Show simple item record

dc.contributor.authorKoch, Williamen_US
dc.contributor.authorMancuso, Renatoen_US
dc.contributor.authorWest, Richarden_US
dc.contributor.authorBestavros, Azeren_US
dc.date.accessioned2018-07-25T13:17:09Z
dc.date.accessioned2020-05-05T14:16:30Z
dc.date.available2018-07-25T13:17:09Z
dc.date.available2020-05-05T14:16:30Z
dc.date.issued2019-02-01
dc.identifierhttp://arxiv.org/abs/1804.04154v1
dc.identifier.citationWilliam Koch, Renato Mancuso, Richard West, Azer Bestavros. 2019. "Reinforcement Learning for UAV Attitude Control." ACM Transactions on Cyber-Physical Systems, Volume 2, Issue Feb 2019, pp. 1 - 21. https://doi.org/10.1145/3301273
dc.identifier.issn2378-962X
dc.identifier.urihttps://hdl.handle.net/2144/40560
dc.description.abstractAutopilot systems are typically composed of an “inner loop” providing stability and control, whereas an “outer loop” is responsible for mission-level objectives, such as way-point navigation. Autopilot systems for unmanned aerial vehicles are predominately implemented using Proportional-Integral-Derivative (PID) control systems, which have demonstrated exceptional performance in stable environments. However, more sophisticated control is required to operate in unpredictable and harsh environments. Intelligent flight control systems is an active area of research addressing limitations of PID control most recently through the use of reinforcement learning (RL), which has had success in other applications, such as robotics. Yet previous work has focused primarily on using RL at the mission-level controller. In this work, we investigate the performance and accuracy of the inner control loop providing attitude control when using intelligent flight control systems trained with state-of-the-art RL algorithms—Deep Deterministic Policy Gradient, Trust Region Policy Optimization, and Proximal Policy Optimization. To investigate these unknowns, we first developed an open source high-fidelity simulation environment to train a flight controller attitude control of a quadrotor through RL. We then used our environment to compare their performance to that of a PID controller to identify if using RL is appropriate in high-precision, time-critical flight control.en_US
dc.format.extentpp. 1 - 21en_US
dc.publisherACMen_US
dc.relation.ispartofACM Transactions on Cyber-Physical Systems
dc.relation.replaceshttps://hdl.handle.net/2144/30036
dc.relation.replaces2144/30036
dc.subjectComputer scienceen_US
dc.subjectRoboticsen_US
dc.titleReinforcement learning for UAV attitude controlen_US
dc.typeArticleen_US
dc.identifier.doi10.1145/3301273
pubs.elements-sourcemanual-entryen_US
pubs.notes13 pages, 9 figuresen_US
pubs.notesEmbargo: Not knownen_US
pubs.organisational-groupBoston Universityen_US
pubs.organisational-groupBoston University, Administrationen_US
pubs.organisational-groupBoston University, College of Arts & Sciencesen_US
pubs.organisational-groupBoston University, College of Arts & Sciences, Department of Computer Scienceen_US
pubs.publication-statusPublisheden_US
dc.date.online2018-04-01
dc.identifier.orcid0000-0003-0798-8835 (Bestavros, Azer)
dc.identifier.mycv363833


This item appears in the following Collection(s)

Show simple item record