Development and Evaluation of an Automated Path Planning Aid

M ODERN air transportation has an excellent flight safety record. When failures do occur in flight, owing to the training and experience of the pilots almost always results in a safe landing. This is evidenced by a rate of only 1.35 accidents per one million hours flown in 2007 by U.S. air carriers [1]. Despite this excellent record, the pilots’ responsibility to land safely in case of an emergency can be very demanding. When an emergency situation occurs during a flight, the pilots’ workload is very high and a number of tasks demand the pilots’ attention. One of the important tasks is the planning and execution of a trajectory resulting in a safe landing. However, this task is complicated by multiple, often conflicting goals, including reducing time to land, staying within the flight envelope limits of the airplane, weather issues, aswell asmeeting any relevant regulatory requirements. Moreover, all of these tasks must be accomplished in a stressful environment, often under severe time pressure [2]. Although fault-tolerant adaptive automation is currently being developed, for the foreseeable future of civil transport aviation, pilots will be the ultimate decision makers, especially in cases of emergencies involving any type of aircraft performance degradation or flight envelope reduction. As a result, current research is being directed at pilot aids that aim at enhancing the pilot’s situation awareness (SA), as well as at supporting the pilot’s decision-making process through the provision of relevant situation-related information. The purpose of this paper is to report on a human factors study related to efforts to develop an automated planning aid (APA) (in terms of both an acceptable interface and control algorithms) that could assist pilots in generating a plan to safely land at alternative landing sites. To do so, the pilots must first determine the “best” landing site and then formulate an expedient and safe trajectory to the ground. This paper presents the results of an evaluation of an APA interface prototype by means of a human-in-the-loop test with commercial airline pilots, focusing on the selection of alternate landing sites during an emergency. Although the implemented APA in the simulator was also able to compute emergency paths to those sites, a detailed description and discussion of this part of the process is omitted in this paper because it had no immediate effect on theAPA interface evaluation. The results of the study are evaluated in comparison to the opinion and judgment of a single subject matter expert. This expert had more than 20,000 h of flight experience in over 20 years of service as a commercial pilot. The authors do acknowledge that this comparison might be improved by incorporating more experts, more test cases, and a larger sample.However, given the fact that these scenarioswere designed in cooperation with that expert to have an unambiguous best solution, it is likely that, given enough time to review each scenario, the vast majority of trained pilots would come to the same conclusion as to which landing site was the best alternative. As such, the authors do believe that the metrics used, and the comparison with the experts ranking of the landing sites, is valid for the evaluation of the APA for selecting an alternate landing site under tight time constraints.