Sascha Lange

About

Sascha Lange is a pioneering researcher in the intersection of reinforcement learning and autonomous robotics, best known for his groundbreaking work on teaching robots to play soccer. His research focuses on developing machine learning algorithms that enable robots to learn complex behaviors directly from reward and punishment signals, rather than through explicit programming. Lange's most influential contribution is his 2009 paper "Reinforcement learning for robot soccer," which has accumulated 266 citations and established a foundational framework for applying RL to real-world robotic tasks. He also developed an efficient approach for robot self-localization, detailed in his 101-citation paper "Calculating the Perfect Match," which significantly improved accuracy in autonomous navigation. Lange's work on the Brainstormers project demonstrated how cognitive concepts could be implemented in soccer-playing robots, while his real-time 3D ball recognition system using stereoscopic cameras advanced computer vision for dynamic environments. His research on learning to dribble through trial and error on physical robots showed that complex motor skills could be acquired without human intervention. Lange's contributions have been instrumental in bridging the gap between theoretical reinforcement learning and practical robotic applications, inspiring a generation of researchers in autonomous systems and robot learning.

Research Focus

Key Achievements

Top Papers

1. 1
   Reinforcement learning for robot soccer

   266 citations · 2009
2. 2
   Calculating the Perfect Match: An Efficient and Accurate Approach for Robot Self-localization

   101 citations · 2006
3. 3
   Real-time 3D Ball Recognition using Perspective and Catadioptric Cameras.

   27 citations · 2007
4. 4
   Modeling Moving Objects in a Dynamically Changing Robot Application

   18 citations · 2005
5. 5
   Making a Robot Learn to Play Soccer Using Reward and Punishment

   16 citations · 2007
6. 6
   Cognitive concepts in autonomous soccer playing robots

   12 citations · 2010
7. 7
   Learning to dribble on a real robot by success and failure

   12 citations · 2008
8. 8
   Die Brainstormers: Entwurfsprinzipien lernfähiger autonomer Roboter

   8 citations · 2006
9. 9
   Evolution of Computer Vision Subsystems in Robot Navigation and Image Classification Tasks

   6 citations · 2005
10. 10
    Appearance-Based Robot Discrimination Using Eigenimages

    5 citations · 2007

Key Collaborators

Contact & Links