I am pursuing my Masters in Robotics and AI at Arizona State University. My research interest lies in 3D Computer Vision, Motion Planning, and Reinforcement Learning for Robotics applications.
Currently I am working with Dr. Hongbin Yu at ASU-EVSTS Lab to develop robust and efficient detection and tracking algorithms with 128 channel LiDARs. My research focus is to improve pedestrian safety at traffic intersections using infrastructure sensor setups.
For my masters thesis I have joined the Logos Lab to work on on Neural Motion Planning for robotic manipulators with Dr. Nakul Gopalan
Earlier I was a software developer and data analyst at Wells Fargo - EGS, where I worked on document layout analysis for numerical data extraction from handwritten forms.
I received my B.Tech degree in Computer Science and Engineering from the National Institute of Technology, Rourkela. During my junior year summer internship, I worked with Dr. Alex Ramirez-Serrano at the UVS Robotarium Lab - University of Calgary and developed efficient perception algorithms for humanoid robots. Before that, I collaborated with Dr. Dilip Kumar Pratihar at the Soft Computing Lab - Indian Institute of Technology, Kharagpur, and developed control algorithms for rehabilitation robots.
Room Automation Module for switching electrical appliances remotely through Web, Bluetooth, and IR.
Autonomous Chess Playing Robot Prabin Kumar Rath, Neelam Mahapatro, Prasanmit Nath and Ratnakar Dash
IEEE ROMAN, 2019
  (Listed under top 10 projects in Quest Ingenium 2018) paper
/
video
/
code
A robot for playing the game of chess physically with an user. Powered with strong chess engines and interactive UI it provides all virtual game features and ensures the authenticity of the original board game.
Software Projects
Shaped-Swarm Arizona State University, 2023 video /
code
Multi-robot swarm pattern formation using the Signed Distance Field (SDF) of hand drawn images.
Beyond-Demonstration Arizona State University, 2023 code
Implementation of T-REX and D-REX Inverse Reinforcement Learning (IRL) algorithm for learning form suboptimal demonstrations.
Deep Q-Learning Arizona State University, 2022 code
Deep Reinforcement Learning with PyTorch and OpenAI-Gym. Implementation of Deep-Q-Learning and Dueling Double Deep-Q-Learning Algorithms.
Reflex-MCTS: A hybrid MCTS agent for playing Pacman Arizona State University, 2022 code
An AI agent that utilizes Monte Carlo Tree Search (MCTS) for exploration and exploitation but switches to customized reflex actions at critical zones.
Traffic Flow Prediction with Spatio-Temporal ResNet Arizona State University, 2022 code
Implementation of ST-ResNet model for predicting traffic flow on BikeNYC and TaxiBJ datasets.
2D SLAM and Navigation Stack in Gazebo NIT Rourkela, 2021 code
A tutorial for integrating ROS navigation and mapping stack with your custom differential drive robot.
The Cartpole Control Problem NIT Rourkela, 2021 video /
code
Trajectory optimization for the cart pole swing-up problem and balancing inverted pendulum on a cart pole system using LQR control.
Robot Command Interpretation NIT Rourkela, 2020 paper /
code
Simple LSTM model for classifying colloquial english sentences onto motion commands. Curated text to command dataset.
End Effector Stabilization of Robotic Arms NIT Rourkela, 2020 paper /
video /
code
Mathematical modelling of end effector pitch. Closed-loop PID control for canceling the base motion effects on the end effector.
Hardware Projects
4-DOF Cube Stacking Arm Arizona State University, 2023 video /
code
Customized replica of Turtlebot arm for pick and place tasks. ROS Control and MoveIt stack. Gym compatible environment for training RL agent in Gazebo.
Autonomous Underwater Vehicle (AUV), Team Tiburon NIT Rourkela, 2019 video
6DOF holonomic robot with integrated vision sensors for carrying out planned missions autonomously in underwater environments.
2D Dot Matrix Printer NIT Rourkela, 2018 video /
code
A 2D Dot matrix printer is a DIY project made from old CD drives.
Semi Autonomous Wireless Bot IIT Kharagpur, 2017 video
A Bluetooth controlled robot with on board FSR (Force Sensitive Resistor) for weight detection. It has indication LEDs to prompt the user about the detected weights.