L1 adaptive dynamic inversion missile autopilot with PI structure and adaptive sampling

An L1 adaptive controller with piecewise continuous (PC) adaptation law augmenting a nonlinear dynamic inversion (NDI) autopilot has been applied to a highly agile, tail-controlled missile. Here proportional integral (PI) structure has been used in inner loop for better tracking and robustness. Adaptive sampling has also been introduced for less computational load. Firstly a two time scale separated NDI autopilot is designed to control the pitch and yaw accelerations in the body-fixed frame while maintaining a desired roll angle. The baseline NDI autopilot is augmented with proposed L1 Adaptive autopilot using PI structure and adaptive sampling to overcome the potential performance degradation due to modeling uncertainties. The advantages of proposed autopilot are demonstrated via simulations of a high-fidelity six-DOF missile model, in which worst case uncertainties are considered.