Online One-Step Parameter Identification Method for a Space Robot with Initial Momentum in Postcapture

Abstract

For precise attitude control of space robots intended to capture a target, the inertial parameters of the target need to be well estimated because the motions of arms and bases are coupled. This paper presents a momentum-based, online, one-step identification method for estimating inertial parameters of tumbling targets with linear momentum by space robots after capture. For obtaining the target’s mass and mass center, the method is the same as the traditional momentum-based method, while two contributions are made for identifying inertia tensor. The first contribution is the second-order term mnrn×vn in the identification equation is written as the product of two functions: function of time and function of mass and mass center of the target. Secondly, the traditional recursive least squares (RLS) method is modified to obtain the corresponding inertia tensor without iteration from the beginning. To verify the validity and feasibility of the proposed identification technique, 2D and 3D models with different targets and initial velocities are both simulated and analyzed. The simulation results show that all the estimated values converge to their ideal values and the method can be easily achieved online via recursive techniques.