Sharif Digital Repository

This paper deals with the problems of attitude determination, parameter identification and reference sensor calibration simultaneously. An LEO satellite's attitude, inertia tensor as well as calibration parameters of Three-Axis-Magnetometer (TAM) including scale factors, misalignments and biases along three body axes are estimated during a maneuver designed to satisfy the condition of persistency of excitation. The advanced nonlinear estimation algorithm of Unscented Kalman Filter (UKF) is a good choice for nonlinear estimation problem of attitude determination, but its computational cost is considerably larger than the widespread low accurate Extended Kalman Filter. Reduced Sigma Point... 

Calibrating the inverse kinematics of complex robots is often a challenging task. Finding analytical solutions is not always possible and the convergence of numerical methods is not guaranteed. The model-free approaches, based on machine learning and artificial intelligence, are fast and easy to work, however, they need a huge amount of experimental training data to provide acceptable results. In this article, we proposed a hybrid method to benefit the advantage of both model-based and model-free approaches. The forward kinematics of the robot is calibrated using a model-based approach, and its inverse kinematics using a neural network. Hence, while there is no need to solve the nonlinear... 

Inspired by algorithms utilized in inertial navigation, an inertial motion capturing algorithm capable of position and heading estimation is introduced. The fusion algorithm is capable of real-time link geometry estimation, which allows for the imposition of biomechanical constraints without a priori knowledge regarding sensor placements. Furthermore, the algorithm estimates gyroscope and accelerometer bias, scaling, and non-orthogonality parameters in real-time. The stationary phases of the links, during which pseudo-measurements such as zero velocity or heading stabilization updates are applied, are detected using optically trained neural networks with buffered accelerometer and gyroscope...