【作者】 那盟； 贾培发； 徐华；

【Author】 NA Meng,JIA Peifa,XU Hua National Key Laboratory of Intelligent Technology and System,Department of Computer Science and Technology,Tsinghua University,Beijing 100084

【机构】 智能技术与系统国家重点实验室,清华大学计算机科学与技术系；

【摘要】 为了实现微型直升机自主着陆,需要使其具有自主识别着陆目标的能力。本文设计了特定的着陆目标及其识别算法,通过模板匹配、数学形态学和投影检测等方法,完成了目标搜索、目标提取以及目标的精确定位,最终将计算出的目标图像形变参数提供给后续飞行控制模块,用来引导直升机自主降落。实验结果表明,该算法在15米范围内具有较高的识别率和识别精度,中心偏差小于4个像素,方向误差小于5°。同时,本文算法具有较高的自适应性、移植性和实时性,能够很好地满足直升机的着陆要求。更多还原

【Abstract】 In order to achieve autonomous landing, a micro helicopter should have the ability to recognize the landing target (a helipad). A special landing target and its recognition algorithm are presented in this paper. The algorithm is integrated with template matching, morphology, project detection and so on to implement the search, extraction and precise location of target in image. Finally the objective parameters are calculated and then sent to the next flight control module to induct helicopter’s autonomous landing. The flight experiment shows the algorithm is accurate to within 4 pixels central position accuracy and to within 5°in orientation. Meanwhile, the results demonstrate that our algorithm has high adaptive, transplantable and real time features which well satisfy the landing request.更多还原