Enabling Visual Action Planning for Object Manipulation Through Latent Space Roadmap

Object manipulation and motion perception: evidence of an influence of action planning on visual processing.

In 3 experiments, the authors investigated the bidirectional coupling of perception and action in the context of object manipulations and motion perception. Participants prepared to grasp an X-shaped object along one of its 2 diagonals and to rotate it in a clockwise- or a counterclockwise direction. Action execution had to be delayed until the appearance of a visual go signal, which induced an...

POMDP based manipulation planning in object composition space

Service robots in domestic environments need the ability to manipulate objects without good prior models in order to cope with the variability of such environments. When multiple measurements can be acquired around an isolated object, standard approaches work satisfactorily as the generated 3-D models can often be used for successful manipulation. However, in cluttered scenes with multiple unkn...

A randomized roadmap method for path and manipulation planning

This paper presents a new randomized roadmap method for motion planning for many dof robots that can be used to obtain high quality roadmaps even when C-space is crowded. The main novelty in our approach is that roadmap candidate points are chosen on C-obstacle surfaces. As a consequence, the roadmap is likely to contain difficult paths, such as those traversing long, narrow passages in C-space...

Object Manipulation Planning for Dextrous Robot Systems

Action-Space Partitioning for Planning

For autonomous artificial decision-makers to solve realistic tasks, they need to deal with searching through large state and action spaces under time pressure. We study the problem of planning in such domains and show how structured representations of the environment’s dynamics can help partition the action space into a set of equivalence classes at run time. The partitioned action space is the...