Dali Wang and Ying Bai
Stewart platforms, Calibration, Modeless
This paper proposes a modeless technique for the calibration of Stewart platforms. Traditional calibration techniques use parametric models of the platform, which typically involve either forward or inverse kinematics. The proposed method does not use a parametric model of the platform for the purpose of error compensation. Instead, it uses a neural network model to approximate the behavior of pose errors. The workspace of the robot is first divided into small subspaces. A neural network is then utilized to model the pose error within each subspace using the data measured at the selected point of surrounding area. After the training process, the neural network model is used for error compensation. It is shown that the proposed method can be used to simplify the calibration process and to improve the pose accuracy.
Important Links:
Go Back
