EVPropNet: Detecting Drones By Finding Propellers For Mid-Air Landing And Following
EVPropNet: Detecting Drones By Finding Propellers For Mid-Air Landing And Following
复制标题
DOI：
10.15607/rss.2021.xvii.074
复制
发表时间：
2021-06
期刊：
影响因子：
--
通讯作者：
N. Sanket;Chahat Deep Singh;Chethan Parameshwara;Cornelia Fermuller;G. D. Croon;Y. Aloimonos
中科院分区：
文献类型：
--
作者：
N. Sanket;Chahat Deep Singh;Chethan Parameshwara;Cornelia Fermuller;G. D. Croon;Y. Aloimonos
无人驾驶飞行器或无人机的可获得性迅速上升，对一般安全和保密构成威胁。大多数商用或定制的无人机都是多旋翼的，由多个螺旋桨组成。由于这些螺旋桨是高速旋转的，它们通常是图像中移动最快的部分，在没有严重运动模糊的情况下，经典相机无法直接“看到”它们。我们利用了一类特别适合于此类场景的传感器，称为事件摄像头，具有高时间分辨率、低延迟和高动态范围。在本文中，我们对螺旋桨的几何形状进行建模，并使用它来生成模拟事件，用于训练一个名为EVPropNet的深度神经网络来从事件摄像机的数据中检测螺旋桨。EVPropNet无需任何微调或再培训即可直接转移到现实世界。我们介绍了我们的网络的两个应用：(A)跟踪和跟踪未标记的无人机和(B)降落在接近悬停的无人机上。在不同螺旋桨形状和尺寸的真实实验中，我们成功地对所提出的方法进行了评估和演示。我们的网络可以在60%的螺旋桨被遮挡的情况下以85.1%的速度检测到螺旋桨，并且在2W的功率预算下可以运行高达35赫兹的频率。据我们所知，这是第一个基于深度学习的螺旋桨探测解决方案(用于探测无人机)。最后，我们的应用程序还显示了令人印象深刻的成功率，跟踪任务和着陆任务的成功率分别为92%和90%。
The rapid rise of accessibility of unmanned aerial vehicles or drones pose a threat to general security and confidentiality. Most of the commercially available or custom-built drones are multi-rotors and are comprised of multiple propellers. Since these propellers rotate at a high-speed, they are generally the fastest moving parts of an image and cannot be directly"seen"by a classical camera without severe motion blur. We utilize a class of sensors that are particularly suitable for such scenarios called event cameras, which have a high temporal resolution, low-latency, and high dynamic range. In this paper, we model the geometry of a propeller and use it to generate simulated events which are used to train a deep neural network called EVPropNet to detect propellers from the data of an event camera. EVPropNet directly transfers to the real world without any fine-tuning or retraining. We present two applications of our network: (a) tracking and following an unmarked drone and (b) landing on a near-hover drone. We successfully evaluate and demonstrate the proposed approach in many real-world experiments with different propeller shapes and sizes. Our network can detect propellers at a rate of 85.1% even when 60% of the propeller is occluded and can run at upto 35Hz on a 2W power budget. To our knowledge, this is the first deep learning-based solution for detecting propellers (to detect drones). Finally, our applications also show an impressive success rate of 92% and 90% for the tracking and landing tasks respectively.