Vision-Based Human Position Estimation and Following Using an Unmanned Hexarotor Helicopter

Abstract

We describe the design and performance of a prototype hexarotor unmanned aerial vehicle (UAV) platform featuring autonomous flying based on an inertial measurement unit (IMU) for use in bluetooth communication environments. A recent trend in studies on free-flying UAVs with hexarotors has been to outfit UAVs with gimbals to support various services. To realize this, it is necessary for UAVs to carry out human tracking’s ability. We present preliminary results of human-following control with this UAV platform and implement a simple following flight strategy, based on a vision-based Kalman filter. The moving object (a walking human) is assumed to be a point object and is projected onto an image plane to form a geometrical constraint equation that provides the position data of the object based on kinematics in the local north-east-down (NED) coordinate system. Computer presents the simulation results of the predicting and following of a walking human using a UAV.