eduzhai > Applied Sciences > Engineering >

FADACS A Few-shot Adversarial Domain Adaptation Architecture for Context-Aware Parking Availability Sensing

  • king
  • (0) Download
  • 20210506
  • Save

... pages left unread,continue reading

Document pages: 10 pages

Abstract: Existing research on parking availability sensing mainly relies on extensivecontextual and historical information. In practice, the availability of suchinformation is a challenge as it requires continuous collection of sensorysignals. In this study, we design an end-to-end transfer learning framework forparking availability sensing to predict parking occupancy in areas in which theparking data is insufficient to feed into data-hungry models. This frameworkovercomes two main challenges: 1) many real-world cases cannot provide enoughdata for most existing data-driven models, and 2) it is difficult to mergesensor data and heterogeneous contextual information due to the differing urbanfabric and spatial characteristics. Our work adopts a widely-used concept,adversarial domain adaptation, to predict the parking occupancy in an areawithout abundant sensor data by leveraging data from other areas with similarfeatures. In this paper, we utilise more than 35 million parking data recordsfrom sensors placed in two different cities, one a city centre and the other acoastal tourist town. We also utilise heterogeneous spatio-temporal contextualinformation from external resources, including weather and points of interest.We quantify the strength of our proposed framework in different cases andcompare it to the existing data-driven approaches. The results show that theproposed framework is comparable to existing state-of-the-art methods and alsoprovide some valuable insights on parking availability prediction.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×