Reinforcement Learning for Robust Athletic Intel-ligence: Lessons Learned From the Second AI Olympics With RealAIGym Competition
Artikel i vetenskaplig tidskrift, 2025
In robotics, many different approaches ranging from classical planning over optimal control to reinforcement learning (RL) are developed and borrowed from other fields to achieve reliable control in diverse tasks. To get a clear understanding of their individual strengths and weaknesses and their applicability in real-world robotic scenarios, it is important to benchmark and compare their performances not only in a simulation but also on real hardware. The second AI Olympics with RealAIGym competition was held at the 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2024) to contribute to this cause and evaluate different controllers according to their ability to solve a dynamic control problem on an underactuated double-pendulum system (Figure 1) with chaotic dynamics. This article describes the four different RL methods submitted by the participating teams; presents their performance in the swing-up task on a real double pendulum, measured against various criteria; and discusses their transferability from simulation to real (sim-to-real) hardware and their robustness to external disturbances.
Entropy
Perturbation methods
Hardware
Torque
Robustness
Automation
Monte Carlo methods
Convolutional neural networks
Robots
Costs
Författare
Felix Wiebe
DFKI
Niccolo Turcato
Università di Padova
Alberto Dalla Libera
Università di Padova
Jean Seong Bjorn Choe
Korea University
Bumkyu Choi
Korea University
Tim Lukas Faust
Technische Universität Darmstadt
Habib Maraqten
Technische Universität Darmstadt
Erfan Aghadavoodi
DFKI
Marco Cali
Università di Padova
Alberto Sinigaglia
Università di Padova
Giulio Giacomuzzo
Università di Padova
Ruggero Carli
Università di Padova
Diego Romeres
Mitsubishi Electric Research Laboratories
Jong-kook Kim
Korea University
Gian Antonio Susto
Università di Padova
Shubham Vyas
DFKI
Dennis Mronga
DFKI
Boris Belousov
DFKI
Jan Peters
DFKI
Frank Kirchner
DFKI
Shivesh Kumar
Chalmers, Mekanik och maritima vetenskaper, Dynamik
IEEE Robotics and Automation Magazine
1070-9932 (ISSN) 1558223x (eISSN)
Vol. In PressÄmneskategorier (SSIF 2025)
Robotik och automation
Reglerteknik
DOI
10.1109/MRA.2025.3631571