A Framework for Multiple Ground Target Finding and Inspection Using a Multirotor UAS

Sensors

Roshan Ragel¹  Felipe Gonzalez²  Jonathan Roberts²  Ajmal Hinas² 

[1] Department of Computer Engineering, University of Peradeniya (UOP), Peradeniya 20400, Sri Lanka;Robotics and Autonomous Systems, Queensland University of Technology (QUT), Brisbane City QLD 4000, Australia;

关键词: unmanned aerial vehicle;    unmanned aerial system;    vision-based navigation;    search and rescue;    vision and action;    ooda;    inspection;    target detection;    remote sensing;   

DOI  :  10.3390/s20010272

来源: DOAJ

【 摘 要 】

Small unmanned aerial systems (UASs) now have advanced waypoint-based navigation capabilities, which enable them to collect surveillance, wildlife ecology and air quality data in new ways. The ability to remotely sense and find a set of targets and descend and hover close to each target for an action is desirable in many applications, including inspection, search and rescue and spot spraying in agriculture. This paper proposes a robust framework for vision-based ground target finding and action using the high-level decision-making approach of Observe, Orient, Decide and Act (OODA). The proposed framework was implemented as a modular software system using the robotic operating system (ROS). The framework can be effectively deployed in different applications where single or multiple target detection and action is needed. The accuracy and precision of camera-based target position estimation from a low-cost UAS is not adequate for the task due to errors and uncertainties in low-cost sensors, sensor drift and target detection errors. External disturbances such as wind also pose further challenges. The implemented framework was tested using two different test cases. Overall, the results show that the proposed framework is robust to localization and target detection errors and able to perform the task.

【 授权许可】

Unknown