Testing the Sensitivity of DNA/RNA Hybridization via Back-Gated Graphene Field Effect Transistors
Advisor: AT Charlie Johnson

A third of all epilepsy patients who experience continuous seizures are treated with surgery to remove epileptogenic brain tissue. Currently, surgeons determine disectable tissue using electrocorticography (ECoG), a technique that detects action potentials, electrical potential difference within a cell, from neurons in the brain. Current ECoG electrodes are not flexible enough to conform to the curved surface of the brain, while depth electrodes cause permanent damage to the brain. Furthermore, the resolution of these sensors is not accurate enough to precisely locate the epileptogenic areas increasing the potential to leave behind epileptic tissue or to extend the resection to healthy tissue. In this project, we want to develop and optimize a microarray of sensors made of materials that are flexible enough, and sensitive enough to increase temporal and spatial resolution. Additionally, a depth electrode delivery device will be developed to mitigate the damage done to neuronal tissue. The sensors are optimized to be used in rat models, which will determine the feasibility and safety of the device for clinical applications.

Nanodiamond Dispersion and the Properties of the Nitrogen Vacancy Center
Mentor: Lee Bassett
Advisor: Yung Huang

Nanodiamonds containing the Nitrogen-Vacancy, or NV, defect are the subject of a great deal of study as of late due to their optical and magnetic properties, which makes these nanodiamonds suitable for use in fields such as biosensing and quantum computation. Our goal is to find a way to isolate individual nanodiamonds on a substrate, to irradiate the nanodiamonds to create NV centers, and finally to characterize the properties of these NV centers. Using a simple method of drop casting a dilute nanodiamond solution and washing off the residue with isopropyl alcohol, we were able to isolate enough single nanodiamonds on a substrate to be irradiated with NV centers. Using these defect containing nanodiamonds, we then probed the optical and magnetic properties of NV centers in nanodiamond.

An Investigation into Self-folding Methods for Centimeter Scale Origami Robots
Advisor: Cynthia Sung

We investigate the possibility of applying two different self folding materials for structures and actuators that could form a centimeter scale robot by applying some external stimulus, such as heat or light. Hand folding robots can take along time, even with a high level of skill. By looking at ways to have systems self-fold and control the fold order, we can create complex designs in a fraction of the time. Based on the work done by a previous REU student, we decided to apply their technique for creating self folding water collectors for our purposes creating structures for our centimeter scale foldable robots. This involves mixing expandable microcapsules in a polymer, laying it on the folds, and later heating the mixture to cause self folding via expansion of the capsules. In addition, we looked at creating actuators by using liquid crystal elastomers, with the eventual goal of being able to use a syringe to lay down the LCEs easily along the 2d design, and have them fold our robot for us.

We have tested these two actuation methods under varying parameters, such as concentration, thickness, curing temperature, etc.
By the end of this summer, we will have determined which of these parameters has the greatest affect on the material’s folding ability and provide design guidelines for future cm-scale origami robots fabricated using these approaches.

An Investigation into Self-folding Methods for Centimeter Scale Origami Robots

Advisor: Cynthia Sung

We investigate the possibility of applying two different self folding materials for structures and actuators that could form a centimeter scale robot by applying some external stimulus, such as heat or light. Hand folding robots can take along time, even with a high level of skill. By looking at ways to have systems self-fold and control the fold order, we can create complex designs in a fraction of the time. Based on the work done by a previous REU student, we decided to apply their technique for creating self folding water collectors for our purposes creating structures for our centimeter scale foldable robots. This involves mixing expandable microcapsules in a polymer, laying it on the folds, and later heating the mixture to cause self folding via expansion of the capsules. In addition, we looked at creating actuators by using liquid crystal elastomers, with the eventual goal of being able to use a syringe to lay down the LCEs easily along the 2d design, and have them fold our robot for us.

We have tested these two actuation methods under varying parameters, such as concentration, thickness, curing temperature, etc.
By the end of this summer, we will have determined which of these parameters has the greatest affect on the material’s folding ability and provide design guidelines for future cm-scale origami robots fabricated using these approaches.

The effect of electric field on Schistosoma and its application
Advisor: Haim Bau

Schistosomiasis, a disease caused by the parasitic flatworm Schistosoma, is one of the most deadliest diseases known worldwide. This disease particularly affects people living in tropical and subtropical areas. People are infected when they are exposed to water bodies that are contaminated by the parasitic flatworms. Due to the fact that Schistosoma live in water bodies, one way of preventing the disease is to clear them out of these habitats. We decided to use the idea of taxis to filter out the schistosomes. Several different taxis experiments have been carried out on different organisms. For example the phenomena of electrotaxis has been observed in some nematode and in particular C.elegans. Though electrotaxis has been examined on C.elegans, it has not been examined in Schistosomes. If Schistosomes do show electrotaxis we can use this behavior to create a filtration device. This device will use electric field as a way to attract these worm to filter them out of infested water bodies. We will run the electrotaxis experiment on the C.elegans as a control and compare the response to the Schistosomes.

