Abstract

In this work, a real-time collision avoidance algorithm was presented for autonomous navigation in the presence of fixed and moving obstacles in building environments. The current implementation is designed for autonomous navigation between waypoints of a predefined flight trajectory that would be performed by an UAV during tasks such as inspections or construction progress monitoring. It uses a simplified geometry generated from a point cloud of the scenario. In addition, it also employs information from 3D sensors to detect and position obstacles such as people or other UAVs, which are not registered in the original cloud. If an obstacle is detected, the algorithm estimates its motion and computes an evasion path considering the geometry of the environment. The method has been successfully tested in different scenarios, offering robust results in all avoidance maneuvers. Execution times were measured, demonstrating that the algorithm is computationally feasible to be implemented onboard an UAV.

Highlights

  • IntroductionThe use of UAVs (unmanned aerial vehicles), known as drones, has increased exponentially

  • In the last years, the use of UAVs, known as drones, has increased exponentially

  • The use of UAVs, known as drones, has increased exponentially. These vehicles have been introduced in many different engineering fields such as civil engineering [1,2] or photogrammetry and remote sensing [3]

Read more

Summary

Introduction

The use of UAVs (unmanned aerial vehicles), known as drones, has increased exponentially. These vehicles have been introduced in many different engineering fields such as civil engineering [1,2] or photogrammetry and remote sensing [3]. In the civil engineering field, UAVs have been used for many different applications [4]. Such as safety monitoring, inspection tasks, or progress monitoring of the construction. They are a powerful tool to capture aerial images used to detect unsafe conditions and for monitoring the well-being of the workforce [5]

Objectives
Methods
Results
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call