Autonomous Underwater Black Box Retrieval Using BlueROV2
In this project, I designed a modular system to enable underwater robots to detect and retrieve black boxes autonomously. Using ROS2 and Unity simulation, I created an environment for testing. The system achieved real-time detection of ArUco markers for locating objects and used PD controllers for precise depth control. To handle complex decision-making, I implemented a behavior tree framework, allowing the robot to switch between manual and autonomous modes seamlessly. This project combined hardware and software integration for real-world applications.
Advancing Map-less Underwater Navigation
This research investigates the efficacy of various reinforcement learning algorithms, including Actor-Critic with Experience Replay (A2C), Proximal Policy Optimization (PPO), and Trust Region Policy Optimization (TRPO), to enhance the navigation capabilities of low-cost underwater robots. The focus is on developing a robust and resilient navigation system integrating advanced sensor fusion techniques for improved depth perception and obstacle avoidance. The project also explores the refinement of reward functions to recover effectively from navigation failures, ensuring consistent progress toward designated goals in diverse underwater environments.
Industry 4.0: Remote Control Secure and Optimize Operations for Industrial Automation
In this project, I created a system that allows you to remotely control any industrial process using your mobile devices. The user may keep track of the process and record the data to any server; in this example, I used MySQL. I also used GRAPHANA to show the data for analysis. Users may also use machine learning to create a prediction model and anticipate their Factory's future output. I picked a Siemens S7 1200 CPU since it has OPC functionality, allowing me to create a secure data exchange between my PLC and the server. I used the Node-Red programming tool to connect many microprocessors and edge devices together. The Raspberry Pi 4 enabled the system to be both robust and cost-effective.
IOT- Based Air Pollution Monitoring and Forecasting System
Air pollution is the major environmental and public health concern in the globe today. Air pollution leads to detrimental consequences on Human health, climate and ecology. Air is increasingly contaminated due to release of Toxic gases by businesses, vehicle emissions and increased concentration of dangerous gases and particulate matter in the environment. Particulate matter is one of the most important indicators having the considerable role to the growth in air pollution. This provides a requirement for measurement and analysis of real-time air quality monitoring so that suitable choices may be done in a timely timeframe. This work proposes a real-time independent air quality monitoring system which incorporates multiple parameters: PM 2.5, carbon monoxide, carbon dioxide, temperature, humidity and air pressure.
UR5 Robot Control and Object Manipulation
This project, I built a framework for controlling a UR5e robotic arm integrated with a Robotiq 3-Finger Gripper. I implemented YOLO-based object detection to identify objects in real-time and enabled advanced manipulation tasks through inverse kinematics and trajectory planning using MoveIt2. The system was optimized for adaptive grasping, making it capable of handling a variety of objects. Tools like ROS2, YOLO models, and MoveIt2 played a crucial role in achieving precise control and object handling.
Abstract art evokes diverse emotional responses, making it challenging to analyze computationally. This project focuses on:
-
Texture Classification: Classifying textures (e.g., chaotic, smooth, rough) in abstract art using a fine-tuned CLIP model.
-
Sentiment Analysis: Extracting polarity and subjectivity from viewer-generated text annotations.
-
Correlation Analysis: Exploring relationships between textures and emotional responses.
Epson LQ310 Printer Integration With S7-1200 PLC
I successfully integrated the Epson LQ310 printer with the S7-1200 PLC, establishing seamless communication via RS232 and configuring automated printing functionality. This project involved developing and optimizing PLC logic to generate and transmit data for accurate, real-time printing in industrial operations. The project began with understanding the communication protocol requirements of both the S7-1200 PLC and the Epson LQ310 printer. The RS232 protocol was selected as the communication medium, and its configurations, including baud rate, parity, stop bits, and data bits, were carefully matched between the devices to ensure compatibility. I developed and tested the PLC logic to prepare and transmit data in the correct format that the printer could interpret. This included programming the PLC to generate control characters and print commands, enabling the automation of tasks such as line breaks, form feeds, and specific text formatting.
Railway Ticket Generator Python and SQL Database Solution
This project involves creating a Railway Ticket Generator software using Python and an SQL database. The software will facilitate ticket booking, management, and storage of passenger information, providing an efficient and user-friendly solution for railway ticketing needs.
Autonomous TurtleBot3 Navigation in Maze Environment
This project focused on creating a simulation for multi-robot navigation in a maze using TurtleBot3 robots. I implemented SLAM, or Simultaneous Localization and Mapping, to enable the robots to build a 2D map of the maze and navigate through it effectively. Additionally, I developed a system for synchronous robot behavior, where multiple robots could start moving simultaneously in response to external signals. To achieve this, I used ROS2 for integration, designed custom configurations in Gazebo for the simulation environment, and leveraged tools like slam_toolbox and the ROS2 Navigation Stack.
Dynamical Modeling of SCARA Robot and Hexabot Robot
This project focuses on the dynamical modeling of two distinct robotic systems: the SCARA Robot and the Hexabot Robot, under the guidance of Prof. Cédric Anthierens at the University of Toulon. Utilizing MATLAB and Simulink, the project delves into the kinematic and dynamic analyses of these robots, employing Denavit-Hartenberg (DH) parameters, Homogeneous Transformation Matrices, Jacobians, and the identification of singularities for the SCARA Robot, as well as the Hexabot Robot's frame analysis and Jacobian calculations.
Simplex 5 Series Converted into PLC System
In this project I have mimic the UI and some working process of Simplex 5 series into complete PLC based. This system can follow some specific quadric equation by following the carve path . User can select 8 different graph for completing their Dyeing process. I have used siemens S7-1200 and Comfort HMI, and for programing I have used STL language as I have good grip on this language.
Mars Rover
This project aims to test microwave imaging's ability to diagnose COVID 19-indueed damage in a lung model and generate microwave images that can be interpreted to identify the damaged area. COMSOL Multiphysics was used to load a Fusion 360 created lung model with many thorax In order to distinguish between a healthy and diseased lung, a microwave image processing method was utilized to examine the output image. The model's output error increased as thc damaged region grew. The specific absorption rate estimated using CST software and used to assess thc safety Of microwave frequencies near tbc human lung. Looking to thc future. microwave imaging technology will become more widely available. allowing medical imaging to tx•come less restricted to large medical institutions. Microwave imaging SYStgtns are expected to become a feasible diagnostic tool for a wide range Of medical applications over the next decade. -ntcy will also be accessible as a monitoring tool, not just in large clinical centers but also at home for sclf•monitoring. 'Ibosc instruments will be at.ailable both in major clinical centers and at home for self-monitoring.