<![CDATA[Unmanned Aerial Vehicle (UAV) has proven to be a useful tool in many fields. UAV or quadcopter is equipped with a number of sensors and imaging systems that can perform various image-based things. This proves that quadcopter utilization can be allocated for a particular purpose, for example the purpose of reaching the landing mechanism. The landing is a phase that can cause fatal damage to the quadcopter if it does not land in its place. Thus the quadcopter is designed to automatically land on an object in various forms using the douglas-peucker algorithm. First convert the RGB color space to grayscale for easy to detect objects and eliminate noise using blur. Then it detects the edges of the object using canny and proceeds to search for the appropriate contour values ​​to get the edges of the object so that the quadcopter can recognize each object shape from the edge obtained and assisted with the object counting using the douglas-peucker method. Then it will be displayed on a frame that is divided into 9 grids and performs automatic movement according to the grid position until perform autnomous landing. Testing is done by testing the movement of quadcopter to detect object with speed 0,3m / s and 0,5m / s and test landing accuracy with height 125cm, 150cm, 175cm and 200cm. The results show that, at a speed of 0.3 m / s obtained the percentage of accuracy of good object detection is 100% and obtained the best landing percentage accuracy is 83.3% at 125cm altitude.]]>
Copyrights © 2018
