Theses and Dissertations from UMD
Permanent URI for this communityhttp://hdl.handle.net/1903/2
New submissions to the thesis/dissertation collections are added automatically as they are received from the Graduate School. Currently, the Graduate School deposits all theses and dissertations from a given semester after the official graduation date. This means that there may be up to a 4 month delay in the appearance of a give thesis/dissertation in DRUM
More information is available at Theses and Dissertations at University of Maryland Libraries.
Browse
3 results
Search Results
Item Unblock: Interactive Perception for Decluttering(2021) govindaraj, krithika; Aloimonos, Yiannis; Systems Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Novel segmentation algorithms can easily identify objects that are occludedor partially occluded, however in highly cluttered scenes the degree of occlusion is so high that some objects may not be visible to a static camera. In these scenarios, humans use action to change the configuration of the environment, elicit more information through perception, process the information before taking the next action. Reinforcement learning models this behavior, however unlike humans, the phase where perception data is understood is not included, as images are directly used as observations. The aim of this thesis is to establish a novel method that indirectly uses perception data for reinforcement learning to address the task of decluttering a scene using a static camera.Item Expanding Constrained Kinodynamic Path Planning Solutions through Recurrent Neural Networks(2019) Shaffer, Joshua Allen; Xu, Huan; Aerospace Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Path planning for autonomous systems with the inclusion of environment and kinematic/dynamic constraints encompasses a broad range of methodologies, often providing trade-offs between computation speed and variety/types of constraints satisfied. Therefore, an approach that can incorporate full kinematics/dynamics and environment constraints alongside greater computation speeds is of great interest. This thesis explores a methodology for using a slower-speed, robust kinematic/dynamic path planner for generating state path solutions, from which a recurrent neural network is trained upon. This path planning recurrent neural network is then used to generate state paths that a path-tracking controller can follow, trending the desired optimal solution. Improvements are made to the use of a kinodynamic rapidly-exploring random tree and a whole-path reinforcement training scheme for use in the methodology. Applications to 3 scenarios, including obstacle avoidance with 2D dynamics, 10-agent synchronized rendezvous with 2D dynamics, and a fully actuated double pendulum, illustrate the desired performance of the methodology while also pointing out the need for stronger training and amounts of training data. Last, a bounded set propagation algorithm is improved to provide the initial steps for formally verifying state paths produced by the path planning recurrent neural network.Item Reflective Reasoning(2006-05-24) Chong, Waiyian; Perlis, Donald R; Computer Science; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)This dissertation studies the role of reflection in intelligent autonomous systems. A reflective system is one that has an internal representation of itself as part of the system, so that it can introspect and make controlled and deliberated changes to itself. It is postulated that a reflective capability is essential for a system to expect the unexpected---to adapt to situations not forseen by the designer of the system. Two principal goals motivated this work: to explore the power of reflection (1) in a practical setting, and (2) as a method for approaching bounded optimal rationality via learning. Toward the first goal, a formal model of reflective agent is proposed, based on the Beliefs, Desires and Intentions (BDI) architecture, but free from the logical omniscience problem. This model is reflective in the sense that aspects of its formal description, comprised of set of logical sentences, will form part of its belief component, and hence be available for reasoning and manipulation. As a practical application, this model is suggested as a foundation for the construction of conversational agents capable of meta-conversation, i.e., agents that can reflect on the ongoing conversation. Toward the second goal, a new reflective form of reinforcement learning is introduced and shown to have a number of advantages over existing methods. The main contributions of this thesis consist of the following: In Part II, Chapter 2, the outline of a formal model of reflection based on the BDI agent model; in Chapter 3, preliminary design and implementation of a conversational agent based on this model; In Part III, Chapter 4, design and implementation of a novel benchmark problem which arguably captures all the essential and challenging features of an uncertain, dynamic, time sensitive environment, and setting the stage for clarification of the relationship between bounded-optimal rationality and computational reflection under the universal environment as defined by Solomonoff's universal prior; in Chapter 5, design and implementation of a computational-reflection inspired reinforcement learning algorithm that can successfully handle POMDPs and non-stationary environments, and studies of the comparative performances of RRL and some existing algorithms.