Modified MRAS-Based Algorithm for Inertia Estimation of Mobile Robotic Chassis Drive Systems

Permanent magnet synchronous motor (PMSM) is typically used to drive the mobile robotic chassis system. In this paper, a modified model reference adaptive system (MRAS) based algorithm is proposed to estimate the moment of inertia of the PMSM drive system. First, an extended state observer (ESO) technique is introduced to reconstruct the adjustable model so that the adverse effects of nonlinear dynamics on its accuracy are removed. Thus, the estimation precision of the moment of inertia can be enhanced. Then, a novel sliding-mode adaptive law is designed to replace the PI adaptive law, which avoids complicated PI parameters adjustment and improves the dynamic estimation performance. The existence and the reachability of the sliding mode are proved with aiding from the Lyapunov function. Experimental results verify the effectiveness of the proposed method.