Human factors evaluations of Free Flight Issues solved and issues remaining

摘要

The Dutch National Aerospace Laboratory (NLR) has conducted extensive human-in-the-loop simulation experiments in NLR's Research Flight Simulator (RFS), focussed on human factors evaluation of Free Flight. Eight years of research, in co-operation with partners in the United States and Europe, has shown that Free Flight has the potential to increase airspace capacity by at least a factor of 3. Expected traffic loads and conflict rates for the year 2020 appear to be no major problem for professional airline crews participating in flight simulation experiments. Flight efficiency is significantly improved by user-preferred routings, including cruise climbs, while pilot workload is only slightly increased compared to today's reference. Detailed results from three projects and six human-in-the-loop experiments in NLR's Research Flight Simulator are reported. The main focus of these results is on human factors issues and particularly workload, measured both subjectively and objectively. An extensive discussion is included on many human factors issues resolved during the experiments, but also open issues are identified. An intent-based Conflict Detection and Resolution (CD&R) system provides "benefits" in terms of reduced pilot workload, but also "costs" in terms of complexity, need for priority rules, potential compatibility problems between different brands of Flight Management Systems and large bandwidth. Moreover, the intent-based system is not effective at solving multi-aircraft conflicts. A state-based CD&R system also provides "benefits" and "costs". Benefits compared to the full intent-based system are simplicity, low bandwidth requirements, easy to retrofit (no requirements to change avionics infrastructure) and the ability to solve multi-aircraft conflicts in parallel. The "costs" involve a somewhat higher pilot workload in similar circumstances, the smaller look-ahead time which results in less efficient resolution manoeuvres and the sometimes falseuisance alerts due to missing intent information. The optimal CD&R system (in terms of costs versus benefits) has been suggested to be state-based CD&R with the addition of intended or target flight level. This combination of state-based CD&R with a limited amount of intent provides "the best of both worlds". Studying this CD&R system is still an open issue.