Adaptive backstepping neural controller for aircraft autolanding

Pashilkar, AA and Sundararajan, N and Saratchandran, P (2005) Adaptive backstepping neural controller for aircraft autolanding. Technical Report. National Aerospace Laboratories, Bangalore, India.

[img] PDF
Restricted to Repository staff only

Download (530kB)
[img] Indexer Terms (Generate index codes conversion from application/pdf to indexcodes)
Restricted to Repository staff only

Download (10kB)


This report presents an adaptive back-stepping neural controller for reconfigurable flight control of aircraft in the presence of changes in the aerodynamic characteristics. Radial Basis Function Neural networks are introduced in an adaptive back stepping architecture13; with full state measurement for aircraft trajectory following. For the RBF neural network a learning scheme in which the network starts with no hidden neurons and adds new hidden neurons based on the trajectory error is proposed. Using Lyapunov theory stable tuning rules are derived for parameter update of the RBF neural networks and proof of stability in the ultimate bounded sense is given for the resulting controller. The longitudinal model of an open loop unstable high performance aircraft in the terminal landing phase subjected to single elevator hard over failures is used to demonstrate the capability of the controller. The resulting controller is able to successfully stabilize and land the aircraft in the presence of severe winds and control surface failures.

Item Type: Monograph (Technical Report)
Uncontrolled Keywords: Neural network;Fault tolerant;Actuator failure; Autolanding; Autolanding;Flight control
Subjects: AERONAUTICS > Aircraft Stability and Control
Depositing User: Ms Indrani V
Date Deposited: 16 Jan 2009
Last Modified: 24 May 2010 04:09

Actions (login required)

View Item View Item