| Abstract | U višeagentskom pristupu projektiranju programske podrške, funkcionalnost se postiže interakcijom autonomnih procesnih entiteta ili agenata. Pristup je posebno prikladan kod problema koji su ili inherentno porazdijeljeni, pa seautonomni čimbenici ističu već u specifikaciji, ili iznimno složeni te ih je pogodno izraziti kroz što autonomnije specijalizirane podsustave. Komunikacija u višeagentskom sustavu obično se modelira prema poznatim obrascima interakcije među ljudima ili životinjama, pa agenti često imaju antropomorfna svojstva poput vjerovanja, želja ili namjera. Analiza praktičnih aspekatavišeagentskogorganiziranja napodručjuračunarskogvidapodrazumijevaeksperimentiranje svrlo složenim programskim sustavima. To djelomično objašnjava slabiju zastupljenost tog područja uznanstvenim publikacijamaodnosnosporijiprodorvišeagentskihpristupanapodručjuindustrijskih primjena. Razmatra se višeagentski sustav iz domene porazdijeljenog računarskog vida, u kojem su agenti organizirani u višerazinskoj hijerarhijskoj strukturi. Cilj sustava je praćenje pokretnih objekata u stvarnom vremenu, temeljeno na kooperaciji skupa upravljivih kamera. Svakoj kameri pridruženo je zasebno umreženo računalo kojim upravlja agent promatrač, tj. program koji analizira pribavljene slike i upravlja smjerom gledanja kamere. Promatrači podešavaju smjer gledanja u smislu ostvarivanja čim boljih rezultata praćenja u skladu sa svojim odgovornostima koje se mogu odnositi ili na pojedine objekte ili na određene dijelove scene. Koordinaciju promatrača obavljaju tzv. agenti koordinatori, koji udružuju pojedinačne poglede u ukupni prikaz scene, te formuliraju globalne ciljeve porazdijeljenog sustava u obliku odgovornosti pojedinih promatrača. Korekcije odgovornosti promatrača obično obezvrijeđuju njihove dotadašnje kontekste praćenja, pa kvaliteta dojavljivanih podataka privremeno pada dok nesigurnost informacija o sceni raste. Učestalost korekcija stoga mora biti pažljivo odmjerena, što upućuje na to da je robustna autonomna prilagodba promatrača stanju scene iznimno važna za ispravan rad sustava. Radipostizanjavećefleksibilnosti, predloženojeproširenjearhitekturespromatračima kojiimaju mogućnost kretanja. Zbog potrebe za njihovom povremenom apsolutnom lokalizacijom, pogodno je postaviti dodatni zahtjev da promatrači, pored detekcije praćenih objekata, budu sposobni i za detekciju pokretnih promatrača. Tako bi se moglo reći da pokretni promatrači imaju dvostruku ulogu, i promatrača i pokretnog objekta. Konačno, predviđeno je i višerazinsko proširenje temeljne arhitekture u kojem podređeni koordinatori imaju ulogu promatrača prema koordinatoru na višoj razini. U takvoj organizaciji, podređeni koordinatori ispunjavaju svoje odgovornosti raspodjelom poslova pridruženim agentima. Sustav za porazdijeljeno praćenje je razmatran u kontekstu lokalizacije jednostavnih pokretnih robota metodom globalnog vida. U takvoj organizaciji, upravljanje pojedinim robotima temelji se na zajedničkoj infrastrukturi koja pruža informacije o položaju svih robota u zadanom području, kao mrežni servis dostupan svim zainteresiranim stranama. Klijenti predložene infrastrukture bi bili nezavisni upravljači pojedinih robota, koji bi tada mogli stvarati autonomne planove akcija robota. Pokazuje se da je takvo sučelje moguće elegantno uklopiti u postojeću arhitekturu, u okviru specijaliziranog koordinatora na vrhu hijerarhije. |
| Abstract (english) | In the multi-agent approach to software design, the functionality of a complex system is expressed through interaction between autonomous processing entities called agents. The approach is particularly suited for problems which are either inherently distributed, where autonomous subjects emerge directly from the specification, or extremely complex, so that it is very beneficial to express the solution through decoupled autonomous specialized subsystems. Communication in a multi-agent system is usually modelled after known interaction patterns between humans or animals, so that agents oftenfeature anthropomorphic qualities such asbeliefs, desiresorintentions. Analysisofpractical aspects of multi-agent architectures in the field of computer vision implies experiments with large scale software designs. That explains, in part, a rather weak attention in scientific publications and a relatively slow progress towards industrial applications. This work considers a multi-agent system in the domain of distributed computer vision, in which the agents are organized within a multilevel hierarchical structure. The objective of the system is to track multiple moving objects in real time using cooperation of several controllable cameras. Each camera is assigned a separate network connected computer which is controlled by an observer agent, i.e. a program which processes acquired images and controls the viewing direction of the camera. Observers adapt their viewing directions in order to achieve optimal tracking results according to their responsibilities, which are related either to individual objects or to particular parts of the scene. The cooperation of observers is managed by the second kind of agents in the system, which are called coordinators. The tasks of a coordinator consist of integrating individual views into the comprehensive representation of the scene, and formulating global objectives of the distributed system into responsibilities of individual observers. Corrections of observer responsibilities often invalidate their tracking contexts which is reflected by a decreasing quality of provided data and an increasing uncertainty of the scene representation. The frequency of corrections therefore must be carefully regulated, which suggests that a robust autonomous adaptation of the observers to the changes in the scene is extremely beneficial for the correct functioning of the system. In order to achieve a greater flexibility, an extension of the basic architecture is proposed, which deals with observers being capable of autonomous movement. Following the need for their occasional absolute localization, it is suitable to introduce an additional requirement, that observers should be able to detect their moving peers as well. Consequently, it can be stated that, in such a system, moving observers have a dual role: as an observer and as a tracked object. Finally, a multilevel extension of the basic architecture is also introduced, in which each inferior coordinator acts as an observer to the assigned superior coordinator. In such an arrangement, inferior coordinators fulfill their observer responsibilities by delegating jobs to assigned agents. The system for distributed visual tracking is considered in the context of localizing a group of simple moving robots by a global vision approach. The approach implies building a common infrastructure capable of providing positions of all robots in the area, as a network service available to general public. The infrastructure could be used by independent robot controllers, which would then be able to make autonomous plans of robot actions. It is shown that such network interface can be fitted into proposed architecture, within a specialized coordinator on the top of the hierarchy. |
