FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2786227318> ?p ?o ?g. }

• W2786227318 abstract "Robot swarms are systems composed of a large number of rather simple robots. Due to the large number of units, these systems, have good properties concerning robustness and scalability, among others. However, it remains generally difficult to design controllers for such robotic systems, particularly due to the complexity of inter-robot interactions. Consequently, automatic approaches to synthesize behavior in robot swarms are a compelling alternative. One of these approaches, Embodied Evolutionary Robotics (EER), opens many possibilities, due to learning taking place in parallel for each robot in the swarm, while deployed for task operation, i.e. online. Parallel evaluations and information exchanges among robots accelerate learning, which is open-ended, thus allowing for potential adaptation to changing conditions.That said, EER approaches are relatively new, and their properties remain to be studied. In this thesis, we focus on online behavior adaptation in a swarm of robots using distributed EER methods. We consider a swarm of robots that coexist in an environment, and must progressively adapt to given tasks. Additionally, since robots may face changing conditions that may repeat over time, retaining acquired knowledge about previous conditions could improve their adaptivity. However, when confronted to new situations, adaptive systems may instantaneously forget what was learned before, thus hindering such adaptivity. The contributions in this thesis aim at investigating and improving the adaptivity of evolving robot swarms. To this end, we provide four main contributions:We investigate the influence of task-driven selection pressure in a swarm of robotic agents using a distributed EER approach. We evaluate the impact of a range of selection operators on the performance of a distributed EER algorithm for a robot swarm. The results show that task-driven selection pressure is necessary when addressing given tasks in such a distributed setup, and the higher the selection pressure, the better the performances obtained.We investigate the evolution of collaborative behaviors in a swarm of robotic agents using a distributed EER approach. We perform a set of experiments for a swarm of robots to adapt to a collaborative item collection task that cannot be solved by a single robot. Our results show that the swarm learns to collaborate to solve the task using a distributed approach. Additionally, some inefficiencies regarding learning to choose actions to collect items are analyzed, and perspectives are discussed to improve action choice.We propose and experimentally validate a completely distributed mechanism that allows to learn the structure and parameters of the robot neurocontrollers in a swarm using a distributed EER approach. This allows for the robot controllers to augment their capacity and expressivity. Our experiments show that our fully-decentralized mechanism leads to similar results as a mechanism that depends on global information.We propose an algorithm to avoid forgetting from the perspective of an evolving population when adapting to changing conditions. In a set of preliminary experiments, we test a centralized version of the algorithm, showing the feasibility of the approach.Finally, we discuss how it can be transferred to the decentralized context of distributed EER." @default.
• W2786227318 created "2018-02-23" @default.
• W2786227318 creator A5053831284 @default.
• W2786227318 date "2017-12-19" @default.
• W2786227318 modified "2023-09-23" @default.
• W2786227318 title "Distributed Embodied Evolutionary Adaptation of Behaviors in Swarms of Robotic Agents" @default.
• W2786227318 hasPublicationYear "2017" @default.
• W2786227318 type Work @default.
• W2786227318 sameAs 2786227318 @default.
• W2786227318 citedByCount "0" @default.
• W2786227318 crossrefType "dissertation" @default.
• W2786227318 hasAuthorship W2786227318A5053831284 @default.
• W2786227318 hasConcept C100609095 @default.
• W2786227318 hasConcept C104317684 @default.
• W2786227318 hasConcept C107457646 @default.
• W2786227318 hasConcept C120314980 @default.
• W2786227318 hasConcept C120665830 @default.
• W2786227318 hasConcept C121332964 @default.
• W2786227318 hasConcept C127413603 @default.
• W2786227318 hasConcept C139807058 @default.
• W2786227318 hasConcept C154945302 @default.
• W2786227318 hasConcept C169337768 @default.
• W2786227318 hasConcept C181335050 @default.
• W2786227318 hasConcept C185592680 @default.
• W2786227318 hasConcept C199505168 @default.
• W2786227318 hasConcept C201995342 @default.
• W2786227318 hasConcept C2780451532 @default.
• W2786227318 hasConcept C34413123 @default.
• W2786227318 hasConcept C41008148 @default.
• W2786227318 hasConcept C48044578 @default.
• W2786227318 hasConcept C55493867 @default.
• W2786227318 hasConcept C63479239 @default.
• W2786227318 hasConcept C77088390 @default.
• W2786227318 hasConcept C90509273 @default.
• W2786227318 hasConceptScore W2786227318C100609095 @default.
• W2786227318 hasConceptScore W2786227318C104317684 @default.
• W2786227318 hasConceptScore W2786227318C107457646 @default.
• W2786227318 hasConceptScore W2786227318C120314980 @default.
• W2786227318 hasConceptScore W2786227318C120665830 @default.
• W2786227318 hasConceptScore W2786227318C121332964 @default.
• W2786227318 hasConceptScore W2786227318C127413603 @default.
• W2786227318 hasConceptScore W2786227318C139807058 @default.
• W2786227318 hasConceptScore W2786227318C154945302 @default.
• W2786227318 hasConceptScore W2786227318C169337768 @default.
• W2786227318 hasConceptScore W2786227318C181335050 @default.
• W2786227318 hasConceptScore W2786227318C185592680 @default.
• W2786227318 hasConceptScore W2786227318C199505168 @default.
• W2786227318 hasConceptScore W2786227318C201995342 @default.
• W2786227318 hasConceptScore W2786227318C2780451532 @default.
• W2786227318 hasConceptScore W2786227318C34413123 @default.
• W2786227318 hasConceptScore W2786227318C41008148 @default.
• W2786227318 hasConceptScore W2786227318C48044578 @default.
• W2786227318 hasConceptScore W2786227318C55493867 @default.
• W2786227318 hasConceptScore W2786227318C63479239 @default.
• W2786227318 hasConceptScore W2786227318C77088390 @default.
• W2786227318 hasConceptScore W2786227318C90509273 @default.
• W2786227318 hasLocation W27862273181 @default.
• W2786227318 hasOpenAccess W2786227318 @default.
• W2786227318 hasPrimaryLocation W27862273181 @default.
• W2786227318 hasRelatedWork W1574344148 @default.
• W2786227318 hasRelatedWork W1902729993 @default.
• W2786227318 hasRelatedWork W2027546131 @default.
• W2786227318 hasRelatedWork W2034722654 @default.
• W2786227318 hasRelatedWork W2073184696 @default.
• W2786227318 hasRelatedWork W2084779340 @default.
• W2786227318 hasRelatedWork W2095900893 @default.
• W2786227318 hasRelatedWork W2122917127 @default.
• W2786227318 hasRelatedWork W2314363168 @default.
• W2786227318 hasRelatedWork W2556481766 @default.
• W2786227318 hasRelatedWork W2587484553 @default.
• W2786227318 hasRelatedWork W2738428716 @default.
• W2786227318 hasRelatedWork W2890021449 @default.
• W2786227318 hasRelatedWork W2916454130 @default.
• W2786227318 hasRelatedWork W3036084968 @default.
• W2786227318 hasRelatedWork W3100717233 @default.
• W2786227318 hasRelatedWork W3114601207 @default.
• W2786227318 hasRelatedWork W3176202740 @default.
• W2786227318 hasRelatedWork W1858981721 @default.
• W2786227318 hasRelatedWork W2154501513 @default.
• W2786227318 isParatext "false" @default.
• W2786227318 isRetracted "false" @default.
• W2786227318 magId "2786227318" @default.
• W2786227318 workType "dissertation" @default.