Papers
30
Total Citations
1,607
H-Index
16
About
Danny Driess is a pioneering robotics and machine learning researcher whose work sits at the intersection of embodied AI, foundation models, and robot learning. Best known for his landmark contributions to multimodal language models and robotic control, Driess has rapidly established himself as one of the most influential figures in modern robotics research. His most celebrated work, PaLM-E (2023, 350 citations), introduced the concept of embodied language models that directly integrate real-world sensor data into large language models, fundamentally advancing how AI systems can reason about and interact with physical environments. This was complemented by his pivotal role in RT-2 (2023, 267 citations), which demonstrated that vision-language models trained on internet-scale data could transfer rich semantic knowledge directly into robotic control. His contributions to the Open X-Embodiment initiative and the π₀ vision-language-action flow model further cemented his leadership in generalizable robot learning. Earlier in his career, Driess explored tactile sensing, physics-based manipulation planning, and task-and-motion planning, showcasing remarkable breadth. His survey on foundation models in robotics (2024, 163 citations) has become an essential reference for the field. With over 1,300 citations across a decade of work, Driess represents a new generation of researchers reshaping how robots understand and act in the world.
Research Focus
Key Achievements
Top Papers
- 1PaLM-E: An Embodied Multimodal Language Model350 citations · 2023
- 2RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control267 citations · 2023
- 3Foundation models in robotics: Applications, challenges, and the future163 citations · 2024
- 4π₀: A Vision-Language-Action Flow Model for General Robot Control127 citations · 2025
- 5
- 6Open X-Embodiment: Robotic Learning Datasets and RT-X Models101 citations · 2023
- 7Long-Horizon Multi-Robot Rearrangement Planning for Construction Assembly86 citations · 2022
- 8Active learning with query paths for tactile object shape exploration52 citations · 2017
- 9
- 10