Research on the Control Method of Unmanned Helicopter Under the Background of Artificial Intelligence
Wenze Li,Xuanzi Zhang,Bo Huang,Yi Chen,Ruiqi Zhang,S. BalaMurugan
DOI: https://doi.org/10.1142/s0219265921430192
2021-12-06
Journal of Interconnection Networks
Abstract:A synopsis of a multidisciplinary research initiative focused on critical strategies for the genuine independent flight of tiny, vertical flying take-off (VTOL) unmanned aerial vehicles (UAVs). The research activities are the flight testbed, a simulation and test environment, and integrated components for onboard navigation, perception, design, and control. The necessity to create an unmanned helicopter system in different new civil applications cannot be overlooked. A highly reliable model may be used in the design, analysis, and implementation. The helicopter is fitted with a reference system for flight test data measurement and recording attitude heading reference system (AHRS) and the accompanying data storage modules. Recently, artificial intelligence-based deep learning (DL) has demonstrated excellent outcomes for a wide range of robotic activities in the areas of perception, planning, location, and management. Its remarkable skills to learn from complex data obtained in actual surroundings make it appropriate for many autonomous robotic applications. At the same time, UHS is currently widely utilized in various civil tasks in security, cinematography, disaster assistance, package delivery, or warehouse management (Unmanned Helicopter System). This paper conducted detailed work on current applications and the most significant advances and their performance and limits for the DL-UHS method. Furthermore, the essential strategies for deep learning are explained in depth — finally, discussing the principal hurdles of applying deep learning for UHS solutions. The proposed DL-UHS enhance outcome to evaluate the control strategies for the unmanned helicopter to achieve the low signal to noise error ratio of 31.3%, the error rate of 33.6%, the high-performance ratio of 91.4%, enhance accurate path planning 97.5%, prediction ratio of 96.3%, less trajectory cost ratio of 17.8% and increased safety tracking rate 93.6% when compared to other popular methods.