LiDAR-based Quadrotor Autonomous Inspection System in Cluttered Environments

TL;DR

This work tackles reliable UAV inspections in GNSS-denied, cluttered environments by coupling a LiDAR-based quadrotor with a dual-phase workflow: a human-in-the-loop phase for 3D mapping and inspection-point recording, followed by autonomous execution of optimized tasks. The system integrates LiDAR-SLAM (FAST-LIO2), vegetation-friendly mapping (ROG-Map), and a two-tier planning stack (IPC for real-time navigation and TSP/A* for task sequencing), enabling untrained operators to initiate inspections that are subsequently performed autonomously. Field experiments across slopes, landslides, agriculture, manufacturing, and forestry demonstrate that autonomous inspection significantly reduces trajectory length and flight time compared with human-in-the-loop operations, while preserving data quality and safety through robust obstacle avoidance and relocalization. The approach offers a practical, scalable solution for UAV-based inspections in safety-critical and resource-constrained settings, with potential applicability to industrial, environmental, and agricultural monitoring.

Abstract

In recent years, autonomous unmanned aerial vehicle (UAV) technology has seen rapid advancements, significantly improving operational efficiency and mitigating risks associated with manual tasks in domains such as industrial inspection, agricultural monitoring, and search-and-rescue missions. Despite these developments, existing UAV inspection systems encounter two critical challenges: limited reliability in complex, unstructured, and GNSS-denied environments, and a pronounced dependency on skilled operators. To overcome these limitations, this study presents a LiDAR-based UAV inspection system employing a dual-phase workflow: human-in-the-loop inspection and autonomous inspection. During the human-in-the-loop phase, untrained pilots are supported by autonomous obstacle avoidance, enabling them to generate 3D maps, specify inspection points, and schedule tasks. Inspection points are then optimized using the Traveling Salesman Problem (TSP) to create efficient task sequences. In the autonomous phase, the quadrotor autonomously executes the planned tasks, ensuring safe and efficient data acquisition. Comprehensive field experiments conducted in various environments, including slopes, landslides, agricultural fields, factories, and forests, confirm the system's reliability and flexibility. Results reveal significant enhancements in inspection efficiency, with autonomous operations reducing trajectory length by up to 40\% and flight time by 57\% compared to human-in-the-loop operations. These findings underscore the potential of the proposed system to enhance UAV-based inspections in safety-critical and resource-constrained scenarios.

Figures (16)

• Figure 1: The system overview of our LiDAR-based quadrotor.
• Figure 2: Different views of our LiDAR-based quadrotor.
• Figure 3: Inspection points optimization in the 2D case. By solving the Traveling Salesman Problem (TSP), the sequence of inspection points under human-in-the-loop inspection (green lines with yellow stars) is optimized, resulting in the shortest total path (purple lines with yellow stars), thereby significantly improving efficiency in autonomous inspection.
• Figure 4: Overview of the quadrotor inspection system in slope task. (a) Point cloud map constructed during the inspection. (b) Image captured of the corrosion of the wire ropes supporting the barriers. (c) Image captured of the accumulation of debris behind barriers. (d) Image captured of the condition of the blocked drainage pipe and corrosion within the concrete layers.
• Figure 5: First-person view photos collected during autonomous inspection. Our quadrotor successfully avoided small obstacles, demonstrating the robustness of the navigation system.