Visual-Inertial Sensor Fusion: Localization, Mapping and Sensor-to-Sensor Self-calibration

By Jonathan Kelly, G. Sukhatme. Int. J. Robotics Res., 2011

570 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition, Geology

View Paper: https://www.semanticscholar.org/paper/6f7892e29b76ba107ca3e6a923ce56dd3a7cd580

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 Geology

Citations in the corpus

Visual-Inertial Navigation: A Concise Review (2019, methodology): “With the similar idea, in [74, 129, 160], the observability of IMU-camera (monocular, RGBD) calibration was analytically studied, which shows that the extrinsic transformation between the IMU and camera is observable given generic motions.”
Robust Stereo Visual Inertial Odometry for Fast Autonomous Flight (2017, methodology): “In contrast, filter-based approaches, which generally use the Extended Kalman Filter (EKF) [1], or the Uncented Kalman Filter [19], are much more efficient while achieving accuracy comparable to optimization-based approaches.”
Vision-based state estimation for autonomous rotorcraft MAVs in complex environments (2013, background): “A nonlinear observability analysis of the estimation problem shows the presence of unobservable modes that can only be eliminated through motions that involve non-zero linear accelerations [4, 5].”
Unified temporal and spatial calibration for multi-sensor systems (2013, methodology): “More recently, recursive approaches were employed to jointly estimate relative rotation and translation from measurements acquired by dynamically moving a camera IMU combination in front of a calibration pattern [10], [11].”
A Tutorial on Quantitative Trajectory Evaluation for Visual(-Inertial) Odometry (2018, background): “This yaw-only rigid body transformation (one DoF rotation plus a translation) corresponds to the four unobservable DoFs for visual-inertial systems [12].”
Self-supervised calibration for robotic systems (2013, background): “[18] or [19]), proves what we will call structural observability—that there exists some dataset for which the parameters are observable—it does not guarantee that the parameters are observable for any dataset.”
Vision based navigation for micro helicopters (2012, methodology): “The gravity vector has been estimated in other work (such as in (Kelly and Sukhatme (2011))) because the reference frame was the drifting and unaligned vision frame.”
Keyframe-based visual–inertial odometry using nonlinear optimization (2015, background): “…cameras (Jia and Evans, 2012; Li et al., 2013), offline (Lobo and Dias, 2007; Mirzaei and Roumeliotis, 2007, 2008) and online (Jones and Soatto, 2011; Kelly and Sukhatme, 2011; Dong-Si and Mourikis, 2012; Weiss et al., 2012) calibration of the relative position and orientation of camera and IMU.”
A framework for maximum likelihood parameter identification applied on MAVs (2018, unknown): “Hx,k = ∂rz(xk,θ, zk, vk) ∂xk (x̄k, ?̄?, zk,0) (20)”
High-precision, consistent EKF-based visual-inertial odometry (2013, methodology): “…the EKF Jacobians are computed, even though the IMU’s rotation about gravity (the yaw) is not observable in VIO (see, e.g., (Jones and Soatto, 2011; Kelly and Sukhatme, 2011; Martinelli, 2012)), it appears to be observable in the linearized system model used by the MSCKF, and the same occurs in…”
Optic-Flow Based Control of a 46g Quadrotor (2013, background): “One of the most successful solution is inertial SLAM (Simultaneous Localization and Mapping) [2], [3], [4], [5],”
Estimation, Control, and Planning for Aggressive Flight With a Small Quadrotor With a Single Camera and IMU (2017, methodology): “The minimal sensor suite for autonomous localization consists of two inexpensive, lightweight and widely available sensors, a monocular camera and an IMU as shown in [12], [13], [14], [15].”
Advanced closed loop visual navigation for micro aerial vehicles (2014, methodology): “The whole system in this work and earlier work (Kelly and Sukhatme, 2011; Weiss et al., 2013) is considered as “self-calibrating power-on-and-go” system, meaning that we only need to initialize inter- and intra-sensor calibration approximately, while the state-estimation framework estimates the exac”
Real-time onboard visual-inertial state estimation and self-calibration of MAVs in unknown environments (2012, methodology): “There exist various methods in the literature to calibrate these unknowns [7], [8].”
Keyframe-Based Visual-Inertial SLAM using Nonlinear Optimization (2013, unknown): “Kelly and Sukhatme [9] provided calibration results and a study of observability in the context of filtering-based visionIMU fusion.”
Opportunities and challenges with autonomous micro aerial vehicles (2012, background): “Algorithms for self-calibration and the elimination of the biases in the IMU facilitate such solutions (Mirzaei and Roumeliotis, 2008; Kelly and Sukhatme, 2011).”