Stephan Wiess (Full Professor - Alpen-Adria Universität Klagenfurt) , Control of Networked Systems :
Research interests:
Fields:
Projects:
Several international partners: universities, research centres..
Mars helicopter scout: project on the development of a self-deployable drone for exploration or mars (am I right with this?)
Research challenges : Self contained airborne system - mix of control, perception, reckoning aspects
Self Calibration
The approach is to estimate state through different sensors exploiting the control algorithm
The approach for state estimation is sensor fusion and redundancy : need for automatic self-calibration.
Observability Awareness
Example : calibrating IMUs using camera optic flow and KF.
Idea : add sensors around a core sensor (IMU)
Some states are observable, others are not : depending on sensor type and dynamics
Nonlinear observability depends on current state (it is thus local).
Quality of observability can be derived by maths : Taylor expansions
Local observability depends on the eigenvalues of the Empirical Local (because NL) Observability Gramian : the goal is to minimise the minimal eigenvalue.
So the idea (see above) is to use inputs to sollecitate modes of the system and make it observable.
stephan.weiss@aau.at
Questions : 
How do you find a trade off between observability and  trajectory quality?
Hardware, real time aspects?

26.10.17 - Invited people to join the document

Welcome people :)

27.10.17 - Minutes from the lab meeting 

NCCR rehearsals : planning, timescale, hardware and projects involved
Carine :
meeting with designers of fly jacket - some improvements are proposed
Yusuke :