Konrad Rawlik

About

Konrad Rawlik’s research lies at the intersection of stochastic optimal control, robotics, and motor control, with a particular focus on temporal and stiffness optimization in movement. His major contributions include pioneering approximate inference frameworks that enable robots to autonomously optimize movement duration, intermediate goals, and stiffness parameters without requiring manual specification. In his most cited work, “An Approximate Inference Approach to Temporal Optimization in Optimal Control” (2010, 32 citations), Rawlik introduced algorithms that treat temporal parameters as latent variables, allowing robots to efficiently plan periodic and goal-directed motions. His follow-up work on stiffness and temporal optimization in periodic movements (2011, 26 citations) demonstrated how exploiting passive dynamics can dramatically reduce actuation costs, a key insight for energy-efficient robotic locomotion. Rawlik’s doctoral thesis (2013, 14 citations) further formalized the deep connections between stochastic optimal control and probabilistic inference, offering a unified perspective that has influenced both robotics and computational neuroscience. Though his later work (2017) has fewer citations, his early contributions remain foundational for researchers seeking to embed optimization into real-time robotic control.

Research Focus

Key Achievements

Top Papers

1. 1
   An Approximate Inference Approach to Temporal Optimization in Optimal Control

   32 citations · 2010
2. 2
   Stiffness and temporal optimization in periodic movements: An optimal control approach

   26 citations · 2011
3. 3
   On probabilistic inference approaches to stochastic optimal control

   14 citations · 2013
4. 4
   An Approximate Inference Approach to Temporal Optimization for Robotics

   2 citations · 2017

Key Collaborators

Contact & Links