Approach and landing guidance using constrained model predictive static programming

Abstract

Listen

Translate article

The paper proposes a guidance scheme during the approach and landing phase of an unpowered vehicle to achieve high precision at touchdown by limiting the control inputs within allowable bounds using constrained Model Predictive Static Programming. The advantages of a single flight phase are incorporated into the design since the guess control history is obtained from a single segment based guidance approach. The requirement of ensuring continuity in states and guidance commands are hence avoided leading to minimization of control deviations using a simple cost function. Constrained input commands are used for determining control surface deflection based on dynamic inversion considering the longitudinal pitch rate dynamics. Comparison is carried out between the results obtained for constrained as well as unconstrained schemes for the nominal trajectory. Comparisons of load factor, dynamic pressure and flight path angle for three segment, two segment, single segment guidance strategies have been carried out with MPSP based guidance to evaluate the relative advantages offered by the scheme. Performance analysis of the proposed scheme is done based on numerous simulations with simultaneous variations in initial states, vehicle parameters and aerodynamic parameter variations. Simulation results indicate that the proposed scheme is suitable for landing on runways of shorter lengths due to minimal touchdown errors.