David Surovik
Papers
8
Total Citations
104
H-Index
6
About
David Surovik is a leading researcher in the field of robotic locomotion, with a primary focus on the control and planning of tensegrity robots—lightweight, compliant structures composed of rods and cables that offer exceptional ruggedness and adaptability for traversing challenging terrain. His major contributions include pioneering the use of symmetry-reduced reinforcement learning to tame the high-dimensional dynamics of tensegrity systems, as demonstrated in his most-cited work (38 citations), and developing kinodynamic planning frameworks that enable effective gait primitives for spherical tensegrity locomotion. Surovik has also advanced legged robotics, notably through reliable trajectory optimization for dynamic quadrupeds, where he introduced analytical costs and learned initializations to improve planning over longer horizons. His work on efficient model identification using Bayesian optimization and off-the-shelf physics engines has practical implications for real-world deployment. With over 90 total citations across his key publications, Surovik’s research bridges data-driven control and classical planning, offering scalable solutions for robots operating in unstructured environments. His achievements include advancing the state of the art in perceptive trajectory optimization and adaptive locomotion on rough terrain, making him a notable figure in the intersection of soft robotics and dynamic control.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4Efficient Model Identification for Tensegrity Locomotion10 citations · 2018
- 5
- 6
- 7
- 8