Objectives
- Theory of Grasp Modelling - We will study the
requirements and effects of the agent's embodiment on the
situatedness, awareness, task and environment understanding and thus
provide the means for adaptation and self-reasoning. Learning from
human grasping examples will be used to predict and derive which
part(s) of the space of all possible grasps can be realised with a
specific artificial hand configuration.
- Self- and Context-Awareness - We will investigate how
an agent benefits from using tutor based and autonomous exploration
together with physical modelling of the world to learn more about
the possibilities and constraints offered by its embodiment. This
will lead to ontology and new representations in the area of systems
integration where we consider sensors, embodiment and the
environment including other agents allowing for faster knowledge
inference based on existing and new competences.
- Curiosity and Surprise Driven Behaviour - We will show
how the detection of an unexpected event or action is exploited to
efficiently add new values, categories or dimensions to the grasping
ontology while at the same time exploiting surprise to bootstrap the
learning process. Expectations are derived from the prediction of
agent behaviour using the experiences gained from self-awareness or
introspection.
- Inferring new Grasping Strategies - We will use the
ontology and acquired general knowledge to generate expectations for
grasping and manipulation tasks as means of correction between the
predicted and the actual state. This will allow adaptation to new
objects and situations without the need for extensive
re-programming.
- Exploitation to Future Prosthesis, Industrial and Service
Markets - Finally, we set out to exploit the theoretical
findings by investigating the grasp mapping to different artificial
hands. The objective is to learn how kinematical design and the
number of DOFs influence dexterity and how to optimize the graspable
sub-set of all possible grasps while minimising DOFs. This shall
ultimately lead to efficient hand designs that are suitable for low
cost production to aid as many handicapped persons as possible. It
shall also influence future manufacturing of robotic hands useful to
industrial as well as service-oriented enterprises.