Nandagopan K

IIT Palakkad

Implementation of reinforcement learning algorithms in hardware using MATLAB and Simulink. Implementation and testing of latest neural network architecture (KAN) and developing its variants. Migration of MLP based neural networks to KAN.

Indian Institute of Technology Palakkad (Ihub)

Improving and Implementing Deep Reinforcement Learning algorithms in Gazebo-simulated environments using PyTorch and ROS. Implementation of optimal control algorithms in hardware. Migration of ROS 1 code to ROS 2.

Luminar Technolab

Collecting, cleaning, and preprocessing data. Exploring and analysing data using statistical methods and tools. Building predictive models and algorithms to solve specific business problems. Communicating insights and results to stakeholders through visualisations, presentations, and reports. Statistical analysis, quantitative analytics, forecasting/predictive analytics.

Deep reinforcement learning algorithms were implemented on a turtlebot3 for autonomous navigation and obstacle avoidance.

Gazebo · ROS · Python · Deep Learning · Pytorch

An Linear Quadratic Regulation (LQR) controller was implemented to stabilize a self balancing robot. It was implemented in a simulated environment(Gazebo) and ROS was used to communicate with the robot.

Python · ROS · Gazebo

Implemented state-of-the-art (SOTA) Deep Reinforcement Learning algorithms in the Gymnasium Classic Control environment and 3D environments. Bench-marking the results.

Python · Pytorch

This is an Amazon review sentiment analysis web app that is built using the Streamlit framework and the VADER model. The application outputs the sentiment of the review as either positive, negative, or neutral.

Python · Scikit-Learn · Streamlit

This Flask-based web app uses NLP techniques and a pre-trained Naive-Bayes model to classify messages as spam or non-spam.

Python · scikit-learn · flask