Abstract
Autonomous landing of an unmanned aerial vehicle or a drone is a challenging problem for the robotics research community. Previous researchers have attempted to solve this problem by combining multiple sensors such as global positioning system (GPS) receivers, inertial measurement unit, and multiple camera systems. Although these approaches successfully estimate an unmanned aerial vehicle location during landing, many calibration processes are required to achieve good detection accuracy. In addition, cases where drones operate in heterogeneous areas with no GPS signal should be considered. To overcome these problems, we determined how to safely land a drone in a GPS-denied environment using our remote-marker-based tracking algorithm based on a single visible-light-camera sensor. Instead of using hand-crafted features, our algorithm includes a convolutional neural network named lightDenseYOLO to extract trained features from an input image to predict a marker’s location by visible light camera sensor on drone. Experimental results show that our method significantly outperforms state-of-the-art object trackers both using and not using convolutional neural network in terms of both accuracy and processing time.
Highlights
The evolution of the self-driving vehicle is about to shift into overdrive thanks to the recent advancements of computer vision and artificial intelligence
Autonomous unmanned aerial vehicles (UAVs) and cars are receiving the most attention because their practicality in daily life
We introduce an enhanced method for remote-marker-based tracking to exploit the trained features extracted from a proposed convolutional neural network (CNN) named lightDenseYOLO and perform marker tracking at very long distances, the furthest being 50 m
Summary
The evolution of the self-driving vehicle is about to shift into overdrive thanks to the recent advancements of computer vision and artificial intelligence. Companies such as Tesla, Google, Intel and Amazon are spending billions of dollars to develop cutting-edge autonomous systems about to transform how we think about transportation in the decade. Autonomous unmanned aerial vehicles (UAVs) and cars are receiving the most attention because their practicality in daily life. Most drones or UAVs are not truly autonomous and are operated remotely by a human controller from the ground. The generation drone requires a self-controlling function to fly in an unstructured environment and perform autonomous landing
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
