Even though grasping has been a central topic in robotics for decades, robots still have great difficulty to pick up arbitrary objects when they operate in open, unknown environments or under uncontrolled conditions. Lifting this limitation would make using robots fit for many repetitive chores: robots could check and restock supermarket aisles, tidy up households, dispatch mail orders at distribution centers, collect ripe fruits, or do ecological pest control by selectively removing bugs.
To make this vision a reality, robots have to learn to grasp as reliable as humans. We try to discover the tricks people unconsciously employ when they cannot rely on their perception, and transfer these insights to robots. One such strategy is to exploit the structure of the environment to guide hand and finger motion quickly and reliably during the grasp.
The lead effort for the grasping projects discussed below are funded by the Soft Manipulation (SOMA) Horizon 2020 EU project. Please visit the project website for more information.
How can we build robots that can exploit constraints to motion rather than avoid them? How do we control such robots? We explore these questions in our research project on soft hands.
Our approach to grasping is motivated by the fact that humans don't avoid contact with the environment but rather exploit it to generate haptic feedback complementing visual feedback. This exploitation of environmental constraints simplifies the grasping problem by converting a high-dimensional configuration search problem into successive local searches guided by these environmental constraints, such as surfaces or edges.
We are developing algorithms that model the environment as a collection of environmental constraints which can be used to generate reactive feedback plans that lead to robust and reliable grasps.
Please note: Once you watch the video, data will be transmitted to Youtube/Google. For more information, see Google Privacy.
Uncertainty is the major obstacle for robots manipulating objects in the real world. A robot can never perfectly know its position in the world, the position of objects, and the outcome of its actions. A particularly hard challenge is motion planning under uncertainty. How should the robot move, if the model of the world might be wrong or incomplete?
However, our approach reasons about uncertainty and contact. A robot can significantly reduce uncertainty if it uses contact sensing to establish controlled contact with the environment. Moreover, the robot's capabilities are increased if it anticipates contact events that can happen during the execution of the plan. Our goal is to develop algorithms that can plan under uncertainty while exploiting contact and reasoning about sensor events during planning for high dimensional motion problems.
Please note: Once you watch the video, data will be transmitted to Youtube/Google. For more information, see Google Privacy.
In classical motion planning the environment is considered to be static. But changing the environment can improve planning e.g. removing an obstacle to free the path to the goal. Part feeders, fixtures and conveyor belt systems are examples of industrial engineered environments. Safety rails, coin slots and traffic signs are examples of our every day life.
We are developing automated design algorithms that engineer the environment to increase grasp robustness. The focus of our research on environmental design lies on design strategies in the context of motion/grasp planning with contact.
Soft Manipulation (SoMa) - funded by the European Commission in the Horizon 2020 program, award number 645599, May 2015 - April 2019. Alexander von Humboldt professorship - awarded by the Alexander von Humboldt foundation and funded through the Ministry of Education and Research, BMBF,
July 2009 - June 2014