Measurement of spatiotemporal varying field remains a challenge today, given existing robotics and sensor technologies, with their constraints in sensor speed and cost.
In order to form an economically viable, heterogenous monitoring system for big data collection in reservoirs, we focus on developing multi-robot adaptive path planning algorithms and cost effective smart sensor packages. This monitoring system consists of smart robots, both drifters and static nodes to improve monitoring efficiency and accuracy.
Additionally, we also extend the path-planning to include simultaneous water collection and field estimation, and work with an industry partner to develop a novel cost-effective crystal-based spectroscopy for chlorophyll a sensing. Thus we hope to upgrade the capability and intelligence of NUSwan and other similar environmental robots.