Wouter Caarls

About

Wouter Caarls is a robotics and control systems researcher whose work sits at the intersection of machine learning, computer vision, and autonomous robotics. He is best known for his contributions to extremum seeking control (ESC) for robotic applications, having authored a widely referenced comparison of ESC algorithms (33 citations) that serves as a practical guide for engineers navigating viewpoint optimization and object grasping challenges. His influential work on active vision via ESC (38 citations) introduced a model-free strategy enabling robots to autonomously optimize sensor viewpoints in unstructured environments — a significant step forward for real-world object recognition and manipulation. Caarls has made notable strides in reinforcement learning for physical systems, addressing the critical challenge of safe exploration to prevent mechanical damage — a concern reflected across multiple publications totaling dozens of citations. His model-plant mismatch compensation using reinforcement learning (37 citations) elegantly bridges learning-based and model-based control paradigms. He has also contributed to agricultural robotics, developing autonomous mobile systems for crop inspection (19 citations), demonstrating a commitment to applied, socially relevant robotics. With expertise spanning embedded image processing, temporal difference learning, and optimal control benchmarking, Caarls represents a versatile and practically minded voice in modern robotics research.

Research Focus

Key Achievements

Top Papers

1. 1
   Active Vision via Extremum Seeking for Robots in Unstructured Environments: Applications in Object Recognition and Manipulation

   38 citations · 2018
2. 2
   Model-Plant Mismatch Compensation Using Reinforcement Learning

   37 citations · 2018
3. 3
   Comparison of extremum seeking control algorithms for robotic applications

   33 citations · 2012
4. 4
   Benchmarking model-free and model-based optimal control

   28 citations · 2017
5. 5
   Design and Development of an Autonomous Mobile Robot for Inspection of Soy and Cotton Crops

   19 citations · 2019
6. 6
   Parallel Online Temporal Difference Learning for Motor Control

   16 citations · 2015
7. 7
   Application Driven Design Of Embedded Real-Time Image Processors

   8 citations · 2003
8. 8
   On the Intelligent Control Design of an Agricultural Mobile Robot for Cotton Crop Monitoring

   7 citations · 2019
9. 9
   Learning while preventing mechanical failure due to random motions

   7 citations · 2013
10. 10
    Evaluation of physical damage associated with action selection strategies in reinforcement learning

    7 citations · 2017

Key Collaborators

Contact & Links