The goal of this project is to implement a Kalman Filter to estimate the rotation dynamics of a satellite, focusing on the angular position and gyro bias using noisy sensor measurements from a star tracker and gyroscope. The underlying model of the satellite’s rotation is described in continuous time, which reflects the natural dynamics of satellite motion, while the measurements from the sensors are sampled in discrete intervals, mirroring real-world data collection scenarios. The filter was designed to handle real-world challenges such as constant gyroscopic bias and Gaussian white noise in measurements. Through systematic simulation, I set the satellite to undergo a constant rotation, with true conditions unknown to the estimator to mimic practical scenarios. 

The project includes detailed visualizations such as the comparison of estimated values with true and measured data, error dynamics over time, and the effectiveness of bias estimation. These plots not only demonstrated the efficacy of the Kalman Filter in reducing estimation errors but also highlighted its critical role in enhancing the reliability of satellite navigation systems. The implementation and subsequent analysis provided valuable insights into the adaptive capabilities of Kalman Filters in complex, noisy environments.

This project primarily focuses on the 'Localization and Control' problem with the holonomic class of mobile robots. It discusses a computationally less intensive with high rate localization and navigation techniques for a three wheeled omnidirectional mobile robot, using wheel encoders and kinematic relations. Insight to an elegant and pragmatic approach to navigate the robot from point A to point B in the defined work-space, enabling the robot to autonomously reach a target position and orientation defined by the user, is given. Implementation of the ‘Go to Pose’ algorithm is done for both single-waypoint and multi-waypoint navigation. The experimental results obtained reinforce the robustness of the algorithm that incorporates PID controller.

This work led to a conference publication in IEEE DISCOVER - doi : https://doi.org/10.1109/DISCOVER52564.2021.9663644 

The challenges include: Design and Construction of three wheeled omnidirectional mobile robot, Odometry sensory network, PID tuning (finding result space). 

Recently we have witnessed a lot of natural disasters in India and across the world. One such natural disaster is Flooding. This monsoon floods have been rampant across our country. Due to this, providing relief and rescue operations have been at the forefront of activities in the country. Most of these operations revolve around evacuating residents and livestock to safer places; over and above providing sacks of food, medical aid and other necessities to affected areas where people are stranded.

With the incessant downpour of rain in recent times and metropolitan cities getting flooded; e-Yantra came up with Flood Disaster Management theme called “Supply Bot”. The idea is to build a robot to provide the affected district or cities with aid. The aid to be received by the district is communicated wireless to the Bot using an overhead imaging camera that emulates a Satellite.

Once the Bot has identified the beacon of help: release of food, medical aid or other necessary goods for surviving floods or evacuation is carried out. The package of supplies required by the district or city is embodied in the beacon (metaphorically). Moving this package as close as possible to the affected district or city is the primary task of the Supply Bot. If the capital of the state is affected, the Bot has to attend to it first.

The challenges in the theme include: designing and building a robot with basic components given, Python Programming, Image Processing, Embedded C Programming.

The team eYRC#95

I (holding the robot) was the design lead in my team , custom designed the entire robot to meet the theme's requirements as efficiently as possible. In addition, I designed a unique relief dispatch mechanism by employing anticipatory control adjustments to the robot's trajectory during dispatch of the relief aids. Also programmed in embedded C for serial data reception using ZigBee module and ISRs.

A health care monitoring system aimed at paralytic patients was developed which included several sensory modules and actuators. The project basically consisted of a glove that housed electronics which continuously monitored patient's vital signs, such as body temperature, and Beats Per Minute (BPM). This data was continuously logged on to real time data base at fixed time intervals, and the same was retrieved from the database and was being visualized in graphs and other formats at a distance and alerted the doctors/nurses/guardians accordingly.

Of all the domains that needs uttermost attention, health care seemed to crave a bit more attention. As Indians, most of the common people cannot afford rich hospital facilities, and if their health issues and conditions are not addressed within the stipulated time, we cannot cure the patient. More so , paralytic patients face a great deal of difficulty and challenge in performing day to day activities. Due to their restricted movement of their body, they struggle and take time to convey a simple message such as conveying their guardian when they are thirsty, or need to use the washroom, etc. This project aims to eliminate this barrier of communication by assigning simple functions based on the hand tilt of by using various electronic sensors and actuators as discussed above. Also in order to keep track of their vital signs such as temperature and heart rate or pulse rate, all of these health signatures are pushed to a local webserver or a cloud, known as firebase , which works in real-time. The same information can be retrieved automatically and is given to the doctors in a timely basis, so that when no immediate check on the patient is required , the doctor or nurse can look on this information to monitor the health of such paralytic patients in real-time.

Control of mobile robots is a very vast and diverse field within Robotics and Automation. We attempted to design a robot that is entirely autonomous and successfully achieved doing so. The project's objective is to develop a robot that can follow an n-shaped structure, or rather a polygon ( n - sided ),  based on the coordinates of the sides  ( x, y ) of the polygon given initially. These coordinates are taken in as system variables and a PID controller governs the traversal of the robot in the desired trajectory. 

This project was developed as a practical exposure to the course "Control of Mobile Robots" by Prof. Magnus Egerstedt from The Georgia Institute of Technology

Two-factor authentication adds an additional layer of security to the authentication process by making it even more difficult for intruders to gain access since knowing the victim's password alone is not sufficient to obtain full access. Two-factor authentication has long been used to control access to sensitive security systems and data, and online service providers are increasingly using 2FA to protect their user’s credentials from being used by hackers who have stolen password, or used phishing campaigns to obtain user credentials. 

Presented a conference paper in International Conference on Recent Trends on Science and Technology - 2020 based on this work.