Max Robotics — Global Service Robot Directory + US Certification

About

Svetoslav Kolev is a roboticist whose research bridges model-based control, state estimation, and reinforcement learning to enable autonomous, contact-rich manipulation. His most influential work, an integrated system for real-time model predictive control of humanoid robots (126 citations), demonstrates how high-level human guidance can be combined with intelligent, autonomous control to generate diverse behaviors. Kolev has made foundational contributions to physically consistent state estimation and system identification for robotic systems under contact (41 citations), developing frameworks that fuse sensor data with physical priors to ensure accurate and robust performance during manipulation tasks. His practical innovations include a low-cost, 3D-printed fingertip force sensor (18 citations), designed for easy modification and maintenance, enabling accessible tactile sensing for robotic hands. More recently, Kolev has explored reinforcement learning for non-prehensile manipulation, successfully transferring policies from simulation to physical systems (9 citations). His work is notable for its emphasis on physical consistency—ensuring that estimation, control, and learning all respect the laws of mechanics—making his contributions valuable for researchers developing robots that must interact reliably with the real world.