Optimizing Deflection-routed Butterfly Fat Tree Networks on FPGAs
Advisor: Andre DeHon

Deflection-routed Butterfly Fat Tree networks (BFTs) implemented on Field Programmable Gate Arrays (FPGAs) can be used as resource and time-efficient communication networks between processing elements at leaf nodes. This work demonstrates how the routing function of switching nodes implementing a localized-deflection scheme can be optimized to minimize packet deflections and ensure an even distribution of packet bandwidth under high traffic. Our network generator program constructs arbitrarily-sized networks with a configurable bisection bandwidth determined by Rent parameter 0<p<1. Generated networks were tested under a range of packet injection rates and patterns. We expect our optimizations and network generator program to add to the robustness of constructing and utilizing BFT networks. We hypothesize that highly configurable deflection-routed BFTs with localized-deflection schemes are ideal candidates to be implemented with partially reconfigurable FPGA regions in rapid-prototyping applications.

Planning and Executing Missions with Multiple Autonomous Quadrotors
Advisor: Rahul Mangharam

Autonomous quadrotor technology has the potential to achieve wide-spread implementation in applications such as urban delivery, search and rescue, and 3D mapping. The efficiency with which such missions are completed can be greatly increased through the use of multiple, cooperative quadrotors rather than a single, independent quadrotor. In order to complete autonomous multi-quadrotor missions, we must ensure that the flight trajectories are planned such that there are no mid-air collisions between quadrotors. We investigated Planning Domain Definition Language (PDDL) as a way to encode the constraints, actions, and goals of a multi-quadrotor mission to formulate a safe autonomous flight plan. Using a PDDL plan solver, we were able to find a flight plan for a simple mission involving two quadrotors flying from two different starting points to two different destinations while avoiding a no-fly zone. However, because of the lack of scalability for missions involving a larger number of quadrotors, we do not recommend that PDDL be adopted as the standard planning language for autonomous multi-quadrotor missions.

Deformations and mechanics behind the cell-free layer
Advisor: Paulo Arratia

In the flow of red blood cells through microchannels, there is a development of a cell free layer around the outer walls of the channels where there is an absence of red blood cells and a high concentration of platelets. From past research, it is known that this formation is due to the tendency of the red blood cells to migrate towards regions of lower shear stress but is also affected by factors like deformability of the red blood cells, the pressure drop induced on the channel, the geometry of the channel, and the properties of the fluid itself. This project looks at understanding what factors most influence the development of this cell free layer in order to better understand the mechanics behind the flow of blood. By simulating blood flow within the body using a blood solution through a microfluidic device we can obtain the velocity and shear profiles for each red blood cell and focus on understanding the development of the cell free layer. It is important to understand this movement of red blood cells in order to better understand how this flow is affected when veins undergo change in their natural geometry in medical procedures and to comprehend its importance in naturally occurring processes like clotting and circulation.

Learning Topology from Spectral Characteristics of Brain Signals
Advisor: Alejandro Ribeiro

The emerging field of graph signal processing (GSP) suggests an interpretation of brain function as a dynamic process supported on top of a graph. The subject of this paper is on identifying a model, which unifies brain function and structure. By expressing brain signals as the sum of eigenvectors, we can match the distribution of functional signals to the density of white matter streamline connections in the brain. This leads to the development of a method for topology inference from brain signals. Using convex optimization, we solve for the optimal topology, which is both sparse and preserves the spectral templates of the functional graph.

Brain Machine Interfaces
Advisor: Drew Richardson

The brain serves as the most critical part of the human neural system, and with this work we are venturing to have the brain also serve as the most critical part of external circuitry that responds to spikes in neural activity. Since the brain uses electrical signals to communicate from neuron to neuron, we can use the brain as a component in our circuit to achieve our goal. Our brain-machine interface (BMI) allows us to read brain waves and look for brain activity that registers above a certain magnitude in voltage; we refer to this high magnitude activity as a “spike”. Upon registering a spike the circuit then turns on the stimulating component of the system and sends small amounts of current into the brain. Testing these theories on rats, preliminary results and past works show that this artificial brain stimulation can act as a “6th sense” of sorts to help rats get through a maze faster than otherwise.