Max Robotics — Global Service Robot Directory + US Certification

About

Kendall Lowrey’s research bridges the critical gap between simulation and reality in humanoid robotics, with a focus on dynamic motion planning, model predictive control, and contact-rich state estimation. His most influential work, “Ensemble-CIO” (145 citations), demonstrated that full-body dynamic motion plans could transfer to physical humanoid robots like the Darwin-OP, overcoming a longstanding barrier in the field. In “An integrated system for real-time model predictive control of humanoid robots” (126 citations), Lowrey developed a framework that enables autonomous task execution from high-level user guidance, significantly advancing real-time control. His contributions to state estimation include a physically-consistent sensor fusion method and a modified Unscented Kalman Filter for whole-body multi-contact dynamics, both critical for robust, contact-rich behaviors. Lowrey also explored reinforcement learning for non-prehensile manipulation, showing how model-based methods can transfer policies from simulation to physical systems. More recently, he introduced Lyceum, a high-performance ecosystem for robot learning built on Julia and MuJoCo, designed to accelerate research with scalable, efficient tools. With over 300 total citations, Lowrey’s work has shaped practical, real-world humanoid control and continues to influence the next generation of autonomous robotic systems.