Design of autonomous systems that can deal with the variations in their surrounding world by exploiting models. Develop world modelling and control methods at all task-relevant levels, from multi-robot collaboration to low-level motion control of drive wh
Group van de Molengraft focusses on the design of autonomous systems that are robust-by-design against the variations that can occur during task execution. The research aims at developing methods for world modelling that exploit the available prior knowledge on task, environment and robot. Such world models allow for robust decision making and control during task execution. A specific aspect of our approach is “laziness”: use feedforward control when possible, use feedback control when needed. Open-loop guarded motion is used where world-model based monitors guard the important conditions and trigger transition events.
Read moreMeet some of our Researchers
Recent Publications
Our most recent peer reviewed publications
-
A Bayesian optimization framework for the automatic tuning of MPC-based shared controllers
(2024) -
Total Turning and Motion Range Prediction for Safe Unicycle Control
(2024) -
Semantic learning from keyframe demonstration using object attribute constraints
Frontiers in Robotics and AI (2024) -
Prediction Horizon Requirements for Automated Driving
(2024) -
A Large Area Soft Robotic Haptic Display
(2024)
Contact
-
Visiting address
P.O. Box 5135600 MB EindhovenNetherlands -
Teamleadm.j.g.v.d.molengraft@ tue.nl
-
Secretarycst@ tue.nl