Papers
14
Total Citations
355
H-Index
8
About
Romeo Orsolino is a robotics researcher specializing in legged locomotion, motion planning, and the integration of model-based and data-driven control strategies for quadrupedal and bipedal robots. His work sits at the intersection of optimal control, reinforcement learning, and biomechanically inspired robot design, with a sustained focus on enabling robots to navigate complex, uneven terrain reliably and efficiently. Among his most influential contributions is RLOC (2022, 122 citations), a hybrid framework that combines reinforcement learning with optimal control to achieve terrain-aware quadrupedal locomotion using both proprioceptive and exteroceptive sensing. This work exemplifies his broader research philosophy: leveraging the complementary strengths of data-driven and model-based approaches. His pioneering development of the "Feasible Region" concept extends classical stability analysis by explicitly accounting for actuator torque limits, a theoretically elegant and practically critical advancement that has shaped how the community reasons about legged robot balance. He has further contributed robust methods for online footstep optimization, real-time trajectory adaptation, and payload identification, collectively amassing over 300 citations. His work consistently bridges theoretical rigor with hardware implementation, making meaningful strides toward deploying agile, robust legged robots in real-world environments.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3Online payload identification for quadruped robots38 citations · 2017
- 4Feasible Region: An Actuation-Aware Extension of the Support Region37 citations · 2020
- 5
- 6
- 7An Efficient Paradigm for Feasibility Guarantees in Legged Locomotion10 citations · 2020
- 8
- 9
- 10