Yash Srivastava

I'm a recent graduate of the MS in Robotics program at Georgia Tech, with a specialization in the intersection of Artificial Intelligence, Control Systems, and Perception for Autonomous Robots.

I'm skilled in Python, C++, MATLAB, ROS/ROS2. I have extensive Prototyping experience, and have worked with Raspberry Pi, Arduino. I'm Familiar with Linux OS and Embedded programming. I have a strong passion for the drone industry and a deep interest in engaging with projects centered around drone autonomy, and I gained valuable experience during the past summer as a Systems Engineering Intern at DroneUp.

My ongoing research is advised by Dr Sean Wilson, where I'm building a remotely accessible testbed for Quadrotor Swarms in the Robotarium.

I Completed my undergrad in Electronics and Computer Engineering from VIT-Chennai (2022). I was the Avionics Lead at Team Aviators International where I built the team's first drone capable of fully autonomous flight.

Outside of the lab I enjoy playing cricket, tennis, alongside biking and reading. Feel free to drop me an e-mail.

Email  /  CV  /  LinkedIn  /  GitHub

Built the foundation for a Swarm Drone testbed using Crazyflie and ROS2, in the Robotarium at Georgia Tech. Developed a Finite State Machine based algorithm for the experiment pipeline, with automated charging. Implemented Barrier Certificates for safety guarantees by minimally-invasive trajectory modification.

Designed and Implemented a state machine based algorithm for navigating to a defined goal in a GPS-denied environment for Differential Drive Autonomous Mobile Robots running ROS2. Used Computer Vision and Support Vector Machines for building a real-time road sign classifier for driving towards the goal.

Developed a low-cost indoor localization solution using Ultra-Wide Band (UWB) and embedded system programming. Added a layer to the Two-Way Ranging protocol to sync 8 UWB beacons, achieving centimeter-level precision in the XY plane, with a 94% average packet response rate and an 8Hz update rate.

Implemented a series of control regimes for a swarm of multi-rotor UAV systems to rendezvous at a location, assemble in the desired formation and transport payload collaboratively to the desired location. Tech Stack: Ardupilot-SITL, ROS Melodic, Gazebo sim, Dronekit-python.

Developed a rule-based algorithm for autonomous navigation for multirotors that enabled the drone to travel to pre-defined waypoints with reactive obstacle avoidance. Validated using Gazebo Sim.
Built a quadrotor and hexarotor for hardware implementation using a Raspberry Pi 4 as the onboard computer for executing the algorithm. Dronekit-python used for communicating with the Pixhawk. Interfaced an Arduino with ultrasonic sensors and integrated it with the Rpi using ROS.

Built a drone-based solution to help forest authorities in preventing poaching activity in forest areas. Combined inputs from a stereo-vision and ultrasonic sensors for developing a robust two-fold obstacle avoidance algorithm to traverse forest cover.
Simulated a Wireless Sensor Network in MATLAB to visualize data routing from triggered node to the base station. Performed poacher and animal detection using a custom-trained YOLO model.

Built an autonomous 2WD bot with machine learning based reactive navigation for indoor environments using a Raspberry Pi 3, Arduino Uno and ultrasonic sensors. Tested in a cluttered indoor setting.
Trained the machine learning model on data collected by the robot. The model generates the best move for navigation in real-time.

Designed a bot for cleaning polluted water bodies. Worked on the navigation and trash detection aspects.

Source code