Towards comfortable cycling: A practical approach to monitor the conditions in cycling paths

Abstract

This is a no brainer. Using bicycles to commute is the most sustainable form of transport, is the least expensive to use and are pollution-free. Towns and cities have to be made bicycle-friendly to encourage their wide usage. Therefore, cycling paths should be more convenient, comfortable, and safe to ride. This paper investigates a smartphone application, which passively monitors the road conditions during cyclists ride. To overcome the problems of monitoring roads, we present novel algorithms that sense the rough cycling paths and locate road bumps. Each event is detected in real time to improve the user friendliness of the application. Cyclists may keep their smartphones at any random orientation and placement. Moreover, different smartphones sense the same incident dissimilarly and hence report discrepant sensor values. We further address the aforementioned difficulties that limit such crowd-sourcing application. We evaluate our sensing application on cycling paths in Singapore, and show that it can successfully detect such bad road conditions.

Inferring Phone Location State

“…In a somewhat different approach, work by Wijerathne et al [13] took advantage of smartphone accelerometers to help monitor road conditions. Similar to our work, the paper attempted to generalize between all positions the phone could be in, but instead of detecting the exact position of the user's phone, the authors attempted to detect rough roads and bumps while cycling.…”

Mining Composite Spatio-Temporal Lifestyle Patterns from Geotagged Social Data

“…Prior works on understanding lifestyles of different demographics have used targeted one-off surveys [9], coupled with location-tracking smartphone apps for residents or tourists [10], [11], required developing custom mobile apps for crowdsourced lifestyle profiling of the older population [12], or the use of proprietary data such as call data records for activitybased pattern mining [13]. A recent work by Hu et al [8] used LBSN data for deriving spatial and temporal patterns analogously.…”

Urban Space Insights Extraction using Acoustic Histogram Information

“…With the recent enhancements of wireless networks, large-scale systems for various sectors can be realized easily, which resides an enormous amount of spatial-temporal data that can be harvested. These data contain many interesting aspects of day-to-day utilization of an urban space or life of citizens [1,2,3,4,5,6]. With the advancement of machine learning and data mining technique, different platforms have been developed to cater different usages of monitoring in a smart city.…”