Robotics¶

MAP-IT : 2D Mapping Iterative Closest Point and Pose Graph on Mobile Robot¶

I applied the Levenberg-Marquardt algorithm for Point to Line Iterative Closest Point (PLICP) and developed occupancy grid mapping from scratch using NumPy. Additionally, I built a web application for mobile robot control, integrating it seamlessly with Robot Operating System 2 (ROS2) to enhance functionality and user interaction. Furthermore, I published a paper on my work at the International Conference of Advanced Technology and Multidiscipline (ICATAM 2024).

Graphical User Interface with MuJoCo Simulation¶

This project involves building a MuJoCo simulation for a mobile robot and implementing PLICP with Graph SLAM. The simulation leverages OpenGL (via Python bindings) to visualize and plot graphs in real-time. The graphical user interface (GUI) is developed using Tkinter, providing an intuitive way to interact with the simulation and monitor the robot's behavior.

MuJoCo Tkinter GUI - Mobile Robot Mapping

WyNDA : Data-Driven Discovery for Mathematical Model¶

I reimplemented the code from the paper “WyNDA: A Method to Discover Mathematical Models of Dynamical Systems from Data” and developed both Python and C++ implementations to extract mathematical models of dynamical systems from observational data.

Quadcopter Trajectory Planning and Tracking 3D Simulation¶

I implemented minimum snap trajectory optimization for both position and orientation, as well as geometric control on \(SE(3)\) for nonlinear control in non-flat spaces. Additionally, I built a 3D simulation environment using MuJoCo to test and validate these control strategies.