Adaptive complementary filter using fuzzy logic and simultaneous perturbation stochastic approximation algorithm

This paper addresses the problem of attitude estimation using low cost, small-sized inertial sensors under dynamic maneuvers. An adaptive complementary filter with fuzzy logic and simultaneous perturbation stochastic approximation (SPSA) algorithm is proposed. By recognizing the situation of dynamic condition via fuzzy logic, the cut-off frequency of the complementary filter is determined adaptively under varying vehicle dynamics. Also, the SPSA algorithm is used to tune the parameters of fuzzy system. Simulation results based on the test data show that the proposed SPSA-based fuzzy complementary filter exhibits a significant performance improvement for attitude estimation during dynamic maneuvers.

► We propose an adaptive complementary filter with fuzzy logic and SPSA tuning. ► Fuzzy logic can identify the dynamics of the vehicle. ► The cut-off frequency is adapted adaptively by the fuzzy logic under dynamics. ► SPSA is an efficient optimization tool for the tuning of the fuzzy parameters. ► The proposed scheme can make the filter more robust to acceleration disturbances.