Follow me on social media:

Copyright @ 2022

Research and Projects

I love tackling complex challenges in computing, automation, and intelligent systems by developing smart algorithms, optimizing performance, and bringing ideas to life. Here are a few selected projects:

SwiftScout

Ever wondered how swarm robotics could transform Search and Rescue? SwiftScout merges multi-robot collaboration, a flexible C++ API, and ROS2-based simulation to autonomously navigate known maps, detect stationary targets using HSV-based image processing, and coordinate effectively even in harsh environments. The project’s lean CI/CD workflow, transparent UML diagrams, and pair programming approach ensure reliability, while GoogleTest and Doxygen maintain code quality and clarity. Phase 1 delivers a structured design with UML stubs, and Phase 2 brings the system to life via ROS2 nodes and object detection. Hungry for more? Future improvements include dynamic robot spawns, refined path planning, and even broader applications pushing the frontiers of collaborative robotics for safer, more efficient missions.
Supervisor: Dr. Tommy Chang

Enhanced Path Planning with Expanding Path RRT* (EP-RRT*)

Ever wondered how robots conquer maze-like corridors, dodge dense clusters of obstacles, or swiftly traverse narrow passageways? Enter Expanding Path RRT* (EP-RRT*), a powerful upgrade to the classic Rapidly-exploring Random Tree Star algorithm that supercharges path planning with strategic, heuristic-driven sampling and focused path expansion. By prioritizing and refining the most promising routes, EP-RRT* slashes exploration time, improves convergence rates, and guarantees optimal solutions as it adapts to dynamic or highly cluttered environments. Its blend of accelerated convergence and asymptotic optimality makes EP-RRT* a game-changer for everything from urban navigation to complex industrial robotics, setting a new standard in efficient and reliable autonomous path planning.
Supervisor: Dr. Reza Monfaredi

Report

A*-algorithm on a Differential Drive Non-Holonomic TurtleBot3 Robot

This project showcases an A* search algorithm implemented for a differential drive, non-holonomic TurtleBot3 robot in two parts: a 2D Python-based planner that calculates an optimal path while factoring in clearance and RPM values, and a Gazebo simulation in ROS2 that visualizes the robot following the planned path in a virtual environment, thereby providing a comprehensive demonstration of path planning and motion execution on a TurtleBot3 platform.
Supervisor: Dr. Reza Monfaredi

Optimization of Rocker-Bogie Mechanism using Heuristic Approaches

The objective of this research was to improve the design of the wheel suspension system of a planetary rover based on well-defined mobility criteria using optimization paradigms in conjunction with a rigorously developed objective function that takes into consideration the stimulatory data for the rover's associated commissions. With the advent of computational technologies, predictive analysis, and the multi-objective use of optimization, seven algorithms were leveraged.
Supervisors: Dr. Pooja Shah and Dr. Paresh Gujarati

Preprint

A comparative study on the modern deep learning architectures for predicting nutritional deficiency in rice plants

Plant nutrient insufficiency identification is the process of detecting the number of nutrients deficient in the soil. In this project, we have used five famous deep learning architectures namely, ResNets, DenseNets, MobileNets, VGG-16, and SqueezeNet to identify insufficiency of nutrients like Nitrogen, Calcium and Boron by this Non-destructive method. We created a hydroponic experiment in which we gathered images for three different types of nutrient deficiencies (N, K, and Ca) and compared them to rice plants growing in full nutrition, for a total of four classes.
Supervisors: Asst. Prof. Kirti Bardhan and Dr. Pooja Shah.

Publication