Visual-Inertial Monocular SLAM With Map Reuse

@article{MurArtal2017VisualInertialMS,
  title={Visual-Inertial Monocular SLAM With Map Reuse},
  author={Raul Mur-Artal and Juan D. Tard{\'o}s},
  journal={IEEE Robotics and Automation Letters},
  year={2017},
  volume={2},
  pages={796-803}
}
In recent years there have been excellent results in visual-inertial odometry techniques, which aim to compute the incremental motion of the sensor with high accuracy and robustness. However, these approaches lack the capability to close loops and trajectory estimation accumulates drift even if the sensor is continually revisiting the same place. In this letter, we present a novel tightly coupled visual-inertial simultaneous localization and mapping system that is able to close loops and reuse… 

Figures and Tables from this paper

Scale Estimation and Refinement in Monocular Visual-Inertial SLAM System
TLDR
This paper presents an approach to estimate scale, gravity and accelerometer bias together, and regard the estimated gravity as an indication for estimation convergence, and proposes a methodology that is able to use weight derived from the robust norm for outliers handling, so that the estimated scale can be refined.
Evaluation of Monocular Visual-Inertial SLAM: Benchmark and Experiment
TLDR
A tightly-coupled and optimization-based monocular Visual-Inertial SLAM system is proposed, which can tackle the scale ambiguity - a problem that arises by poor initialization, providing better accuracy in some sequences owing to the improved initialization.
Relocalization, Global Optimization and Map Merging for Monocular Visual-Inertial SLAM
TLDR
A monocular visual-inertial SLAM system, which can relocalize camera and get the absolute pose in a previous-built map and can reuse a map by saving and loading it in an efficient way and validate the accuracy on public datasets and compare against other state-of-the-art algorithms.
Accurate Initial State Estimation in a Monocular Visual–Inertial SLAM System
TLDR
Experimental results show that the proposed methods can achieve good initial state estimation, the gravity refinement approach is able to efficiently speed up the convergence process of the estimated gravity vector, and the termination criterion performs well.
Robust visual-inertial SLAM: combination of EKF and optimization method
TLDR
A novel tightly-coupled monocular visual-inertial Simultaneous Localization and Mapping (SLAM) algorithm combining the advantages of both filtering methods and optimization methods that achieves unprecedented accuracy and robustness.
Map-Based Visual-Inertial Monocular SLAM using Inertial assisted Kalman Filter
TLDR
This paper presents a novel tightly-coupled monocular visual-inertial Simultaneous Localization and Mapping algorithm following an inertial assisted Kalman Filter and reusing the estimated 3D map to achieve an efficient motion tracking bearing fast dynamic movement in the front-end.
Robust initialization of monocular visual-inertial estimation on aerial robots
  • Tong Qin, S. Shen
  • Computer Science
    2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)
  • 2017
TLDR
A robust on-the-fly estimator initialization algorithm to provide high-quality initial states for monocular visual-inertial systems (VINS) and makes the implementation open source, which is the initialization part integrated in the VINS-Mono1.
Monocular Visual-Inertial SLAM: Continuous Preintegration and Reliable Initialization
TLDR
A new visual-inertial Simultaneous Localization and Mapping (SLAM) algorithm that is able to achieve robust and real-time estimates of the sensor poses in unknown environment and is comparable to the state-of-art method.
Direct Visual-Inertial Ego-Motion Estimation Via Iterated Extended Kalman Filter
TLDR
Investigation suggests that the accuracy of the flight velocity estimates from the proposed approach is comparable to those of two state-of-the-art Visual Inertial Systems (VINS) while the proposed framework is 10 times faster thanks to the omission of reconstruction and mapping.
An Improved Monocular Visual-Inertial Navigation System
TLDR
A new tightly-coupled visual-inertial concurrent localization and mapping approach is proposed with precise and real-time motion estimating and map reconstruction capabilities and is more accurate with lower computational cost compared with existing VISLAM systems.
...
...

References

SHOWING 1-10 OF 30 REFERENCES
Robust visual inertial odometry using a direct EKF-based approach
TLDR
A monocular visual-inertial odometry algorithm which achieves accurate tracking performance while exhibiting a very high level of robustness by directly using pixel intensity errors of image patches, leading to a truly power-up-and-go state estimation system.
Keyframe-based visual–inertial odometry using nonlinear optimization
TLDR
This work forms a rigorously probabilistic cost function that combines reprojection errors of landmarks and inertial terms and compares the performance to an implementation of a state-of-the-art stochastic cloning sliding-window filter.
Monocular Visual–Inertial State Estimation With Online Initialization and Camera–IMU Extrinsic Calibration
  • Zhenfei Yang, S. Shen
  • Computer Science
    IEEE Transactions on Automation Science and Engineering
  • 2017
TLDR
This paper proposes a methodology that is able to initialize velocity, gravity, visual scale, and camera–IMU extrinsic calibration on the fly and shows through online experiments that this method leads to accurate calibration of camera-IMU transformation, with errors less than 0.02 m in translation and 1° in rotation.
Scale Drift-Aware Large Scale Monocular SLAM
TLDR
This paper describes a new near real-time visual SLAM system which adopts the continuous keyframe optimisation approach of the best current stereo systems, but accounts for the additional challenges presented by monocular input and presents a new pose-graph optimisation technique which allows for the efficient correction of rotation, translation and scale drift at loop closures.
Visual-Inertial-Aided Navigation for High-Dynamic Motion in Built Environments Without Initial Conditions
TLDR
A novel method to fuse observations from an inertial measurement unit (IMU) and visual sensors, such that initial conditions of the inertial integration can be recovered quickly and in a linear manner, thus removing any need for special initialization procedures.
IMU Preintegration on Manifold for Efficient Visual-Inertial Maximum-a-Posteriori Estimation
TLDR
This paper addresses the issue of increased computational complexity in monocular visual-inertial navigation by preintegrating inertial measurements between selected keyframes by developing a preintegration theory that properly addresses the manifold structure of the rotation group and carefully deals with uncertainty propagation.
Simultaneous State Initialization and Gyroscope Bias Calibration in Visual Inertial Aided Navigation
TLDR
It is shown that the gyroscope bias, not accounted for in [1], significantly affects the performance of the closed-form solution and a new method is introduced to automatically estimate this bias and is robust to it.
Visual-inertial direct SLAM
TLDR
This paper proposes for the first time a direct, tightly-coupled formulation for the combination of visual and inertial data and evaluates the algorithm in several real sequences with ground truth trajectory data, showing a state-of-the-art performance.
Direct visual-inertial odometry with stereo cameras
TLDR
This work proposes a novel direct visual-inertial odometry method for stereo cameras that outperforms not only vision-only or loosely coupled approaches, but also can achieve more accurate results than state-of-the-art keypoint-based methods on different datasets, including rapid motion and significant illumination changes.
ORB-SLAM2: An Open-Source SLAM System for Monocular, Stereo, and RGB-D Cameras
TLDR
ORB-SLAM2, a complete simultaneous localization and mapping (SLAM) system for monocular, stereo and RGB-D cameras, including map reuse, loop closing, and relocalization capabilities, is presented, being in most cases the most accurate SLAM solution.
...
...