no code implementations • 21 Sep 2023 • Panagiotis Petropoulakis, Ludwig Gräf, Josip Josifovski, Mohammadhossein Malmir, Alois Knoll
The results show that RL agents using numerical states can perform on par with non-learning baselines.
no code implementations • 15 Jun 2023 • Mohammadhossein Malmir, Josip Josifovski, Noah Klarmann, Alois Knoll
We introduce a disturbance-augmented Markov decision process in delayed settings as a novel representation to incorporate disturbance estimation in training on-policy reinforcement learning algorithms.
no code implementations • 13 Jun 2022 • Josip Josifovski, Mohammadhossein Malmir, Noah Klarmann, Bare Luka Žagar, Nicolás Navarro-Guerrero, Alois Knoll
Fully randomized simulations and fine-tuning show differentiated results and translate better to the real robot than the other approaches tested.
no code implementations • 22 Mar 2022 • Nicolás Navarro-Guerrero, Sibel Toprak, Josip Josifovski, Lorenzo Jamone
The object perception capabilities of humans are impressive, and this becomes even more evident when trying to develop solutions with a similar proficiency in autonomous robots.
no code implementations • 11 Mar 2022 • Marco Oliva, Soubarna Banik, Josip Josifovski, Alois Knoll
We derive a graph representation that models the physical structure of the manipulator and combines the robot's internal state with a low-dimensional description of the visual scene generated by an image encoding network.