Reactive obstacle avoidance of monocular quadrotors with online adapted depth prediction network
Reactive obstacle avoidance of monocular quadrotors with online adapted depth prediction network
复制标题
DOI：
10.1016/j.neucom.2018.10.019
复制
发表时间：
2019-01
期刊：
影响因子：
6
通讯作者：
Xin Yang;Hongcheng Luo;Yuhao Wu;Yang Gao;Chunyuan Liao;K. Cheng
中科院分区：
文献类型：
--
作者：
Xin Yang;Hongcheng Luo;Yuhao Wu;Yang Gao;Chunyuan Liao;K. Cheng
由于单目四旋翼飞行器缺乏 3D 信息，基于单目相机的避障是一项基本但极具挑战性的任务。由于深度学习的巨大进步，最近基于卷积神经网络（CNN）[1]的单目深度估计和障碍物检测方法变得越来越流行。然而，预训练 CNN 的深度估计通常会因训练数据中不同类型的场景而导致精度大幅下降，这对于无人机在未知环境中避障来说很常见。在本文中，我们提出了一种反应式避障系统，该系统采用在线自适应 CNN，逐步改进单目相机在不熟悉的环境中的深度估计。基于与 CNN 并行运行的直接单目 SLAM，动态收集运动立体图像对作为训练数据。引入了新方法，用于从 SLAM 和高效在线 CNN 调整提供的噪声数据中选择高度可靠的训练样本。通过将四旋翼飞行器的动态运动约束和深度估计误差嵌入到空间深度图中，将 CNN 计算的深度图转换为自我动态空间 (EDS)。 EDS 会自动计算考虑相机视场约束的可穿越航路点，并据此为四轴飞行器生成适当的控制输入。公共数据集、模拟环境和看不见的杂乱室内环境的实验结果证明了我们系统的有效性。
Obstacle avoidance based on a monocular camera is a fundamental yet highly challenging task due to the lack of 3D information for a monocular quadrotor. Recent methods based on convolutional neural networks (CNNs) [1] for monocular depth estimation and obstacle detection become increasingly popular due to the considerable advances in deep learning. However, depth estimation by pre-trained CNNs usually suffers from large accuracy degradation for scenes of different types from the training data which are common for obstacle avoidance of drones in unknown environments. In this paper, we present a reactive obstacle avoidance system which employs an online adaptive CNN for progressively improving depth estimation from a monocular camera in unfamiliar environments. Pairs of motion stereo images are collected on-the-fly as training data based on a direct monocular SLAM running in parallel with the CNN. Novel approaches are introduced for selecting highly reliable training samples from noisy data provided by SLAM and efficient online CNN tuning. The depth map computed from the CNN is transformed into Ego Dynamic Space (EDS) by embedding both dynamic motion constraints of a quadrotor and depth estimation errors into the spatial depth map. Traversable waypoints with consideration of the camera’s field of view constraint are automatically computed in EDS based on which appropriate control inputs for the quadcopter are produced. Experimental results on both public datasets, simulated environments and unseen cluttered indoor environments demonstrate the effectiveness of our system.