Riding quality evaluation through mobile crowdsensing

Public transport plays an importation role in our daily life. The information related to passengers satisfaction is very beneficial for optimizing the transportation service. The comfort experienced by passengers in a given environment can be treated as a subjective assessment, because it is possible to find a considerable difference in responses of different people to the same situation. Besides, the factors on which the opinions of passengers on comfort level are based are objective variables that characterize the surroundings, e.g. temperature, noise, driving behabior, and passengers crowding rate. This thesis investigates an application of mobile crowd sensing to detect and analyze the riding quality of public transport vehicles. The lightweight system leverage sensors equipped on participator’s smartphones to collect surrounding information. By analyze the uploaded data at server, we are able to estimate both aggressive driving behaviors and environment contexts. Series of data processing methods are exploited to overcome the affection of body movement and road condition, and crowd sourcing is applied to improve the robustness of the results. We have tested this system in several vehicles in multiple cities. The results indicate that the system can provide sufficient accuracy to identify dozens of riding-comfort metrics.