Precise Robot Localization in Architectural 3D Plans

  title={Precise Robot Localization in Architectural 3D Plans},
  author={Hermann Blum and Julian Stiefel and C{\'e}sar Cadena and Roland Y. Siegwart and Abel Gawel},
This paper presents a localization system for mobile robots enabling precise localization in inaccurate building models. The approach leverages local referencing to counteract inherent deviations between as-planned and as-built data for locally accurate registration. We further fuse a novel image-based robust outlier detector with LiDAR data to reject a wide range of outlier measurements from clutter, dynamic objects, and sensor failures. We evaluate the proposed approach on a mobile robot in a… 

Figures and Tables from this paper

Construction Robot Localization System Based on Multi-sensor Fusion and 3D Construction Drawings

This paper presents a localization system which uses multi-sensor fusion and 3D construction drawings to conduct high-precision localization in construction sites and shows that the system achieves higher accuracy than LiDAR SLAM localization.

Towards BIM-based robot localization: a real-world case study

This paper converts BIM data intolocalization-oriented point cloud maps into iterative closest point (ICP)-based localization on the metric map using a laser scanner, and shows that ICP-based localization can track the robot pose with low errors, demonstrating the feasibility of BIM-based robot localization.

Model Aware LiDAR Odometry and Mapping (MA-LOAM): Improving Simultaneous Localization and Mapping accuracy by robustly leveraging a Building Information Model

A methodology that is resilient to outliers and scan model differences is proposed and added on top of Lidar Odometry And Mapping (LOAM) and also made open source to the public.

Semantic localization in BIM using a 3D LiDAR sensor

The experimental results show that the semantic localization of a mobile 3D LiDAR sensor in BIM can track the pose with only scan matching and present centimeter-level errors over 340 meters traveling, thus demonstrating the feasibility of the proposed mapping-free localization framework.


Abstract. In contrast to stationary systems, mobile robots have an arbitrarily expandable workspace. As a result, the spatial dimensioning of the task to be mastered plays only a subordinate role and

Online Synchronization of Building Model for On-Site Mobile Robotic Construction

This research presents a novel method for a data flow that synchronizes building information with the robot map and updates building components to their "as-built" states, in order to facilitate an

Self-Improving Semantic Perception for Indoor Localisation

A framework in which semantic models are continuously updated on the robot to adapt to the deployment environments by combining continual learning with self-supervision is proposed, which improves its perception during deployment and maintains this improvement while adapting to further environments.

Self-Improving Semantic Perception on a Construction Robot

This work proposes a framework in which semantic models are continuously updated on the robot to adapt to the deployment environments, and tightly couples multi-sensor perception and localisation to continuously learn from self-supervised pseudo labels.

LaLaLoc: Latent Layout Localisation in Dynamic, Unvisited Environments

LaLaLoc enables fine-grained pose estimation in a scene without the need for prior visitation, as well as being robust to dynamics, such as a change in furniture configuration.

Robotic Plaster Spraying: Crafting Surfaces with Adaptive Thin-Layer Printing

Embedded in a long tradition of craftsmanship, inside or outside building surfaces, is often treated with plaster, which plays both functional and ornamental roles. Today, plasterwork is predominan...



On the position accuracy of mobile robot localization based on particle filters combined with scan matching

This paper analyze the accuracy of an integrated laser-based robot pose estimation and positioning system for mobile platforms and shows that by combining standard components such as Monte-Carlo localization, KLD sampling, and scan matching, an accuracy of a few millimeters at taught-in reference locations can be achieved.

Robust LiDAR-based localization in architectural floor plans

This paper proposes a system for robot localization in architectural CAD drawings that employs a simultaneous localization and mapping approach to online augment the floor plan with a map represented as a pose-graph with LiDAR measurements and achieves sub-centimeter accuracy.

The normal distributions transform: a new approach to laser scan matching

  • P. BiberW. Straßer
  • Computer Science
    Proceedings 2003 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2003) (Cat. No.03CH37453)
  • 2003
First results on real data demonstrate, that the normal distributions transform algorithm is capable to map unmodified indoor environments reliable and in real time, even without using odometry data.

SuMa++: Efficient LiDAR-based Semantic SLAM

An extension of a recently published surfel-based mapping approach exploiting three-dimensional laser range scans by integrating semantic information to facilitate the mapping process, which enables us to reliably filter moving objects, but also improve the projective scan matching via semantic constraints.

Real-time loop closure in 2D LIDAR SLAM

This work presents the approach used in the backpack mapping platform which achieves real-time mapping and loop closure at a 5 cm resolution and provides experimental results and comparisons to other well known approaches which show that, in terms of quality, this approach is competitive with established techniques.

A Fully-Integrated Sensing and Control System for High-Accuracy Mobile Robotic Building Construction

It is shown that the system can achieve sub-cm end-effector positioning accuracy during fully autonomous operation using solely onboard sensing and a novel method for high-accuracy localization updates relative to the known building structure is proposed.

Autonomous repositioning and localization of an in situ fabricator

This paper introduces a newly developed machine, called the In situ Fabricator (IF), whose goal is to bring digital fabrication to the construction site and demonstrates that the IF can build with sub-centimeter accuracy over long building sequences.

LOAM: Lidar Odometry and Mapping in Real-time

The method achieves both low-drift and low-computational complexity without the need for high accuracy ranging or inertial measurements and can achieve accuracy at the level of state of the art offline batch methods.

SeDAR - Semantic Detection and Ranging: Humans can Localise without LiDAR, can Robots?

This work presents a global localisation approach that relies solely on the semantic labels present in the floorplan and extracted from RGB images, and demonstrates that range measurements are unnecessary as it can achieve results comparable to state-of-the-art without them.

Fast Global Registration

An algorithm for fast global registration of partially overlapping 3D surfaces that provides the accuracy achieved by well-initialized local refinement algorithms, without requiring an initialization and at lower computational cost.