根據Lissaman 與 Shollenberger於1970年的Science文章，提到鳥類的編隊飛行帶來了節能的效果，因此進一步思考：假若使用仿生拍翼機進行群體飛行，是否也能得到類似的節能效果。 本文執行兩種實驗，第一種為動態實驗，將三隻拍翼機固定於框架上懸掛，進行水平迴旋飛行，模擬鳥類在巡航時的群體飛行，探討功率消耗是否減少。因為三隻拍翼機飛行懸掛飛行，本身並不穩定，經常造成最長滯空時間之量測中斷，故實驗所得的有限功耗數據缺乏一致性，而暫告實驗失敗。 第二種為靜態實驗，改用淡江大學風工程中心的大型風洞，進行拍翼機編隊之氣動力量測。本文製作了可移動式拍翼機群夾具，量測 1~3隻拍翼機，在不同前後距離、不同風速、不同攻角條件下的升力與推力變化，進而探討是否有達到節能的效果。實驗數據雖然複雜，但初步分析已有節能效果在風速3m/s、攻角10˚效果最好，功率減少消耗了1.2W。比對鳥類在編隊飛行時，會自行彈性調整相位和頻率，以達最佳節能效果，為本文後續研究之未來可努力改善方向。 According to Lissaman and Shollenberger’s article published in Science in 1970, birds fly in formation to reduce the overall energy expenditure for flight. Thus this work adopted the same principle of flight formation to flapping micro-air-vehicles (MAVs) to save energy. Two experiments are described herein. Firstly, a rigid frame housing three flapping MAVs was made and a tethered flight of it around a fixed suspension point was performed to evaluate the cruise performance of the MAV formation and check for energy saving by monitoring the endurance against a known reference value. Due to the intrinsic instabilities of the MAVs, lack of data consistency about in power saving in formation flight concludes less promising results from the 1st experiment. In the second experiment, a jig housing three flapping MAVs was subject to a wind tunnel test at the Wind Engineering Center of Tamkang University. The aerodynamic force evaluation of the MAV formation was done. The tests were performed under different wind speeds, and angle of attacks, and the lift and thrust force produced by the MAVs was measured via a force gauge. Due to the mechanical vibrations from the jig itself and the wind tunnel structure, interpretation of the complicated lift/thrust data was processed by the help of Fast Fourier Transform. From the preliminary examination, it is found that at 3 m/s, 10˚ angle of attack, the performance was the best to save power up to 1.2W. When compared to the real birds flying in a formation, the energy saving of flapping MAVs may be further improved by replicating the dynamic adjustments of frequency, phase change and separation among neighboring MAVs, to obtain best energy-saving results in the future.