Projects
Project Mahoraga
Advancing game performance through advanced AI algorithms and reinforcement learning for unmatched efficiency and adaptability.
The project’s core is built around AlphaZero, a self-learning AI system that uses reinforcement learning to master game strategies, paired with Monte Carlo Tree Search (MCTS) for optimizing decision-making in real-time.
Enhancing Spacecraft Autonomy
Enhancing Spacecraft Autonomy With Transfer Learning and Generative AI.
A research project exploring reinforcement learning (RL)-based autonomous navigation across planetary surfaces. Enhanced learning by augmenting terrain data using generative models such as WGANs, and Diffusion Models — all enforced with scientific constraints to improve generalization and realism.
VorteX
VorteX is a versatile, multi-functional robot designed to adapt to various tasks.
VorteX is a versatile, multi-functional robot designed to adapt to various tasks, from navigation to advanced processing. Built with a focus on speed, intelligence and flexibility.
Enhancing_Urban_Traffic_Management
Enhancing Urban Traffic Management through Intelligent Control Systems.
Urban traffic congestion is a pervasive problem, leading to significant delays, increased fuel consumption, and environmental pollution. This project introduces an innovative approach to urban traffic management by leveraging Long Short-Term Memory (LSTM) networks for predictive analysis and Reinforcement Learning (RL) agents for adaptive traffic signal control. Our intelligent control system aims to dynamically optimize traffic flow, reduce congestion, and improve overall urban mobility.
Snake Q-Learning
An experimental Snake game where a Deep Q-Network (DQN) agent learns to navigate the environment, avoid obstacles, and find food using reinforcement learning techniques.
This repository is a deep dive into the evolution of the Snake Game using Reinforcement Learning (RL). It features three progressively complex models—Classic, Advanced, and Expert—each exploring unique gameplay mechanics and RL strategies.
Design and Analysis of Flat Radomes for MSPA
This project investigates the interaction between flat radome structures and Microstrip Patch Antennas.
The analysis focuses on minimizing signal distortion while ensuring structural protection. The analysis also involves parametric studies of radome placement, material properties, and their effect on antenna gain, radiation pattern, and polarization.
Star Cluster Analysis NGC1300
Star Cluster Analysis in nearby Galaxies.
This project involves the analysis of star clusters within the NGC 1300 galaxy, using photometric data and theoretical evolutionary tracks to classify clusters by age and symmetry. The analysis also includes overlaying young, middle-aged, and old-aged clusters over CO moment maps from ALMA to study their distribution within the galaxy.
Models Architecture
Model Archive.
Welcome to my Model Archive! This project contains a collection of machine learning and deep learning models.
UV-Vis_and_Excitation_Energy
Theoretical UV-Vis and Excitation Spectrum using DFT and TDDFT with PySCF.
This project involves the theoretical calculation and plotting of UV-Vis and excitation spectra for neutral molecules. It utilizes Density Functional Theory (DFT) and Time-Dependent Density Functional Theory (TDDFT) implemented via the PySCF library.
