Andreas Persson
Papers
7
Total Citations
32
H-Index
4
About
Andreas Persson’s research lies at the critical intersection of robotics, artificial intelligence, and human-robot interaction, with a central focus on bridging the symbolic-sub-symbolic gap. His most influential work, “Symbolic Learning and Reasoning With Noisy Data for Probabilistic Anchoring” (2020, 11 citations), proposes a novel framework that enables robotic agents to learn from noisy sensor data while simultaneously reasoning about objects and communicating with humans at a symbolic level. This contribution addresses a fundamental challenge in AI: grounding abstract symbols in perceptual reality. Persson’s earlier work on perceptual anchoring, including his 2017 study on learning actions to improve object anchoring (5 citations), demonstrates how robots can track and maintain object entities over time using artificial neural networks. His research on fluent human-robot dialogues about grounded objects in home environments (2014, 4 citations) and ontology-based symbol grounding systems (2014, 3 citations) has practical implications for assistive robotics. Additionally, his work on fast matching of binary descriptors for large-scale robot vision (2016, 4 citations) addresses computational efficiency challenges in real-world applications. Through these contributions, Persson has advanced the development of robots that can perceive, reason, and communicate more naturally with humans.
Research Focus
Key Achievements
Top Papers
- 1Symbolic Learning and Reasoning With Noisy Data for Probabilistic Anchoring11 citations · 2020
- 2Learning Actions to Improve the Perceptual Anchoring of Objects5 citations · 2017
- 3Fluent Human–Robot Dialogues About Grounded Objects in Home Environments4 citations · 2014
- 4
- 5An Ontology-based Symbol Grounding System for Human-Robot Interaction3 citations · 2014
- 6
- 73D Scan-based Navigation using Multi-Level Surface Maps2 citations · 2009