Off-Site Indoor Localization Competitions Based on Measured Data in a Warehouse

Area of Science:

• Robotics and Automation
• Geomatics Engineering
• Computer Science

Background:

• Indoor localization performance varies significantly with application context.
• Existing competitions lack evaluation of practical industrial scenarios.
• Controlled environments do not fully represent real-world challenges.

Purpose of the Study:

• Introduce two novel indoor localization competitions: PDR Challenge 2017 and xDR Challenge 2018.
• Evaluate practical worker and vehicle tracking performance in warehouse environments.
• Establish a benchmark for indoor localization in industrial settings.

Main Methods:

• Utilized real-world data from actual warehouse picking operations for the PDR Challenge 2017.
• Introduced 'xDR' (eXtended Dead Reckoning), combining Pedestrian Dead Reckoning (PDR) and Vehicle Dead Reckoning (VDR).
• The xDR Challenge 2018 focused on smartphone-based VDR for forklift tracking, a world-first.

Main Results:

• Successfully organized competitions with international participation, demonstrating significant interest.
• Collected and utilized actual measured data for realistic performance evaluation.
• Applied multi-faceted evaluation metrics to assess localization accuracy and robustness.

Conclusions:

• The PDR and xDR Challenges provide unique capabilities for evaluating practical indoor localization in warehouses.
• These competitions offer a more realistic assessment than those using simulated or controlled data.
• Findings from the competitions offer valuable insights for advancing industrial indoor tracking technologies.