Information-driven source search using cooperative mobile sensor platforms

Information-driven source search strategies are widely employed to search for invisible hazardous gas sources in turbulent environments due to their robustness against noisy measurements. Based on the Bayesian framework, these strategies estimate source terms (e.g., source location, release strength) and enable mobile sensors to minimize uncertainty in these estimations. Various information-driven search strategies for outdoor and indoor environments, as well as for multiple agents and multiple sources, have been developed and validated through real-world experiments.

Gaussian Process-based active & adaptive sampling strategies for mobile sensor networks

Gaussian Processes (GPs) allow for effective modeling and prediction of spatial and temporal phenomena. By leveraging GPs, the active & adaptive sampling methods help sensors collect high-quality data efficiently by adapting to their environment.

Efficient 3D Reconstruction Strategy Utilizing UAV

UAVs provide a versatile platform for acquiring visual data for 3D modeling. By utilizing information gain and optimization, UAV-based path planning enables efficiently capturing detailed spatial information and accurately reconstructing structures while reducing operational costs and time.