Alexander Schperberg
University of California, Los Angeles, Mitsubishi Electric (United States), UCLA Medical Center
Papers
6
Total Citations
95
H-Index
5
About
Alexander Schperberg is a leading roboticist whose work pushes the boundaries of legged locomotion and dexterous manipulation, with a particular focus on enabling robots to conquer extreme, contact-rich environments. His most celebrated contribution is the **SCALER** robot, a tough, versatile quadruped introduced in 2022 (34 citations) that stands as one of the first high-degree-of-freedom machines capable of free-climbing bouldering walls, overhangs, and ceilings under Earth’s gravity—a monumental leap for field robotics. Schperberg’s research uniquely bridges motion planning and control, as demonstrated by his framework for simultaneously solving locomotion and multi-finger grasping (26 citations), allowing robots to climb with unprecedented autonomy. He is also a pioneer in adaptive control, developing auto-tuning algorithms using Unscented Kalman Filters (UKF) to robustly calibrate controllers and trajectory planners online (16 citations). His more recent work, **OptiState** (2024, 9 citations), fuses Kalman filtering with transformer-based vision for state estimation in highly dynamic motion. Through innovations in force control and real-to-sim error prediction, Schperberg is systematically closing the gap between simulation and reality, making his robots not only more capable but also more reliable in the wild.
Research Focus
Key Achievements
Top Papers
- 1SCALER: A Tough Versatile Quadruped Free-Climber Robot34 citations · 2022
- 2
- 3Auto-Tuning of Controller and Online Trajectory Planner for Legged Robots16 citations · 2022
- 4
- 5
- 6