Year: 2025
Type: article
Abstract: The paper aims to simulate UAVs in complex network environments to evaluate their performance in obstacle avoidance using stereo sensors. Since there are strict security regulations limiting real-world UAV flight testing, this research addresses the critical need for comprehensive simulation-based approaches.. The system integrates real-time obstacle detection and avoidance algorithms, enabling UAVs to navigate safely in environments with obstacles. Developed on the ROS framework with PX4 SITL and Gazebo, the system supports comprehensive end-to-end testing, from stereo image acquisition to UAV navigation using the 3DVFH* algorithm. It utilizes the MAVLink protocol for control and QGroundControl for monitoring. Simulation results confirm the system’s effectiveness, establishing a solid foundation for advanced autonomous UAV navigation in real-world applications.
Institution: University of Da Nang
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4415040235
FWCI:
Cites:
Cited by:
Related to: