{"id":{"repo_id":"rgu","oai_identifier":"oai:rgu-repository.worktribe.com:1993222"},"canonical_url":"https://search.dev.ndltd.org/etd/rgu/oai:rgu-repository.worktribe.com:1993222","repository":{"repo_id":"rgu","name":"Robert Gordon University","base_url":"https://rgu-repository.worktribe.com/oaiprovider"},"display":{"title":"Synthesis of stochastic learning automata.","abstract":"Over the past two decades, considerable interest has developed in the field of stochastic learning automata theory and, consequently, the application areas for learning systems. In control engineering, they are viewed as a means to implement optimal adaptive controllers for situations where little or no a priori information on the plant is available. Stochastic automata with a variable structure operate by means of a global random search, interacting with the environment to improve the action strategy towards optimum performance. They represent therefore a novel and attractive solution to a large class of problems involving high order uncertainties. At the same time, research has progressed in digital stochastic computing, in which variables are represented by random pulse trains, enabling analogue functions, effectively transformed into Boolean logic operations, to be performed at high speed by conventional digital hardware. These techniques were seen as ideally suited to the practical implementation of stochastic learning automata. This project is seen as the convergence of these two lines of activity, developing hardware automaton designs and devising applications to simulated and real system parameter optimisation problems. To provide continuity with previous theoretical studies and also lay the necessary foundations of hardware system design experience, basic two-state systems were designed and constructed initially. A standard modular design evolved which was incorporated in a hierarchical structure. This design philosophy enabled large state order automata to be implemented, providing a powerful tool for the optimisation of multivariable, multimodal systems. A prototype hierarchical structure 128-state automaton has been constructed and tested in both static experiments and a real process control application, based on a small-scale thermal system. The hardware learning automaton approach has been shown here to permit the effective, economic realisation of high-speed real-time system controllers.","abstract_html":"Over the past two decades, considerable interest has developed in the field of stochastic learning automata theory and, consequently, the application areas for learning systems. In control engineering, they are viewed as a means to implement optimal adaptive controllers for situations where little or no a priori information on the plant is available. Stochastic automata with a variable structure operate by means of a global random search, interacting with the environment to improve the action strategy towards optimum performance. They represent therefore a novel and attractive solution to a large class of problems involving high order uncertainties. At the same time, research has progressed in digital stochastic computing, in which variables are represented by random pulse trains, enabling analogue functions, effectively transformed into Boolean logic operations, to be performed at high speed by conventional digital hardware. These techniques were seen as ideally suited to the practical implementation of stochastic learning automata. This project is seen as the convergence of these two lines of activity, developing hardware automaton designs and devising applications to simulated and real system parameter optimisation problems. To provide continuity with previous theoretical studies and also lay the necessary foundations of hardware system design experience, basic two-state systems were designed and constructed initially. A standard modular design evolved which was incorporated in a hierarchical structure. This design philosophy enabled large state order automata to be implemented, providing a powerful tool for the optimisation of multivariable, multimodal systems. A prototype hierarchical structure 128-state automaton has been constructed and tested in both static experiments and a real process control application, based on a small-scale thermal system. The hardware learning automaton approach has been shown here to permit the effective, economic realisation of high-speed real-time system controllers.","abstract_has_math":false,"creators":["Neville, Richard Graham"],"institution":"Robert Gordon's Institute of Technology","degree_name":"PhD","degree_level":"Doctoral","degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":["P. Mars"],"committee_chairs":[],"committee_members":[],"year":1980,"date_issued":"1980","date_published":"1980","updated_at":"2026-07-24T04:10:03Z","subjects":["Stochastic learning","Control engineering","Performance optimisation","Boolean logic","Hardware design","Multimodal systems"],"languages":["en"],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["oai:rgu-repository.worktribe.com:1993222","https://doi.org/10.48526/rgu-wt-1993222"],"render_values":[{"text":"oai:rgu-repository.worktribe.com:1993222","href":null,"code":true},{"text":"https://doi.org/10.48526/rgu-wt-1993222","href":"https://doi.org/10.48526/rgu-wt-1993222","code":true}]}]},"links":{"outbound_url":"https://rgu-repository.worktribe.com/1993222/1/NEVILLE%201980%20Synthesis%20of%20stochastic%20learning","outbound_label":"Repository record","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["P. Mars"]},{"key":"dc:contributor.sponsor","label":"Sponsor","values":["Social Science Research Council, USA"]},{"key":"dc:creator","label":"Author","values":["Neville, Richard Graham"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["1980-10-31"]},{"key":"dc:date.issued","label":"Date","values":["1980"]},{"key":"dc:publisher.institution","label":"Dc Publisher Institution","values":["Robert Gordon's Institute of