Ground Stereo Vision-Based Navigation for Autonomous Take-off and Landing of UAVs: A Chan-Vese Model Approach

Abstract

This article aims at flying target detection and localization of a fixed-wing unmanned aerial vehicle (UAV) autonomous take-off and landing within Global Navigation Satellite System (GNSS)-denied environments. A Chan-Vese model-based approach is proposed and developed for ground stereo vision detection. Extended Kalman Filter (EKF) is fused into state estimation to reduce the localization inaccuracy caused by measurement errors of object detection and Pan-Tilt unit (PTU) attitudes. Furthermore, the region-of-interest (ROI) setting up is conducted to improve the real-time capability. The present work contributes to real-time, accurate and robust features, compared with our previous works. Both offline and online experimental results validate the effectiveness and better performances of the proposed method against the traditional triangulation-based localization algorithm.

ROS-based ground stereo vision detection: implementation and experiments

“…2, the detection algorithm extracts a pair of pixel points ( x l , y l ) and ( x r , y r ) from the captured sequential images, while the localization algorithm integrates the calibration data, a pair of detected pixel points and the feedback angles ( P l , T l ) and ( P r , T r ) of pan–tilt units into calculating the spatial coordinates at each time step. Mathematical models of the stereo localization were developed and illustrated in [9] at length.

Fig. 2Algorithm workflow of the ground stereo vision guidance and localization system.

…”

Stereo Vision Guiding for the Autonomous Landing of Fixed-Wing UAVs: A Saliency-Inspired Approach

“…For example, local point feature-based detection can receive many local features in a cluttered environment. To solve this problem, Chan-Vese algorithms were applied to flying object detection on the ground-captured sequential images in our previous works [16,17]. In this paper, we mainly focus on inspiration gained from the HVS.…”

“…To verify its effectiveness, we mainly compared the saliencybased calculation results with the methods proposed in [17]. A Chan-Vese (CV) model-based approach is proposed and developed for ground stereo vision detection and a region-of-interest (ROI) set-up is presented to improve real-time capabilities.…”

“…On the other hand, in our previous work [17], different visual algorithms were applied to this guiding system. To verify its effectiveness, we mainly compared the saliencybased calculation results with the methods proposed in [17].…”

“…They are also much more expensive compared to a visual system. Therefore, taking into account the above considerations, we primarily focused on developing a ground stereo vision guidance system [15][16][17]. By using a ground binocular camera system to guide the UAV's landing, many researchers have expanded work in the area of developing vision systems [14][15][16][17][18].…”