AIP921-FA26-WE-NTN-AIWaterMonitoringSys
G7+, Wednesdays, 3:30 PM – 5:30 PM, Newton


Program Description
Course Details: • Name: AI Powered Water Monitoring System • Date: August 19, 2026 – November 18, 2026 • Time: Wednesdays, 3:30 PM – 5:30 PM • Location: Newton Classroom • Grades: G7+ • Prerequisite: None This 1-on-1 research and engineering course guides students through the development of an AI-powered embedded water monitoring system for early Harmful Algal Bloom (HAB) detection. Over approximately three months, the student will investigate HABs and water-quality indicators, conduct scientific research, formulate a hypothesis, and design an experimental methodology. The student will develop a prototype integrating an underwater camera, embedded AI platform, and environmental sensors such as temperature and turbidity, with optional pH and dissolved oxygen sensing. Using collected and curated data, the student will train and evaluate computer-vision models and investigate whether combining imagery with environmental sensor data improves bloom-risk assessment. The course follows an authentic research and engineering workflow emphasizing independent problem solving, experimentation, technical decision-making, data analysis, and scientific communication. Final deliverables include an embedded AI prototype, experimental dataset, research report, science fair poster, and technical presentation. Learning Outcomes: Students will be able to: Explain HABs and key water-quality indicators including temperature, turbidity, pH, and dissolved oxygen. Conduct a scientific literature review and formulate a research question, hypothesis, variables, and experimental methodology. Collect, label, preprocess, and augment image and sensor datasets. Train and evaluate computer-vision models using accuracy, precision, recall, F1 score, and confusion matrices. Compare image-only and multimodal image-plus-sensor approaches. Design and integrate an embedded system incorporating cameras, environmental sensors, and an AI computing platform. Acquire, synchronize, and analyze experimental data and perform systematic error analysis. Apply an iterative design-build-test-improve engineering process and troubleshoot hardware, software, and AI issues. Evaluate system performance, limitations, uncertainty, and engineering tradeoffs. Communicate results through a scientific report, research poster, technical presentation, and judge-style Q&A.
Program Location
Vinci Newton
1121 Washington Street ste 3, West Newton, MA 02465, USA
.png)