Skip to main content

A Balking Queue Approach for Modeling Human-Multi-Robot Interaction for Water Monitoring

  • Conference paper
  • First Online:

Part of the book series: Lecture Notes in Computer Science ((LNAI,volume 10621))

Abstract

We consider multi-robot scenarios where robots ask for operator interventions when facing difficulties. As the number of robots increases, the operator quickly becomes a bottleneck for the system. Queue theory can be effectively used to optimize the scheduling of the robots’ requests. Here we focus on a specific queuing model in which the robots decide whether to join the queue or balk based on a threshold value. Those thresholds are a trade-off between the reward earned by joining the queue and cost of waiting in the queue. Though such queuing models reduce the system’s waiting time, the cost of balking usually is not considered. Our aim is thus to find appropriate balking strategies for a robotic application to reduce the waiting time considering the expected balking costs. We propose using a Q-learning approach to compute balking thresholds and experimentally demonstrate the improvement of team performance compared to previous queuing models.

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

Buying options

Chapter
USD   29.95
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
eBook
USD   39.99
Price excludes VAT (USA)
  • Available as EPUB and PDF
  • Read on any device
  • Instant download
  • Own it forever
Softcover Book
USD   54.99
Price excludes VAT (USA)
  • Compact, lightweight edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Learn about institutional subscriptions

Notes

  1. 1.

    SJF stands for Shortest Job First.

  2. 2.

    In this model, the arrivals to the system are customers. However our work applies this model into a robotic application, so the arrivals are robots with different requests.

  3. 3.

    Assuming a fully-observable setting works for this application, since the only global state variable is the queue size, which can be obtained easily.

  4. 4.

    We estimate the dynamic variables of the domain such as the average arrival rate, average service time, probability of failures, etc. based on some the data from field.

  5. 5.

    During the training phase in Q-learning approach, we used a small range around the estimated values for each of the arrival and service rate.

  6. 6.

    In reinforcement learning, an episode means a run of the algorithm beginning from a start state to a final state.

  7. 7.

    In our model, failures only happen for balking. This assumption is in favor of non-balking models. For example, if a boat waits too long for the operator the battery might run out, thus the mission fails just because time passes. Hence, in practice the results will probably be even more in favor of our approach.

References

  1. Chien, S.Y., Lewis, M., Mehrotra, S., Brooks, N., Sycara, K.: Scheduling operator attention for multi-robot control. In: Intelligent Robots and Systems (IROS), pp. 473–479. IEEE (2012)

    Google Scholar 

  2. Rosenthal, S., Veloso, M.: Using symbiotic relationships with humans to help robots overcome limitations. In: Workshop for Collaborative Human/AI Control for Interactive Experiences (2010)

    Google Scholar 

  3. Scerri, P., Pynadath, D.V., Tambe, M.: Towards adjustable autonomy for the real world. J. Artif. Intell. Res. 17(1), 171–228 (2002)

    MathSciNet  MATH  Google Scholar 

  4. Chien, S.Y., Lewis, M., Mehrotra, S., Han, S., Brooks, N., Wang, H., Sycara, K.: Task switching for supervisory control of multi-robot teams. IEEE Trans. Hum. Mach. Syst. (2016)

    Google Scholar 

  5. Rosenfeld, A.: Human-multi-robot team collaboration using advising agents: (doctoral consortium). In: Proceeding of the International Conference on Autonomous Agents and Multiagent Systems, pp. 1516–1517 (2016)

    Google Scholar 

  6. Naor, P.: The regulation of queue size by levying tolls. J. Econom. Soc. 37(1), 15–24 (1969)

    Article  MATH  Google Scholar 

  7. Farinelli, A., Raeissi, M.M., Brooks, N., Scerri, P.: Interacting with team oriented plans in multi-robot systems. J. Auton. Agents Multi-Agent Syst. 31(2), 332–361 (2017)

    Article  Google Scholar 

  8. Rosenfeld, A., Agmon, N., Maksimov, O., Azaria, A., Kraus, S.: Intelligent agent supporting human-multi-robot team collaboration. In: IJCAI, pp. 1902–1908 (2015)

    Google Scholar 

  9. Dai, T., Sycara, K., Lewis, M.: A game theoretic queueing approach to self-assessment in human-robot interaction systems. In: IEEE International Conference on Robotics and Automation, Shanghai, pp. 58–63 (2011)

    Google Scholar 

  10. Buşoniu, L., Babuška, R., De Schutter, B.: Multi-agent reinforcement learning: an overview. In: Srinivasan, D., Jain, L.C. (eds.) Innovations in Multi-Agent Systems and Applications-1, pp. 183–221. Springer, Heidelberg (2010). doi:10.1007/978-3-642-14435-6_7

    Google Scholar 

  11. Hu, Y., Gao, Y., An, B.: Multiagent reinforcement learning with unshared value functions. IEEE Trans. Cybern. 45(4), 647–662 (2015)

    Article  Google Scholar 

  12. Tan, M.: Multi-agent reinforcement learning: independent vs. cooperative agents. In: Proceedings of the Tenth International Conference on Machine Learning, pp. 330–337 (1993)

    Google Scholar 

  13. Sutton, R.S., Barto, A.G.: Reinforcement Learning: An Introduction, vol. 1. MIT press, Cambridge (1998). No. 1

    Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Masoume M. Raeissi .

Editor information

Editors and Affiliations

Rights and permissions

Reprints and permissions

Copyright information

© 2017 Springer International Publishing AG

About this paper

Cite this paper

Raeissi, M.M., Brooks, N., Farinelli, A. (2017). A Balking Queue Approach for Modeling Human-Multi-Robot Interaction for Water Monitoring. In: An, B., Bazzan, A., Leite, J., Villata, S., van der Torre, L. (eds) PRIMA 2017: Principles and Practice of Multi-Agent Systems. PRIMA 2017. Lecture Notes in Computer Science(), vol 10621. Springer, Cham. https://doi.org/10.1007/978-3-319-69131-2_13

Download citation

  • DOI: https://doi.org/10.1007/978-3-319-69131-2_13

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-319-69130-5

  • Online ISBN: 978-3-319-69131-2

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics