dc.contributor.advisor | Assaleh, Khaled | |
dc.contributor.advisor | Jaradat, Mohammad Abdel Kareem Rasheed | |
dc.contributor.author | Avdeev, Alexander | |
dc.date.accessioned | 2015-03-04T10:15:41Z | |
dc.date.available | 2015-03-04T10:15:41Z | |
dc.date.issued | 2014-12 | |
dc.identifier.other | 35.232-2014.33 | |
dc.identifier.uri | http://hdl.handle.net/11073/7726 | |
dc.description | A Master of Science thesis in Mechatronics Engineering by Alexander Avdeev entitled, "Artificial Intelligence Based Identification of the Attitude Dynamics for a Quadroto UAV," submitted in December 2014. Thesis advisor is Dr. Khaled Assaleh and thesis co-advisor is Dr. Mohammad Abdel Kareem Rasheed Jaradat. Available are both soft and hard copies of the thesis. | en_US |
dc.description.abstract | Quadrotor UAVs have become very popular, recently. At the same time, having a model of a system proves rather useful in almost any engineering task. In the case of quadrotors this becomes a challenging task, because they are inherently unstable, exhibit nonlinear behavior and a lot of coupling. In addition to this, quadrotors' behavior is greatly influenced by characteristics and coefficients, which are very hard to measure directly or determine analytically, such as: aerodynamic coefficients of the propellers and inertia of the frame. However, all the difficulties listed above are known to be successfully overcome by use of artificial intelligence. This thesis presents a process of building a setup suitable for data gathering and identification of pitch, roll and yaw dynamics through the use of several data-driven techniques. First of all, transfer functions describing the system were found numerically to establish a base line for comparison. Then, distributed time delay neural networks (DTDNN), nonlinear autoregressive neural networks (NARX) followed by an adaptive neural fuzzy inference system (ANFIS) and polynomial regression were used to identify the system. A comparison was based on several criteria to provide an adequate evaluation of the obtained models. | en_US |
dc.description.sponsorship | College of Engineering | en_US |
dc.description.sponsorship | Department of Mechanical Engineering | en_US |
dc.language.iso | en_US | en_US |
dc.relation.ispartofseries | Master of Science in Mechanical Engineering (MSME) | en_US |
dc.subject | Quadrotor | en_US |
dc.subject | System Identification | en_US |
dc.subject | Distributed Time Delay Neural Networks | en_US |
dc.subject | Nonlinear Autoregressive Neural Networks | en_US |
dc.subject | Adaptive Neural Fuzzy Inference System | en_US |
dc.subject | Polynomial Classifiers | en_US |
dc.subject.lcsh | Artificial intelligence | en_US |
dc.subject.lcsh | Quadrotor helicopters | en_US |
dc.subject.lcsh | Flight control | en_US |
dc.subject.lcsh | Drone aircraft | en_US |
dc.subject.lcsh | Control systems | en_US |
dc.title | Artificial Intelligence Based Identification of the Attitude Dynamics for a Quadrotor UAV | en_US |
dc.type | Thesis | en_US |