und dem ersten Gelenkwert entspricht, und einem zweiten Koordinatenwert, der in dem extrahierten Element von Aktionsinformationen über den Übertragungsquellenroboter enthalten ist und dem zweiten Gelenkwert entspricht, und Verbreiten des Fehlers an den Koordinatenwert eines anderen Elements von Aktionsinformationen über den Übertragungsquellenroboter, wobei

die Anzahl der Elemente der ersten Aktionsinformationen über den Übertragungszielroboter kleiner ist als die Anzahl der Elemente der Aktionsinformationen über den Übertragungsquellenroboter, und

die Anzahl der Elemente der erzeugten zweiten Aktionsinformationen über den Übertragungszielroboter größer ist als die Anzahl der Elemente der ersten Aktionsinformationen über den Übertragungszielroboter;

eine Ausgabeeinheit (104), die so gestaltet ist, dass sie die erzeugten Elemente zweiter Aktionsinformationen nach außen abgibt.


 
2. Aktionsübertragungsvorrichtung (100) nach Anspruch 1, wobei
die Aktionsinformationen über den Übertragungsquellenroboter, die ersten Aktionsinformationen über den Übertragungszielroboter und die zweiten Aktionsinformationen über den Übertragungszielroboter jeweils eine Mehrzahl von Datenelementen aufweisen, die durch eine Mehrzahl verschiedener Räume definiert sind, und
die Aktualisierungsformel alle einschränkenden Bedingungen der verschiedenen Räume erfüllt.
 
3. Aktionsübertragungsvorrichtung (100) nach Anspruch 1 oder 2, wobei die Aktionsinformationen über den Übertragungsquellenroboter, die ersten Aktionsinformationen über den Übertragungszielroboter und die zweiten Aktionsinformationen über den Übertragungszielroboter zwei Datenmengen über einen Robotergelenkwert und einen Koordinatenwert aufweisen, der einem roboterspezifischen Teil entspricht.
 
4. Aktionsübertragungsverfahren, umfassend:

einen Schritt der Übertragungsquellen-Aktionsinformationserfassung der Erfassung einer Mehrzahl von Elementen von Aktionsinformationen über einen Übertragungsquellenroboter, wobei eine Roboteraktion eine Operation zur Bewegung eines bestimmten Teils eines Roboters an eine wahlfreie durch dessen Koordinatenwert definierte Position ist;

einen Schritt der Übertragungsziel- Aktionsinformationserfassung der Erfassung einer Mehrzahl von Elementen erster Aktionsinformationen über einen Übertragungszielroboter, wobei die Anzahl der Gelenke des Übertragungsquellenroboters und die Anzahl der Gelenke des Übertragungszielroboters bekannt sind und jeweils gleich sind; und

einen Korrekturschritt der Erzeugung einer Mehrzahl von Elementen zweiter Aktionsinformationen über den Übertragungszielroboter, indem die Aktionsinformationen über den Übertragungsquellenroboter um eine vorgeschriebene Aktualisierungsformel unter Verwendung der ersten Aktionsinformationen über den Übertragungszielroboter korrigiert werden, wobei die Korrektur durch die Aktualisierungsformel für jedes Element der ersten Aktionsinformationen über den Übertragungszielroboter durchgeführt wird, indem aus der Mehrzahl von Elementen von Aktionsinformationen über den Übertragungsquellenroboter ein Element von Aktionsinformationen extrahiert wird, das einen zweiten Gelenkwert umfasst, der sich nahe einem ersten Gelenkwert befindet, der in dem Element erster Aktionsinformationen über den Übertragungszielroboter enthalten ist, Berechnen eines Fehlers zwischen einem ersten Koordinatenwert, der in dem Element erster Aktionsinformationen über den Übertragungszielroboter enthalten ist und dem ersten Gelenkwert entspricht, und einem zweiten Koordinatenwert, der in dem extrahierten Element von Aktionsinformationen über den Übertragungsquellenroboter enthalten ist und dem zweiten Gelenkwert entspricht, und Verbreiten des Fehlers an den Koordinatenwert eines anderen Elements von Aktionsinformationen über den Übertragungsquellenroboter, wobei

die Anzahl der Elemente der ersten Aktionsinformationen über den Übertragungszielroboter kleiner ist als die Anzahl der Elemente der Aktionsinformationen über den Übertragungsquellenroboter, und

die Anzahl der Elemente der erzeugten zweiten Aktionsinformationen über den Übertragungszielroboter größer ist als die Anzahl der Elemente der ersten Aktionsinformationen über den Übertragungszielroboter;

einen Schritt der Ausgabe der erzeugten Elemente zweiter Aktionsinformationen nach außen.


 
5. Nichtflüchtiges, computerlesbares Medium, das ein Programm speichert, das dazu dient, zu bewirken, dass ein Computer das Verfahren nach Anspruch 4 ausführt.
 


