微型飛行器降落地面平台之視覺伺服控制器設計

本論文提出一個以視覺伺服控制導引微型飛行器，降落於室內環境之地面平台上。微型飛行器之續航力一般較為有限，透過降落於可移動之地面平台上來充電補給，將可以有效提升微型飛行器之巡航範圍。由於室內環境狹小，降落平台與飛行器的尺寸均受到限制，因此必須以對位控制準確導引微型飛行器進行降落。微型飛行器上搭載之前置攝影機與機腹攝影機，可觀測前方與下方的視野，以擷取出降落平台目標特徵。首先，我們將設計一特徵模板裝置於降落平台上，以便微型飛行器在各種不同的姿態下，於環境中偵測與萃取出對位降落所需之影像特徵。藉由規劃地面目標搜尋策略，搭配交互切換兩台攝影機觀測環境資訊，將可操控微型飛行器進行搜索並靠近降落平台。本論文將設計考量兩種降落情境，其一為降落平台靜止不動時，經由模糊控制來進行平台之影像特徵模板的目標對位，採用垂直降落的方式以停泊於降落平台上。而對於在移動中的降落平台，將採以滑行的方式降落。由降落平台上之模板的影像特徵點，估測出降落平台與微型飛行器間之相對姿態，預測降落平台之移動軌跡。以設計模糊控制器來縮小兩者間之相對姿態誤差，使微型飛行器朝降落平台之相同移動方向前進，進而滑行降落在目標平台上。本論文的視覺伺服降落控制器，經由多次實驗與不同情境結果獲得驗證。

關鍵字

微型飛行器；降落；對位控制；視覺伺服

並列摘要

This thesis proposed a visual servoing for controlling and guiding a micro aerial vehicle to land on a ground platform in an indoor environment. The battery life of a micro aerial vehicle is often limited, but the micro aerial vehicle can recharge by landing on a mobile platform to extend its flight range. However, the space in indoor environment is usually narrow, and the sizes of the landing platform and the aerial vehicle are restricted. Using the alignment control can lead the micro aerial vehicle to perform its landing precisely. Here, the micro aerial vehicle is equipped with front and belly pin-hole cameras to observe the features of the target or landing platform. In the beginning, we set up a unique pattern on the landing platform so that the micro aerial vehicle can be recognized. The visual feature for landing also can be extracted to navigate the micro aerial vehicle. The landing platform searching and approaching algorithms with cameras switching are also provided to observe the environment information and draw the distance between micro aerial vehicle and landing platform. On the other hand, this thesis concerns two landing modes: vertical landing and gliding landing. When the landing platform is still, the vertical landing mode uses the fuzzy control to make the micro aerial vehicle align with the landing platform equipped with the designed visual pattern. Under a circumstance in which the landing platform is moving, the gliding landing mode is utilized after estimating the yaw orientation and distance related to the landing platform. By using the fuzzy control, the differences in the orientation and distance between the two moving platforms can be diminished, and the micro aerial vehicle is able to glide onto the mobile landing platform. The proposed visual servoing have been proved by several tests to validate the performance of landing.

並列關鍵字

micro aerial vehicle ； landing ； aligning control ； visual servoing

參考文獻

[8]F. Lin, B. M. Chen, K. Y. Lum, and T. H. Lee, "A Robust Real-Time Embedded Vision System on an Unmanned Rotorcraft for Ground Target Following," IEEE Transactions on Industrial Electronics, vol. 59,NO. 2, pp. 1038-1049, February, 2012.

[12]T. K. Venugopalan, T. Taher, and P. G. Barbastathis, "Autonomous Landing of an Unmanned Aerial Vehicle on an Autonomous Marine Vehicle," Oceans, Hampton Roads VA, pp. 1-9, Oct 14-19, 2012.

[13]D. Lee, T. Ryan, and H. J. Kim, "Autonomous Landing of a VTOL UAV on a Moving Platform Using Image-Based Visual Servoing," 2012 IEEE International Conference on Robotics and Automation (ICRA), Saint Paul MN, pp. 971-976, May 14-18, 2012.

[17]S. Gao, Y. Cheng, and C. Song, "Drogue Detection for Vision-Based Autonomous Aerial Refueling Via Low Rank and Sparse Decomposition with Multiple Features," Infrared Physics & Technology, vol. 60, pp. 266-274, September, 2013.

[19]R. S. McEwen, B. W., Hobson, L. McBride, and J. G. Bellingham, "Docking Control System for a 54-cm-Diameter (21-in) AUV," IEEE Journal of Oceanic Engineering, vol. 33,NO. 4, pp. 550-562, October, 2008.

國際替代計量

微型飛行器降落地面平台之視覺伺服控制器設計

全文下載

主題瀏覽