Fast and Robust Initialization for Visual-Inertial SLAM
Authors:
- Carlos CamposCCCarlos Campos
- J. M. M. MontielJMJ. M. M. Montiel
- Juan D. TardósJTJuan D. Tardós
Preprint
Abstract
Visual-inertial SLAM (VI-SLAM) requires a good initial estimation of the initial velocity, orientation with respect to gravity and gyroscope and accelerometer biases. In this paper we build on the initialization method proposed by Martinelli and extended by Kaiser et al. , modifying it to be more general and efficient. We improve accuracy with several rounds of visual-inertial bundle adjustment, and robustify the method with novel observability and consensus tests, that discard erroneous solutions. Our results on the EuRoC dataset show that, while the original method produces scale errors up to 156%, our method is able to consistently initialize in less than two seconds with scale errors around 5%, which can be further reduced to less than 1% performing visual-inertial bundle adjustment after ten seconds.Keywords
All Related Versions
- Version 1, 2019-08-28, ArXiv
- Published version: , 1288.
This publication has 8 references indexed in Scilit:
- Visual-Inertial Monocular SLAM With Map ReuseIEEE Robotics and Automation Letters, 2017
- Simultaneous State Initialization and Gyroscope Bias Calibration in Visual Inertial Aided NavigationIEEE Robotics and Automation Letters, 2016
- The EuRoC micro aerial vehicle datasetsThe International Journal of Robotics Research, 2016
- ORB-SLAM: A Versatile and Accurate Monocular SLAM SystemIEEE Transactions on Robotics, 2015
- Keyframe-based visual–inertial odometry using nonlinear optimizationThe International Journal of Robotics Research, 2014
- Closed-Form Solution of Visual-Inertial Structure from MotionInternational Journal of Computer Vision, 2013
- High-precision, consistent EKF-based visual-inertial odometryThe International Journal of Robotics Research, 2013
- Visual-Inertial-Aided Navigation for High-Dynamic Motion in Built Environments Without Initial ConditionsIEEE Transactions on Robotics, 2012