Vision based Precision Landing of UAVs
The foremost, yet challenging problem of an Unmanned Aerial Vehicle (UAV) is the autonomous landing. Most of the proposed methods rely on expensive equipment or do not satisfy the high precision needs of some UAV applications, such as package retrieval and delivery, indoor flights, charging stations, surveillance, etc.
Here, we present a solution for high precision landing based on the use of ArUco markers. In our solution, an UAV equipped with a simple rpi camera is able to detect ArUco markers from an altitude of 10m(can vary upto 100-500m based on camera and LiDar range). Once the marker is detected, the UAV changes its flight behavior in order to land on the exact position where the marker is located. We evaluated our proposal using SITL vehicle & the simulation platform Gazebo and ROS. The results show an average offset of only 2-5 centimeters, which vastly improves the landing accuracy compared to the traditional GPS-based landing, that typically deviates from the intended target by 1 to 3 meters.
Passive Liveness Detection Module
Face Recognition systems have been widely adopted for user authentication in security systems due to their simplicity and effectiveness. However, the main drawback of such systems is that they are vulnerable to spoof attacks. It is an easy way to fool face recognition systems by facial pictures such as print media, repetitive frames, replayed videos etc.
To overcome such problem, RapidAI Vision designed an in-house passive Liveness Detection module that adds an additional layer of security. It is trained on our own created dataset by using the combination of different methodology of Machine Learning & Deep Learning with >99% of accuracy.
