Elaborating on learned demonstrations with temporal logic specifications
C. Innes, S. Ramamoorthy - RSS-20
Surfing on an uncertain edge: Precision cutting of soft tissue using torque-based medium classification
A. Straizys, M. Burke, S. Ramamoorthy - ICRA-20
Disentangled Relational Representations for Explaining and Learning from Demonstration
Y. Hristov, D. Angelov, A.Lascarides, M. Burke, S. Ramamoorthy - CORL-19
From explanation to synthesis: Compositional program induction for learning from demonstration
M. Burke, S. Penkov, S. Ramamoorthy - RSS-19
Robust Autonomy and Decisions Group
We investigate learning, adaptation and control mechanisms to enable autonomous robots to cope with the uncertain and the unknown. Our research focus areas include:
- Dexterous manipulation skills: understanding what humans/animals do, in order to achieve similar robotic skills
- Human-robot interaction, including paradigms such as machine teaching
- Active sensing and incremental, compositional model learning
- Physics-informed machine learning, including geometric/topological representations and multi-scale modelling
- Safety and trustworthiness of autonomous systems
We develop techniques that are relevant to a broad range of applications, ranging from advanced manufacturing to mobility and field robotics. An application domain of particular current interest is healthcare technologies, including biomedical devices and biophysical discovery platforms.