Localization for autonomous construction vehicles using monocular camera and AprilTag

Sammanfattning: Global Navigation Satellite System (GNSS) is commonly used to provide thelocation to navigate the vehicles in outdoor environment. However, GNSS is unreliablein some complex environment such as an open-pit mining environment.Although localization methods based on vision sensors have been proposed tosolve this problem, most of them use on-board cameras easily obscured by mudand dirt for autonomous construction vehicle applications.This thesis proposes to estimate the position of the autonomous construction vehicleswith o-board monocular cameras using the vision ducial marker April-Tags. This localization system is tested in indoor experiments and on an outdoortest track and the approach is evaluated using both static and dynamictests. The results show that the localization system successfully locate the autonomousconstruction vehicle with centimeter precision when the vehicle standsstill. However, the accuracy deteriorates considerably if there are sudden movementsof the vehicle.