A Network-Based Visual Analytics Approach for Performance Evaluation of Swarms of Robots in the Surveillance Task

Autor(es)

Linhares, Claudio D. G. ; Tinoco, Claudiney R. ; Ponciano, Jean R. ; Oliveira, Gina M. B. ; Travencolo, Bruno A. N. ; Xavier-Junior, JC ; Rios, RA

• Tipo de documento: Artigo Científico
• Fonte: INTELLIGENT SYSTEMS, PT I; v. 13653, p. 16-pg., 2022-01-01.

Resumo

Effectiveness in swarm robotics relies on aspects such as coordination and collective knowledge about the environment. By considering the evolution of intra-swarm communications over time as a temporal network, different strategies can be used in the data analysis. Information visualisation techniques are useful in this context because they can enhance the analysis of individual and global performances by including the user in the data exploration. This work proposes a visual analytics approach that considers a new matrix-based layout and other well-established ones to assess the swarm's efficiency. To analyse this approach, we also propose a temporal network dataset that models the evolution of the communications of a swarm of robots in the surveillance task, including eventual failures. We performed visual analyses in this network and demonstrated that the proposed approach allows easy identification of patterns, trends, and anomalies related to communication and task evolution. As a consequence, the decision-making process and eventual adjustments become faster and more reliable. (AU)

• Processo FAPESP: 20/07200-9 - Analisando dados complexos vinculados a COVID-19 para apoio à tomada de decisão e prognóstico
• Beneficiário: Agma Juci Machado Traina
• Modalidade de apoio: Auxílio à Pesquisa - Regular
• Processo FAPESP: 16/17078-0 - Mineração, indexação e visualização de Big Data no contexto de sistemas de apoio à decisão clínica (MIVisBD)
• Beneficiário: Agma Juci Machado Traina
• Modalidade de apoio: Auxílio à Pesquisa - Temático
• Processo FAPESP: 20/10049-0 - Redes complexas e recuperação de imagens por conteúdo apoiadas por características de atenção visual seletiva
• Beneficiário: Cláudio Douglas Gouveia Linhares
• Modalidade de apoio: Bolsas no Brasil - Pós-Doutorado