Error analysis and compensation of an Inductosyn-based angular measurement system

The inductosyn-based angular measurement system exhibits measurement errors due to factors such as installation, manufacturing, and circuit errors, which severely affect the performance of Control Moment Gyroscope. These errors must be compensated, and the prerequisite for compensation is the analysis and modeling of error components. This paper first analyzes the impact of errors on the control performance of the Control Moment Gyroscope gimbal servo system to verify the correctness of the theoretical error modeling, and presents a polynomial model of the errors. Then, to solve issues of traditional compensation methods such as low universality and unsatisfactory compensation for uncertain model errors, a compensation method based on Radial Basis Function network is proposed. Finally, data are collected and verified on an experimental platform. The experimental results show that Radial Basis Function network compensation method can compensate the angular measurement system with a root mean square error of 0.88° to 0.0181°. Compared with the 0.1788° achieved by the traditional least squares compensation method, the error is reduced by an order of magnitude, and the compensation effect is improved by 81.4%. Meanwhile, the compensation effect is almost consistent across angular measurement system of different specifications, indicating that this method has good feasibility and universality.