You are here : Home > INDUSTRIAL INNOVATION > DEMOS > Automatic fine-grained transport mode recognition for wearable-based mobility applications

Articles & files | Focus | Article

Mobility Observer

Published on 23 March 2023

Automatic fine-grained transport mode recognition for wearable-based mobility applications



Transport Mode Recognition is a data fusion process, which:

  • Classifies automatically a person's or object's transport mode while on the go
  • Leverages various measurements provided by sensors typically integrated into smartphones and wearables

Classification level refinement allows:

  • Differentiation between similar usage cases, e.g. detailed rail or road transport modes, while conserving device autonomy


Fine-Grained Transportation Mode Recognition is a compulsory tool to improve applications for intermodality, social and urban sensing uses and energetic efficiency

  • Carbon footprint estimation
  • Real-time door-to-door smart planning
  • Smart mobility surveying
  • Mobility behavior analysis for specific social groups
  • Driving analysis
  • Road user analysis and collision prevention
  • Goods mobility tracking
  • Mode-centric services and applications

   What’s new?

  • Fine road, rail and airplane transport recognition
  • Qualitative walking information

 How does it work?

mobility observer.PNG

  What’s Next?

The "Bon Voyage" cooperation project, funded by the EU Horizon 2020 research and innovation program (Grant 635867), has successfully developed fine transport mode recognition and this will soon enrich a real-time journey planning application.

Leti researchers continue to pioneer affordable, innovative, smart solutions for users and operators in the global mobility market by fusing sensors, increasing Observer performances, device autonomy and developing crowd sensing functionality.

Leti's approach embraces
  • Specification of requirements (latency, autonomy, etc.)
  • Adjustment of fine-grained recognition (classified mode output)
  • Integration, testing and transfer to industry.


Lorintiu, O., & Vassilev,  A. (novembre 2016).
Transportation mode recognition based on smartphone embedded sensors for carbon footprint estimation in Intelligent Transportation Systems (ITSC), 2016 IEEE 19th International Conference (pp. 1976-1981). IEEE.
  •     FLYER