Entity

Time filter

Source Type


Lampton A.,Systems Technology Inc. | Niksch A.,Texas A&M University | Niksch A.,Vehicle Systems and Control Laboratory | Valasek J.,Texas A&M University | Valasek J.,Vehicle Systems and Control Laboratory
Journal of Aerospace Computing, Information and Communication | Year: 2010

Casting the problem of morphing a microair vehicle as a reinforcement-learning problem to achieve desired tasks or performance is a candidate approach for handling many of the unique challenges associated with such small aircraft. This paper presents an early stage in the development of learning how and when to morph a micro air vehicle by developing an episodic unsupervised learning algorithm using the Q-learning method to learn the shape and shape change policy of a single morphing airfoil. Reinforcement is addressed by reward functions based on airfoil properties, such as lift coefficient, representing desired performance for specified flight conditions. The reinforcement learning as it is applied to morphing is integrated with a computational model of an airfoil. The methodology is demonstrated with numerical examples of an NACA type airfoil that autonomously morphs in two degrees of freedom, thickness and camber, to a shape that corresponds to specified goal requirements. Because of the continuous nature of the thickness and camber of the airfoil, this paper addresses the convergence of the learning algorithm given several discretizations. Convergence is also analyzed with three candidate policies: 1) a fully random exploration policy, 2) a policy annealing from random exploration to exploitation, and 3) an annealing discount factor in addition to the annealing policy. The results presented in this paper show the inherent differences in the learned action-value function when the state-space discretization, policy, and learning parameters differ. It was found that a policy annealing from fully explorative to almost fully exploitative yielded the highest rate of convergence as compared to the other policies. Also, the coarsest discretization of the state-space resulted in convergence of the action-value function in as little as 200 episodes. Copyright © 2010 by Amanda Lampton, Adam Niksch and JohnValasek. Source


Valasek J.,Texas A&M University | Valasek J.,Center for Autonomous Vehicles and Sensor Systems | Valasek J.,Vehicle Systems and Control Laboratory | Shryock K.J.,Texas A&M University
ASEE Annual Conference and Exposition, Conference Proceedings | Year: 2015

Systems Engineering (SE) has long been a staple in the aerospace engineering industry, but it has been slow to gain traction in academia. This is due to both the challenge of incorporating SE content in the traditional framework of engineering curricula and the lack of experience with SE by academic practitioners. This paper presents the results of a 17 month project between two large public institutions to investigate and incorporate educational tools and practical experiences in the teaching of SE in existing design courses, to be later transitioned into a broad range of courses within the curricula. The main objective of the project introduced students to the practical applications of the fundamentals of SE without displacing other course content. The target courses at Texas A&M University included three senior-level courses, of which two were required capstone design courses and one an optional technical design elective. For the capstone design courses, this content was added primarily to the early part of the semester and consisted of identifying a customer need, conducting a requirements definition study, developing a Concept of Operations, and subsequently translating it into a Request for Proposal. For the design elective, this project enhanced prior SE content in the course but then integrated the course into the collective instruction within the broad Systems Engineering Design Initiative effort at the institution. The paper will present the modifications to course content, pedagogy used in the project, results from assessment of the project, lessons learned by the instructors, and comments from both students in the course and industry advisory board members who reviewed the course deliverables. In summary, project outcomes were achieved, and the students felt their experiences were particularly rewarding. Students enjoyed teaming on a project involving another university, and the external industry advisory board members remarked on the valuable real-world experience students received through the project. This approach to incorporating SE content within current courses with minimal disruption to other content should be applicable to most engineering programs. © American Society for Engineering Education, 2015. Source


Valasek J.,Vehicle Systems and Control Laboratory | Kirkpatrick K.,Vehicle Systems and Control Laboratory | May J.,Vehicle Systems and Control Laboratory
AIAA Infotech at Aerospace Conference and Exhibit 2012 | Year: 2012

Advances in unmanned flight have led to the development of Unmanned Air Systems that are capable of carrying state-of-the-art video capturing systems for the intended purpose of surveillance and tracking. These vehicles have the capability to fly through a target area with a mounted camera and allow humans to operate both the UAS and the camera to attempt to survey any objects that are deemed targets. These systems have worked well when controlled by humans, but having them operate autonomously to reduce operator workload and manpower is even more challenging when the camera is fixed to the airframe instead of being mounted on a gimbal, so that the aircraft must be steered in order to steer the camera. The presence of winds must also be accounted for. This paper develops an algorithm for surveillance of ground targets by UAS with fixed pan and tilt cameras, in the presence of winds. This paper develops an algorithm for surveillance of ground targets by UAS. The specific RL algorithm used is Q-learning, and the objective of the approach is to bring any target located in an image captured by a camera into the center of the image using the learned control policy. The learning agent determines offline (initially) how to control the UAS and camera to get a target from any point in the image to the center and hold it there. A feature of this approach is that the learning agent will continue to learn and refine and update the previously offline learned control policy, during actual operation. Results presented in the paper demonstrate that the approach has merit for autonomous surveillance of ground targets. © 2012 by the American Institute of Aeronautics and Astronautics, Inc. All rights reserved. Source


Siddarth A.,Texas A&M University | Siddarth A.,Vehicle Systems and Control Laboratory | Valasek J.,Texas A&M University | Valasek J.,Vehicle Systems and Control Laboratory
AIAA Guidance, Navigation, and Control Conference 2011 | Year: 2011

This paper develops a general control algorithm for the exact output tracking of nonlinear systems with non-minimum phase dynamics. The control technique is causal and does not require preview or knowledge of the desired reference beforehand. Additionally, the control is independent of the operating condition and the desired reference. The main idea of the paper is to convert the output tracking problem into a slow state tracking problem for singularly perturbed systems. Previous work on singularly perturbed systems have shown asymptotic tracking of slow states only for a class of nonlinear systems that are linear in the fast states. However, this paper develops a control technique that does not have this restriction and is applicable to a general class of nonlinear singularly perturbed systems. The procedure is to compute the desired internal state trajectory and the control scheme that stabilizes the nonlinear system online, thereby guaranteeing asymptotic output tracking. Performance of this approach is demonstrated in simulation for two benchmark problems: the beam-ball example that is slightly non-minimum phase and fails to have a well-defined relative degree, and the Conventional Take-off and Landing (CTOL) non-minimum phase aircraft. Results presented in the paper show that the approach is able to accomplish perfect tracking while stabilizing the closed-loop system, while keeping all closed-loop signals bounded. © 2011 by Anshu Siddarth and John Valasek. Published by the American Institute of Aeronautics and Astronautics, Inc. Source


Henrickson J.V.,Texas A&M University | Henrickson J.V.,Vehicle Systems and Control Laboratory | Skelton R.E.,Texas A&M University | Valasek J.,Texas A&M University | Valasek J.,Vehicle Systems and Control Laboratory
2016 AIAA Guidance, Navigation, and Control Conference | Year: 2016

This paper develops and applies tensegrity concepts to the design of shape-controllable 2D airfoils. After introducing tensegrity systems and dynamics, a tension-driven shape control strategy is outlined, and a method of generating variable complexity tensegrity airfoils is developed. The described shape control strategy is then applied to the task of transforming a given tensegrity airfoil from some initial shape to a desired final shape. Results show the generation of tensegrity systems that approximate various NACA 4-series airfoil profiles, and simulation results demonstrate successful shape control of both symmetric and asymmetric tensegrity airfoils. © 2016, American Institute of Aeronautics and Astronautics Inc, AIAA. All rights reserved. Source

Discover hidden collaborations