Implementation of Collective Perception Strategy in a Self-Organizing Swarm System Using Bayesian Decision Rule

Relevance. Improving collective perception strategies in swarm systems is a key challenge for enhancing the efficiency of autonomous robotic groups in complex and dynamic environments. Existing approaches, such as DMMD, DMVD, and DC, have limited capabilities in classifying objects with non-obvious features, necessitating the development of new methods.

Objective. Increasing the accuracy of perceiving specific characteristics of an object investigated by a multi-agent robotic system.

Methods. The proposed criterion employs a Bayesian decision rule to update the posterior probabilities of alternatives based on data collected by the robots. The validity of the proposed solutions was confirmed through simulation of a typical collective perception task on a defined tested.

Results. A comparison was made with established collective perception strategies: DMMD, DMVD, and DC. It was shown that these strategies have limited applicability in classifying complex objects. A software implementation of the collective perception scenario was tested in a swarm robotic system consisting of 20 robots inspecting a scene composed of multicolored tiles. The experimental results demonstrated that the authors' approach endowed the robot swarm with previously unattainable functional capabilities in collective perception for complex scenarios.

Novelty. A method for detecting object properties using a statistical criterion was proposed. The strategy quantifies the consensus-building process among swarm members over sequential time steps, followed by intra- and inter-period processing of information generated by the swarm's robots. The results expand the theoretical foundations of swarm intelligence by introducing a new method for processing distributed information. Practical significance lies in improving the efficiency of swarm systems for monitoring, search, and classification tasks in medicine, ecology, and other fields.

1. Dorigo M., et al. Swarm robotics. Scholarpedia. 2014;9(1):1463. DOI:10.4249/scholarpedia.1463

2. Campo A., Garnier S., Dedriche O., Zekkri M., Dorigo M. Self-Organized Discrimination of Resources. PLoS ONE. 2011;6(5): e19888. DOI:10.1371/journal.pone.0019888

3. Sailor M.J., Link J.R. “Smart dust”: nanostructured devices in a grain of sand. Chemical Communications. 2005;11;1375. DOI:10.1039/b417554a. EDN:MHMXGT

4. Montes de Oca M.A., Ferrante E., Scheidler A., Pinciroli C., Birattari M., Dorigo M. Majority-rule opinion dynamics with differential latency: a mechanism for self-organized collective decision-making. Swarm Intelligence. 2011;5(3-4):305–327. DOI:10.1007/s11721-011-0062-z. EDN:GHZZLS

5. Gorodetsky V.I. Behavioral Model for Cyber-Physical System and Group Control: The Basic Concepts. Izvestiya SFedU. Engineering Sciences. 2019;1(203):144–162. (in Russ.) DOI:10.23683/2311-3103-2019-1-144-162. EDN:LYUZBR

6. Karpov V.E. Social communities of robots: from reactive to cognitive agents. Soft Measurements and Computing. 2019;2(15):61‒78. (in Russ.) EDN:SEFEFV

7. Zikratov I.A., Viksnin I.I., Zikratova T.V. Multiagent Planning of Intersection Passage by Autonomous Vehicles. Scientific and Technical Journal of Information Technologies, Mechanics and Optics. 2016;16(5):839‒849. (in Russ.) EDN:WRPHSP

8. Yuryeva R.A., Komarov I.I., Viksnin I.I. Immunological principles of decision-making in multiagent robotic systems. Global Scientific Potential. 2015;5(50):87‒91. (in Russ.) EDN:UKOVSB

9. Strobel V., Ferrer E.C., Dorigo M. Managing Byzantine Robots via Blockchain Technology in a Swarm Robotics Collective Decision Making Scenario. Proceedings of the 17th International Conference on Autonomous Agents and Multiagent Systems, AAMAS '14, 10‒15 July 2018, Stockholm Sweden. IFAAMAS; 2018. p.541‒549.

