Skip to main content
Log in

Unavoidable deadends in deterministic partially observable contingent planning

  • Published:
Autonomous Agents and Multi-Agent Systems Aims and scope Submit manuscript

Abstract

Traditionally, a contingent plan, branching on the observations an agent obtains throughout plan execution, must reach a goal state from every possible initial state. However, in many real world problems, no such plan exists. Yet, there are plans that reach the goal from some initial states only. From the other initial states, they eventually reach a deadend—a state from which the goal can not be achieved. Deadends that cannot be avoided by resorting to a different plan, are called unavoidable deadends. In this paper we study planning with unavoidable deadends in belief space. We distinguish between two types of such deadends, and adapt offline and online contingent planners to identify and handle unavoidable deadends, using two approaches—an active approach that begins by distinguishing between the solvable and deadend states, and a lazy approach, that plans to achieve the goal, identifying deadends as they occur. We empirically analyze how each approach performs in different cases.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7

Similar content being viewed by others

Notes

  1. This implies that \(a_i\) is applicable in \(a_{i-1}(\ldots (a_1(I)))\) for every \(1\le i\le n\), because otherwise \(a_n(a_{n-1}(\ldots (a_1(I))))\) would be undefined.

  2. Belief carries a connotation of being subjective. Hence, knowledge state might be a better term, but we follow the standard terminology.

  3. This definition applies also if we were to allow actions that both affect the world state and provide informative observations.

References

  1. Albore, A., & Geffner, H. (2009). Acting in partially observable environments when achievement of the goal cannot be guaranteed. In Workshop on planning and plan execution for real-world systems, at 19th international conference on planning and scheduling (ICAPS’09).

  2. Albore, A., Palacios, H., & Geffner, H. (2009). A translation-based approach to contingent planning. In IJCAI 2009, Proceedings of the 21st international joint conference on artificial intelligence, Pasadena, California, USA, July 11-17, 2009 (pp. 1623–1628).

  3. Albore, A., Ramírez, M., & Geffner, H. (2011). Effective heuristics and belief tracking for planning with incomplete information. In Proceedings of the 21st international conference on automated planning and scheduling, ICAPS 2011, Freiburg, Germany June 11-16, 2011.

  4. Aminof, B., De Giacomo, G., Lomuscio, A., Murano, A., & Rubin, S. (2021). Synthesizing best-effort strategies under multiple environment specifications. In Proceedings of the international conference on principles of knowledge representation and reasoning (Vol. 18, pp. 42–51).

  5. Aminof, B., De Giacomo, G., & Rubin, S. (2021). Best-effort synthesis: Doing your best is not harder than giving up. IJCAI.

  6. Bonet, B., & Geffner, H. (2011). Planning under partial observability by classical replanning: Theory and experiments. In IJCAI 2011, Proceedings of the 22nd international joint conference on artificial intelligence, Barcelona, Catalonia, Spain, July 16-22, 2011 (pp. 1936–1941).

  7. Bonet, B., & Geffner, H. (2014). Belief tracking for planning with sensing: Width, complexity and approximations. Journal of Artificial Intelligence Research, 50, 923–970. https://doi.org/10.1613/jair.4475.

    Article  MathSciNet  MATH  Google Scholar 

  8. Botea, A., Müller, M., & Schaeffer, J. (2002). Using abstraction for planning in sokoban. In International conference on computers and games (pp. 360–375). Springer.

  9. Brafman, R.I., & Shani, G. (2012). A multi-path compilation approach to contingent planning. In Proceedings of the twenty-sixth AAAI conference on artificial intelligence, July 22-26, 2012, Toronto, Ontario, Canada.

  10. Brafman, R. I., & Shani, G. (2012). Replanning in domains with partial information and sensing actions. Journal of Artificial Intelligence Research, 45, 565–600. https://doi.org/10.1613/jair.3711.

    Article  MathSciNet  MATH  Google Scholar 

  11. Brafman, R. I., & Shani, G. (2016). Online belief tracking using regression for contingent planning. Artificial Intelligence, 241, 131–152. https://doi.org/10.1016/j.artint.2016.08.005.

    Article  MathSciNet  MATH  Google Scholar 

  12. Camacho, A., Muise, C., & McIlraith, S. A. (2016). From fond to robust probabilistic planning: Computing compact policies that bypass avoidable deadends. In Twenty-sixth international conference on automated planning and scheduling.

