An Intelligent Observer

This paper describes an integrated mobile robotic system dubbed the intelligent observer (IO). The IO is a mobile robot which moves through an environment (such as an office building or a factory) while autonomously observing moving targets selected by a human operator. The robot carries one or more cameras which allow it to track objects while at the same time sensing its own location. It interacts with a human user who issues task-level commands, such as indicating a target to track by clicking in a camera image. The user could be located far away from the observer itself, communicating with the robot over a network. As the IO performs its tasks, the system provides real-time visual feedback to the user. We have implemented a prototype of the IO which integrates basic versions of five major components: landmark detection, target tracking, motion planning, motion control, and user interface. We have performed initial experiments using this prototype, which demonstrate the successful integration of these components and the utility of the overall systems.

[1]  Daniel P. Huttenlocher,et al.  Tracking non-rigid objects in complex scenes , 1993, 1993 (4th) International Conference on Computer Vision.

[2]  Jean-Claude Latombe,et al.  Reliable navigation using landmarks , 1995, Proceedings of 1995 IEEE International Conference on Robotics and Automation.

[3]  D.J. Kriegman,et al.  Stereo vision and navigation in buildings for mobile robots , 1989, IEEE Trans. Robotics Autom..

[4]  John F. Canny,et al.  A Computational Approach to Edge Detection , 1986, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[5]  Seth Hutchinson Exploiting visual constraints in robot motion planning , 1991, Proceedings. 1991 IEEE International Conference on Robotics and Automation.