10. Ivanov D.Ya. Methods of Swarm Intelligence for Control of Groups Of Small-Sized Unmanned Aerial Vehicles. Izvestiya SFedU. Engineering Sciences. 2011;3(116):221–229. (in Russ.) EDN:NPKHEP

11. Kalyaev I.A., Gaiduk A.R., Kapustyan S.G. Models and algorithms for collective control in robot groups. Moscow: FIZMATLIT Publ.; 2009. 280 p. (in Russ.) EDN:MUWSIT

12. Saenko I., Sokolov A., Lauta O., Gubsky P. Method of Target Distribution During Group Flight of Mini-UAVs to Targets. Information and Space. 2024;2:113‒120. (in Russ.) EDN:CGIZBA

13. Valentini G., Brambilla D., Hamann H., Dorigo M. Collective Perception of Environmental Features in a Robot Swarm. Proceedings of the 10th International Conference on Swarm Intelligence, 7‒9 September 2016, Brussels, Belgium. Lecture Notes in Computer Science, vol.9882. Cham: Springer; 2016. p.65‒76. DOI:10.1007/978-3-319-44427-7_6

14. Fagiolini A., Pellinacci M., Valenti G., Dini G., Bicchi A. Consensus-based distributed intrusion detection for multirobot systems. Proceedings of the International Conference on Robotics and Automation, ICRA 2008, 19‒23 May 2008, Pasadena, USA. p.120‒127. DOI:10.1109/ROBOT.2008.4543196

15. Valentini G., Hamann H. Time-variant feedback processes in collective decision-making systems: influence and effect of dynamic neighborhood sizes. Swarm Intelligence. 2015;9:153–176. DOI:10.1007/s11721-015-0108-8

16. Reina A., Valentini G., Hamann H., Dorigo M. A Design Pattern for Decentralised Decision Making. PLoS ONE. 2015;10. DOI:10.1371/journal.pone.0140950

17. Zikratov I.A., Zikratova T.V. Using Behavioral Models to Study Robot Societies. Information and Space. 2022;4:170‒174. (in Russ.) EDN:DQASLC

18. Zikratova T.V. The Method of Group Control in Multi-Agent Robotic Systems Under the Influence of Destabilizing Factors. Proceedings of Telecommunication Universities. 2021;7(3):92‒100. (in Russ.) DOI:10.31854/1813-324X-2021-7-3-92-100. EDN:JFMYBF

19. Valentini G. Indirect Modulation of Majority-Based Decisions. Studies in Computational Intelligence. 2017;706:55‒66. DOI:10.1007/978-3-319-53609-5_4

20. Valentini G., Hamann H., Dorigo M. Efficient Decision-Making in a Self-Organizing Robot Swarm: On the Speed Versus Accuracy Trade-Off. Proceedings of the 14th International Conference on Autonomous Agents and Multiagent Systems, AAMAS '15, 4‒8 May 2015, Istanbul, Turkey. IFAAMAS; 2015. p.1305–1314

21. Valentini G., Hamann H., Dorigo M. Self-organized collective decision making: the weighted voter model. Proceedings of the 13th International Conference on Autonomous Agents and Multiagent Systems, AAMAS '14, 5‒9 May 2014, Paris, France. IFAAMAS; 2014. p.45–52.

22. Valentini G. Achieving Consensus in Robot Swarms. Studies in Computational Intelligence. 2017;706. DOI:10.1007/978-3-319-53609-5

23. Valentini G., Ferrante E., Hamann H., Dorigo M. Collective decision with 100~Kilobots: speed versus accuracy in binary discrimination problems. Autonomous Agents and Multi-Agent Systems. 2015;30(3):553–580. DOI:10.1007/s10458-015-9323-3

24. Valentini G., Hamann H., Dorigo M. Self-Organized Collective Decision-Making in a 100-Robot Swarm. Twenty-Ninth AAAI Conference on Artificial Intelligence. 2015;29(1). DOI:10.1609/aaai.v29i1.9720

25. Ryabtsev S.S. A Method for Detecting Byzantine Robots Based on Data from the Collective Decision-Making Process in Swarm Robotic Systems. Control, Communication and Security Systems. 2022;3:105‒137. (in Russ.) DOI:10.24412/2410-9916-2022-3-105-137. EDN:SVSCHG

26. Zikratov I.A., Zikratova T.V., Novikov E.A. Swarm Robotics System Algorithm for Defense against Coordinated Behavior Strategy Attacks. Proceedings of Telecommunication Universities. 2024;10(3):75‒86. (in Russ.) DOI:10.31854/1813-324X-2024-10-3-75-86. EDN:XUDVOR