SAR optimization and Convolutional Neural Network based fault estimations and for auto-landing control model

Abstract

During auto landing, the aircraft flies at a significantly low altitude and low speed. So the consequent accidents and flight crashes are highly possible. Several constrained space and foremost external interruption while landing is considered as one of the most complex phases of the aircraft. Therefore it is necessary to recover the aircraft from various major disturbances and to estimate the rate of fault during aircraft landing. In addition to this, for the effective design of an aircraft, it is essential in determining the fault that affects the aircraft. To overcome the issues, this article aim to propose a novel CNNLSTM-SAR-based fault estimation approach to estimate the fault rate from various state trajectories of the aircraft. Here we employed a Convolution Neural Network (CNN) and Long Short-Term Memory (LSTM) model integrated with the Search and Rescue (SAR) optimization algorithm to react instantaneously to the broad range of failures such as actuator failure and failure due to the wind. Then the performances of the proposed CNNLSTM-SAR based fault estimation approach are compared and the results demonstrated that the proposed approach provide a smooth landing with minimum fault and error.