  13. Chrpa, L., Lipovetzky, N., & Sardina, S. (2017). Handling non-local dead-ends in agent planning programs. In IJCAI (pp. 971–978).

  14. Cimatti, A., Pistore, M., Roveri, M., & Traverso, P. (2003). Weak, strong, and strong cyclic planning via symbolic model checking. Artificial Intelligence, 147(1–2), 35–84.

    Article  MathSciNet  MATH  Google Scholar 

  15. Cserna, B., Doyle, W. J., Ramsdell, J. S., & Ruml, W. (2018). Avoiding dead ends in real-time heuristic search. In Thirty-second AAAI conference on artificial intelligence.

  16. Eriksson, S., Röger, G., & Helmert, M. (2017). Unsolvability certificates for classical planning. In Twenty-seventh international conference on automated planning and scheduling.

  17. Falcone, F., Gerevini, A. E., & Saetti, A. (2017). On realizing planning programs in domains with dead-end states. In Tenth annual symposium on combinatorial search.

  18. Fu, J., Ng, V., Bastani, F., & Yen, I.-L. (2011). Simple and fast strong cyclic planning for fully-observable nondeterministic planning problems. In Twenty-second international joint conference on artificial intelligence.

  19. Hoffmann, J., & Nebel, B. (2001). The FF planning system: Fast plan generation through heuristic search. Journal of Artificial Intelligence Research, 14, 253–302. https://doi.org/10.1613/jair.855.

    Article  MATH  Google Scholar 

  20. Komarnitsky, R., & Shani, G. (2016). Computing contingent plans using online replanning. In Proceedings of the thirtieth AAAI conference on artificial intelligence, February 12-17, 2016, Phoenix, Arizona, USA (pp. 3159–3165).

  21. Komarnitsky, R., Maliah, S., & Shani, G. (2021). Computing contingent plan graphs using online planning. ACM Transactions on Autonomous and Adaptive Systems, 1(1), 1.

    Google Scholar 

  22. Lipovetzky, N., Muise, C. J., & Geffner, H. (2016). Traps, invariants, and dead-ends. In Proceedings of the twenty-sixth international conference on automated planning and scheduling, ICAPS 2016, London, UK, June 12-17, 2016 (pp. 211–215).

  23. Muise, C. J., McIlraith, S. A., & Beck, J. C. (2012). Improved non-deterministic planning by exploiting state relevance. In Proceedings of the twenty-second international conference on automated planning and scheduling, ICAPS 2012, Atibaia, São Paulo, Brazil, June 25-19, 2012.

  24. Muise, C. J., Belle, V., McIlraith, S. A. (2014). Computing contingent plans via fully observable non-deterministic planning. In Proceedings of the twenty-eighth AAAI conference on artificial intelligence, July 27 -31, 2014, Québec City, Québec, Canada (pp. 2322–2329).

  25. Shmaryahu, D., Shani, G., Hoffmann, J., & Steinmetz, M. (2018). Simulated penetration testing as contingent planning. In Proceedings of the twenty-eighth international conference on automated planning and scheduling, ICAPS 2018, Delft, The Netherlands, June 24-29, 2018 (pp. 241–249).

  26. Steinmetz, M., & Hoffmann, J. (2017). State space search nogood learning: Online refinement of critical-path dead-end detectors in planning. Artificial Intelligence, 245, 1–37.

    Article  MathSciNet  MATH  Google Scholar 

  27. To, S. T., Son, T. C., & Pontelli, E. (2011). Contingent planning as and/or forward search with disjunctive representation. In Twenty-first international conference on automated planning and scheduling.

Download references

Acknowledgements

We thank the reviewers, who invested a substantial effort, delving into the theoretical contributions, and suggesting ways to simplify and clarify the definitions, theorems, proofs, and algorithms, as well as suggesting additional experimental results that shed further light on our methods. Your efforts are much appreciated. This paper was partially supported by the ISF fund, under Grant Number 1210/18.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Guy Shani.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Shtutland, L., Shmaryahu, D., Brafman, R.I. et al. Unavoidable deadends in deterministic partially observable contingent planning. Auton Agent Multi-Agent Syst 37, 3 (2023). https://doi.org/10.1007/s10458-022-09570-w

Download citation

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1007/s10458-022-09570-w

Keywords

Navigation