Proceedings

An Architecture for Vision and Action

Download BibTeX Cite

Proceedings of the Fourteenth International Joint Conference on Artificial Intelligence. Action and Perception. Pages 72-80.

© 1995 International Joint Conferences on Artificial Intelligence

Abstract

Vision systems that have successfully supported nontrivial tasks have invariably taken advantage of constraints derived from the task and environment to increase reliability and lower the complexity of perception. We propose that it is possible to build a general purpose vision system, that is, one that can support a wide variety of tasks, and take advantage of such constraints. The central idea within our proposed architecture is the reactive skill. Skills are concurrent control routines assembled at run time using instructions from a symbolic execution system. Visual modules are used as resources in the construction of these skills. Skills control the agent as continuous feedback loops but are constructed using discrete, symbolic instructions. The key to general-purpose vision is the ability to parametrize the primitive elements of the vision system and to compose visual and control routines in a variety of ways. We demonstrate the architecture in the context of an implemented example task of a robot collecting trash off a floor and depositing it in a garbage can.

  1. [Agre and Chapman, 1987] Philip E. Agre and David Chap- man. Pengi: An implementation of a theory of activity. In Sixth National Conference on Artificial Intelligence, Seat- tle, WA, July 1987. AAAI.
  2. [Aloimonos, 1990] John Aloimonos. Purposive and qualita- tive active vision. In International Conf erence on Pattern Recognition, pages 346-360, 1990.
  3. [Aloimonos, 1994] Yiannis Aloimonos. What I have learned. CVGIP: Image Understanding, 60:74-85, 1994.
  4. [Bajcsy, 1988] R. Bajcsy. Active perception. Proceedings of the IEEE, 76:996-1005, 1988.
  5. [Brooks, 1986] Rodney A. Brooks. A robust layered control system for a mobile robot. IEEE Journal of Robotics and Automation, RA-2:14-23, 1986.
  6. [Brown, 1994] Christopher M. Brown. Toward general vi- sion. CVGIP: Image Understanding, 60:89-91, 1994.
  7. [Dickmanns and Graefe, 1988] Ernst Dieter Dickmanns and Volker Graefe. Dynamic monocular machine vision. Ma- chine Vision and Applications, 1:223-240, 1988.
  8. [Firby et al., 1995] R. James Firby, Peter N. Prokopowicz, and Michael J. Swain. Plan representations for picking up trash. Technical report, Department of Computer Science, University of Chicago, 1995.
  9. [Firby, 1987] R. James Firby. An investigation into reac- tive planning in complex domains. In Sixth National Con- ference on Artificial Intelligence, Seattle, WA, July 1987.
  10. A A AI.
  11. [Firby, 1989a] R. James Firby. Adaptive execution in com- plex dynamic worlds. Technical Report YALEU/CSD/RR 672, Department of Computer Science, Yale University, 1989.
  12. [Firby, 1989b] R. James Firby. Adaptive execution in com- plex dynamic worlds. Technical Report YALEU / CSD /RR #672, Computer Science Department, Yale University, January 1989.
  13. [Firby, 1994] R. James Firby. Task networks for controlling continuous processes. In Proceedings of the Second Inter- national Conference on AI Planning Systems, pages 49-54, 1994.
  14. [Horswill, 1993] Ian D. Horswill. Specialization of Perceptual Processes. PhD thesis, Dept. of EE & CS, Massachusetts Institute of Technology, 1993.
  15. [Huttenlocher and Rucklidge, 1992] Daniel P. Huttenlocher and William J. Rucklidge. A multi-resolution technique for comparing images using the hausdorff distance. Tech- nical Report CUCS TR 92-1321, Department of Computer Science, Cornell University, 1992.
  16. [Kahn et al., 1994] Roger E. Kahn, R. James Firby, and Michael J. Swain. The message hub. Animate Agent Project Working Note 4, University of Chicago, April1994.
  17. [Kahn, 1993] Ari Kahn. Using the datacube client class. The University of Chicago, Computer Science Department, 1993.
  18. [Kosecka and Bajcsy, 1993] Jana Kosecka and Ruzena Ba- jcsy. Cooperation of visually guided behaviors. In IEEE International Conf erence on Computer Vision, pages 502- 506, 1993.
  19. [Marr, 1982] D.C. Marr. Vision. Freeman, 1982.
  20. [Nelson, 1989] R. C. Nelson. Using flow field divergence for obstacle avoidance in visual navigation. IEEE Trans- actions on Pattern Analysis and Machine Intelligence, 11:1102-1106, 1989.
  21. [Prokopowicz et al., 1994] Peter N. Prokopowicz, Michael J. Swain, and Roger E . Kahn. Task and environment- sensitive tracking. In Proceedings of the !APR/ IEEE Workshop on Visual Behaviors, 1994.
  22. [Slack, 1990] Marc G. Slack. Situationally driven local nav- igation for mobile robots. Technical Report JPL Publica- tion 90-17, Jet Propulsion Laboratory, April1990.
  23. [Slack, 1992] M .G . Slack. Sequencing formally defined reac- tions for robotic activity: Integrating raps and gapps. In Vol. 1828 Sensor Fusion V: Simple Sensing for Complex Action, Boston, MA, November 1992. SPIE.
  24. [Swain and Ballard, 1991] Michael J. Swain and Dana H. Ballard. Color indexing. International Journal of Com- puter Vision, 7:11-32, 1991.
  25. [Tarr and Black, 1994] Michael J. Tarr and Michael J. Black. A computational and evolutionary perspective on the role ofrepresentation in vision. CVGIP: Image Understanding, 60:65-73, 1994.
  26. [Ullman, 1984] Shimon Ullman. Visual routines. Cognition, 18:97-159, 1984.

References were extracted automatically from the PDF and may contain errors.

Welcome Future Conferences AI Journal Division IJCAI Proceedings About IJCAI IJCAI Awards