Egbert van der Wal - Object Grasping with the NAO
With autonomous robots becoming more and more common, the interest in applications of mobile robotics increases. Many applications of robotics include the grasping and manipulation of objects. As many robotic manipulators have several degrees of freedom, controlling these manipulators is not a trivial task. The actuator needs to be guided along a proper trajectory towards the object to grasp, avoiding collisions with other objects and the surface supporting the object. In this project, the problem of learning a proper trajectory towards an object to grasp, located in front of a humanoid robot, the Aldebaran NAO, is solved by using machine learning. Three algorithms were evaluated. Learning from demonstration using a neural network trained on a training set of recorded demonstrations was not capable of learning this task. Using Nearest Neighbor on the same training set yielded much better results in simulation but had more problems picking up objects on the real robot. A form of Reinforcement Learning (RL) tailored to continuous state and action spaces, the Continuous Actor Critic Learning Automaton (CACLA), proved to be an effective way to learn the problem by exploring the action space to obtain a good trajectory in a reasonable amount of time. This algorithm also proved to be robust against the additional complexity of operating on the real robot after being trained in simulation, bridging the reality gap.
Laatst gewijzigd: | 13 juni 2019 13:40 |
Meer nieuws
-
01 juli 2025
‘Geef zeehonden de ruimte’
De Waddenzee verandert voortdurend. De lokale dieren hebben adaptaties nodig om te overleven in een omgeving die is vormgegeven door de getijden. Door de zeehonden in het gebied te onderzoeken zijn PhD-onderzoekers Margarita Méndez-Aróstegui en...
-
30 juni 2025
David Lentink partner in internationaal onderzoeksproject naar diernavigatie
Prof. dr. David Lentink is partner in het NaviSense project dat van de Duitse overheid 54,7 miljoen euro toegekend heeft gekregen voor onderzoek naar de mechanismen die dieren gebruiken om te navigeren en hoe deze mechanismen technologie kunnen...
-
26 juni 2025
Prof. Adri Minnaard ontvangt RSC Chemistry Biology Interface Horizon Prize met Lipidomics team
Prof. dr. Adri Minnaard is met het ‘Lipidomics Team’ waar hij deel van uitmaakt, uitgeroepen tot winnaar van de Chemistry Biology Interface Horizon Prize van de Royal Society of Chemistry (RSC).