This paper comprehensively reviews Robot Fleet Management (RFM), detailing current practices, challenges, and advancements in this critical domain. With the expansion of automation across industries such as manufacturing, logistics, and healthcare, effective coordination of robot fleets has become pivotal. This review focuses on key aspects such as task allocation, trajectory planning, and communication strategies, identifying prominent applications in industrial and delivery settings while exploring emerging sectors, such as marine operations and hospital environments. Challenges, such as dynamic interaction management and scalability, were examined, emphasizing the practical validation of algorithms and the integration of heterogeneous fleets. The literature was selected following the PRISMA methodology based on studies retrieved from Scopus, Web of Science, and the IEEE Xplore digital library. This study provides actionable insights into RFM methodologies and identifies research gaps, highlighting the need for innovative solutions for scalability, communication robustness, and real-world application testing. The findings contribute significantly to the advancement of autonomous systems and the optimization of logistical operations, offering a foundational resource for academics and practitioners driving innovation in multi-robot systems.

Unmanned Aerial Vehicle (UAV) has been increasingly employed in several missions with a pre-defined path. Over the years, UAV has become necessary in complex environments, where it demands high computational cost and execution time for traditional algorithms. To solve this problem meta-heuristic algorithms are used. Meta-heuristics are generic algorithms to solve problems without having to describe each step until the result and search for the best possible answer in an acceptable computational time. The simulations are made in Python, with it, a statistical analyses was realized based on execution time and path length between algorithms Particle Swarm Optimization (PSO), Grey Wolf Optimization (GWO) and Glowworm Swarm Optimization (GSO). Despite the GWO returns the paths in a shorter time, the PSO showed better performance with similar execution time and shorter path length. However, the reliability of the algorithms will depend on the size of the environment. PSO is less reliable in large environments, while the GWO maintains the same reliability.

O presente artigo apresenta, analisa e discute por que e como pode se pensar a inserção da tecnologia atrelada às ciências e à computação para crianças de 4 e 5 anos na educação infantil. As reflexões aqui exibidas amparam o trabalho realizado desde 2018 no Projeto Aprendendo e Experimentando Tecnologia e Ciências na Educação de Crianças de 4 e 5 Anos. Procura-se deixar de lado o pensamento que reduz o contato com a tecnologia a tablets, celulares ou mesmo montagem de kits de robótica. Nesse sentido, calcados pelos campos da neurociência e da teoria histórico-cultural para pensar o desenvolvimento infantil e pelo conceito de usuário ativo, i.e., aquele que utiliza e cria tecnologias computacionais de forma crítica, significativa, reflexiva e ética, compreendemos que a tecnologia envolve todas as criações humanas que alteram os ambientes natural, cultural e social no qual nos encontramos. Reconhecemos os perigos para o desenvolvimento pleno das crianças que envolvem ao mesmo tempo o reducionismo operado no contato com a tecnologia e propomos um trabalho que não desvincule ciências, matemática, design, engenharia e computação centrado em elementos do cotidiano das crianças.