Maciej Krupka
Papers
1
Total Citations
4
H-Index
1
About
Maciej Krupka is at the forefront of legged robotics and deep reinforcement learning, with a particular focus on creating versatile locomotion controllers that can operate across diverse robotic platforms. His most cited work, "One Policy to Run Them All: an End-to-end Learning Approach to Multi-Embodiment Locomotion" (2024), introduces a groundbreaking framework that enables a single learned policy to control quadruped, humanoid, and hexapod robots—a significant departure from the traditional approach of designing separate controllers for each morphology. This research addresses a critical gap in the field, demonstrating that end-to-end learning can produce robust, transferable locomotion skills without platform-specific engineering. With 4 citations already in its publication year, the work signals strong early impact and has the potential to reshape how researchers approach multi-embodiment robotics. Krupka’s contributions advance the goal of generalist robot controllers, reducing the need for manual tuning and accelerating the deployment of legged robots in real-world environments. His work is essential reading for anyone interested in scalable, adaptive locomotion systems.
Research Focus
Key Achievements
Top Papers
- 1