Projects
My life's motto is pretty simple: I may not know a lot, but I'm great at figuring things out.
Here are some topics I knew nothing about but worked it out.
Visualized Kinematics for a Motoman Robot
In the rapidly advanced realm of robotics, comprehending motion mechanisms in pivotal for optimizing performace and adaptability. This research delves into the intricate kinematics of the Motoman HC10DPT robot by employing the Denavit-Hartenberg (DH) parameters, using Python and C for computational precision. We systematically addressed both inverse and forward kinematic models: the former to ascertian joint parameters from a known end-effector position and the latter to predict the end-effector's location based on provided joint parameters. Furthermore, through advanced computational visualization techniques, the robot's trajectory was visualized, offering invaluable insights into its motion efficiency and path patterns. This interdisciplinary study provides a comprehensive understanding of the robot's movement mechanics.
Application of Mutual Information QUBO for Feature Selection
This study investigates the use of mutual information and Quadratic Unconstrained Binary Optimization (QUBO) on the D-wave quantum computer for predicting the top performing sectors in the S&P 500. The analysis begins by using mutual information for feature selection to help identify the mutual information between sharpe ratio in the sector and the S&P 500 index, using daily price data over a period of three years from 2020 to 2022. The resulting mutual information matrix is then converted into a QUBO problem, which is solved on the D-wave quantum computer using its quantum annealing algorithm. The experimental results show that the mutual information QUBO approach can accurately predict the performance of the sector, calculated by the sharpe ratio, with a high degree of correlation between the predicted and actual returns. The study also demonstrates the scalability of the method, as it can handle a large number of stocks and can be applied to other sectors or markets.
Synthesis, Phase Analysis, and Superconducting Properties of Bi-2223 via Solid-State Reaction
The stable preparation of bismuth-based high-temperature superconductors, either as Bi-2223 or Bi/Pb-2223, has been a subject of extensive research. The pursuit for better performance has led to in-depth studies of thermodynamics, phase formation mechanisms, and diverse methods to obtain pure Bi-2223 samples. In this study, we synthesized Bi-2223 using the solid- state reaction method and analyzed its phase patterns, concluding the presence of superconductive phases.
