Showing papers by "Gitakrishnan Ramadurai published in 2017"
TL;DR: In this article, the authors developed representative driving cycles for passenger cars and motorcycles which reflect the real-world driving conditions in Chennai, India, using on-board diagnostic (OBD) reader and Global Positioning System (GPS) receivers.
65 citations
TL;DR: An existing second-order continuum model of traffic flow is extended, using area occupancy for traffic concentration instead of density, to suggest area occupancy as concentration measure of heterogeneous traffic lacking in lane discipline.
53 citations
TL;DR: In this article, the authors quantified the emissions from the buses during peak and off-peak periods using second-by-second activity data collected using global positioning system (GPS) receivers.
Abstract: Intra-city buses provide essential transportation and mobility services in cities. However, most of the buses in Indian cities run on diesel fuel causing significant emissions and air quality issues. This study aims to quantify the emissions from the buses during peak and off-peak periods using second-by-second activity data collected using global positioning system (GPS) receivers. Four-bus routes plying within the city of Chennai in India was selected and second-by-second speed and acceleration was used to determine the operating mode of the bus. Corresponding emissions were estimated using vehicle specific power (VSP). Results show that the average speed of the bus during peak and off-peak periods were 17.8 kmph and 21.5 kmph, respectively; the corresponding percentages of time idling were 27% and 22%. Further, the percentage increase in total emissions of CO2, CO, HC and NOx from the bus during peak periods with respect to off-peak periods was 17%, 16%, 37% and 21%, respectively. This study provides useful insights regarding the operating and emission characteristics of buses that will be valuable to policy makers in improving their service and efficiency.
7 citations
31 Oct 2017
TL;DR: This work shows that by properly augmenting an large general (non-traffic) dataset with a small low-resolution heterogeneous traffic dataset the authors can obtain state-of-the-art vehicle detection performance and is expected to further encourage the wide-spread use of deep learning for traffic video image processing.
Abstract: Video image processing of traffic camera feeds is useful for counting and classify1 ing vehicles, estimating queue length, traffic speed and also for tracking individual 2 vehicles. Even after over three decades of research, challenges remain. Vehicle 3 detection is especially challenging when vehicles are occluded which is common 4 in heterogeneous traffic. Recently Deep Learning has shown remarkable promise 5 in solving many computer vision tasks such as object recognition, detection, and 6 tracking. We explore the promise of deep learning for vehicle detection and classifi7 cation. However, training deep learning architectures require huge labeled datasets 8 which are time-consuming and expensive to acquire. We circumvent this problem 9 by data augmentation. In particular, we show that by properly augmenting an exist10 ing large general (non-traffic) dataset with a small low-resolution heterogeneous 11 traffic dataset (that we collected) we can obtain state-of-the-art vehicle detection 12 performance. This result is expected to further encourage the wide-spread use of 13 deep learning for traffic video image processing. 14
1 citations