Sensor Fusion for Autonomous Indoor UAV Navigation in Confined Spaces

TL;DR

This paper presents a sensor fusion system combining depth, IMU, and LiDAR data for autonomous indoor UAV navigation, achieving high accuracy and efficiency in GPS-denied confined environments.

Contribution

The work introduces a novel sensor fusion approach integrated with ROS and RTAB-Map, demonstrating improved navigation accuracy and resource efficiency in indoor UAV operations.

Findings

Navigation error as low as 0.4 meters

Mapping RMSE of 0.13 meters

Flight orientation error of 0.1%

Abstract

In this paper, we address the challenge of navigating through unknown indoor environments using autonomous aerial robots within confined spaces. The core of our system involves the integration of key sensor technologies, including depth sensing from the ZED 2i camera, IMU data, and LiDAR measurements, facilitated by the Robot Operating System (ROS) and RTAB-Map. Through custom designed experiments, we demonstrate the robustness and effectiveness of this approach. Our results showcase a promising navigation accuracy, with errors as low as 0.4 meters, and mapping quality characterized by a Root Mean Square Error (RMSE) of just 0.13 m. Notably, this performance is achieved while maintaining energy efficiency and balanced resource allocation, addressing a crucial concern in UAV applications. Flight tests further underscore the precision of our system in maintaining desired flight orientations, with a remarkable error rate of only 0.1%. This work represents a significant stride in the development of autonomous indoor UAV navigation systems, with potential applications in search and rescue, facility inspection, and environmental monitoring within GPS-denied indoor environments.