Real-time onboard visual-inertial state estimation and self-calibration of MAVs in unknown environments

By Stephan Weiss, Markus Achtelik, Simon Lynen, Margarita Chli, Roland Yves Siegwart. IEEE International Conference on Robotics and Automation, 2012

435 citations recorded by the source corpus.

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

View Paper: https://www.semanticscholar.org/paper/4e0da7fddfb5cdbe6acb139bfe247f1d2fba3adc

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 Ocean Engineering

Citations in the corpus

A robust and modular multi-sensor fusion approach applied to MAV navigation (2013, methodology): “In our previous work [17], we proposed to use a ring buffer for the states so that we could apply state updates in the past.”
Keyframe-based visual–inertial odometry using nonlinear optimization (2015, methodology): “, 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.”
Automatic re-initialization and failure recovery for aggressive flight with a monocular vision-based quadrotor (2015, background): “This research was supported by the Swiss National Science Foundation through project number 200021-143607 (Swarm of Flying Cameras) and the National Centre of Competence in Research Robotics (NCCR). control) are required by the operators, which renders reinitialization during flight very difficult o”
Robust Stereo Visual Inertial Odometry for Fast Autonomous Flight (2017, background): “There is extensive literature on visual odometry, ranging from pure vision-based methods [8]–[12], loosely-coupled VIO solutions [13], [14], to the recently more popular tightlycoupled VIO solutions [1], [4], [5], [15]–[20] which will be the major focus of this paper.”
Vision-Controlled Micro Flying Robots: From System Design to Autonomous Navigation and Mapping in GPS-Denied Environments (2014, methodology): “Later, the system was extended [16] to incorporate data from an Inertial Measurement Unit (IMU) and, thus, estimate the absolute scale automatically while self-calibrating all the sensors (this approach will be outlined in Section IV).”
Direct visual-inertial odometry with stereo cameras (2016, background): “In early works, visual-inertial fusion has been approached as a pure sensor-fusion problem: Vision is treated as an independent, black-box 6-DoF sensor which is fused with inertial measurements in a filtering framework [30] [23] [6].”
VINet: Visual-Inertial Odometry as a Sequence-to-Sequence Learning Problem (2017, methodology): “For comparison, we instead mimic the system of (Weiss et al. 2012) by fusing pose estimates from LIBVISO2 (Geiger et al. 2013) with the inertial data using an EKF.”
Keyframe-Based Visual-Inertial SLAM using Nonlinear Optimization (2013, methodology): “The estimates of the vision-only approach are then used as input to the loosely-coupled baseline algorithm of [20] (with fixed scale and inter-sensor calibration).”
Optic-Flow Based Control of a 46g Quadrotor (2013, methodology): “A simpler approach consists of fusing inertial data directly with the epipolar constraint [8], [9], [10], [11].”
A framework for maximum likelihood parameter identification applied on MAVs (2018, neutral): “rz,k(xk,θ, vk) ≈ r̄z,k + Hx,kδxk + Hθ,kδθk + Hv,kvk (19)”
High-precision, consistent EKF-based visual-inertial odometry (2013, methodology): “For instance, (Mourikis and Roumeliotis, 2007 and Tardif et al., 2010) employ Runge–Kutta numerical integration, (Weiss and Siegwart, 2011; Weiss et al, 2012) use a closed-form, approximate solution to the differential equation, while several algorithms (especially in the GPSaided inertial…”
Visual-Inertial Navigation: A Concise Review (2019, methodology): “led fusion, in either ﬁltering or optimization-based estimation, processes the visual and inertial measurements separately to infer their own motion constraints and then fuse these constraints (e.g., [27, 87, 88, 89, 90, 91]). Although this method is computationally efﬁcient, the decoupling of visua”
Advanced closed loop visual navigation for micro aerial vehicles (2014, background): “Our modifications (Weiss et al. (2012b) and Section 3.3.2 in this dissertation) enabled its use not only in small work-spaces, but also in large outdoor environments.”