Andres Rico

The increasing expansion of ubiquitous sensing systems has created large streams of time-series data that are difficult for non-technical users to interpret. Large Language Models (LLMs) offer a promising interface for transforming sensor data into natural language insights, particularly in distributed environments where users may lack familiarity with data analysis. However, models optimized for text generation often struggle to interpret raw time-series signals, producing responses that are generic, inaccurate, or poorly grounded in the data. This study evaluates a prompt structure based on the Retrieval-Augmented Generation (RAG) framework for interpreting sensor-derived time-series data from water-consumption monitoring systems installed in household storage tanks. The prompt integrates statistical summaries, sensor metadata, and contextual information about household water-use practices. Performance is evaluated using synthetic datasets representing a year of tank water-consumption measurements and a rubric-based evaluation framework applied by three independent language-model evaluators. Results show that augmenting prompts with structured contextual information improves the clarity and grounding of language model responses to sensor time-series data, increasing evaluation scores and reducing failure modes such as hallucination, contradiction with the data, and misuse of contextual information, as assessed by independent evaluator models. These findings highlight the potential of structured contextual prompting to support locally deployed language models that produce reliable and actionable interpretations of sensor time-series data.

This work presents the design and fabrication of MAIA, an open-source, modular, low-cost, and portable bioreactor for democratizing the development of synthetic biology based projects for urban settings. The integration of open-source synthetic biology (synbio) tools in a city’s infrastructure planning and design is crucial for addressing the great challenges related to urbanization. Synbio tools have great potential to help us complement our current sensing and actuating urban infrastructure. The MAIA reactor controls bacterial growth variables, making it suitable for cell-based experiments while reducing the need for expensive laboratory equipment. Its low-cost and open-source design allow for easy replication and modification, making it accessible to a broader audience. Its portability makes it suitable for use outside of traditional laboratory settings. We qualitatively and quantitatively validated the reactor’s capability to support cell growth, stimulate gene expression, and act as a creative tool for students and users.

Rapid urbanization has intensified pressures on global water systems, particularly impacting informal settlements. Understanding water usage patterns within these settlements is of importance for better addressing water scarcity issues. Current methods for gaining information about water within these settings tend to lack spatio-temporal granularity and miss complex patterns of behavior related to water usage. As a consequence, there is a shortage of the reliable quantitative measurements needed to improve water management processes and modeling. Here we introduce a low-cost sensing platform for water assessment in informal settlements. Households within these types of settlements, lacking water utility connections and piping, often use storage tanks and buckets to distribute, store, and consume water; hence, the platform consists of four distinct sensor modules that can be placed on these types of water infrastructure. Evaluated in controlled settings, the sensors prove to be reliable for measuring water quantity, quality, and usage. Field testing within an informal community in Mexico reveals that the system can comprehensively track multiple tank storage levels, assess water quality, and capture bucket usage patterns without disrupting a household’s common activities or infrastructure. Our validation shows the technique’s potential to improve water management in informal communities, while opening opportunities for enhancement of water-related research and policy making through combinations of top-down and bottom-up interventions.

Rapid urbanization has intensified pressures on global water systems, particularly impacting informal settlements. Understanding water usage patterns within these settlements is of importance for better addressing water scarcity issues. Current methods for gaining information about water within these settings tend to lack spatio-temporal granularity and miss complex patterns of behavior related to water usage. As a consequence, there is a shortage of the reliable quantitative measurements needed to improve water management processes and modeling. Here we introduce a low-cost sensing platform for water assessment in informal settlements. Households within these types of settlements, lacking water utility connections and piping, often use storage tanks and buckets to distribute, store, and consume water; hence, the platform consists of four distinct sensor modules that can be placed on these types of water infrastructure. Evaluated in controlled settings, the sensors prove to be reliable for measuring water quantity, quality, and usage. Field testing within an informal community in Mexico reveals that the system can comprehensively track multiple tank storage levels, assess water quality, and capture bucket usage patterns without disrupting a household's common activities or infrastructure. Our validation shows the technique's potential to improve water management in informal communities, while opening opportunities for enhancement of water-related research and policy making through combinations of top-down and bottom-up interventions.

Diversity in building types and evolving usage patterns within homes, offices and schools can be a big hurdle for smart city infrastructure. Smart sensors are stifled by long calibration procedures, fixed pre-programmed decision making and the need for large labeled data sets. These problems introduce trade offs between system accuracies, scalability and adaptability for Smart Building IoT infrastructure. This paper presents Chameleon, an adaptive sensor fusion and hybrid machine learning architecture that is able to classify room activity states. We test the performance in a classroom environment and an office environment with completely distinct geographical and layout characteristics. We evaluate the system with a total of eight weeks of experiment data and a total of 16 training and clustering sequences. Experiments show that the system is able to correctly classify activity states with accuracies in the range of 87%-99% on test sets. We show that the system is able to keep its high accuracies in a way that makes it adaptable to different rooms and scalable due to minimal hardware and software requirements.