A machine learning based system for multi-sensor 3D localization of stationary objects

  • Everton Berz, Deivid Tesch, Fabiano Hessel
  • IET Cyber-Physical Systems Theory & Applications, January 2018, the Institution of Engineering and Technology (the IET)
  • DOI: 10.1049/iet-cps.2017.0067

Indoor localization system for stationary objects

What is it about?

Indoor object localization technique. In a warehouse is very expensive locate a object or a set of objects. You can not use a GPS (not work in indoor environments). The market solutions are based in signal processing and need an expensive hardware infraestructure. Our solution is based on visual techniques combined with machine learning techniques, producing a less expensive and more accurate solution for indoor object localization in warehouses.

Why is it important?

Combine visual techniques with machine learning techniques, producing a less expensive and more accurate solution for indoor object localization in warehouses, allowing to reduce the time spent to find objects in warehouses. Also, the proposed solution save computer processing, savng energy consumption.

Perspectives

Fabiano Hessel
Pontificia Universidade Católica do Rio Grande do Sul (PUCRS)

We hope that this article can contribute to improve the indoor object localization systems. Our focus is not the warehouses of big industries, like Amazon for example. These industries have your owned indoor located systems. However, exist a big market to be explored. To medium and smal industries the market solutions are very expensive or not enough accurate. We think that medium and small industries can have real gains with our solution.

Read Publication

http://dx.doi.org/10.1049/iet-cps.2017.0067

The following have contributed to this page: Everton Luís Berz and Fabiano Hessel