Visual-Inertial Navigation: Challenges and Applications

Updates

• (11/16) Speakers’ slides are avaialble now in the program (see the bottom).

• (10/25) The workshop program is available now (at the end of this page). Note: All posters will have (up to) 5-min spotlight oral presentation and 45-min poster presentation.

• (10/1) Nine papers were finally selected to present at the workshop.

• (7/28) We are pleased to annouce that there will be a limited number of IET-CSR Travel Grants to (partitially) support students to travel to workshop to present their work. Information about how to apply coming later.

• (7/26) Upon many requests, paper submission deadline is extended to: August 15, 2019

• (7/26) Paper limits do not include references (say n pages), i.e., 6+n for research papers, 4+n for field reports, and 2+n for demo papers.

• (7/3) We are pleased to announce that LORD Sensing will sponsor the LORD Best Paper Award (with MicroStrain IMUs as prize)!

• (7/2) Please prepare papers following the IROS template available in LaTeX and MS Word.

• (7/1) We are pleased to announce that there will be a Special Issue in the IET Cyber-systems and Robotics, which will invite some of best papers presented at this workshop.

Overview

As cameras and IMUs are becoming ubiquitous, visual-inertial navigation systems (VINS) that provide high-precision 3D motion estimation, hold great potentials in a wide range of applications from augmented reality (AR) and aerial navigation to autonomous driving, in part because of the complementary sensing capabilities and the decreasing costs and size of these sensors. While visual-inertial navigation, alongside with SLAM, has witnessed tremendous progress in the past decade, yet certain critical aspects in the design of visual-inertial systems remain poorly explored, greatly hindering the widespread deployment of these systems in practice. For example, many VINS algorithms are yet not robust to high dynamics and poor lighting conditions; they are yet not accurate enough for long-term, large-scale operations, in particular, in life-critical scenarios; and yet they are unable to provide semantic and cognitive understandings to support high-level decision making. This workshop brings together researchers in robotics, computer vision and AI, from both academia and industry, to share their insights and thoughts on the R&D of VINS. The goal of this workshop is to bring forward the latest breakthroughs and cutting-edge research on visual-inertial navigation and beyond, to open discussions about technical challenges and future research directions for the community, and to identify new applications of this emerging technology.

Call for Contributions

We welcome submissions of papers describing VINS-related work in progress, preliminary results, novel concepts, and industry experiences. All submitted papers will be reviewed by at least two experts (see Program Committee below) on the basis of technical quality, relevance, significance, and clarity. Topics of interest to this workshop include, but are not necessarily limited to:

• Visual-inertial odmetry

• Visual-inertial perception

• Visual SLAM

• Sensor calibration

• High-speed visual control and estimation of aerial vehicles

• Deep learning for visual SLAM

• Cooperative visual-inertial navigation

• Multi-sensor fusion

• Co-design of hardware and software of VINS

• Simulations and benchmarking of visual-inertial navigation

• Visual perception in challenging and dynamic environments

• Human motion modeling

• Field robotics

• AR/VR

We accept the following forms of contributions:

• Research papers (up to 6 pages)

• Field reports (2-4 pages)

• Live demos of working systems (up to 2 pages)

All accepted papers will appear on the workshop website. Note that the authors retain all the intellectual properties of their contributions to the workshop. We will also be exploring the possibility of a journal special issue for the best contributions at the workshop.

Important Dates

• August 1, 2019 –> August 15, 2019: Deadline of paper submission

  • Please submit your papers/reports/demos via email: iros2019-vins-workshop@udel.edu

• September 1, 2019 –> September 8, 2019: Notification of acceptance

• October 1, 2019: Submission of final version

• November 8, 2019: Workshop at IROS 2019

Organizers

• Guoquan (Paul) Huang, University of Delaware

• Shaojie Shen, Hong Kong University of Science and Technology

• Michael Kaess, CMU

• Stergios Roumeliotis, University of Minnesota

• John Leonard, MIT

Invited Speakers (confirmed)

