A real-time waypoint-based 3D local path planning algorithm is proposed for obstacle avoidance using the optical flow obtained by a frontal monocular camera mounted on a quadrotor UAV. The algorithm accounts for vertical and horizontal obstacle avoidance, as well as for avoidance of frontally approaching obstacles. Implementation and testing are carried out in the ROS environment and the algorithm effectiveness is demonstrated via Gazebo simulations. Realtime algorithm performance is also assessed through software profiling and in terms of worst case execution time using the NVIDIA Jetson TX1 and RaspberryPi 4 for hardware-in-the-loop (HIL) tests.
Quadrotor UAV 3D Path Planning with Optical-Flow-based Obstacle Avoidance
15.06.2021
4336902 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Optical Flow Based Obstacle Avoidance and Path Planning for Quadrotor Flight
British Library Conference Proceedings | 2018
|Quadrotor Research Platform with Obstacle Avoidance
Trans Tech Publications | 2016
|Laser-Guided Quadrotor Obstacle Avoidance
AIAA | 2015
|