On-street parking sensing

On-street parking sensing refers to technologies used to obtain and present real-time parking availability information. Approaches to the problem of finding a free parking spot include analytics and advanced hardware systems.

The advent of autonomous cars is creating a second wave of demand for such data.[1]

Approaches

Smartphone data analytics

Adoption of smartphones by drivers allows tracking vehicle movement through accelerometer and GPS data and thus derive parking session start and end time. Combined with parking capacity data, this allows an estimation of real-time parking availability. Google Maps announced this functionality in 2017 based on historical data gathered from phones with location services turned on.[2][3] Although zero infrastructure cost and general principle is appealing, lack of penetration among drivers and smartphone sensor data interpretation uncertainty limits the prediction accuracy.

Camera analytics

One approach is to use computer vision to derive automated parking occupancy estimates from on-street cameras.[4][5] Solutions include video and snapshot analytics with cloud and onboard processing. This approach offers higher accuracy and potential cost-sharing given multi-purpose camera use for somewhat higher infrastructure cost and limited accuracy at night and in bad weather conditions. Cloud processing solutions have data traffic requirements starting from 10Gb/month. Onboard processing systems have lower data requirements but require more powerful (and expensive) hardware.

Lamp post radars

Ultrasonic and electromagnetic rangefinders mounted on lamp posts and other street infrastructure can detect parking events. The sensing unit cost is lower than for cameras but shorter sensing range, making the overall system metrics close to camera analytics in terms of accuracy and cost.

On-stall road sensors

San Francisco and Los Angeles have tried pilot programs to gather parking space occupancy data via sensors and smart parking meters, as part of matching parking prices with demand.[6] Battery powered sensing unit installed flush or surface mount in parking spaces using magnetic, infrared and/or ultrasound. High spatial resolution and close proximity provide higher accuracy than camera analytics and radars. Road works and wireless infrastructure deployment make overall system roll out more difficult to manage.

See also

References

  1. ↑ "This is How Much Time Self-Driving Cars Will Save You". Inverse. Retrieved 2017-10-14.
  2. ↑ "Google Maps will now help you find parking". The Verge. Retrieved 2017-10-14.
  3. ↑ "Google Maps Can Now Tell You How Utterly Horrible Parking Is Gonna Be". Inverse. Retrieved 2018-11-02.
  4. ↑ Matsuda, Akihiro; Matsui, Tomokazu; Matsuda, Yuki; Suwa, Hirohiko; Yasumoto, Keiichi (March 2021). "A System for Real-time On-street Parking Detection and Visualization on an Edge Device". 2021 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events (PerCom Workshops). pp. 227–232. Bibcode:2021perc.conf...95M. doi:10.1109/PerComWorkshops51409.2021.9431076. ISBN 978-1-6654-0424-2.
  5. ↑ Morell, José Ángel; Luque, Gabriel; Alba, Enrique (July 2025). "On-street parking space localization with deep learning using low-quality images from public cameras". Internet of Things. 32 101619. doi:10.1016/j.iot.2025.101619.
  6. ↑ Chieppo, Charles (2014-05-22). "Playing the Slots: Technology's Growing Role in Bringing Efficiency to Parking". GovTech. Retrieved 2026-08-06.
This page is issued from Wikipedia. The text is available under Creative Commons Attribution-Share Alike 4.0 unless otherwise noted. Additional terms may apply for the media files.