Technology"]},{"key":"dc:relation.isreferencedby","label":"Dc Relation Isreferencedby","values":["https://rgu-repository.worktribe.com/output/1993222"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"dc:type.qualificationlevel","label":"Dc Type Qualificationlevel","values":["Doctoral"]},{"key":"dc:type.qualificationname","label":"Dc Type Qualificationname","values":["PhD"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Stochastic learning","Control engineering","Performance optimisation","Boolean logic","Hardware design","Multimodal systems"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["oai:rgu-repository.worktribe.com:1993222","https://doi.org/10.48526/rgu-wt-1993222"]},{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://rgu-repository.worktribe.com/1993222/1/NEVILLE%201980%20Synthesis%20of%20stochastic%20learning"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Over the past two decades, considerable interest has developed in the field of stochastic learning automata theory and, consequently, the application areas for learning systems. In control engineering, they are viewed as a means to implement optimal adaptive controllers for situations where little or no a priori information on the plant is available. Stochastic automata with a variable structure operate by means of a global random search, interacting with the environment to improve the action strategy towards optimum performance. They represent therefore a novel and attractive solution to a large class of problems involving high order uncertainties. At the same time, research has progressed in digital stochastic computing, in which variables are represented by random pulse trains, enabling analogue functions, effectively transformed into Boolean logic operations, to be performed at high speed by conventional digital hardware. These techniques were seen as ideally suited to the practical implementation of stochastic learning automata. This project is seen as the convergence of these two lines of activity, developing hardware automaton designs and devising applications to simulated and real system parameter optimisation problems. To provide continuity with previous theoretical studies and also lay the necessary foundations of hardware system design experience, basic two-state systems were designed and constructed initially. A standard modular design evolved which was incorporated in a hierarchical structure. This design philosophy enabled large state order automata to be implemented, providing a powerful tool for the optimisation of multivariable, multimodal systems. A prototype hierarchical structure 128-state automaton has been constructed and tested in both static experiments and a real process control application, based on a small-scale thermal system. The hardware learning automaton approach has been shown here to permit the effective, economic realisation of high-speed real-time system controllers."]},{"key":"dc:title","label":"Title","values":["Synthesis of stochastic learning automata."]}]}],"canonical_facts":{"dc:contributor.advisor":["P. Mars"],"dc:contributor.sponsor":["Social Science Research Council, USA"],"dc:creator":["Neville, Richard Graham"],"dc:date":["1980-10-31"],"dc:date.issued":["1980"],"dc:description.abstract":["Over the past two decades, considerable interest has developed in the field of stochastic learning automata theory and, consequently, the application areas for learning systems. In control engineering, they are viewed as a means to implement optimal adaptive controllers for situations where little or no a priori information on the plant is available. Stochastic automata with a variable structure operate by means of a global random search, interacting with the environment to improve the action strategy towards optimum performance. They represent therefore a novel and attractive solution to a large class of problems involving high order uncertainties. At the same time, research has progressed in digital stochastic computing, in which variables are represented by random pulse trains, enabling analogue functions, effectively transformed into Boolean logic operations, to be performed at high speed by conventional digital hardware. These techniques were seen as ideally suited to the practical implementation of stochastic learning automata. This project is seen as the convergence of these two lines of activity, developing hardware automaton designs and devising applications to simulated and real system parameter optimisation problems. To provide continuity with previous theoretical studies and also lay the necessary foundations of hardware system design experience, basic two-state systems were designed and constructed initially. A standard modular design evolved which was incorporated in a hierarchical structure. This design philosophy enabled large state order automata to be implemented, providing a powerful tool for the optimisation of multivariable, multimodal systems. A prototype hierarchical structure 128-state automaton has been constructed and tested in both static experiments and a real process control application, based on a small-scale thermal system. The hardware learning automaton approach has been shown here to permit the effective, economic realisation of high-speed real-time system controllers."],"dc:identifier":["oai:rgu-repository.worktribe.com:1993222","https://doi.org/10.48526/rgu-wt-1993222"],"dc:identifier.uri":["https://rgu-repository.worktribe.com/1993222/1/NEVILLE%201980%20Synthesis%20of%20stochastic%20learning"],"dc:language":["en"],"dc:publisher.institution":["Robert Gordon's Institute of Technology"],"dc:relation.isreferencedby":["https://rgu-repository.worktribe.com/output/1993222"],"dc:subject":["Stochastic learning","Control engineering","Performance optimisation","Boolean logic","Hardware design","Multimodal systems"],"dc:title":["Synthesis of stochastic learning automata."],"dc:type":["Thesis"],"dc:type.qualificationlevel":["Doctoral"],"dc:type.qualificationname":["PhD"]},"updated_at":"2026-07-24T04:10:03Z"}