The autonomous driving of robots is coming and requires precise and reliable positioning information with low-cost sensors for the mass market. In this paper, we propose a tightly coupled sensor fusion of multiple complementary sensors including Global Navigation Satellite System (GNSS) receivers with Real-Time Kinematics (RTK), Inertial Measurement Units (IMUs), wheel odometry, Local Positioning System (LPS) and Visual Positioning. The focus of this paper is on the integration of LPS and vision since the coupling of GNSS-RTK, INS and wheel odometry is already state of the art. We include the positions of the LPS anchors and the bearing vectors and distances from the robot's camera towards the patch features as state vectors in our Kalman filter, and show the achievable positioning accuracies.
Precise Positioning of Robots with Fusion of GNSS, INS, Odometry, LPS and Vision
2019-03-01
1647709 byte
Conference paper
Electronic Resource
English
GNSS Precise Point Positioning Based on Multipath Signal Determination for Outdoor Mobile Robots
British Library Online Contents | 2012
|Precise Positioning for Automotive with Mass Market GNSS Chipsets
British Library Conference Proceedings | 2018
|The New Search Method in Precise GNSS Positioning
IEEE | 2018
|