• Luca Carlone, MIT

• Jakob Engel, Facebook Oculus

• Maurice Fallon, Oxford

• Jonathan Kelly, University of Toronto

• Giuseppe Loianno, NYU

• Laurent Kneip, Shanghai Tech

• Davide Scaramuzza, University of Zurich

• Guofeng Zhang, Zhejiang University

• Ross Hartley, Amazon (University of Michigan)

Program Committee

• Kevin Eckenhoff, University of Delaware / Facebook

• Chao Guo, Google

• Shoudong Huang, University of Technology Sydney

• Mingyang Li, Alibaba

• Yasir Latif, University of Adelaide

• Yong Liu, Zhejiang University

• Agostino Martinelli, INRIA Rhone Alpes

• Wisely Babu Benzun Pious, Bosch

• Yue Wang, Zhejiang University

Poster Papers

• Dengshen Chen, Yuanlong Yu, and Xiang Gao: Semi-Supervised Deep Learning Framework for Monocular Visual Odometry

• Geoff Fink, and Claudio Semini: Proprioceptive Sensor Dataset for Quadrupeds

• Wanlong Li, Yu Tang, Chao Ding, Xueshi Li, and Feng Wen: Visual-Inertial Ego-Motion Estimation using Rolling-Shutter Camera in Autonomous Driving

• Jiajun Lyu, Jinhong Xu, Xingxing Zuo, and Yong Liu: An Efficient LiDAR-IMU Calibration Method Based on Continuous-Time Trajectory

• Yongseok Lee, Hanbyeol Yoon, Jinuk Heo, WonHa Lee, and Dongjun Lee: Wearable Visual-Inertial Hand Tracking Interface Regardless of Environment and Occlusion

• Patrick Geneva, Kevin Eckenhoff, Woosik Lee, Yulin Yang, and Guoquan Huang: OpenVINS: A Research Platform for Visual-Inertial Estimation

• Ziqiang Wang, Chengcheng Guo, Lin Zhao, Mei Li, and Xinyu Qi: Direct Sparse Visual-Inertial Odometry with Stereo Cameras

• He Zhang, Lingqiu Jin, and Cang Ye: A Depth-Enhanced Visual Inertial Odometry for a Robotic Navigation Aid for Blind People (LORD Best Paper)

• Joshua Jaekel, and Michael Kaess: Robust Multi-Stereo Visual-Inertial Odometry

Program

• 9:00-9:05AM: Welcome and Introduction

• 9:05-9:45AM: Stergios Roumeliotis (UMN): A Short Tutorial on VINS (slides)

• 9:45-10:15AM: Davide Scaramuzza (Zurich): Visual Inertial SLAM: Current Status and the Road Ahead (slides)

• 10:15-10:45AM: Laurent Kneip (Shanghai Tech): Dimensionality reduction in visual-inertial SLAM (slides)

• 10:45-11:15AM: Maurice Fallon (Oxford): VILENS - the Challenge of Visual Navigation on Quadruped Robots (slides)

• 11:15-11:45AM: COFFEE BREAK

• 11:45-12:15PM: Luca Carlone (MIT): Chasing a Chimera: from VIN to real-time high-level understanding (slides)

• 12:15-1:00PM: Poster Spotlight (5 minutes each paper)

• 1:00-2:00PM: LUNCH; Poster Setup

• 2:00-2:30PM: Guofeng Zhang (ZJU): Robust VI-SLAM and HD-Map Reconstruction for Location-based Augmented Reality (slides)

• 2:30-3:00PM: Giuseppe Loianno (NYU): Challenges and Opportunities for Visual Inertial Navigation of Aerial Robots (slides)

• 3:00-3:30PM: Paloma Sodhi / Michael Kaess (CMU): Robust Multi-Stereo Visual-Inertial Odometry (slides)

• 3:45-4:15PM: COFFEE BREAK

• 4:15-4:45PM: Ross Hartley (Amazon): Contact-aided Invariant Extended Kalman Filtering for Legged Robot State Estimation (slides)

• 4:45-5:30PM: Poster Session

• 5:30-5:45PM: Concluding Remarks (incl. LORD best paper award annoucement)