UMD Theses and Dissertations

Permanent URI for this collectionhttp://hdl.handle.net/1903/3

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 given thesis/dissertation in DRUM.

More information is available at Theses and Dissertations at University of Maryland Libraries.

Browse

Search Results

Now showing 1 - 5 of 5
  • Thumbnail Image
    Item
    DYNAMICS AND CONTROL OF AN ELASTIC ROD IN AIR AND WATER
    (2019) Burch, Travis Taylor; Paley, Derek A; Aerospace Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    This thesis investigates the modeling and control of bio-inspired flexible structures for robotics applications. Many animals move through complicated natural environments and perform complex tasks by exploiting soft structures. Soft structures are highly versatile and are a growing area of interest in robotics because they can have decreased weight, size, and mechanical complexity relative to more traditional rigid robotics. This work uses planar discrete elastic rod (PDER) theory for modeling two types of flexible structures. First, a flexible airfoil is modeled using PDER theory, including the Improved Lighthill model (ILM) of hydrodynamic forces to study the propulsion thrust. The propulsion thrust generated by rigid and flexible foils are also measured experimentally and compared to the model. Second, a state-space description of a flexible pendulum with torque input is presented. Linear state-and output-feedback hybrid controllers stabilize the inverted flexible pendulum starting from the down equilibrium.
  • Thumbnail Image
    Item
    BIO-INSPIRED PUMPING MECHANISMS IN AN INTERMEDIATE REYNOLDS NUMBER
    (2018) Saffaraval, Farhad; Kiger, Kenneth; Mechanical Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    Pumps are important to applications across a wide range of scales. Most of traditional applications occur within a range where inertia is the dominating factor influencing the pump performance, and hence many practical designs are based on mechanisms that rely on this assumption. As one moves towards smaller devices, however, the increasing effect of viscosity renders these traditional mechanisms ineffective. The current work looks towards a bio-inspired system consisting of an array of oscillating plates to contend with this challenge. The plates are placed within a channel, and the pumping performance generated is examined for a small range of Reynolds numbers intermediate between inertial and viscous regimes (0.1 < Re < 10). The goal of this work is to observe the effect of how different plate kinematics can be utilized to break the symmetry the system to produce a net pumped flow. Rigid and flexible plates are studied, using both sinusoidal and triangle wave actuation kinematics. The tests are first conducted with a single appendage, and then repeated with an array of 5 closely spaced plates to observe the effect of their interaction on the overall performance. The results of the single plate tests indicate that increased asymmetry introduced in the triangle wave actuation results in increased pumping performance as well as energy consumption. Tests were conducted at two Reynolds number conditions, Re = 0.6 and 6. The pumping performance was found to be an order of magnitude higher for the Re = 6 case. In the case of flexible plates, the results show that a mass specific pumping efficiency was higher for the flexible case with a higher frequency at the same Reynolds numbers. For the plate array, the results indicate five flexible plates with 〖∆θ〗_i=-90 will generate more than 4 times the flow rate in comparison to the single flexible plate. Asymmetric triangle actuation in conjunction with symplectic metachronal motion (〖∆θ〗_i=30) exhibits pumping performance more than 10 times of using a single rigid plate. Total work is noticeably higher for multiple plate system and will result in a reduced overall pumping efficiency in comparison to the single appendage.
  • Thumbnail Image
    Item
    BIO-INSPIRED DISTURBANCE REJECTION WITH OCELLAR AND DISTRIBUTED ACCELERATION SENSING FOR SMALL UNMANNED AIRCRAFT SYSTEMS
    (2015) Gremillion, Gregory; Humbert, James Sean; Aerospace Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    Rapid sensing of body motions is critical to stabilizing a flight vehicle in the presence of exogenous disturbances as well as providing high performance tracking of desired control commands. This bandwidth requirement becomes more stringent as vehicle scale decreases. In many flying insects three simple eyes, known as the ocelli, operate as low latency visual egomotion sensors. Furthermore many flying insects employ distributed networks of acceleration-sensitive sensors to provide information about body egomotion to rapidly detect external forces and torques. In this work, simulation modeling of the ocelli visual system common to flying insects was performed based on physiological and behavioral data. Linear state estimation matrices were derived from the measurement models to form estimates of egomotion states. A fully analog ocellar sensor was designed and constructed based on these models, producing state estimation outputs. These analog state estimate outputs were characterized in the presence of egomotion stimuli. Feedback from the ocellar sensor, with and without complementary input from optic flow sensors, was implemented on a quadrotor to perform stabilization and disturbance rejection. The performance of the closed loop sensor feedback was compared to baseline inertial feedback. A distributed array of digital accelerometers was constructed to sense rapid force and torque measurements. The response of the array to induced motion stimuli was characterized and an automated calibration algorithm was formulated to estimate sensor position and orientation. A linear state estimation matrix was derived from the calibration to directly estimate forces and torques. The force and torque estimates provided by the sensor network were used to augment the quadrotor inner loop controller to improve tracking of desired commands in the presence of exogenous force and torque disturbances with a force-adaptive feedback control.
