Back
A Pilot Intent Estimator for Haptic Support Systems in Helicopter Maneuvering Tasks
{Haptic support systems are commonly designed as feedback systems that track a known trajectory. However, when a pilot performs a maneuver in a free airspace, the desired trajectory is not known a priori. Thus, an estimation of the pilot intended trajectory is needed. This paper investigates the estimation of the pilot desired trajectory for designing a haptic support system that helps pilots perform helicopter maneuvers in a free airspace. A probabilistic algorithm is implemented to estimate the pilot intended trajectory according to the actions made on the control devices. The haptic aid is developed as a feedback controller able to follow the estimated intended trajectory. Two simulations were conducted to test the effectiveness of the proposed algorithm. First, filtered step-wise pilot control inputs were considered. Then, real pilot inputs were employed, which were recorded in a lateral reposition maneuver performed on a motion simulator. The developed algorithm was able to recognize the intended path according to the given pilot inputs in both scenarios.}
@inproceedings{D039IntinoOGFBP2018, title = {{A Pilot Intent Estimator for Haptic Support Systems in Helicopter Maneuvering Tasks}}, booktitle = {{2018 AIAA Modeling and Simulation Technologies Conference}}, abstract = {{Haptic support systems are commonly designed as feedback systems that track a known trajectory. However, when a pilot performs a maneuver in a free airspace, the desired trajectory is not known a priori. Thus, an estimation of the pilot intended trajectory is needed. This paper investigates the estimation of the pilot desired trajectory for designing a haptic support system that helps pilots perform helicopter maneuvers in a free airspace. A probabilistic algorithm is implemented to estimate the pilot intended trajectory according to the actions made on the control devices. The haptic aid is developed as a feedback controller able to follow the estimated intended trajectory. Two simulations were conducted to test the effectiveness of the proposed algorithm. First, filtered step-wise pilot control inputs were considered. Then, real pilot inputs were employed, which were recorded in a lateral reposition maneuver performed on a motion simulator. The developed algorithm was able to recognize the intended path according to the given pilot inputs in both scenarios.}}, pages = {132--140}, publisher = {Curran}, address = {Kissimmee, FL, USA}, year = {2018}, slug = {d039intinoogfbp2018}, author = {D\textquotesingleIntino, G and Olivari, M and Geluardi, S and Fabbroni, D and B\"ulthoff, HH and Pollini, L} }