Small UAV camera gimbal stabilization using digital filters and enhanced control algorithms for aerial survey and monitoring

Abstract

Aerial photography, monitoring and survey using small Unmanned Aerial Vehicles (UAVs) is a modern, cheap, simple, helpful and still developing and improving area. For these purposes the ongoing research is focused mainly on the cameras and image processing methods and software. However, as it was confirmed in the article, a stabilized camera gimbal is also very necessary to obtain quality and bright pictures or video records and to allow the operator or the tracking computer to track the camera’s line of sight to the point of an interest. Because the camera stabilization is a key factor influencing the quality of the pictures or videos and considering the application on the UAVs performing the flights in the low altitudes and often also in the mountain terrain, the wind conditions, turbulences, wind shears, which can vary in their magnitudes and directions significantly, the convenient stabilization of the camera gimbal can have a significant influence on the quality of the obtained results, which are very important for the creation of the precise 2D or 3D models. Furthermore, to increase the UAV payload, it is important to use lightweight solutions. Due to the onboard electronics of small UAVs, regarding the limited memory and computational performance, a small microcontroller including a convenient, simple, and still fast enough control algorithm needs to be designed and implemented. In order to stabilize a camera gimbal, it is needed to design a model of the actuators as well as the gearings, to propose an effective control algorithm and to implement the control algorithms into the on-board microcontrollers. This article deals with the modelling of the actuator, conventional commercial servomotor used for a camera gimbal stabilization and with the design and verification of the improved control algorithm based on the inverse characteristics of the actuator model. Due to the requirement of the high-quality images, where the fast stabilization is needed, a dynamic correction feedback was implemented. And as the gyroscopes are very sensitive to the UAVs vibrations, the vibrations of the camera gimbal were eliminated by the digital low pass filter. The theoretical background was experimentally verified by the geological survey of the stone pits in Sedlice, Vechec and Klatov in the Eastern Slovakia.