Documentdetail
ID kaart
oai:arXiv.org:2409.07091
Onderwerp
Computer Science - RoboticsAuteur
Baert, Mattijs Leroux, Sam Simoens, PieterCategorie
Computer Science
Jaar
2024
vermelding datum
18-09-2024
Trefwoorden
tasks learningBeschrijving
Specifying tasks for robotic systems traditionally requires coding expertise, deep domain knowledge, and significant time investment.
While learning from demonstration offers a promising alternative, existing methods often struggle with tasks of longer horizons.
To address this limitation, we introduce a computationally efficient approach for learning probabilistic deterministic finite automata (PDFA) that capture task structures and expert preferences directly from demonstrations.
Our approach infers sub-goals and their temporal dependencies, producing an interpretable task specification that domain experts can easily understand and adjust.
We validate our method through experiments involving object manipulation tasks, showcasing how our method enables a robot arm to effectively replicate diverse expert strategies while adapting to changing conditions.
Baert, Mattijs,Leroux, Sam,Simoens, Pieter, 2024, Learning Task Specifications from Demonstrations as Probabilistic Automata
Evaluating Image Hallucination in Text-to-Image Generation with Question-Answering
evaluation computer dataset images tti generation i-halla
Role of Repetitive Transcranial Magnetic Stimulation in Treatment of Fibromyalgia: A Randomized Controlled Trial
significant following stimulation fms treatment
