Resumen de Conops for a safe integration of multi-rpas operations in civil airspace
The gradual integration of remotely piloted aircraft systems (RPAS) in civil airspace, sharing airways with commercial flights, is expected to be completed once the legal issues and those regarding the unmanned traffic management are solved. This will open the floodgates to a myriad of new services, with a demand that will probably face a lack of pilots, a situation already present in the manned case. Multi-RPAS operations could be a solution, if correctly addressed.
The thesis proposes a framework that pursues the feasibility, from a human factors perspective, of having a single pilot/aircrew controlling several RPAS concurrently in non-segregated airspace. Such feasibility implies that this multitasking should be safe, and not interfere with the job of the air traffic controllers due to delays or errors associated to the parallel piloting. To this extent, a set of tools and measures are suggested, which include workload balance and prediction, action monitoring and interface usability.
The management of the workload takes into account the cognitive profile of each pilot to determine their limitations and time requirements while executing the tasks. Based on these profiles, the scheduling seeks the concurrent piloting while providing a safe margin of total workload, but also the flexibility in the strategies chosen for executing the tasks. The balance of the workload among pilots, a necessary safety measure, requires the prediction of it; this would be done based on the aggregation of different sources of information like aircraft readings, scheduled tasks, external reports and a suggested map of expected workload drivers.
Monitorings are justified by the fact that concurrent piloting could be misleading, so the system provides a safety check, acting as a kind of first officer. The reason behind the focus put on usability is that a handy interface providing an appropriate awareness is an essential requirement in multi-RPAS operations.
While multiplying the productivity is the main goal, it also offers benefits for a one-pilot-one-aircraft ratio, as it provides extra safety measures or the possiblity to parallelise other roles in the flightcrew. To illustrate its potential, a prototype was implemented and some experiments with pseudo-pilots conducted to compare the performance with or without some of the features. These showed a decrease in the number of errors, oversights, and subjective stress, and were useful to inspire improvements.
Some components of the framework could be leveraged outside of it. For instance, it relies on the exploitation of the potential of Controler-Pilot Data Link Communications (CPDLC), anticipating a future widespread implementation and full use. A CPDLC display was designed to reduce the head-down of current implementations and provide a more descriptive status of the communications, both key aspects for the quick response that a multi-RPAS pilot could require. As a standalone application, it could be used by pilots and controllers to train the CPDLC phraseology and composition rules, or as an inspiration for future implementations. Its connectivity middleware allows the simulation of different scenarios of Quality of Service for the link, which can be used to train the procedures when related problems arise. Also, the implementation of the suggested map of workload could serve to air traffic management analysts to get the perspective of the pilots during the operations and detect hot spots.
The safety measures and monitorings were positively evaluated by the pilots and controllers surveyed, even when some constituted redundant checks to existing air traffic control monitoring tools. The CPDLC display was quite well considered also to be used by controllers, who found it intuitive, quick, and that the information was clearly displayed and at hand. Finally, the handover procedure suggested was well evaluated to avoid the errors arising during the process of control migration.
