Mission Planning and Decision Support for Underwater Glider Networks: A Sampling on-Demand Approach

Area of Science:

• Oceanography
• Marine Robotics
• Geospatial Analysis

Background:

• Optimal sampling strategies are crucial for efficient underwater glider missions.
• Traditional methods focus on minimizing overall uncertainty, which may not align with specific mission objectives.
• Current mission planning for glider fleets is complex and labor-intensive.

Purpose of the Study:

• To develop a flexible optimal sampling approach for underwater glider fleet mission planning.
• To enable scientists to define specific uncertainty targets for data collection.
• To introduce a new optimality criterion (A η) and a supporting decision support system.

Main Methods:

• Proposed a "sampling on-demand" approach with user-defined objective maps for uncertainty.
• Introduced a novel optimality criterion, A η.
• Utilized a Simulated Annealing-based optimizer to solve the minimization problem.
• Incorporated glider navigation constraints, path geometry, and ocean current effects.
• Developed the Sampling on-Demand and Decision Support (SoDDS) MATLAB toolbox.

Main Results:

• The SoDDS toolbox effectively implements the sampling on-demand strategy.
• Demonstrated successful application of the approach in three distinct scenarios.
• The A η criterion and optimizer successfully addressed mission constraints.
• SoDDS automates data acquisition and simplifies user input.

Conclusions:

• The sampling on-demand approach offers a significant advancement in systematic glider fleet mission planning.
• SoDDS reduces the effort required for mission planning and enhances data collection efficiency.
• This method supports achieving mission-specific scientific and operational goals through tailored uncertainty reduction.