GATSBI: An Online GTSP-Based Algorithm for Targeted Surface Bridge Inspection
This work addresses the problem of efficient and thorough autonomous bridge inspection for infrastructure maintenance, which is significant for civil engineering and robotics communities. It offers an incremental improvement over existing exploration methods.
This paper presents GATSBI, an online path planning algorithm for UAVs to inspect bridge surfaces for defects without prior geometric models. GATSBI uses online LiDAR scans to build a 3D occupancy map, semantically segments bridge voxels, and then formulates a Generalized Traveling Salesperson Problem (GTSP) to plan an inspection path. The algorithm replans in a receding horizon fashion as new environmental data becomes available, leading to a more efficient and thorough inspection compared to a classical exploration algorithm.
We study the problem of visual surface inspection of a bridge for defects using an Unmanned Aerial Vehicle (UAV). We do not assume that the geometric model of the bridge is known beforehand. Our planner, termed GATSBI, plans a path in a receding horizon fashion to inspect all points on the surface of the bridge. The input to GATSBI consists of a 3D occupancy map created online with LiDAR scans. Occupied voxels corresponding to the bridge in this map are semantically segmented and used to create a bridge-only occupancy map. Inspecting a bridge voxel requires the UAV to take images from a desired viewing angle and distance. We then create a Generalized Traveling Salesperson Problem (GTSP) instance to cluster candidate viewpoints for inspecting the bridge voxels and use an off-the-shelf GTSP solver to find the optimal path for the given instance. As the algorithm sees more parts of the environment over time, it replans the path to inspect novel parts of the bridge while avoiding obstacles. We evaluate the performance of our algorithm through high-fidelity simulations conducted in AirSim and real-world experiments. We compare the performance of GATSBI with a classical exploration algorithm. Our evaluation reveals that targeting the inspection to only the segmented bridge voxels and planning carefully using a GTSP solver leads to a more efficient and thorough inspection than the baseline algorithm.