VINS-Mono: A Robust and Versatile Monocular Visual-Inertial State Estimator

By Tong Qin, Peiliang Li, Shaojie Shen. IEEE Transactions on robotics, 2017

4,417 citations recorded by the source corpus.

One camera and one low-cost inertial measurement unit (IMU) form a monocular visual-inertial system (VINS), which is the minimum sensor suite (in size, weight, and power) for the metric six degrees-of-freedom (DOF) state estimation. In this paper, we present VINS-Mono: a robust and versatile monocular visual-inertial state estimator. Our approach starts with a robust procedure for estimator initialization. A tightly coupled, nonlinear optimization-based method is used to obtain highly accurate visual-inertial odometry by fusing preintegrated IMU measurements and feature observations. A loop detection module, in combination with our tightly coupled formulation, enables relocalization with minimum computation. We additionally perform 4-DOF pose graph optimization to enforce the global consistency. Furthermore, the proposed system can reuse a map by saving and loading it in an efficient way. The current and previous maps can be merged together by the global pose graph optimization. We validate the performance of our system on public datasets and real-world experiments and compare against other state-of-the-art algorithms. We also perform an onboard closed-loop autonomous flight on the microaerial-vehicle platform and port the algorithm to an iOS-based demonstration. We highlight that the proposed work is a reliable, complete, and versatile system that is applicable for different applications that require high accuracy in localization. We open source our implementations for both PCs (https://github.com/HKUST-Aerial-Robotics/VINS-Mono) and iOS mobile devices ( https://github.com/HKUST-Aerial-Robotics/VINS-Mobile).

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition, Electrical and Electronic Engineering

View Paper: https://www.semanticscholar.org/paper/9da965529ee3da77178ed99cf13d97be0fa85f6e

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

In Computer Vision and Pattern Recognition

In Electrical and Electronic Engineering

Citations in the corpus

Adaptive VIO: Deep Visual-Inertial Odometry with Online Continual Learning (2024, methodology): “Then, the preintegration until i + 1 th IMU frame can be expressed in the following form[24]: where ai, wi are acceleration and angular velocity mea surements of i th IMU frame, after subtracting the network predicted bias ba, b 9, respectively.”
FUEL: Fast UAV Exploration Using Incremental Frontier Structure and Hierarchical Planning (2020, methodology): “state estimation [36] and mapping modules [34] are also crucial to fullfil the real-world tasks.”
Contact-aided invariant extended Kalman filtering for robot state estimation (2019, methodology): “Section 12 itemizes additional sensor measurements that ﬁt within the InEKF framework and describes the relation between the developed ﬁlter and landmark-based SLAM.”
Events-To-Video: Bringing Modern Computer Vision to Event Cameras (2019, methodology): “Here, we go one step further and directly apply an off-the-shelf VIO system (specifically, VINS-Mono [36], which is state-of-the-art [11]) to videos reconstructed from events using either our approach, MR, or HF, and evaluate against UltimateSLAM.”
F-LOAM : Fast LiDAR Odometry and Mapping (2021, neutral): “In additional, the visual SLAM such as VINS-MONO [29] is also compared.”
Collaborative Multi-Robot Search and Rescue: Planning, Coordination, Perception, and Active Vision (2020, background): “This is the case, for instance, of visual-inertial odometry with either monocular cameras [113], or multiple sensors [114].”
A Comprehensive Survey of Visual SLAM Algorithms (2022, background): “This is despite the fact that, since it only considers pose and velocity from the latest IMU states during the optimization process, this algorithm still demonstrated its suitability in embedded implementations [73].”
FAST-LIO: A Fast, Robust LiDAR-Inertial Odometry Package by Tightly-Coupled Iterated Kalman Filter (2020, background): “Visual (-inertial) odometry (VO), such as Stereo VO [1], [2] and Monocular VO [3], [4] are commonly used on mobile robots due to their lightweight and low-cost.”
Are We Ready for Service Robots? The OpenLORIS-Scene Datasets for Lifelong SLAM (2019, background): “Then ATE of each matched ORB_SLAM2 [16] (stereo fisheye) ORB_SLAM2 [16] (RGB-D) DS-SLAM [18] (RGB-D) DSO [17] (RGB) VINS-Mono [19] (RGB+IMU) VINS-Mono [19] (fisheye+IMU) InfiniTAMv2 [20] (RGB-D) ElasticFusion [ The percentage value on the top left of each scene is average Correct Rate (CR ε , φ as…”
Extending Monocular Visual Odometry to Stereo Camera Systems by Scale optimization (2019, background): “To solve the scale problem without intensive computational cost, authors [4]–[6] have fused monocular VO with an inertial measurement unit (IMU) to create visual-inertial navigation systems (VINS).”
LDSO: Direct Sparse Odometry with Loop Closure (2018, background): “VINS-Mono [9] in turn is a featurebased monocular visual-inertial SLAM system that employs a marginalizing odometry front-end very similar to okvis.”
A Tutorial on Quantitative Trajectory Evaluation for Visual(-Inertial) Odometry (2018, methodology): “We can see the “raw” estimation from VINS-Mono is in a different reference frame as the groundtruth, and therefore cannot be directly compared.”
GVINS: Tightly Coupled GNSS–Visual–Inertial Fusion for Smooth and Consistent State Estimation (2021, methodology): “Note that the formulation of visual and inertial factors are inherited from [6] [26] [27] thus not the contribution of this work.”
A General Optimization-based Framework for Global Pose Estimation with Multiple Sensors (2019, methodology): “There are many impressive VO/VIO algorithms, such as [4, 6]–[11].”
Visual-Inertial Navigation: A Concise Review (2019, methodology): “Most have a two-thread system that optimizes a small window of “local” keyframes and features limiting drift in the short-term, while a background process optimizes a long-term sparse pose graph containing loopclosure constraints enforcing long-term consistency [31, 83, 99, 100].”
R3LIVE: A Robust, Real-time, RGB-colored, LiDAR-Inertial-Visual tightly-coupled state Estimation and mapping package (2021, neutral): “III), “LVI-SAM” (with modified its LiDAR front-end for Livox Avia LiDAR), “R(2)LIVE” [12], “VINSMono” (IMU+Camera) [26], “Fast-LIO2” (IMU+LiDAR) [22] with the ground-truth in Fig.”
D3VO: Deep Depth, Deep Pose and Deep Uncertainty for Monocular Visual Odometry (2020, methodology): “VIO methods [44, 56, 71, 72] dominate this benchmark by integrating IMU measurements to get a pose or motion prior and meanwhile estimating the absolute scale.”
The TUM VI Benchmark for Evaluating Visual-Inertial Odometry (2018, methodology): “ROVIO is more prone to drift and diverges on several sequences, which might be explained by it’s use of a Kalman filter compared to computationally more demanding non-linear least squares optimization employed by OKVIS, VINS-Mono and BASALT.”