Localization Limitations of ARCore, ARKit, and Hololens in Dynamic Large-scale Industry Environments

  title={Localization Limitations of ARCore, ARKit, and Hololens in Dynamic Large-scale Industry Environments},
  author={Tobias Feigl and Andreas Porada and Steve Steiner and Christoffer Loeffler and Christopher Mutschler and Michael Philippsen},
Augmented Reality (AR) systems are envisioned to soon be used as smart tools across many Industry 4.0 scenarios. The main promise is that such systems will make workers more productive when they can obtain additional situationally coordinated information both seemlessly and hands-free. This paper studies the applicability of today’s popular AR systems (Apple ARKit, Google ARCore, and Microsoft Hololens) in such an industrial context (large area of 1,600m2, long walking distances of 60m between… 

Figures and Tables from this paper


A new Mobile Augmented Reality (MAR) methodology that exploits OPEN-V-SLAM to extend the application range of Unity AR Foundation and better handle large-scale environments is proposed.

VI-SLAM2tag: Low-Effort Labeled Dataset Collection for Fingerprinting-Based Indoor Localization

This work presents VI-SLAM2tag, a system for auto-labeling of dynamically collected fingerprints using the visual-inertial simultaneous localization and mapping (VI- SLAM) module of ARCore, and shows how it can be used to train neural network based models that achieve a positioning accuracy of roughly 2 m.

A Method for Location Initialization of Handheld Devices using Autonomous Driving Vehicles for Interactive Systems

This work proposes a method for acquiring the relative position of smartphones by exploiting the localization information of mobility systems and their user interaction functionality, using a vehicle's own position and mapping information as an initialization reference for estimating a smartphone’s relative position.

A Quest for Co-Located Mixed Reality: Aligning and Assessing SLAM Tracking for Same-Space Multi-User Experiences

The utility of aligned SLAM solutions is explored by surveying approaches toward aligning disparate device coordinate spaces, formalizing their theoretical accuracy and limitations, and detailing how to assess the accuracy and safety of 6DoF/SLAM tracking solutions for any arbitrary device and dynamic environment without the need for an expensive ground truth optical tracking.

Comparison of Three Off-the-Shelf Visual Odometry Systems

A single odometry system might still not be sufficient, so using multiple instances and sensor fusion approaches are necessary while waiting for additional research and further improved products.

Overcoming the problem of uncertain tracking errors in an AR navigation application

This thesis proposes four different Augmented Reality display techniques that can overcome the problem of uncertain tracking errors and uses a mix between spatially registered information and turn instructions to achieve that result.

A Sense of Quality for Augmented Reality Assisted Process Guidance

A Machine Learning (ML) classifier is developed that predicts quality metrics from a 9-degrees of freedom inertial measurement unit while it simultaneously guide and track the work processes with a HoloLens AR system to enrich modern assistance systems with a sense of process quality.

Benchmarking Built-In Tracking Systems for Indoor AR Applications on Popular Mobile Devices

A benchmark evaluation of the tracking performances of some of the most popular AR handheld devices, which can be regarded as a representative set of devices for sale in the global market.


Abstract. Augmented Reality/Virtual Reality (AR/VR) technology offers novel and excited experiences to visitors of cultural sites. Extended recent technological advances in smart communication

RoVaR: Robust Multi-agent Tracking through Dual-layer Diversity in Visual and RF Sensor Fusion

RoVaR is an embodiment of a dual-layer diversity approach that intelligently attends to cross-modal information using algorithmic and data-driven techniques that jointly share the burden of accurately tracking multiple agents in the wild.



Visual SLAM and Structure from Motion in Dynamic Environments

This article presents for the first time a survey of visual SLAM and SfM techniques that are targeted toward operation in dynamic environments and identifies three main problems: how to perform reconstruction, how to segment and track dynamic objects, and how to achieve joint motion segmentation and reconstruction.

Evaluating and enhancing google tango localization in indoor environments using fiducial markers

A thorough evaluation of the localization accuracy of the Tango platform in different conditions is presented; a fiducial marker-based extension of theTango localization system is presented, which improves the localization estimates in certain conditions and a solution based on the use of additional visual markers is proposed to improve tracking accuracy in dynamic environments where spatial and/or illumination changes may occur.

Monocular Visual-Inertial State Estimation for Mobile Augmented Reality

This work proposes a tightly-coupled, optimization-based, monocular visual-inertial state estimation for robust camera localization in complex indoor and outdoor environments and develops a lightweight loop closure module that is tightly integrated with the state estimator to eliminate drift.

ORB-SLAM2: An Open-Source SLAM System for Monocular, Stereo, and RGB-D Cameras

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.

Parallel Tracking and Mapping for Small AR Workspaces

A system specifically designed to track a hand-held camera in a small AR workspace, processed in parallel threads on a dual-core computer, that produces detailed maps with thousands of landmarks which can be tracked at frame-rate with accuracy and robustness rivalling that of state-of-the-art model-based systems.

Vision meets robotics: The KITTI dataset

A novel dataset captured from a VW station wagon for use in mobile robotics and autonomous driving research, using a variety of sensor modalities such as high-resolution color and grayscale stereo cameras and a high-precision GPS/IMU inertial navigation system.

The TUM VI Benchmark for Evaluating Visual-Inertial Odometry

The TUM VI benchmark is proposed, a novel dataset with a diverse set of sequences in different scenes for evaluatingVI odometry, which provides camera images with 1024×1024 resolution at 20 Hz, high dynamic range and photometric calibration, and evaluates state-of-the-art VI odometry approaches on this dataset.

Keyframe-based visual-inertial online SLAM with relocalization

This paper presents a keyframe-based approach to visual-inertial simultaneous localization and mapping (SLAM) for monocular and stereo cameras based on a real-time capable visual- inertial odometry method that provides locally consistent trajectory and map estimates.

Supervised Learning for Yaw Orientation Estimation

This novel yaw orientation estimation combines a single inertial sensor located at the human's head with inaccurate positional tracking to enable low-cost long-time stable head orientation in NP tracking on 100 $m\times 100$ m.

Detecting position using ARKit II: generating position-time graphs in real-time and further information on limitations of ARKit

ARKit is a framework which allows developers to create augmented reality apps for the iPhone and iPad. In a previous study, we had shown that it could be used to detect position in educational