Gritsenko I.S.   Seidakhmet A.Z.   Bekbagambetov A.   Gritsenko P.S.  

Dynamical model of humanoid robot biolod type A

Reporter: Seidakhmet A.Z.

Dynamical model of humanoid robot biolod type A

A. Seidakhmet1*, I. Gritsenko2, A. Bekbaganbetov, P. Gritsenko
1Kazakh National University, Almaty, Kazakhstan
*e-mail address:seydakhmet@mail.ru
2Kazakh National University, Almaty, Kazakhstan
*e-mail address:bkopnq@yandex.kz

The goal of this work was to get precise model of kinematic and dynamic characteristics of small humanoid robot bioloid type A.  Obtained data was used for computer model development.  Created model can be used for robot locomotion dynamics and load analysis, also for motion planning and control research.It should be mentioned that obtained data and model can be used for analysis of various types of robots with different kinematic structure.
Our results are:
• Robot kinematics data is obtained and this data used for robot kinematics model development. Every robot limb is described as serial manipulator.  Robot model is described as 4 connected serial manipulators. This data and model can be easily reused for analysis of locomotion of other robots with different kinematic scheme.
• Dynamic characteristics – mass, centers of masses and inertia tensors are obtained. For plastic parts with uniform mass distribution dynamic characteristics are obtained with computer computations based on parts 3d models. This data was verified with dynamics measures of real parts.  Servo motors could not be analyzed as uniform mass distribution objects and dynamics data of motors ax-12A is obtained with measures.
• Obtained data is verified with number ax-12a motors loads measures.Robot Dynamics model is developed on basis of robotics toolbox MATLAB instruments. Model can connect and get data from Real robot online with help of our robot interface class which is based on USB2Dynamixel hardware and data transfer library. This solutions can be used for robot motion analysis and control strategies investigations and developments. Model provides ability to test control algorithms with virtual and real robot using same interface.

References
[1] P.I. Corke, “Robotics, Vision & Control”, Springer 2011, ISBN 978-3-642-20143-1.
[2] Marc H. Raibert,  “Dynamically Stable Legged Locomotion” 1983
[3] Bruno Siciliano, OussamaKhatib (Eds.) “Springer Handbook of Robotics” 2008 ISBN: 978-3-540-23957-4
[4] Frank L.Lewis “Robot Manipulator Control Theory and Practice” 2004 ISBN: 0-8247-4072-6