Machine Learning, Game Theory, Algorithms, Mobile Applications
Crowdsourcing Based Bike-Sharing System, 2016 - 2018
Bike-Sharing System (BSS) is a convenient service deployed in many big cities to alleviate the last-mile problem. The explosion of users in BSS causes imbalance of resources in the system and inconvenience of bike utilization for user. Now, it is quite often that some users are not able to borrow or return bikes from or to stations because of the serious imbalance of resources in BSS.
To tackle the problem, we built a mobile crowdsourcing platform to collect real-time bike resources information at bike stations by crowdsourcing to provide users more accurate information of bike stations nearby. It also provides users the full bike utilization guidance, which composes of three segments: from user's start point to a start station, from the start station to a target station and from the target station to user’s terminal point.
Internet of Things Based Fire Escaping System, 2014 - 2018
Fire hazard puts health of millions at risk and results in billions of loss all around the world every year. When fire hazard occurs in a building, especially the building with complex structure, people in the building are unable to find the feasible or better escaping route without guidance so that the precious surviving time is wasted, which causes unnecessary injury or death. Therefore, a fire escaping system that provides real time fire and routing information during fire escaping can help people escape as fast and safe as possible.
In this work, we design a fire spread prediction based fire escaping route planning approach for people. We design a smartphone based neighbor discovery method so that the crowd information could be considered in fire escaping route planning. We also build a network of TelosB nodes, Android smartphones and Arduino suite (Arduino Mega 2560, XBee and Wifi Bee).
Related paper is published in PPNA [pdf].