Pierre Desreumaux
Papers
1
Total Citations
31
H-Index
1
About
Pierre Desreumaux is a leading researcher in robot adaptation and learning, with a focus on repertoire-based online adaptation—a data-efficient approach that equips robots to respond to unforeseen changes, such as physical damage. His key contributions center on developing methods where a diverse set of policies is first learned in simulation, then intelligently selected in real-time to handle novel situations. His most-cited paper, "Adaptive Prior Selection for Repertoire-Based Online Adaptation in Robotics" (2020, 31 citations), demonstrates how prior knowledge can be leveraged to accelerate adaptation, reducing the need for extensive real-world trials. This work has influenced the growing field of lifelong robot learning, offering a practical bridge between simulation and deployment. Desreumaux’s research is notable for its emphasis on robustness and efficiency, making it highly relevant for autonomous systems operating in unpredictable environments. His achievements include advancing the theoretical foundations of repertoire-based learning and inspiring further work on resilient robotic control.
Research Focus
Key Achievements
Top Papers
- 1Adaptive Prior Selection for Repertoire-Based Online Adaptation in Robotics31 citations · 2020