  • Thumbnail Image
    Item
    Bio-Inspired Polymer Microparticles for Targeted Recognition and Response
    (2014) Arya, Chandamany; Raghavan, Srinivasa R.; Chemical Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    Microbeads and microcapsules are container structures that are frequently used in biomedical applications. In this dissertation, we have sought to impart new functionalities to these particles, inspired by phenomena observed with biological cells. We have engineered polymer microparticles that recognize and respond to specific species from the surroundings (e.g. cells, polymer chains, metal ions). Three classes of new microparticles are reported, which are each reminiscent of a different type of biological cell in terms of recognition capabilities and response. In the first part of this dissertation, we create functionalized microbeads from the biopolymer, chitosan, and use these to selectively recognize and capture Circulating Tumor Cells (CTCs) from blood. The microbeads are functionalized with a protein (streptavidin) and packed into an array within a microfluidic device. Blood samples with biotin-labeled CTCs are flowed over the packed bed of chitosan beads. Similar to how macrophages adhere to foreign bacteria (i.e. antibody-antigen interactions), the streptavidin-labeled chitosan beads can selectively recognize and adhere to the biotin-labeled CTCs. We show that such a packed bed of chitosan beads could serve as an inexpensive platform for customized capture of different rare cells (cancer cells, stem cells etc) from blood. In the next study, we develop a class of microbeads that undergo clustering (aggregation) in the presence of specific polymers. The inspiration for this comes from the cells (e.g., platelets) and polymers involved during the formation of blood clots. Our system consists of chitosan microbeads coated with cyclodextrins (sugar molecules with a hydrophobic binding pocket), which are then exposed to a polymer that is decorated with hydrophobic units. The particles bind to the polymer chains via hydrophobic interactions and in turn, the particles are induced to form clusters. Subsequently, the polymer precipitates and forms a matrix around the particle clusters, leading to a structure that is reminiscent of a blood clot (platelets enveloped by a mesh of fibrin chains). Lastly, we develop a class of microparticles that have the ability to selectively destroy other microparticles. The inspiration here is from the body's immune system, where cells like the killer T cells selectively destroy cancer and virus infected cells without harming healthy cells. Towards this end, we synthesize two types of microparticles: chitosan capsules that contain the enzyme glucose oxidase (GOx), and beads of a different biopolymer, alginate that are crosslinked with copper (Cu2+) ions. The chitosan capsules enzymatically convert glucose from the surroundings into gluconate ions. When these capsules approach the alginate/copper beads, the gluconate ions chelate the copper ions, leading to the disintegration of the alginate beads. Other beads that do not contain Cu2+ are not affected in this process.
  • Thumbnail Image
    Item
    FLY-EAR INSPIRED MINIATURE SENSOR SYSTEM FOR TWO-DIMENSIONAL SOUND SOURCE LOCALIZATION
    (2011) Lisiewski, Andrew Paul; Yu, Miao; Mechanical Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    A micro-scale sound localization sensor is developed and studied in this thesis to address the fundamental challenge of miniaturizing sound localization systems. When miniaturizing a microphone array, there is a critical size limitation at which the array will be unable to localize the sound source in a discernible manner. However, a solution to this dilemma came about when studying the hearing mechanisms of a particular fly, known as Ormia ochracea. Background research into the hearing mechanisms of the fly found that it can accurately locate a sound source even though its eardrums are separated by a distance of only 500 μm. The fly's exceptional directional hearing capability has been linked to a distinct mechanical coupling between its two eardrums, which helps amplify minute directional cues. Inspired by the remarkable hearing capabilities of the fly's micro-scale ear, researchers have sought to develop micro-scale sensors to mimic the fly's ear. One limitation of simply imitating the fly's ear is that the fly is only capable of localizing a sound source in one dimension. In this thesis work, the knowledge gained from understanding the fly ear mechanism is applied to achieve the goal of developing a micro-scale sound localization sensor capable of sound source localization in two dimensions. In this thesis, for the first time, micro-scale fly-ear inspired sensor devices employing three or four coupled membranes have been designed. Reduced-order models have been developed to achieve a fundamental understanding of the performance of each sensor design. Furthermore, a micro-scale sensor device incorporating three mechanically coupled membranes arranged in an equilateral triangular configuration has been successfully developed. Experimental study of the sensor device incorporated with a low coherence fiber optic interferometric detection system has suggested that the micro-scale fly-ear inspired sensor can achieve a much improved performance in terms of phase differences and directional sensitivities when compared to a similar sized microphone array constructed with separate microphones. In addition, localization techniques have been developed to best use the fly-ear inspired sound localization sensors. Future work is suggested to incorporate this sensor system with a fully autonomous robot to improve robot homing and navigation.