Expert in Deep Reinforcement Learning for Robotic Snap-Fit Assembly in Isaaclab

In this Project one have to train a robot (like franka) with Deep Reinforcement Learning to do a snap-fit assembly of a Snap-fit Buckle in isaacLab.

(I) Development of a simulation of the assembly unit – with particular focus on the elastic properties of the components (e.g., Via FEM, simplified mechanical models, or kinematic chains; depending on the chosen simulation tool)
(ii) Development of a robust reinforcement learning policy that can reliably perform the joining (insertion of male part into female part) process under varying conditions (e.g., Different starting positions or different configurations in gripping and positioning components)

The part (i) is already done and I have a simulation in isaaclab with the CAD files inside and the deformability is already handled. I also have defined RL reward function Logics in it to perform the task."