Neural Network Learning of Robot Arm Impedance in Operational Space
スポンサーリンク
概要
- 論文の詳細を見る
lmpedance control is one of the most effective controlmethods for the manipulators in contact with their environments.The characteristics of force and motion control, however, isdetermined by a desired impedance parameter of a manipulator'send-effector that should be carefully designed according to agiven task and an environment. The present paper proposesa new method to regulate the impedance parameter of theend-effector through learning of neural networks. Three kindsof the feed-forward networks are prepared corresponding toposition, velocity and force control loops of the end-effector beforelearning. First, the neural networks for position and velocitycontrol are trained using iterative learning of the manipulatorduring free movements. Then, the neural network for forcecontrol is trained for contact movements. During learning ofcontact movements, a virtual trajectory is also modified to reducecontrol error. The method can regulate not only stiffness andviscosity but also inertia and virtual trajectory of the end-effector.Computer simulations show that a smooth transition from freeto contact movements can be realized by regulating impedanceparameters before a contact.
- IEEEの論文
- 1996-00-00
IEEE | 論文
- Magnetic and Transport Properties of Nb/PdNi Bilayers
- Supersonic Ion Beam Driven by Permanent-Magnets-Induced Double Layer in an Expanding Plasma
- Surfactant Adsorption on Single-Crystal Silicon Surfaces in TMAH Solution: Orientation-Dependent Adsorption Detected by In Situ Infrared Spectroscopy
- Extended-range FMCW reflectometry using an optical loop with a frequency shifter
- Teachingless spray-painting of sculptured surface by an industrial robot