Estimating Fuel Consumption and Emissions via Traffic Data from Mobile Sensors

Mobile sensing enabled by on-board GPS or smart phones has become the primary source of traffic data. For sufficient coverage of the traffic stream, it is important to maintain a reasonable penetration rate of probe vehicles. From the standpoint of estimating higher-order traffic quantities such as acceleration/deceleration, emission rate and fuel consumption rate, it is desirable to examine the effectiveness of sampling frequency of current sensing technology in capturing higher-order variations inherent in traffic stream. Of the two concerns raised above, the latter is rarely studied in the literature. In this paper, we study the two characteristics of mobile sensing: penetration rate and sampling frequency, and their impacts on the quality of traffic estimation. We utilize a second-order hydrodynamic model known as the phase transition model [Colombo, 2002a] and the Next Generation SIMulation [NGSIM, 2006] dataset containing high time-resolution vehicle trajectories. It is demonstrate through extensive numerical study that while first-order traffic quantities can be accurately estimated using prevailing sampling frequency at a reasonably low penetration rate, higher-order traffic quantities tend to be misinterpreted due to insufficient sampling frequency of current mobile devices. We propose, for estimating emission and fuel consumption rates, a correction factor approach which is proven to yield improved accuracy via statistical validation.