Revendications

1. Dispositif de transfert d'action (100) comprenant :

une unité d'acquisition d'informations d'action de source de transfert (101) conçue pour acquérir une pluralité d'éléments d'informations d'action concernant un robot source de transfert, une action de robot étant une opération pour déplacer une partie spécifique d'un robot vers un emplacement arbitraire défini par sa valeur de coordonnées ;

une unité d'acquisition d'informations d'action de destination de transfert (102) conçue pour acquérir une pluralité d'éléments de premières informations d'action concernant un robot destination de transfert qui a des caractéristiques physiques différentes de celles du robot source de transfert, le nombre d'articulations du robot source de transfert et le nombre d'articulations du robot destination de transfert étant connus et égaux entre eux ; et

une unité de correction (103) conçue pour générer une pluralité d'éléments de secondes informations d'action concernant le robot destination de transfert en corrigeant les informations d'action concernant le robot source de transfert par une formule de mise à jour prescrite utilisant les premières informations d'action concernant le robot destination de transfert, la correction étant effectuée par la formule de mise à jour pour chaque élément de premières informations d'action concernant le robot destination de transfert en extrayant, de la pluralité d'éléments d'information d'action concernant le robot source de transfert, un élément d'informations d'action comprenant une seconde valeur d'articulation située près d'une première valeur d'articulation incluse dans l'élément de premières informations d'action concernant le robot destination de transfert, en calculant une erreur entre une première valeur de coordonnée incluse dans l'élément de premières informations d'action concernant le robot destination de transfert et correspondant à la première valeur d'articulation et une seconde valeur de coordonnée incluse dans l'élément extrait d'informations d'action concernant le robot source de transfert et correspondant à la seconde valeur d'articulation, et en propageant l'erreur à la valeur de coordonnées d'un autre élément d'informations d'action concernant le robot source de transfert,

le nombre d'éléments des premières informations d'action concernant le robot destination de transfert étant inférieur au nombre d'éléments des informations d'action concernant le robot source de transfert, et

le nombre d'éléments des secondes informations d'action générées concernant le robot destination de transfert étant supérieur au nombre d'éléments des premières informations d'action concernant le robot destination de transfert ;

une unité de sortie (104) conçue pour transmettre à l'extérieur les éléments générés des secondes informations d'action.


 
2. Dispositif de transfert d'action (100) selon la revendication 1,
les informations d'action concernant le robot source de transfert, les premières informations d'action concernant le robot destination de transfert et les secondes informations d'action concernant le robot destination de transfert comprenant chacune une pluralité d'éléments de données définis par une pluralité d'espaces différents, et
la formule de mise à jour satisfaisant à toutes les conditions restrictives pour les différents espaces.
 
3. Dispositif de transfert d'action (100) selon la revendication 1 ou 2, les informations d'action concernant le robot source de transfert, les premières informations d'action concernant le robot destination de transfert et les secondes informations d'action concernant le robot destination de transfert comprenant deux ensembles de données sur une valeur d'articulation de robot et une valeur de coordonnées correspondant à une partie spécifique du robot.
 
4. Procédé de transfert d'action comprenant :

une étape d'acquisition d'informations d'action de source de transfert consistant à acquérir une pluralité d'éléments d'informations d'action concernant un robot source de transfert, une action de robot étant une opération pour déplacer une partie spécifique d'un robot vers un emplacement arbitraire défini par sa valeur de coordonnées ;

une étape d'acquisition d'informations d'action de destination de transfert consistant à acquérir une pluralité d'éléments de premières informations d'action concernant un robot destination de transfert, le nombre d'articulations du robot source de transfert et le nombre d'articulations du robot destination de transfert étant connus et égaux entre eux ; et

une étape de correction consistant à générer une pluralité d'éléments de secondes informations d'action concernant le robot destination de transfert en corrigeant les informations d'action concernant le robot source de transfert par une formule de mise à jour prescrite utilisant les premières informations d'action concernant le robot destination de transfert, la correction étant effectuée par la formule de mise à jour pour chaque élément de premières informations d'action concernant le robot destination de transfert en extrayant, de la pluralité d'éléments d'informations d'action concernant le robot source de transfert, un élément d'informations d'action comprenant une seconde valeur d'articulation située près d'une première valeur d'articulation incluse dans l'élément de premières informations d'action concernant le robot destination de transfert, en calculant une erreur entre une première valeur de coordonnée incluse dans l'élément de premières informations d'action concernant le robot destination de transfert et correspondant à la première valeur d'articulation et une seconde valeur de coordonnée incluse dans l'élément extrait d'informations d'action concernant le robot source de transfert et correspondant à la seconde valeur d'articulation, et en propageant l'erreur à la valeur de coordonnées d'un autre élément d'informations d'action concernant le robot source de transfert,

le nombre d'éléments des premières informations d'action concernant le robot destination de transfert étant inférieur au nombre d'éléments des informations d'action concernant le robot source de transfert, et

le nombre d'éléments des secondes informations d'action générées concernant le robot destination de transfert étant supérieur au nombre d'éléments des premières informations d'action concernant le robot destination de transfert ;

une étape de transmission à l'extérieur des éléments générés des secondes informations d'action.


 
5. Support non transitoire lisible par ordinateur stockant un programme permettant à un ordinateur d'exécuter le procédé selon la revendication 4.
 




Drawing






























































Cited references

REFERENCES CITED IN THE DESCRIPTION



This list of references cited by the applicant is for the reader's convenience only. It does not form part of the European patent document. Even though great care has been taken in compiling the references, errors or omissions cannot be excluded and the EPO disclaims all liability in this regard.

Patent documents cited in the description




Non-patent literature cited in the description