This article presents the design of a control strategy for the Calais canal, a navigation canal located in a lowland area in northern France that is affected by tides. Moreover, the available actuators are discrete-valued and the hierarchy of operational objectives is time-varying. All these circumstances render water level regulation of the Calais canal a challenging problem. In view of this situation, the design of the overall control architecture is divided into a sequence of structured tasks, which are distributed among layers. The upper layer determines the current operating mode based on the analysis of several environmental and operational aspects. Information regarding the current mode is taken into account at the intermediate layer to select the appropriate optimization-based control problem, which is solved using lexicographic minimization. The optimal control setpoints are determined and sent to the lower layer, where scheduling problems are solved to select low-level control actions from a finite set to minimize the mismatch with respect to the optimal setpoints. Different realistic simulation scenarios are tested to demonstrate the effectiveness of the proposed approach.

This work was supported in part by the Researchlab Autonomous Shipping (RAS), Delft University of Technology; and in part by the Spanish State Research Agency (AEI) and the European Regional Development Fund (ERDF) through the Project Safe Coordination of Autonomous Vehicles (SaCoAV) under Grant MINECO PID2020-114244RB-I00. Recommended by Associate Editor G. Papafotiou

Peer reviewed