http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Giancarlo Eder Guerra Padilla,Seong-Hwan Kim,Kee-Ho Yu 제어로봇시스템학회 2018 제어로봇시스템학회 국제학술대회 논문집 Vol.2018 No.10
In the occurrence of natural disasters or any other event that causes damage to the communication system’s infrastructure of a city, logistics and rescue works tend to lack in countermeasures to assure a stable communication when needed. This study is aimed to developing a collision-free optimal flight path of a solar-powered UAV, exploiting its capability for long endurance flights, using it as a Low Altitude Platform (LAP) for communication relay. The model considers the solar incidence angle, aircraft attitude, influence of Artificial Potential Fields, and Free-space transmission signal. Apart from this considerations, a flight altitude restriction is defined and bounded to maximum antenna range. As a result of the simulation, a collision-free flight path considering antenna specifications for communication relay was tested resulting in the acquisition of significant data regarding the feasibility of the approach.
Reinforcement Learning based Flight Path Planning of a Solar Powered UAV
Giancarlo Eder Guerra Padilla,Kun-Jung Kim,Kee-Ho Yu 제어로봇시스템학회 2021 제어로봇시스템학회 각 지부별 자료집 Vol.2021 No.12
This paper presents the analysis of the flight path of a solar-powered Unmanned Aerial Vehicle (UAV) using Reinforcement Learning (RL) algorithm as a preliminary study for the optimal flight path planning considering weather conditions. The RL algorithm has the advantage of being able to learn from online simulations as well as offline hardware equipment. For the flight path generation a RL algorithm was trained using the UAV"s equations of motion as environment parameters. First, the UAV"s kinematic and dynamic equations of motion as environment parameters. First, the UAV"s kinematic and dynamic equations were specified in order to train the model. Then, using a critic-actor learning method for the agent model, the policy was obtained as a function of the state observations and rewards. The learning process was carried out with a waypoint follower problem, and the results analyzed.
도심 기류를 고려한 A* 알고리즘 기반 UAM 비행경로계획 기법
김민창,Giancarlo Eder Guerra Padilla,유기호 제어·로봇·시스템학회 2023 제어·로봇·시스템학회 논문지 Vol.29 No.11
Urban air mobility (UAM) is widely anticipated as a solution to urban traffic challenges, and numerous companies, including Joby Aviation and Airbus, are actively developing aircraft for this purpose. However, there are several challenges associated with commercializing UAM, such as obtaining airworthiness certification and establishing the necessary systems and policies. Moreover, UAM operates, especially with low-weight electric Vertical Take-off and Landing (eVTOL) aircraft are susceptible to the influence of winds. To address the issue of wind influence, this study presents a method for predicting urban airflow and flight path planning that circumvents wind-hazard areas. Since real-time measurements often fall short in predicting urban airflow accurately, computational fluid dynamics software, OpenFOAM was used to obtain urban airflow data. We have adapted the A* algorithm to create flight paths that avoid wind-hazard areas. The path planning based on A* was conducted according to the defined simulation scenario. Finally, the generated flight paths were compared, and their distinctive characteristics were discussed.
박형준,Giancarlo Eder Guerra Padilla,유기호 제어·로봇·시스템학회 2024 제어·로봇·시스템학회 논문지 Vol.30 No.2
The aquatic UAV is a kind of unmanned system capable of operating in air and on the water surface. To overcome the problem caused by different media, designing a hybrid mechanism that can adapt its morphology for flying and sailing is necessary. In addition, an efficient sail and rudder control system is essential because locomotion on the water surface is performed using only wind kinetic energy. In this study, we simulated the sailing of a solar-powered hybrid aquatic UAV under fixed wind conditions. First, the operating environment and system modeling for wind, wing-sail, and rudder force for water surface locomotion were implemented. In addition, a no-go zone was derived by running a velocity prediction program to confirm the sailing performance of the designed UAV. Then, low-level wing-sail controller and a fuzzy interface system based rudder controller were designed to control the behavior of the 3-DOF dynamic model. An objective function was applied in the low-level wing-sail controller to derive the wind direction needed to produce the maximum thrust, and fifteen fuzzy rules were used in the rudder controller by defining a membership function for heading angle and rotational speed. Finally, the feasibility of the control system in downwind and upwind scenarios was verified by the waypoint tracking simulation based on the nonlinear guidance logic.