Real-Time Adaptive Dynamics Based State Estimation Scheme for Unmanned Aircrafts

In this paper we present a state estimation scheme for Unmanned Aircrafts (UAs) utilizing dynamics based models and multi-sensor data fusion. Employing the UA in can substantially enhance estimator performance, but obtaining accurate parameters each is computationally costly complex. To eliminate these issues, propose two decoupled Extended Kalman Filters (EKFs), namely Rotational Decoupled Filter (RDEKF) Translational (TDEKF). The filters are identified real-time using Deep Neural Network Modified Relay Feedback Test (DNN-MRFT) approach. This approach doesn’t demand prior knowledge of physical parameters, requiring only an Inertial Measurement Unit (IMU) positioning system model classification. Our provides position, velocity attitude estimates, addition to smooth lag-free inertial acceleration estimates. We show experimentally advantages our on trajectory tracking problems that uses low rate position sensors. also demonstrate how estimated feedback control reduce error optimally tuned by 43%. Moreover, proposed produces estimates leads reduction controller action 6.6%, when compared kinematic estimators. compare achieved results against other methods require full or noise models, performance capability.