Visual Motor Control of a 7 DOF Robot Manipulator Using Function Decomposition and Sub-Clustering in Configuration Space

Swagat Kumar, Naman Patel, Laxmidhar Behera

    Research output: Contribution to journalArticle

    10 Citations (Scopus)

    Abstract

    This paper deals with real-time implementation of visual-motor control of a 7 degree of freedom (DOF) robot manipulator using self-organized map (SOM) based learning approach. The robot manipulator considered here is a 7 DOF PowerCube manipulator from Amtec Robotics. The primary objective is to reach a target point in the task space using only a single step movement from any arbitrary initial configuration of the robot manipulator. A new clustering algorithm using Kohonen SOM lattice has been proposed that maintains the fidelity of training data. Two different approaches have been proposed to find an inverse kinematic solution without using any orientation feedback. In the first approach, the inverse Jacobian matrices are learnt from the training data using function decomposition. It is shown that function decomposition leads to significant improvement in accuracy of inverse kinematic solution. In the second approach, a concept called sub-clustering in configuration space is suggested to provide multiple solutions for the inverse kinematic problem. Redundancy is resolved at position level using several criteria. A redundant manipulator is dexterous owing to the availability of multiple configurations for a given end-effector position. However, existing visual motor coordination schemes provide only one inverse kinematic solution for every target position even when the manipulator is kinematically redundant. Thus, the second approach provides a learning architecture that can capture redundancy from the training data. The training data are generated using explicit kinematic model of the combined robot manipulator and camera configuration. The training is carried out off-line and the trained network is used on-line to compute the joint angle vector to reach a target position in a single step only. The accuracy attained is better than the current state of art.
    LanguageEnglish
    Pages17-33
    JournalNeural Processing Letters
    Volume28
    Issue number1
    DOIs
    Publication statusPublished - 2008

    Fingerprint

    Biomechanical Phenomena
    Manipulators
    Cluster Analysis
    Inverse kinematics
    Robots
    Decomposition
    Redundancy
    Learning
    Psychomotor Performance
    Redundant manipulators
    Robotics
    Jacobian matrices
    End effectors
    Clustering algorithms
    Joints
    Kinematics
    Cameras
    Availability
    Feedback

