On the Modeling and Construction of an Artificial Landing Safety Officer (LSO) for Conducting Pilot Training in a Helicopter Deck Landing Simulator

Abstract

Using concepts and techniques from User Modeling, Fuzzy Logic and Intelligent Agents, a Self- Adaptive Landing Signals Agent (SALSA) was developed at the Royal Military College of Canada for integration in the helicopter simulation and training facility of The Defense Research and Development Canada (DRDC) centre in Toronto. In the context of this work, selfadaptive means the ability to assess the skill level of the pilot, the ship motion and the environmental conditions inside the simulator, and adapt its behavior accordingly, just as a human landing safety officer (LSO) would. Furthermore, SALSA can recall its last assessment of a pilot it has worked with, or quickly make an estimate of the skills of a new pilot, and use that information to commence training at the appropriate skill level. Preliminary tests indicate that SALSA is able to simulate a human LSO and assist in providing progressive training in the simulator.