Unmanned Aerial Vehicle (UAV)-based autonomous damage inspection has become a promising technique to replace certain human-performed inspections due to both safety and cost considerations. While current UAV-based damage inspection methods have been successful in minimizing the length of the flying path considering potential defect locations, the impact of flying path on the quality of structural health monitoring (SHM) has been overlooked. This paper proposes a novel Bayes risk-based mission planning method for UAV-based damage inspection by minimizing not only the UAV path length, but also the associated SHM costs. A functional relationship is first established between the UAV flying path length and damage inspection parameters, such as overlap ratio and inspection distance between the UAV and the target structure. The impact of UAV inspection parameters on associated SHM costs is then analyzed based on Bayes risk. Building upon the formulated functions, a multi-objective optimization model is developed to optimize the inspection parameters and thereby achieve a tradeoff between path length of UAV flight and associated SHM costs (e.g., consequence costs of SHM decisions resulting from the UAV inspection). An example of damage detection on a miter gate is employed to demonstrate the proposed method. The effect of different weighting factors on flying path and SHM costs is also evaluated. The results show that the proposed approach can effectively perform UAV mission planning while accounting for the impact of UAV flying path on SHM costs.
Get the paper here.
