Advanced Speed Control Algorithm using Kalman Filter and RLS

Required Skills: Kalman Filter, Recursive Least Squares algorithm, Adaptive Control, PMSM/induction/DC motor modeling I am designing a motor drive using some adaptive speed control algorithms and currently in simulation stage. I already developed the motor model, Kalman filter, and RLS algorithm. The speed controller using Kalman filter is working fine with estimated angular speed by the Kalman filter. The Kalman filter is also estimating disturbance torque from motor current and rotor position. The RLS then calculates the inertia of the motor rotor and connected load. The problem is when I feed the estimated J into the Kalman filter, the system becomes unstable. But I have tested with a constant J (motor's nominal inertia) and the RLS calculated the J correctly and the speed control works fine. I need to make the system stable by proper tuning of the Kalman filter when the J is fed back to the Kalman filter. This estimated J will be used to auto-tune the speed controller in a wide range of speed and load condition. I have the details and the running code in Matlab and Simulink 2016. My requirement is to make the system stable for a wide range of motor parameters (motor size) or identify constraints that causes instability in a specific condition. If you are an expert in this field and think it is a trivial job for you then please send your estimate. I'll provide you the details after the initial discussion.