Vision-Based State Estimation Using Tracked Landmarks

摘要

In this paper, we develop vision-based state estimation algorithms to enable autonomous vehicles to navigate within GPS-denied environments. To accomplish this objective, we utilize a priori information about the environment. In particular, the algorithms leverage recognizable 'landmarks' in the environment, the positions of which are known in advance, to stabilize the state estimate. Measurements of the positions of one or more landmarks in the image plane of a monocular camera are then filtered using an extended Kalman filter (EKF) with data from a traditional IMU to produce the state estimate. Additionally, the EKF algorithm is adapted to accommodate a stereo setup of two cameras to measure the distance to a landmark using parallax. The performance of the state estimation algorithms for the monocular and stereo implementations is tested and compared using simulation studies with a quadcopter UAV model. State estimation results are then generated and analyzed using flight data from a quadcopter UAV instrumented with an IMU and a monocular camera.