An Automated Approach for Objective Evaluation of Practice Autorotations Using Available Aircraft Data

An autorotation is a maneuver unique to rotary wing aircraft that better ensures the pilot can make a survivable landing after experiencing a loss of engine power in-flight. Timely and decisive action by the pilot is required for the successful execution of the maneuver, which makes a certain level of proficiency essential to the safe conduct of flight operations. Practice autorotations, therefore, are a critical part of initial training for new pilots as well as refresher and proficiency training for experienced pilots. In order to ensure Navy and Marine Corps helicopter pilots achieve and maintain an acceptable level of proficiency at performing autorotations, their ability to perform practice autorotations is observed and evaluated by experienced instructor pilots throughout their continual training regimen. Significant effort is made to ensure objectivity and standardization among different instructors; however, subjectivity, ambiguity, and inconsistency cannot be completely eliminated from human assessments. In addition, the complex dynamics associated with a practice autorotation, and the instructor's responsibility to ensure safety is not compromised, make detailed and objective assessments of pilot performance a challenge for even the most experienced instructor. An objective evaluation methodology using data recorded on the aircraft can provide an independent, unbiased assessment of a pilot's performance, and enhance pilot training and improve standardization in the evaluation process. This paper presents an automated methodology developed to detect and assess practice autorotations performed in U. S. Marine Corps CH-53E heavy-lift helicopters utilizing flight data collected, processed, and downloaded from the onboard Integrated Mechanical Diagnostics System (IMDS). It includes an overview of the determination of requisite assessment criteria, development of analytical techniques used, and the presentation format for analytical results. Examples of results from autonomous detections and analyses of practice autorotations are also presented, and current usage and potential future applications of the automated capability are discussed.