|Article title||APPLICATION OF MULTI-AGENT TECHNOLOGY FOR MANAGING A GROUP OF UNMANNED UNDERWATER VEHICLES|
|Authors||A.I. Mashoshin, P.O. Skobelev|
|Section||SECTION II. MARINE ROBOTICS|
|Month, Year||01, 2016 @en|
|Abstract||The paper considers the principles of developing a distributed multi-agent system for managing a group (“swarm”) of unmanned underwater vehicles (UUV) for distributed solution of the task of patrolling a given area. The paper proposes a new methodological framework for development of distributed smart systems for collective management of the new generation mobile robotic objects. This framework provides the ability to solve important tasks of coordinated actions within a group (swarm) of robots and ensures such important benefits as flexibility and efficiency, productivity and scalability, reliability and viability of the system. Self-sufficiency of UUVs is ensured by an individual intelligent control system (ICS) for each vehicle. This system is capable of reacting to events, planning its performance in order to achieve the objectives and communicating with other devices. For this purpose, multi-agent technology acts as the basis for the ICS, providing the possibility to create plans which are open to change, as well as rebuild them quickly, flexibly and efficiently according to the events and to enforce a given mission in spite of any disturbing events. Software agents of the mission, tasks, routes, surveillance squares, vehicles and other resources within the developed system are designed to solve their local problems: surveillance of a certain square of the waters, choosing a safe route (without collisions), maintaining an adequate supply of battery power, maintenance of communication sessions with the database and identifying specific goals (tasks for each UUV). Each UUV agent performs coordination of its actions during flexibly scheduled time slots, which are mobile and can be reallocated among the resources and shifted, depending on the situation, in order to resolve conflicts. If it is impossible to perform the task, it is transferred to other agents, with possible loss of quality or efficiency. The proposed method of adaptive planning is based on introduction of satisfaction functions for each agent in the system. Such functions describe deviations of parameter attributes from the desired ideal values. Each agent receives a description of the domain from the local ontology, which is part of the region ontology. The current state of the scene is formed by correcting ontological scenes of individual agents. At the same time, consistency of the given state of agent system is supported on the instrumental level. The general trend of agent behavior is defined in the ontology and is directed at achieving locally optimal parameters. Temporary deterioration of the states of individual agents is possible in the current scene, however, dynamic evolution of the whole system aims to automatically adapt to the desired direction. The loss of one or more vehicles does not lead to disruption of the mission, and only increases the running time, possibly with lower quality and a smaller probability. The general principle of constructing the distributed control system for UUV groups requires a network-centric approach to building a multi-agent "system of systems". These subsystems interact with each other on the basis of р2р principles via a common bus. In this case, it is possible to create a self-organizing system, in which the directorial command decision-making center does not give orders to the processes but uses direct interaction among the systems, at the same time catalyzing it. Meanwhile, the command center maintains all the capabilities for task assignment, coordination and evaluation of results. Such an approach gives the possibility to simulate the appearance of a new type of phenomenon – "command (team) intelligence" in groups ("swarm intelligence"). The main principle of this kind of network-centric systems can be formulated as follows: the task must be solved as locally as possible, but at the same time as globally, as is required in the given situation. The paper proposes distributed architecture of network-centric systems for management of a UUV swarm and considers the basic architecture modules and their interaction. The main benefits generated by the system are analyzed and the ways of their implementation and application are outlined.|
|Keywords||Unmanned underwater vehicle; multi-agent technology; real-time planning; network-centric system.|
|References||1. Ageev M.D. Avtonomnye podvodnye roboty. Sistemy i tehnologii. [Autonomous underwater robots. Systems and technology]. Moscow: Nauka, 2005, 400 p.
2. Inzartsev A.V. Primenenie avtonomnogo neobitaemogo podvodnogo apparata dlya nauchnykh issledovaniy v Arktike [Use of autonomous underwater vehicle for research in the Arctic], Podvodnye issledovaniya i robototehnika [Underwater research and robotics], 2007, No. 2 (4), pp. 5-14.
3. Gizitdinova M.R, Kuz’mitsky M.A. Mobilnye podvodnye roboty v sovremennoy okeanografii i gidrofizike [Mobile underwater robots in modern oceanography and hydrophysics], Fundamental’naja i prikladnaja gidrofizika [Fundamental and applied hydrophysics], 2010, Vol. 3, No. 1, pp. 4-13.
