Crossmap (Web Tool)
Interactive web tool for visualizing and optimizing campus coverage using grid-based optimization and comparative result views.
Explore the Project
Project Overview
The Crossmap web tool was developed as part of my role in the Crossmap research group at UMD's FIRE (First-Year Innovation and Research Experience) program. Our team implemented and expanded upon a quantum optimization paper that lacked publicly available code, building the algorithm from scratch using simulated annealing as an alternative to the quantum annealer used in the original research. You can view more of the code implementation in depth by visiting the repository here.
This web interface was my independent contribution to the research group, designed to help generate input matrices for the optimization algorithm. The interface, inspired by Photoshop's layout, allows users to place red circles with adjustable radii on campus maps, which are then exported as binary matrices (1s for marked areas, 0s for unmarked areas) for algorithm input.
Research Achievements
From-Scratch Implementation: Successfully implemented the optimization algorithm from scratch, adapting quantum annealing concepts to work with simulated annealing due to hardware limitations.
Algorithm Adaptation: Modified the original quantum approach to work with classical simulated annealing, demonstrating the flexibility of optimization techniques across different computing paradigms.
Prototype Web Tool: Developed a prototype interface for generating input matrices and visualizing campus Wi-Fi coverage patterns for future algorithm testing.
Technical Implementation
Research Algorithm
Implemented the optimization algorithm from scratch using Python and simulated annealing, adapting the original quantum annealing approach to work with classical computing resources. The algorithm successfully optimizes resource placement in constrained grid environments.
Core Algorithm
Python, Simulated Annealing, Algorithm Analysis
Research Skills
Problem Solving, Analytical Skills, Research Skills
Prototype Web Tool
Developed crossmap.xyz as a prototype interface for generating input matrices and visualizing campus Wi-Fi coverage. The tool allows users to mark existing Wi-Fi access points to identify areas needing additional coverage.
Frontend
HTML/CSS, JavaScript
UI Design
Photoshop-inspired interface, Layer management, Export functionality
Note: The web tool's source code is kept private. The GitHub repository contains only the research documentation and algorithm implementation.
Research Impact
While our current implementation is limited by the computational constraints of simulated annealing (we could only work with small-scale grids due to the exponential qubit requirements), this work serves as a proof of concept for what's possible when quantum computing technology advances.
The real value of this research lies in laying the groundwork for future optimization applications. When quantum hardware becomes more accessible and we can afford to use more qubits, these algorithms could potentially outperform classical computing for large-scale resource placement problems like campus-wide Wi-Fi optimization, emergency response planning, or urban infrastructure design.
Our work in FIRE was essentially a test to see if we could recreate the paper's results from scratch, since no code was provided. Successfully implementing the algorithm from the ground up demonstrates that these optimization techniques are viable and ready for real-world applications once the hardware catches up to the theory.