Line-of-sight-stabilization and tracking control for inertial platforms

Abstract

Nowadays, line of sight stabilization and tracking using inertially stabilized platforms (ISPs) are still challenging engineering problems. With a growing demand for high-precision applications, more involved control techniques are necessary to achieve better performance. In this work, kinematic and dynamic models for a three degrees-of-freedom ISP are presented. These models are based in the vehicle-manipulator system (VMS) framework for modeling of robot manipulators operating in a mobile base (vehicles). The dynamic model follows the Euler-Lagrange formulation and is implemented by numeric simulations using the iterative Newton-Euler method. Two distinct control strategies for both stabilization and tracking are proposed: (i) computed torque control and (ii) sliding mode control using the recent SuperTwisting Algorithm (STA) combined with a High-Order Sliding Mode Observer (HOSMO). Simulations using data from a simulated vessel allow us to compare the performance of the computed torque controllers with respect to the commonly used P-PI controller. Besides, the results obtained for the sliding mode controllers indicate that the Super-Twisting algorithm offers ideal robustness to the vehicle motion disturbances and also to parametric uncertainties, resulting in a stabilization precision of approximately 0,8 mrad.