4. Bozhenov Ju.A. Ispol'zovanie avtonomnykh neobitaemykh podvodnykh apparatov dlya issledovaniya Arktiki i Antarktiki. [Use of autonomous underwater vehicles for research in the Arctic and Antarctica], Fundamental’naya i prikladnaya gidrofizika [Fundamental and applied
hydrophysics], 2011, Vol. 4, No. 1, pp.4-68. 5. Millar G., Mackay L. Maneuvering under the ice, Sea technology, 2015, Vol. 56, No. 4, pp. 35-38.
6. Illarionov G.Ju., Sidenko K.S., Bocharov L.Ju. Ugroza iz glubiny: XXI vek [Danger from the deep: XXI century], Khabarovskaya kraevaya tipografiya [Khabarovsk regional printing office], 2011, 304 p.
7. Ghabcheloo R., Pascoal A., Silvestre C., Kaminer I. Nonlinear Coordinated Path Following Control of Multiple Wheeled Robots with Bidirectional Communication Constraints, International Journal of adaptive control and signal processing, 2007. No. 20. P.133– 157.
8. Vel'tishhev V.V., Egorov S.A., Kropotov A.N., Kuleshov V.I., Gur'ev A.V. Osobennosti razrabotki navigatcionnogo obespecheniya gruppirovki ANPA [Peculiarities of navigational support development of AUV group], Fundamental’naya i prikladnaya gidrofizika [Fundamental and applied hydrophysics], 2014, Vol. 7, No. 2, pp. 41-45.
9. Saharov V.V., Chertkov A.A., Tormashev D.S. Algoritm optimal'nogo planirovaniya gruppovogo vzaimodeystviya robotov [Algorithm of optimal planning of group interaction of robots], Morskoy vestnik [Morskoy Vestnik], 2014, No. 4 (52), pp. 119-122.
10. Ducatelle F., Di Caro C.P., Gambardella L.M. Self-organized cooperation between robotic swarms, Swarm Intelligence, 2011, Vol. 5, No. 2, pp. 73-96.
11. Khaldi B., Cherif F. An Overview of Swarm Robotics: Swarm Intelligence Applied to Multi-robotics, International Journal of Computer Applications (0975–8887), 2015, Vol. 126, No. 2, pp. 31-37.
12. Trianni V. and Nolfi S. Engineering the evolution of self-organizing behaviors in swarm robotics: A case study, Artificial Life, 2011, Vol. 17, No. 3, pp. 183-202.
13. Khamis A., Ahmed Hussein and Ahmed Elmogy. Multi-robot Task Allocation: A Review of the State-of-the-Art, In: Cooperative Robots and Sensor Networks 2015, Studies in Computational Intelligence. A. Koubвa and J.R. Martнnez-de Dios (eds.). Springer, Switzerland – 2015, pp. 31-51.
14. Kamrani F. Using On-line Simulation in UAV Path Planning, DS-RT '07 Proceedings of the 11th IEEE International Symposium on Distributed Simulation and Real-Time Applications. EEE Computer Society 2007, pp. 167-174.
15. Ergezer H., Leblebicioğlu K. 3D path planning for multiple UAVs for maximum information collection, Journal of Intelligent & Robotic Systems, 2014, Vol. 73, No. 1-4, pp. 737-762.
16. Austin R. Unmanned aircraft systems UAVs design, development and deployment. 1st ed. Wiley Aerospace Series, United Kingdom, 2010, pp. 221-226.
17. Kuremoto T., Obayashi M., Kobayashi K. Neuro-Fuzzy Systems for Autonomous Mobile Robots // Horizons in Computer Science Research (ed. T. S. Clary). 2013. N.Y., USA, Vol. 8, pp. 67-90.
18. Soares J.M., Aguiar A. P., Pascoal A.M., Martinoli A. Joint ASV/AUV Range-Based Formation Control: Theory and Experimental Results, Proc. of ICRA 2013 - IEEE International Conference on Robotics and Automation, Karlsruhe, Germany, pp. 5579-5585.
19. Antonelli G., Arrichiello F., Chiaverini S. Experiments of Formation Control with Collisions Avoidance using the Null-Space-Based Behavioral Control, Intelligent Service Robotics, 2008, Vol. 1, No. 1, pp. 27-39.
20. Bшrhaug E., Pavlov A., Ghabcheloo R., Pettersen K.Y., Pascoal A., Silvestre C. Formation Control of Underactuated Marine Vehicles with Communication Constraints, Proceedings 7th IFAC Conference on Manoeuvring and Control of Marine Craft. Lisbon. Portugal. 2006.
