Water quality monitoring through remote sensing with UAVs is best conducted using multispectral sensors; however, these sensors are expensive. We aimed to predict multispectral bands from a low-cost sensor (R, G, B bands) using artificial neural networks (ANN). We studied a lake located on the campus of Unisinos University, Brazil, using a low-cost sensor mounted on a UAV. Simultaneously, we collected water samples during the UAV flight to determine total suspended solids (TSS) and dissolved organic matter (DOM). We correlated the three bands predicted with TSS and DOM. The results show that the ANN validation process predicted the three bands of the multispectral sensor using the three bands of the low-cost sensor with a low average error of 19%. The correlations with TSS and DOM resulted in R2 values of greater than 0.60, consistent with literature values.

No abstract

Smart cities emergence has allowed a wide variety of technological services to metropolitan areas. These services can improve life quality, minimize environmental impacts, improve health service, improve security, and bear the increasing number of people in the cities. Life quality encompasses many subjects, and accessibility for People with Disabilities (PwD) is one. In this article, smart cities focused on helping PwD are called Assistive Smart Cities (ASCs). In this sense, the article proposes a Model for Assistive Smart Cities called MASC. Related works do not cover geographically broad areas, such as cities and metropolitan regions. Moreover, they are not generic in terms of disabilities and are usually intended only for one type of disability. Given this scenario, the MASC covers large regions and supports various disabilities, such as hearing, visual impairment, and limitation of lower limb movements. Unlike the related works, MASC uses the interactions of PwD to compose histories of contexts offered as services. MASC proposes an ontology-based on ubiquitous accessibility concepts. The model evaluation focused on performance, functionality, and usability. Performance and functionality evaluations were performed using data generated by a context simulator called Siafu and data from the Open Street Maps (OSM) platform. Usability was evaluated using a smart wheelchair prototype. The results of usability show 96% acceptance regarding ease of use and 98% regarding system utility. The results indicate that the model supports massive applications, managing information to generate trails. Besides, MASC provides services for different types of users, namely PwD, healthcare professionals, and public administration.

No abstract

The Shortest Path (SP) problem resembles a variety of real-world situations where one needs to find paths between origins and destinations. A generalization of the SP is the Dynamic Shortest Path (DSP) problem, which also models changes in the graph at any time. When a graph changes, DSP algorithms partially recompute the paths while taking advantage of the previous computations. Although the DSP problem represents many real situations, it leaves out some fundamental aspects of decision-making. One of these aspects is the existence of multiple, potentially conflicting objectives that must be optimized simultaneously. Recently, we performed a first incursion on the so-called Multi-Objective Dynamic Shortest Path (MODSP), presenting the first algorithm able to take the MODM perspective into account when solving a DSP problem. In this paper, we go beyond and formally define the MODSP problem, thus establishing and clarifying it with respect to its simpler counterparts. In particular, we start with a brief overview of the related literature and then present a complete formalization of the MODSP problem class, highlighting its distinguishing features as compared to similar problems and representing their relationship through a novel taxonomy. This work also motivates the relevance of the MODSP problem by enumerating real-world scenarios that involve all its ingredients, such as multiple objectives and dynamically updated graph topologies. Finally, we discuss the challenges and open questions for this new class of shortest path problems, aiming at future work directions. We hope this work sheds light on the theme and contributes to leveraging relevant research on the topic.