    Cite this

    @article{e73fc0ec199e46c49270dc4aaf826f70,
    title = "Visual Motor Control of a 7 DOF Robot Manipulator Using Function Decomposition and Sub-Clustering in Configuration Space",
    abstract = "This paper deals with real-time implementation of visual-motor control of a 7 degree of freedom (DOF) robot manipulator using self-organized map (SOM) based learning approach. The robot manipulator considered here is a 7 DOF PowerCube manipulator from Amtec Robotics. The primary objective is to reach a target point in the task space using only a single step movement from any arbitrary initial configuration of the robot manipulator. A new clustering algorithm using Kohonen SOM lattice has been proposed that maintains the fidelity of training data. Two different approaches have been proposed to find an inverse kinematic solution without using any orientation feedback. In the first approach, the inverse Jacobian matrices are learnt from the training data using function decomposition. It is shown that function decomposition leads to significant improvement in accuracy of inverse kinematic solution. In the second approach, a concept called sub-clustering in configuration space is suggested to provide multiple solutions for the inverse kinematic problem. Redundancy is resolved at position level using several criteria. A redundant manipulator is dexterous owing to the availability of multiple configurations for a given end-effector position. However, existing visual motor coordination schemes provide only one inverse kinematic solution for every target position even when the manipulator is kinematically redundant. Thus, the second approach provides a learning architecture that can capture redundancy from the training data. The training data are generated using explicit kinematic model of the combined robot manipulator and camera configuration. The training is carried out off-line and the trained network is used on-line to compute the joint angle vector to reach a target position in a single step only. The accuracy attained is better than the current state of art.",
    author = "Swagat Kumar and Naman Patel and Laxmidhar Behera",
    note = "Reference text: 1 Tsai RY (1987) A versatile camera calibration technique for high-accuracy 3D machine vision metrology using off-the-shelf TV cameras and lenses. IEEE J Rob Autom RA-3(4): 323-344 2 Seraji H, Long MK, Lee TS (1993) Motion control of 7-DOF arms: the configuration control approach. IEEE Trans Robot Autom 9(2): 125-139 3 Dubey RV, Euler JA, Babcock SM (1991) Real-time implementation of an optimization scheme for 7DOF redundant manipulators. IEEE Trans Robot Autom 7(5): 579-588 4 Tevatia G, Schaal S (2000) Inverse kinematics of humanoid robots. In: Proceedings of IEEE International Conference on robotics and automation. San Francisco, CA, pp 294-299 5 Patel RV, Shadpey F (2005) Control of redundant manipulators. Springer 6 T. Kohonen, Self-organization and associative memory: 3rd edition, Springer-Verlag New York, Inc., New York, NY, 1989 7 Martinetz TM, Ritter HJ, Schulten KJ (1990) Three-dimensional neural net for learning visual motor coordination of a robot arm. IEEE Trans Neural Netw 1(1): 131-136 8 Walter JA, Schulten KJ (1993) Implementation of self-organizing neural networks for visual-motor control of an industrial robot. IEEE Trans Neural Netw 4(1): 86-95 9 Behera L, Kirubanandan N (1999) A hybrid neural control scheme for visual-motor coordination. IEEE Control Syst Mag 19(4): 34-41 10 Nimit Kumar , Laxmidhar Behera, Visual–Motor Coordination Using a Quantum Clustering Based Neural Control Scheme, Neural Processing Letters, v.20 n.1, p.11-22, August 2004 [doi>10.1023/B:NEPL.0000039429.89321.07] 11 Walter JA (1998) PSOM network: learning with few examples, In: International Conference on robotics and automation. Leuven, Belgium, pp 2054-2059 12 Walter J, Nolker C, Ritter H (2000) The PSOM algorithm and applications. In: Proceeding of International ICSC symposium on neural computation, pp 758-764 13 Angulo VR, Torras C (2005) Speeding up the learning of robot kinematics through function decomposition. IEEE Trans Neural Netw 16(6): 1504-1512 14 PowerCube Manipulators, Amtec Robotics. http://www.amtec-robotics.com/. Accessed 06 June 2008 15 Mark W. Spong, Robot Dynamics and Control, John Wiley & Sons, Inc., New York, NY, 1989 16 John J. Craig, Introduction to Robotics: Mechanics and Control, Addison-Wesley Longman Publishing Co., Inc., Boston, MA, 1989 17 Open Source Computer Vision Library. http://www.intel.com/technology/computing/opencv/Accessed 06 June 2008 18 Wilson R Tsai camera calibration software, http://www.cs.cmu.edu/-rgw/TsaiCode.html Accessed 06 June 2008",
    year = "2008",
    doi = "10.1007/s11063-008-9079-8",
    language = "English",
    volume = "28",
    pages = "17--33",
    journal = "Neural Processing Letters",
    issn = "1370-4621",
    number = "1",

    }

    Visual Motor Control of a 7 DOF Robot Manipulator Using Function Decomposition and Sub-Clustering in Configuration Space. / Kumar, Swagat; Patel, Naman; Behera, Laxmidhar.

    In: Neural Processing Letters, Vol. 28, No. 1, 2008, p. 17-33.

    Research output: Contribution to journalArticle

    TY - JOUR

    T1 - Visual Motor Control of a 7 DOF Robot Manipulator Using Function Decomposition and Sub-Clustering in Configuration Space

    AU - Kumar, Swagat

    AU - Patel, Naman

    AU - Behera, Laxmidhar

    N1 - Reference text: 1 Tsai RY (1987) A versatile camera calibration technique for high-accuracy 3D machine vision metrology using off-the-shelf TV cameras and lenses. IEEE J Rob Autom RA-3(4): 323-344 2 Seraji H, Long MK, Lee TS (1993) Motion control of 7-DOF arms: the configuration control approach. IEEE Trans Robot Autom 9(2): 125-139 3 Dubey RV, Euler JA, Babcock SM (1991) Real-time implementation of an optimization scheme for 7DOF redundant manipulators. IEEE Trans Robot Autom 7(5): 579-588 4 Tevatia G, Schaal S (2000) Inverse kinematics of humanoid robots. In: Proceedings of IEEE International Conference on robotics and automation. San Francisco, CA, pp 294-299 5 Patel RV, Shadpey F (2005) Control of redundant manipulators. Springer 6 T. Kohonen, Self-organization and associative memory: 3rd edition, Springer-Verlag New York, Inc., New York, NY, 1989 7 Martinetz TM, Ritter HJ, Schulten KJ (1990) Three-dimensional neural net for learning visual motor coordination of a robot arm. IEEE Trans Neural Netw 1(1): 131-136 8 Walter JA, Schulten KJ (1993) Implementation of self-organizing neural networks for visual-motor control of an industrial robot. IEEE Trans Neural Netw 4(1): 86-95 9 Behera L, Kirubanandan N (1999) A hybrid neural control scheme for visual-motor coordination. IEEE Control Syst Mag 19(4): 34-41 10 Nimit Kumar , Laxmidhar Behera, Visual–Motor Coordination Using a Quantum Clustering Based Neural Control Scheme, Neural Processing Letters, v.20 n.1, p.11-22, August 2004 [doi>10.1023/B:NEPL.0000039429.89321.07] 11 Walter JA (1998) PSOM network: learning with few examples, In: International Conference on robotics and automation. Leuven, Belgium, pp 2054-2059 12 Walter J, Nolker C, Ritter H (2000) The PSOM algorithm and applications. In: Proceeding of International ICSC symposium on neural computation, pp 758-764 13 Angulo VR, Torras C (2005) Speeding up the learning of robot kinematics through function decomposition. IEEE Trans Neural Netw 16(6): 1504-1512 14 PowerCube Manipulators, Amtec Robotics. http://www.amtec-robotics.com/. Accessed 06 June 2008 15 Mark W. Spong, Robot Dynamics and Control, John Wiley & Sons, Inc., New York, NY, 1989 16 John J. Craig, Introduction to Robotics: Mechanics and Control, Addison-Wesley Longman Publishing Co., Inc., Boston, MA, 1989 17 Open Source Computer Vision Library. http://www.intel.com/technology/computing/opencv/Accessed 06 June 2008 18 Wilson R Tsai camera calibration software, http://www.cs.cmu.edu/-rgw/TsaiCode.html Accessed 06 June 2008