21. Marino A., Antonelli G., Aguiar A. Pedro., Pascoal A., Chiaverini S. A decentralized strategy for multi-robot sampling/patrolling: theory and experiments, IEEE Transactions on Control Systems Technology, 2015, Vol. 23, No. 1, pp. 313-322.
22. Gorodeckij V.I., Grushinskij M.S., Habalov A.V. Mnogoagentnye sistemy (obzor) [Multi-agent systems (review)], Novosti iskusstvennogo intellekta, [News of artificial intelligence], 1998, No. 2, pp. 64-116.
23. Skobelev P. Multi-Agent Systems for Real Time Adaptive Resource Management, Industrial Agents: Emerging Applications of Software Agents in Industry. Paulo Leitгo, Stamatis Karnouskos (Ed.). Elsevier, 2015, pp. 207-230.
24. Skobelev P.O. Mul'tiagentnye tehnologii v promyshlennyh primeneniyah: k 20-letiyu osnovaniya Samarskoy nauchnoy shkoly mul'tiagentnykh sistem [Multi-agent technology in industrial application: dedicated to the 20th anniversary of Samara scientific school of multi-agent systems], Mehatronika, avtomatizatciya, upravlenie [Mechatronics, automation, control], 2010, No. 12, pp. 33-46.
25. Skobelev P.O. Intellektual'nye sistemy upravleniya resursami v real'nom vremeni: printcipy razrabotki, opyt promyshlennykh vnedreniy i perspektivy razvitiya [Smart systems for realtime resource management: development principles, industrial implementation and future development], Prilozhenie k teoreticheskomu i prikladnomu nauchno-tehnicheskomu zhurnalu
«Informatcionnye tehnologii» [Appendix to theoretical and applied scientific-technical journal “Information technology”], No. 1, pp. 1-32.
26. Rzevski G.А., Skobelev P.O. Kak upravljat' slozhnymi sistemami? Mul'tiagentnye tehnologii dlya sozdaniya intellektual'nyh sistem upravleniya predpriyatiyami [How to manage complex systems? Multi-agent technology for developing smart enterprise control systems]. Samara: Ofort, 2015, 290 p.
27. Vittih V.А., Skobelev P.O. Metod sopryazhennyh vzaimodeystviy dlya upravleniya raspredeleniem resursov v real'nom masshtabe vremeni [Method of conjugated interactions for distributed real-time resource management], Avtometriya [Autometering], 2009, No. 2, pp. 78-87.
28. Vittih V.А., Skobelev P.O. Mul'tiagentnye modeli vzaimodeystviya dlya postroeniya setey potrebnostey i vozmozhnostey v otkrytykh sistemah [Multi-agent models of interaction for developing resource and demand networks in open systems], Avtomatika i telemehanika [Automatics and telemechanics], 2003, No. 1, pp. 177-185.
29. Budaev D.S., Voshhuk G.Ju., Gusev N.A., Majorov I.V., Mochalkin A.N. Razrabotka sistemy soglasovannogo upravleniya gruppoy bespilotnyh apparatov s primeneniem mul'tiagentnykh tehnologiy [Developing a system for coordinated management of a group of unmanned vehicles using multi-agent technology], Izvestija YuFU. Tehnicheskie nauki [Izvestiya SFedU.
Engineering Sciences], 2015, No. 10 (171), pp. 18-28.
30. Belousov A.A., Skobelev P.O., Stepanov M.E. Network-centric approach to adaptive real-time train scheduling in large-scale railway systems, Y. Tan et all. (Eds.), Proceedings of the Sixth International Conference on Swarm Intelligence (ICSI 2015), 26-29 June, 2015, Beijing, China. Part II. LNCS 9141. Springer, 2015, pp. 290-299.
31. Skobelev P., Simonova E. & Zhilyaev A. Using multi-agent technology for distributed management of cluster of remote sensing satellites, International Journal of Design & Nature and Ecodynamics, 2016, Vol. 11, Issue 2, pp. 127-134.
32. Skobelev P., Simonova E., Zhilyaev A. Application of Ontology in Smart System for Distributed Management of Small Satellites Groups, Proceedings of the PAAMS 2016 International Workshops, 1-3 June, 2016, Sevilla, Spain. 2016. Communications in Computer and Information Science series, vol. 0430. Spinger, Switzerland, pp. 1-12.