01663320

Component

Microscopic vehicle fuel/emissions models have been well developed in the past decades. Those models require second-by-second vehicle trajectory data as an input to perform vehicle energy/emissions estimation. Due to the omnipresence of mobile sensors such as smartphones, real-world vehicle trajectory data can be collected in a large scale. However, most large-scaled mobile sensor data in practice are sparse in terms of sampling rate due to the consideration in implementation cost. In this paper, a modal activity framework for vehicle energy/emissions estimation using sparse mobile sensor data is presented. The valid vehicle dynamic states are identified including four driving modes, named acceleration, deceleration, cruising, and idling. The best valid vehicle dynamic state with the largest probabilities is selected to reconstruct the second-by-second vehicle trajectory between consecutive sampling times. Then vehicle energy/emissions factors are estimated based on operating mode distributions. The proposed model is calibrated and validated using the NGSIM’s dataset, and shows good performance in vehicle energy/emissions estimation compared with ground truth. Sensitivity analysis is performed to show the model accuracy with different time intervals. This research provides a new methodology for vehicle fuel/emissions estimation and extends the application of sparse mobile sensor data.

This paper was sponsored by TRB committee ADC20 Standing Committee on Transportation and Air Quality.

18-05700

English

4p

2018

Transportation Research Board 97th Annual Meeting

Digital/other

Figures; Tables

Data collection; Fuels; Pollutants; Sensors; Trajectory

Energy; Environment; Highways

Transportation Research Board Annual Meeting 2018 Paper #18-05700

TRIS, TRB, ATRI

Jan 8 2018 11:27AM