Napat Karnchanachari

About

Napat Karnchanachari is a robotics researcher whose work sits at the intersection of reinforcement learning, model predictive control, and autonomous racing. His major contributions center on making complex control systems practical for real-world deployment. In his highly cited work "Practical Reinforcement Learning For MPC" (2020, 14 citations), Karnchanachari demonstrated how to combine reinforcement learning with model predictive control to enable robots to learn from sparse objectives in under an hour—a breakthrough that dramatically reduces the expert tuning typically required for MPC cost functions. This work has been recognized for its potential to democratize advanced control techniques. As a key contributor to the AMZ Driverless team, Karnchanachari co-authored "The full autonomous racing system" (2020, 11 citations), which presented the complete software stack for an autonomous racecar capable of navigating unknown tracks at high speeds. This achievement showcases his ability to translate theoretical advances into robust, real-time systems. With a total of over 30 citations across his most prominent papers, Karnchanachari's research continues to influence how robots learn and control themselves in dynamic, unstructured environments.

Research Focus

Key Achievements

Top Papers

1. 1
   Practical Reinforcement Learning For MPC: Learning from sparse objectives in under an hour on a real robot

   14 citations · 2020
2. 2
   AMZ Driverless: The full autonomous racing system

   11 citations · 2020
3. 3
   Practical Reinforcement Learning For MPC: Learning from sparse\n objectives in under an hour on a real robot

   7 citations · 2020

Key Collaborators

Contact & Links