Toward Enabling Convenient Urban Transit through Mobile Crowdsensing

Abstract

Abstract-The smart cities of the future are expected to be serviced by advanced, personalized multimodal transit systems, charged with timely transport of citizens. Optimizing routes on such networks is a complex problem, in part due to the fact that simple metrics such as latency by themselves are not sufficient to find the best routes. In this paper, we focus on the problem of providing commuters with personalized routes with the most convenience. We present our mathematical model of user convenience during a multi-leg journey, and the overview of a middleware for enabling convenient transit (including ensuring acceptable network connectivity to mobile apps) by using crowdsourcing. We also report on initial insights obtained through empirical studies on network connectivity and user-perception of convenience in Delhi, India, and Paris, France.

Towards building real-time, convenient route recommendation system for public transit

“…This can be done using a suitable navigation service. Some well-known navigation services include OpenStreetMap 6 and Google Maps Directions API 7 . Note that, other navigation services are also available, but due to limited space we are not listing them here.…”

“…• Communication component: Intermittent network connectivity in public transport [7] may result in message delays or drops. This is undesirable in our application as the transport network is dynamic in nature, and a message may be crucial (requiring zero drop rate), or no longer relevant after some time (after acceptable delay).…”

“…Further, the use of Bayesian incremental learning makes their system computationally intensive, and may result in delays in routeplanning. To overcome these challenges, we have defined convenience in public transit based on three parameters [7] -time, crowdedness, and seat availability. In this paper, we extend this work to propose a system -Sarathi 5 , that identifies users' perception of convenience based on these parameters and uses it to personalize their public transit experiences.…”

“…), the recommendations must be provided to the commuters in real-time. Owing to intermittent and varying network connectivity within public transport [7], we identify the need for a middleware to facilitate the communication of these recommendations to the commuters in real-time.…”

“…We present the initial choice of tools used to develop our system followed by the implementation details (Section IV). Our system is based on our previous work [7] where we have defined convenience based on three parameters -i) seat availability, ii) time delay, and iii) comfort experienced. Section V finally concludes our work.…”