Automated Feature-Driven Flight Planning for Airborne Inspection of Large Linear Infrastructure Assets

Abstract

Inspecting large-scale physical infrastructure assets such as powerlines from the air involves capturing data of thousands of features of an asset and its nearby surroundings, which makes inspection flight planning challenging. The objective of this article is to address the problems of automatically generating and optimizing flight paths and mission plans for airborne inspection of large linear infrastructure assets. This article proposes an automated feature-driven flight planning method using metaheuristic optimization with consideration of aircraft dynamic, sensor, terrain, and airfield location constraints. Results for flight planning the inspection of one of the world’s largest above-ground powerline networks are shown. Automated planning can reduce the human planning workload and enable the cost-saving benefits of automated aircraft to be realized through efficient deployment. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —This work extends previous research and development of an automated aircraft guidance system for infrastructure inspection, by addressing the problem of automated flight planning. Often, deploying manned or automated aircraft reliably and efficiently for large-scale inspection is a challenge. Designing efficient flight paths and missions by hand is complex and time-consuming. Traditional “top-down” planning approaches use region or waypoint-based planning, neglecting the spatial layout of the assets to be inspected, possibly leading to more missions than required, inefficient flight paths, or missed data capture. Since inspecting a large asset with few aircraft can take many days, a solution was sought that can plan across multiple time scales (from second to second path planning to daily mission planning). A solution was designed that can build flight plans automatically from the “bottom-up” while optimizing performance objectives and satisfying constraints. Flight plans can either be used to aid pilots conducting an aerial inspection, as input to an automated line-tracking system for aircraft or UAVs or for estimating the cost of large-scale inspection operations. The presented method and results are motivated by the specific problem of large-scale powerline inspection and the approach may be applied to the inspection of other above-ground infrastructures, such as roads and railways.