Brent Komer

About

Brent Komer is a researcher at the intersection of robotics and machine learning, with a primary focus on reinforcement learning (RL) and biologically inspired control systems. His most cited work, "Setting up a Reinforcement Learning Task with a Real-World Robot" (2018, 24 citations), addresses a critical bottleneck in the field: the notorious difficulty of deploying RL algorithms on physical hardware. By providing a practical framework for bridging the simulation-to-reality gap, Komer has helped make real-world robotic learning more accessible and reliable for the research community. His earlier thesis, "Biologically Inspired Adaptive Control of Quadcopter Flight" (2015), explores how principles from natural neural systems can be applied to stabilize and control drones, demonstrating a novel approach to adaptive flight. While his citation counts reflect a focused, early-career impact, Komer’s contributions are notable for their emphasis on practical implementation—turning theoretical RL into tangible robotic behaviors. His work serves as a valuable resource for students and engineers seeking to move beyond simulated environments and into the messy, rewarding world of real-world robot learning.

Research Focus

Key Achievements

Top Papers

1. 1
   Setting up a Reinforcement Learning Task with a Real-World Robot

   24 citations · 2018
2. 2
   Biologically Inspired Adaptive Control of Quadcopter Flight

   2 citations · 2015

Key Collaborators

Contact & Links