Trajectory annotation using sequences of spatial perception

@article{Feld2018TrajectoryAU,
  title={Trajectory annotation using sequences of spatial perception},
  author={Sebastian Feld and Steffen Illium and Andreas Sedlmeier and Lenz Belzner},
  journal={Proceedings of the 26th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems},
  year={2018}
}
  • Sebastian FeldSteffen Illium Lenz Belzner
  • Published 6 November 2018
  • Computer Science
  • Proceedings of the 26th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems
In the near future, more and more machines will perform tasks in the vicinity of human spaces or support them directly in their spatially bound activities. In order to simplify the verbal communication and the interaction between robotic units and/or humans, reliable and robust systems w.r.t. noise and processing results are needed. This work builds a foundation to address this task. By using a continuous representation of spatial perception in interiors learned from trajectory data, our… 

Figures from this paper

Learning indoor space perception

It is shown that isovist measures do reflect the recurring structures found in different buildings, that these recurring patterns are encoded in the data in a way that unsupervised machine learning can identify them and that the identified structures are meaningful as they represent human relatable concepts.

References

SHOWING 1-10 OF 49 REFERENCES

Discovering and Learning Recurring Structures in Building Floor Plans

A framework is presented to create a data set containing 2D isovist measures calculated along geospatial trajectories that traverse a 3D simulation environment that reflect the recurring structures found in buildings and the recurring patterns are encoded in a way that unsupervised machine learning is able to identify meaningful structures like rooms, hallways and doorways.

A virtual sensor for room detection

  • P. BuschkaA. Saffiotti
  • Computer Science
    IEEE/RSJ International Conference on Intelligent Robots and Systems
  • 2002
A local technique that uses range data to detect spaces during navigation and segmentation, which isolates room-like spaces and detects when the robot has entered a new one; and feature extraction, which associates each space with a set of geometric features useful for navigation or recognition.

Isovists as a Means to Predict Spatial Experience and Behavior

The general outcomes suggest that isovist measurands are indeed a promising means to predict the experience of space and spatial behavior for the chosen experimental tasks.

Approximated environment features with application to trajectory annotation

This paper defines discrete isovists as an approximation of exact isovism using a simple ray casting approach, and defines, create, and use isovist feature cubes for semantic evaluations of floor plans as well as trajectory annotation.

Semantic Labeling of Places with Mobile Robots

This book presents several approaches to enable a mobile robot to categorize places in indoor environments and describes how these categories are indicated by terms which represent the different regions in these environments.

Supervised Learning of Topological Maps using Semantic Information Extracted from Range Data

This paper presents an approach to create topological maps from geometric maps obtained with a mobile robot in an indoor-environment using range data. Our approach utilizes AdaBoost, a supervised

Analyzing architectural space: identifying salient regions by computing 3D isovists

The use of 3D isovists are described to develop a salient region estimation technique for architectural and urban environments and a quantitative technique for detecting regions with strong visual characteristics is proposed.

Detecting and modeling doors with mobile robots

A probabilistic framework for detection and modeling of doors from sensor data acquired in corridor environments with mobile robots is described, which achieves better results than models that only capture behavior, or only capture appearance.

Building Semantic Annotated Maps by Mobile Robots

A framework to build semantic annotated maps from laser range measurements of a mobile robot and two alternative methods to aggregate the resulting class membership vectors into a grid map have been developed.

Simultaneous map building and localization for an autonomous mobile robot

  • J. LeonardH. Durrant-Whyte
  • Computer Science
    Proceedings IROS '91:IEEE/RSJ International Workshop on Intelligent Robots and Systems '91
  • 1991
Discusses a significant open problem in mobile robotics: simultaneous map building and localization, which the authors define as long-term globally referenced position estimation without a priori