A method for decentralized control of adaptive data collection processes in autonomous distributed systems

: pp. 1 - 7
Lviv Polytechnic National University, Computer Engineering Department

The problem of organizing data collection processes in autonomous distributed systems has been considered, in particular, in autonomous mobile cyber-physical systems and autonomous distributed environmental monitoring systems. A model of decentralized control of adaptive data collection processes based on the principle of equilibrium has been proposed. Using this model, the problem of coordinating joint collective actions of adaptive data collection processes is studied from the point of view of finding an effective scheme for their complementarity in the absence of a control center.

A method has been developed for decentralized control of adaptive data collection processes using the principle of equilibrium and reinforcement learning using the normalized exponential function method. The developed method allows for efficient operation of autonomous distributed systems in conditions of dynamic changes in the number of data collection processes and limited information interaction between them.

As a result of research and modeling of the developed decentralized control method, it was found that the use of the normalized exponential function method provides a more efficient search for a solution compared to the adaptive random search method (on average by 28.3%). The dependence of the work of the developed decentralized control method on changes in the number of data collection processes and changes in the information interaction pattern between them was studied using the efficiency retention coefficient.

  1. Shi, Peng & Yan, Bing. (2020). A Survey on Intelligent Control for Multiagent Systems. IEEE Transactions on Systems, Man, and Cybernetics: Systems. pp.1-15. DOI: 10.1109/TSMC.2020.3042823.
  2. Niu, Y., Miao, K., Liu, T., Wu, L. (2023). Survey on Coordination Problems of Multi-agent System and Application in Unmanned Systems. In: Proceedings of 2022 International Conference on Autonomous Unmanned Systems (ICAUS 2022). ICAUS 2022. Lecture Notes in Electrical Engineering, vol. 1010. Springer, Singapore. DOI: 10.1007/978-981-99-0479-2_180
  3. Dorri, A., Kanhere, S., Jurdak, R. (2018) Multi-Agent Systems: A Survey, in IEEE Access, vol. 6. – pp. 28573-28593, DOI: 10.1109/ACCESS.2018.2831228.
  4. Rizk, Y., Awad, M., Tunstel, E. (2018) Decision Making in Multi-Agent Systems: A Survey, in IEEE Transactions    on    Cognitive    and    Developmental    Systems,    vol.     10,     no.     3.     –     pp.     514-529, DOI: 10.1109/TCDS.2018.2840971.
  5. Michael G. Rabbat, Alejandro Ribeiro (2018) Multiagent Distributed Optimization, in Cooperative and Graph Signal Processing, Petar M. Djurić, Cédric Richard (eds.), Academic Press, pp. 147-167. DOI: 10.1016/B978- 0-12-813677-5.00005-5
  6. Richard S. Sutton, Andrew G. Barto (2018) Reinforcement Learning: An Introduction, 2nd Ed., A Bradford Book. - 532 p.  ISBN: 978-026-203-924-6
  7. Dimitri Bertsekas (2019) Reinforcement Learning and Optimal Control, Athena Scientific. - 388 p. ISBN: 978-188-652-939-7
  8. Prudencio, Rafael & Maximo, Marcos & Colombini, Esther (2023) A Survey on Offline Reinforcement Learning: Taxonomy, Review, and Open Problems. IEEE Transactions on Neural Networks and Learning Systems. PP. 1. DOI: 10.1109/TNNLS.2023.3250269.
  9. Howard M. Schwartz (2014) Multi-Agent Machine Learning: A Reinforcement Approach, Wiley. - 256 p. ISBN: 978-111-836-208-2
  10. Gronauer, S., Diepold, K. (2022) Multi-agent deep reinforcement learning: a survey. Artificial Intelligence Review, 55, pp. 895–943. DOI: 10.1007/s10462-021-09996-w
  11. Botchkaryov, A., Golembo, V., Paramud, Y., Yatsyuk, V. (2019) Cyber-physical systems: data collection technologies, A. Melnyk (ed.), Lviv, «Magnolia 2006». – 176 p. (in Ukrainian) ISBN: 98-617-574-139-9
  12. Ansa Shermin S., Sarang C. Dhongdi (2022) Review of Underwater Mobile Sensor Network for ocean phenomena   monitoring,   Journal   of   Network   and   Computer    Applications,    Volume    205,    103418. DOI: 10.1016/j.jnca.2022.103418.
  13. Wang, Z., Li, H.X., Chen, C. (2020) Reinforcement Learning-Based Optimal Sensor Placement for Spatiotemporal     Modeling.     IEEE     Transactions     on     Cybernetics;     50(6).      –      pp.      2861-2871. DOI: 10.1109/TCYB.2019.2901897.
  14. Serge Kernbach (2008) Structural Self-Organization in Multi-Agents and Multi-Robotic Systems, Logos Verlag. - 250 p. ISBN: 978-383-252-048-9
  15. Botchkaryov А. (2020) Structural adaptation of data collection processes in autonomous distributed systems using reinforcement learning methods, Computer Systems and Networks, Lviv Polytechics, Issue 2, Num.1, pp.13-26. (in Ukrainian) DOI: 10.23939/csn2020.01.013
  16. Arup Kumar Sadhu, Amit Konar (2020) Multi-Agent Coordination: A Reinforcement Learning Approach, Wiley. – 320 p. ISBN: 978-111-969-903-3
  17. Zhou, L., Zheng, Y., Zhao, Q., Xiao, F., Zhang, Y. (2022). Game-based coordination control of multi-agent systems. Systems & Control Letters. 169, pp. 1 24. DOI: 10.1016/j.sysconle.2022.105376.