Abstract
Severe weather conditions pose a large threat to the safety of aircraft, since they are responsible for a large percentage of aviation related accidents. With the advent of the free flight environment, the exigency for an autonomous severe weather avoidance capability has increased. In this thesis, an intelligent weather agent is developed for general aviation aircraft. Using a radar image from an onboard weather radar, the intelligent weather agent determines the safest path around severe weather with a minimum detour in distance. The method used is the Simplified Memory Bounded A* search technique that reduces computation time and memory requirements. The algorithm for A* search using the heuristic function is initially developed and evaluated in non-real-time form using Matlab, and demonstrated to be reliable and fast in flight path re-routing for stationary thunderstorms as well as moving thunderstorms. It is extended into a real-time version coded in Borland C++, and integrated with the existing Simplified Flight Management System and coupled with the heading command and hold autopilot of the nonlinear, non-real-time, six degrees-of-freedom Engineering Flight Simulator. Test cases consisting of stationary and moving thunderstorms are used to evaluate the intelligent weather agent online, in real-time. Results demonstrate that the new path suggested by the algorithm developed in this thesis is about 1% to 25% longer than the original path depending upon the size of the thunderstorm that lies in the original path. The detours in the test cases ranged from about 1 mile to 11 miles over and above the original path length. The algorithm did not exhibit any critical failures in the test cases and proved to be robust. The intelligent weather agent, when integrated with the Simplified Flight Management System and coupled with the heading command and hold autopilot, provides an effective and reliable guidance and navigation system for generating safe, alternate flight paths around thunderstorms and squall lines. It also forms the basis for the severe weather agent component of a broader hierarchical intelligent agent based system for free-flight guidance.
Bokadia, Sangeeta (2002). Intelligent weather agent for aircraft severe weather avoidance. Master's thesis, Texas A&M University. Available electronically from
https : / /hdl .handle .net /1969 .1 /ETD -TAMU -2002 -THESIS -B64.