Recently Updated Projects

This project aims to simulate hologram technology through the project of laser-generated images into a fine water mist, creating the appearance of screenless, floating visuals. A laser mounted on an optics bench is shaped using diffraction grading and focusing optics to project patterns into the mist, where the light scattering among the suspended droplets makes the image visible from various angles. A mist generator provides a stable projection volume, creating the illusion of a static or rigid physical body. Using motion sensing, this design strives to enable user interaction by modifying projection patterns in response to human movement. The goal of this project is to demonstrate and create a controllable, interactive display that visually mimics holographic effects using optical components and illusion.

This project aims to optimize a user’s flight itinerary by combining fare forecasting with flight route optimization. This project examines two independent methods to reduce air travel costs. The first leverages a classical approach: machine learning models forecast fares, and graph search methods such as A* build multistop itineraries under user preferences. The second uses quantum computing for both forecasting and route selection, including a Quantum Approximate Optimization Algorithm to formulate an itinerary. The study measures how each approach performs in terms of accuracy, runtime, and usability, and reports where quantum methods provide an advantage and where classical computing does the job. The goal is to provide an accurate comparison of both approaches when applied to fare-optimized itinerary building.

"update"

Introduction: Cell manipulation is an essential part of medicine because of its ability to aid in cancer treatment, genetic disease therapy, and more. It is typically performed using glass pipettes. However, because of the fragility of the cell and the difficulty in achieving a stable position, it is often difficult for glass pipettes to accurately manipulate cells. To solve this, a proposed solution is the use of optical tweezers. Optical tweezers are highly focused laser beams capable of trapping and moving microscopic objects. By integrating optical tweezers with glass pipettes, this project aims to enhance the precision, stability, and overall success rate of single-cell manipulation. This system not only minimizes mechanical stress and cell damage, but also opens new possibilities for advancing research in areas ranging from stem cell biology to targeted genetic modification.

Intellectual Merit: The intellectual merit of this project comes from its effort to integrate two fundamentally different manipulation techniques into a single experimental platform. Traditional glass pipettes are based on suction and mechanical positioning, but this approach introduces shear forces that can destabilize the cell or rupture its membrane. Optical tweezers, in contrast, exploit the radiation pressure of a tightly focused laser to generate gradient forces that trap particles near the beam waist with nanometer-scale precision. However, tweezers alone are often limited by trap stiffness when applied to larger or irregularly shaped cells. By combining the localized suction control of pipettes with the non-contact trapping forces of optical tweezers, this project seeks to overcome the limitations of both methods. The technical goal is to establish a dual-manipulation system capable of stabilizing a cell in three dimensions while minimizing mechanical stress. Such an approach provides a unique testbed for quantifying the limits of laser trapping forces, calibrating pipette suction under controlled conditions, and ultimately expanding the toolkit for single-cell manipulation in ways that could not be achieved by either technology in isolation.

Broader Impact: This project combines optical tweezers with glass pipettes to manipulate single cells, such as amoeba, with the goal of improving precision while reducing stress on the cells. This approach could make experiments in cancer research, stem cell biology, and regenerative medicine more reliable, as these cells are often fragile and difficult to control. It could also help in gene therapy, drug development, and synthetic biology by providing a more consistent way of handling cells. Beyond these immediate applications, the system offers a way to connect physics, engineering, and biology in practice. Developing it could lead to new laboratory techniques, improve reproducibility between experiments, and expand the tools available for high-precision cell studies. Overall, this work has the potential to benefit both ongoing research and training in biophysics and biomedical engineering.

My project aims to create a learning framework for large multi-agent systems that combines mean field game theory with model-free quantum reinforcement learning methods. This means formulating a representative agent model, where the agent would interact with the rest of the population, and their interaction would be described by the reduced quantum states or the expectation-level dynamics, which would then allow for the use of the quantum abstractions such as the quantum channels and the DMFT reduction. In addition, the use of the two-timescale update method would allow for both policy learning and the agent-level mean-field updates. This will also involve the usage of stabilizing components such as smooth stochastic policy, replay buffers, and target networks, coming from the classical and the quantum reinforcement learning. This would help develop an approximate model for the mean field equilibrium without needing any information regarding the external environment, providing an efficient method for learning control and simulations involving the use of the hybrid quantum-classical multi-agent populations. This would hopefully be applicable for the wide range of the quantum decision-making processes, specifically for finance and the economics, as well as the new quantum-enabled simulations involving the use of the many agents.

The proposed research will examine the application of a Quantum-Inspired Evolutionary Optimization (QIEO) algorithm in solving a sample rocket trim state problem, a non-linear, complex aerospace engineering problem. This contribution extends numerical optimization toolkits like CasADi and csdl_alpha with quantum-inspired numerical computation. The project shall proceed by first developing an operational QIEO algorithm in Python to prove feasibility and then creating the actual rocket trim model. A QIEO approach may offer a more efficient alternative to traditional gradient-based solvers.

The intellectual value of this research lies in the integration of quantum-inspired algorithms and computation with aerospace vehicle dynamics. Though it has been demonstrated to have promise in other optimization areas, combining QIEOs with design libraries like CasADi or csdl_alpha to rocket flight dynamics is an uncharted area. This research borrows from ideas in evolutionary algorithms and quantum physics. This project will establish the feasibilty of applying quantum-inspired approaches to practical problems.

The broader implications of this study are important for the aerospace and defense industries. A successful deployment would pave the way for more efficient methods of studying vehicle stability and control. This might potentially be employed to optimize the design of autonomous flight systems, and enhance aerospace vehicle safety and performance. Furthermore, this project serves as a theoretical bridge, between quantum concepts and conventional design frameworks.

This project investigates whether a simple quantum two-level system can provide an informative alternative to classical models of bull–bear regimes in financial markets. Standard regime-switching models treat the market as occupying one of two states, bull or bear, with fixed transition prob- abilities estimated from historical return data. Here, we instead model the market as an effective quantum two-state system with basis states Bull and Bear that evolve under a Hamiltonian gen- erating Rabi-like oscillations, supplemented by a tunable decoherence process that damps purely coherent behavior. Using daily price data from broad market indices, we will label returns as bull or bear using a transparent threshold rule, fit a classical two-state Markov chain as a baseline, and then construct and fit a quantum two-level model whose parameters are chosen to match empirical one-day and multi-day transition statistics.