Autonomous Reactive LiDAR-based Mapping for Powerline Inspection

Abstract

There is a strong demand in the automation of powerline inspection, which is currently performed in a two-stage process: 1) data collection by an aerial vehicle equipped with LiDARs, cameras, and other sensors, and 2) offline data analysis using processing and artificial intelligence algorithms. This procedure is very inefficient and in many cases requires repeating the flights if during the offline analysis the collected data is found to be of insufficient quality for the inspection. This paper proposes a reactive quadrotor-based online powerline inspection system. The proposed method: i) builds online an onboard an accurate preliminary 3D map of the environment, ii) performs an online analysis of the quality of the inspection data being obtained, and iii) in case the data has insufficient quality, commands the quadrotor to stop moving so that more data of that part of the environment can be integrated to increase the resolution and precision until the required quality is fulfilled. Hence, the proposed system maximizes the area inspected ensuring that the gathered data fulfils the specified quality metrics. The proposed system has been implemented onboard and aerial robot and validated in powerline inspection experiments conducted in environments with different types conditions and vegetation.