Robots: Humans' Dependable Helpers

Research and Development in Robotic Systems

Visual Odometrie

© Fraunhofer IFF, Peter Förster

Precise Autonomous Robot Localization in Large Environments such as Factory Buildings by Visual Odometry

Continuous localization of mobile units such as automatic guided transport systems or mobile assistance robots is already a basic technology in automation and logistic. At the same time, existing localization systems entail a substantial investment in infrastructure and/or mobile sensor systems. Visual self-localization on the basis of inexpensive digital camera systems provide more significant potential to cut costs and opportunities to develop new users and fields of application, especially among small and medium-sized enterprises.

In light of these considerations, the content of this subproject of the joint project AVILUS+ was the development of a visual position tracking systems specifically for environments such as factories and warehouses:

  • as a subcomponent for visual self-localization of mobile autonomous systems and
  • as an autonomous system that records other mobile units' routes and travel times

Camera-based navigation and localization of a mobile robot through the use of existing natural features in factory buildings.

Concrete emphases of development were the following technological aspects:

  • Hybrid deep structure of the environment (attributes are both at close range and extend quasi infinitely)
  • Work under poor lighting conditions
  • Handling of blurring caused by camera movement
  • Incorporation of dynamics in the environment

The developed systems were implemented as a prototype of a compact, integrated sensor unit for installation on mobile systems. The system includes two cameras for markerless and marker-supported localization as well as additional LED lighting and computer technology. Industrial ethernet can be used for data integration.

Project data AVILUSplus (overall project):

  • Support program: IKT 2020/Research for Innovation
  • Program focus: Virtual and Augmented Reality
  • Project Ref. No.: 01IM08002
  • Support volume: 10.1 million euros