trical

Design and implementation of control systems for a Triangle Balance Robot in the dc&e club. The goal is to first start with conventional full-state feedback control and then experiment with different control approaches, such as RL and MPC, while making the most out of control and estimation theory in the process.

RESULTS

Imitation Learning for RWP using dataset from NMPC generated trajectories

MUJOCO Simulation------------------------------- Sample NMPC Trajectory

Branches:
Main - For Notes/Documentation and README
Sim - Simulation Results(Mujoco/ Numerical Simulation(RK4), NMPC trajectory generation and Imitation Learning)
Hardware - Hardware Implementation of Control Laws on STM32F103C8T6

Team

1. Himanshu Paudel

References

1. Fantoni, Lozano - Non-linear Control for Underactuated Mechanical Systems (2002)
2. Block, Åström, Spong - The Reaction Wheel Pendulum (2007)