Accurate monitoring of urban waterlogging contributes to the city’s normal operation and the safety of residents’ daily travel. A common method is to forecast the city's global waterlogging status using its partial waterlogging data. However, This method has two challenges: first, existing predictive algorithms are either driven by knowledge or data alone; and second, the partial waterlogging data is not collected selectively.
To solve the problems, a research team led by Jingbin Wang published their new research on 15 August 2024 in Frontiers of Computer Science co-published by Higher Education Press and Springer Nature.
The team proposed a framework for large-scale and fine-grained spatiotemporal waterlogging monitoring based on the opportunistic sensing of limited bus routes. This framework follows the Sparse Crowdsensing and mainly comprises a pair of iterative predictor and selector. The experimental results on real waterlogging and bus routes in Shenzhen show that the proposed framework could easily perform urban waterlogging monitoring with low cost, high accuracy, wide coverage, and fine granularity.
In the research, they follow the idea of sparse crowdsensing, which exploits the spatiotemporal correlation between data sensed from different subregions. Their framework consists of two main modules, predictor and selector, and an iterative process.
Firstly, the predictor is comprised of a continuously evolving graph convolutional network. The predictor uses the waterlogging data in the past time period to predict the probability of waterlogging in the future. As for selector, it has two stages: areas set selection based on uncertainty and representativeness as evaluation criteria; bus routes set selection based on maximal greedy coverage.
As shown in Fig.2, the iterative process includes the selection of bus routes at each time period to assign perceptual tasks, and update the predictor according to the predicted state of the unperceived region and part of the real state of the perceived region.
Future work can focus on the further analysis of the influencing factors of the waterlogging state between each area in order to improve the monitoring effect.
DOI: 10.1007/s11704-023-2714-8