Mobile sensing for estimation of hydro-dynamic parameters for minimally gauged open channels

To ensure fair distribution of surface water among open water channels, it is important to monitor these channels at a high spatial and temporal resolutions. However, channel monitoring is a challenging task if the channels are minimally gauged, i.e. they have limited water level and water flow sensors to describe channel dynamics. Such sensors are generally installed at the control structures (gates) along a channel. If control structures are distant apart from one another then these sensors cannot properly describe the water levels and water velocities in a channel at high spatial resolution. Moreover, it is economically infeasible to install these sensors throughout the channels as they require permanent installation and infrastructure costs. Therefore, inspired by the low-cost mobile sensing and its portability aspect, this paper provides the formulation of a framework that accommodates the GPS measurements of mobile sensing to provide the channel dynamics at a higher resolution. We demonstrate the effectiveness of the proposed framework in comparison to the frameworks with fixed sensors using an experimental dataset obtained from an irrigation channel in Lahore, Pakistan. The proposed framework employs the state-dependent interacting multiple models (SD-IMM) technique and the estimation results show improvements in comparison to the results obtained by the fixed sensors along the channel.