You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Topics include function approximation, learning dynamics, using learned dynamics in control and planning, handling uncertainty in learned models, learning from demonstration, and model-based and model-free reinforcement learning.
University group project concerning the sensorless estimation of the contact forces between a needle mounted on the end-effector of a robot manipulator and a penetrated tissue, and subsequent prediction of layer ruptures using Recursive Least Squares algorithm.
Python scripts for Blender's scripting API to generate logic geometric puzzles utilizing the Phobos (https://github.com/dfki-ric/phobos) add-on for Blender.
Project concerning the design of trajectories approaching (boundary or internal) singularities, in order to compare the effects of the kinematic controllers based on pseudo-inversion and on Damped Least-Squares.
University group project concerning the comparison between the classical and the modified Denavit-Hartenberg convention in Newton-Euler algorithm, and the design of persistently exciting trajectories for the dynamic coefficients identification for the KUKA LWR IV+.