本研究利用海上箱網養殖為基礎來探討路徑規劃，因為箱網不只一個且散佈在海上，所以我們先利用旅行推銷員問題來決定飛行的先後順序，接著探討箱網間的路徑規劃。基於改良型粒子群演算法做些微改善，加入已知障礙物的斜率來讓計算速度更快，路徑也可以更短，並利用假設已知障礙物的狀況下，跟幾種演算法做比較優缺點。而未知障礙物的方面則搭配光達來進行偵測，偵測到障礙物之後，如果障礙物在行進路線中，則會重新計算新路線進行避障，最後在旋翼機上實現這兩種狀況。 This study focuses on the path planning with cage nets on the ocean. We considered the traveling salesman problem first to decide the order of flight path. There are two types of environments to be considered. One is known obstacles, the other is unknown obstacles. And the algorithm that we used is based on improved particle swarm optimization. In the environment with known obstacles, compared to other algorithms, the experimental results show that the proposed method has better performance. In the environment with unknown obstacles, we use the LiDAR to detect the obstacles. If there is any obstacle on the route, the control scheme will generate a new path to avoid the obstacle. Finally, we use the hexacopter to realize this study.