RGB-D camera-based quadrotor navigation in GPS-denied and low light environments using known 3D markers

Abstract

This paper presents an original approach for autonomous navigation based on RGB-D data and known 3D markers, where the basic concept is to detect and recognize the markers and then to use them for a straightforward pose estimation solution. The developed algorithms can allow a quadrotor to autonomously fly in (cooperative) GPS denied environments and/or when there is no natural or artificial illumination of the scene, by following a predetermined path consisting of successive targets having a well defined shape and/or color. Algorithms for target detection and recognition based on depth data are described which are optimized for real time use, paying particular attention to the on-board computational load. Experimental tests have been carried out by integrating a RGB-Depth sensor (ASUS Xtion Pro Live) on-board a custom-built quadrotor. First results confirm the potential of the proposed approach. The technique can be applied to different types of unmanned aerial vehicles (UAVs), as well as unmanned ground vehicles (UGVs)

Survey of computer vision algorithms and applications for unmanned aerial vehicles

“…These methods differ on the type of the visual information used; such as horizons detection ( Dusha, Boles, & Walker, 2007;Grelsson, Linköpings universitet, & Institutionen för systemteknik, 2014 ), landmarks tracking ( Amor-Martinez, Ruiz, Moreno-Noguer, & Sanfeliu, 2014; Eberli, Scaramuzza, Weiss, & Siegwart, 2011 ), or edges detection ( Kim, 2006;Wang, 2011 ). Furthermore, they can be differentiated based on the structure of the vision system: it can be monocular ( Milford, Schill, Corke, Mahony, & Wyeth, 2011;Yang, Scherer, & Zell, 2013;Zeng, Wang, Liu, Chen, & Deng, 2014 ), binocular ( Vetrella, Savvaris, Fasano, & Accardo, 2015;Warren, 2015 ), trinocular ( Jeong, Mulligan, & Correll, 2013;Martínez, Campoy, Mondragón, & Olivares Mendez, 2009 ), or omnidirectional ( Amorós, Paya, Valiente, Gil, & Reinoso, 2014;Killpack, Deyle, Anderson, & Kemp, 2010;Scaramuzza & Siegwart, 2008;Wang, Zhao, Davoine, & Zha, 2012b ) camera system. Some of the early experimental works that use visual information; in order to estimate the aircraft attitude were presented in Todorovic, Nechyba, and Ifju (2003) , Cornall and Egan (2004) , Cornall, Egan, and Price (2006) , Thurrowgood, Soccol, Moore, Bland, and Srinivasan (2009) .…”

Switchable Glass Enabled Contextualization for a Cyber-Physical Safe and Interactive Spatial Augmented Reality PCBA Manufacturing Inspection System

“…Additionally, we employ the retrieved information to calibrate the switchable glass position for subsequent interpolation during human activity recognition. Kinect is a low-cost solution able to that directly provides color and depth information [ 42 ] generating 3D position with demonstrated precision [ 30 ] rendering it as a practical and accessible solution without interfering with its projection [ 43 ].…”

Motion Planning for a UAV with a Straight or Kinked Tether

“…One important advantage brought in by a tether to UAVs is an alternative localization and navigation methodology, which would impact implementation of a tether planner. In lieu of traditional UAV sensory inputs including GPS [15], Inertia Measurement Units (IMUs) [16]- [18], laser range finder [16], [17], stereo [17], [18] and RGB-D cameras [19], the new approach comes with a tether-based UAV controller: tether length encoder, tether azimuth and elevation angle computed by fusing onboard IMU and tether angle sensor [1]. The localization of the UAV could then be realized by measuring the tether length and angles with respect to the global frame.…”