Mapping-Guided Task Discovery and Allocation for Robotic Inspection of Underwater Structures
Marina Ruediger, Ashis Banerjee
AI summary
Problem
Autonomous underwater inspections struggle with unreliable communications, limited visibility, and the need for prior environmental knowledge to plan coverage tasks effectively.
Approach
The method automatically generates and distributes inspection tasks by analyzing real-time SLAM meshes through a geometric scoring and pruning process, coordinated via a decentralized, communication-aware auction framework.
Key results
- Automatically discovers inspection tasks from real-time SLAM meshes without prior environmental knowledge
- Achieves superior spatial coverage while thoroughly inspecting geometrically complex regions
- Requires equal or fewer inspection tasks compared to Voronoi and boustrophedon coverage methods
- Validated in saline and freshwater tank experiments with a heterogeneous multi-robot team
Why it matters
Enables robust, low-cost autonomous underwater inspections in communication-constrained environments, improving defect detection for critical maritime infrastructure.
Abstract
This paper introduces Mapping-based Tasks for Inspection: Discovery and Allocation (Map-TIDAL), a method for generating environmentally informed tasks and distribut- ing them in a heterogeneous multi-robot system for visual inspection of underwater structures. Map-TIDAL leverages the individual robot maps generated during SLAM (without prior knowledge of the environment) and tasks from all the robots through a communication-aware auction process to determine additional inspection locations as the structures are further explored by the robots. This allows the method to adaptively focus on geometrically interesting areas that need detailed inspection while still maintaining good overall coverage with a reasonably small number of inspection tasks. Experiments on both saline and fresh water tanks show that Map-TIDAL yields better coverage while inspecting areas with interesting geometric features more thoroughly, using equal or fewer inspection locations compared to prevalent coverage methods using Voronoi distributions and boustrophedon patterns.