Max Robotics — Global Service Robot Directory + US Certification

About

Bexultan Rakhim is a robotics researcher whose work centers on advancing the control and sensing of variable stiffness actuated (VSA) and variable impedance actuated (VIA) robots—a paradigm that promises safer, more energy-efficient physical human-robot interaction. His most influential contribution, "Successive linearization based model predictive control of variable stiffness actuated robots" (2017, 26 citations), pioneered the use of nonlinear model predictive control (NMPC) to handle the constrained, nonlinear dynamics of these systems, enabling high performance while preserving inherent safety. Rakhim further extended this line of inquiry by developing a deep learning-based approximate optimal control approach for a reaction-wheel-actuated spherical inverted pendulum (2020, 9 citations), showcasing how neural networks can overcome the low motion bandwidth typical of impedance-based actuators. To address the practical challenge of sensor reduction in VSA robots—which require multiple actuators per joint—he proposed a moving horizon estimation framework (2019, 7 citations) that minimizes sensor count without sacrificing state estimation accuracy. His work on optimal sensor placement for VIA robots (2019, 2 citations) provides a systematic method for reducing hardware complexity while maintaining full state observability. Collectively, Rakhim’s research bridges advanced control theory and practical sensor design, offering scalable solutions for next-generation, human-safe robotic systems.