    PY - 2008

    Y1 - 2008

    N2 - This paper deals with real-time implementation of visual-motor control of a 7 degree of freedom (DOF) robot manipulator using self-organized map (SOM) based learning approach. The robot manipulator considered here is a 7 DOF PowerCube manipulator from Amtec Robotics. The primary objective is to reach a target point in the task space using only a single step movement from any arbitrary initial configuration of the robot manipulator. A new clustering algorithm using Kohonen SOM lattice has been proposed that maintains the fidelity of training data. Two different approaches have been proposed to find an inverse kinematic solution without using any orientation feedback. In the first approach, the inverse Jacobian matrices are learnt from the training data using function decomposition. It is shown that function decomposition leads to significant improvement in accuracy of inverse kinematic solution. In the second approach, a concept called sub-clustering in configuration space is suggested to provide multiple solutions for the inverse kinematic problem. Redundancy is resolved at position level using several criteria. A redundant manipulator is dexterous owing to the availability of multiple configurations for a given end-effector position. However, existing visual motor coordination schemes provide only one inverse kinematic solution for every target position even when the manipulator is kinematically redundant. Thus, the second approach provides a learning architecture that can capture redundancy from the training data. The training data are generated using explicit kinematic model of the combined robot manipulator and camera configuration. The training is carried out off-line and the trained network is used on-line to compute the joint angle vector to reach a target position in a single step only. The accuracy attained is better than the current state of art.

    AB - This paper deals with real-time implementation of visual-motor control of a 7 degree of freedom (DOF) robot manipulator using self-organized map (SOM) based learning approach. The robot manipulator considered here is a 7 DOF PowerCube manipulator from Amtec Robotics. The primary objective is to reach a target point in the task space using only a single step movement from any arbitrary initial configuration of the robot manipulator. A new clustering algorithm using Kohonen SOM lattice has been proposed that maintains the fidelity of training data. Two different approaches have been proposed to find an inverse kinematic solution without using any orientation feedback. In the first approach, the inverse Jacobian matrices are learnt from the training data using function decomposition. It is shown that function decomposition leads to significant improvement in accuracy of inverse kinematic solution. In the second approach, a concept called sub-clustering in configuration space is suggested to provide multiple solutions for the inverse kinematic problem. Redundancy is resolved at position level using several criteria. A redundant manipulator is dexterous owing to the availability of multiple configurations for a given end-effector position. However, existing visual motor coordination schemes provide only one inverse kinematic solution for every target position even when the manipulator is kinematically redundant. Thus, the second approach provides a learning architecture that can capture redundancy from the training data. The training data are generated using explicit kinematic model of the combined robot manipulator and camera configuration. The training is carried out off-line and the trained network is used on-line to compute the joint angle vector to reach a target position in a single step only. The accuracy attained is better than the current state of art.

    U2 - 10.1007/s11063-008-9079-8

    DO - 10.1007/s11063-008-9079-8

    M3 - Article

    VL - 28

    SP - 17

    EP - 33

    JO - Neural Processing Letters

    T2 - Neural Processing Letters

    JF - Neural Processing Letters

    SN - 1370-4621

    IS - 1

    ER -