Real-Time Active Vision with Fault Tolerance - Pattern Recognition, 1996., Proceedings of the 13th International Conference on

The active vision paradigm couples perception and action at several different levels. The effective use of active vision in complex robotic tasks requires that these levels operate both independently and cooperatively, reliably and in real-time. In this paper; we present a system for real-time active vision with fault tolerance. The system provides a vocabulary of active vision routines along with the means for composing the routines into continuously running perception-action processes. A novel architecture which enables the integration of the perceptive capabilities of real-time active vision with the active capabilities of other robotic devices is presented. We then enhance the architecture with a unified approach to fault tolerance and present results from experiments and simulations. specified timing constraints. Missing a deadline in a such a system can lead to high-cost consequences, for example when autonomous robots handle radio-active materials or when human life is involved. Therefore, services offered must be available and provide guarantees despite the occurrence of faults vault tolerance). In particular. for real-time systems, fault tolerance techniques must guarantee that timing constraints are met, in addition to functional correctness. In this paper we present a system which provides a powerful interface to AVD's (section 3), and a distributedreal-time architecture for perception and action tasks with several levels of fault-tolerance (section 4). We present a complex task which uses active vision for object capture and show how the system can provide solutions both for the tasks real-time requirements and fault tolerance (section 5 ) . In section 6 our tests and results are shown. 2 Related Work