SURF 2011: Implementation of Control Protocols for Camera Networks

From Murray Wiki
Jump to navigationJump to search

2011 SURF Project Description

  • Mentor: Richard Murray
  • Co-Mentor: Necmiye Ozay

In a recent preprint [3], we proposed a method for distributed control protocol synthesis for smart camera networks where the goal is to guarantee that certain linear temporal logic (LTL) specifications related to a given surveillance task are met. We consider a hierarchical visual surveillance system that supplements a stationary camera network used for tracking with additional active cameras (either pan-tilt-zoom cameras or mobile robots with vision sensors). This is a hands-on project where the goal is to build a setup to implement such embedded controllers on a real camera network. Possible steps of the project are:

  • To come up with a couple of simple surveillance/service scenarios that could potentially be implemented in a lab equipped with a tracking system.
  • To abstract finite state automata that describe the dynamic motion model of the cameras/robots (e.g. using the discretization method proposed in [2]).
  • To synthesize a control protocol that assigns targets to cameras or jobs to robots and that generates a high level motion plan using the method described in [3] with TuLiP ([1]). A tracking system is being installed in the robotics lab at CDS that can be used to obtain target tracks. The track information should be communicated to the controllers that implement the control protocol.
  • To implement the controllers and test their performance.

Required Skills: To make decent progress over the summer, the student should have experience with Python, or otherwise has software engineering knowledge that suggests learning other languages, environments, etc. will not be difficult. For using the tracking system familiarity or willingness to learn ROS (http://www.ros.org/wiki/ ) is crucial. At least elementary experience of working in Unix-like environments would be very helpful.

References

  1. T. Wongpiromsarn, U. Topcu, N. Ozay, H. Xu, and R. M. Murray, TuLiP: a software toolbox for receding horizon temporal logic planning, International Conference on Hybrid Systems: Computation and Control, 2011 (to appear). code available at http://www.cds.caltech.edu/tulip/
  2. T. Wongpiromsarn, U. Topcu, and R. M. Murray, Receding Horizon Temporal Logic Planning, IEEE Transactions on Automatic Control, 2010 (submitted).
  3. N. Ozay, U. Topcu, T. Wongpiromsarn, and R. M. Murray, Distributed Synthesis of Control Protocols for Smart Camera Networks, (submitted). available from http://www.cds.caltech.edu/~utopcu/images//b/b6/OTWM-iccps11.pdf