Characteristics of lateral vehicular interactions in heterogeneous traffic with weak lane discipline

Abstract

Heterogeneous traffic conditions prevail in developing countries. Vehicles maintain weak lane discipline which increases lateral interactions of vehicles significantly. It is necessary to study these interactions in the form of maintained lateral gaps for modeling this traffic scenario. This paper aims at determining lateral clearances maintained by different vehicle types while moving in a heterogeneous traffic stream during overtaking. These data were collected using an instrumented vehicle which runs as a part of the stream. Variation of obtained clearance with average speed of interacting vehicles is studied and modeled. Different instrumented vehicles of various types are developed using (1) ultrasonic sensors fixed on both sides of vehicle, which provide inter-vehicular lateral distance and relative speed; and (2) GPS device with cameras, which provides vehicle type and speed of interacting vehicles. They are driven on different roads in six cities of India, to measure lateral gaps maintained with different interacting vehicles at different speeds. Relationships between lateral gaps and speed are modeled as regression lines with positive slopes and beta-distributed residuals. Nature of these graphs (i.e., slopes, intercepts, residuals) are also evaluated and compared for different interacting vehicle-type pairs. It is observed that similar vehicle pairs maintain less lateral clearance than dissimilar vehicle pairs. If a vehicle interacts with two vehicles (one on each side) simultaneously, lateral clearance is reduced and safety of the vehicles is compromised. The obtained relationships can be used for simulating lateral clearance maintaining behavior of vehicles in heterogeneous traffic.

Evaluating Lateral Interactions of Motorized Two-Wheelers using Multi-Gene Symbolic Genetic Programming

“…The lateral clearance estimated from the developed model is comparable with the results of Minh et al ( 45 ) (that is, 1.416 m versus 1.370 m), and slightly higher than that of Budhkar and Maurya ( 21 ) at 60 km/h speed (that is, 1.669 m versus 1.312 m), but for the remaining cases, the obtained value is lower than that obtained in other studies: Minh et al ( 45 ) for exclusive MTW lane, Chunchu et al ( 22 ) and Budhkar and Maurya ( 21 ) for 30 km/h speed. Conversely, the critical lateral clearance in the current study is found to be higher than the results of Nakatsuji and Nguyen ( 16 ) (that is, 1.447 m versus 0.520 m), but slightly lower than the value reported in Minh et al ( 20 ) (1.145 m versus 1.772 m).…”

“…Based on the performance measures, it can be observed that the lateral clearance model developed by Minh et al ( 45 ) for undivided carriageway performs reasonably well compared with the linear models developed in the same work ( 45 ) for exclusive motorcycle lane, and with Budhkar and Maurya ( 21 ). Conversely, the critical lateral clearance symbolic model was compared with the dynamic lane width model ( 20 ) and the performance measure for the latter resulted in higher errors.…”

“…Understanding that precise estimation of lateral clearance in paired riding events is essential for several applications, the performances of the MGGP symbolic lateral clearance models developed in the current study are compared with the existing lateral clearance models ( 20, 21 , 45 ). Accordingly, the MSE and RMSE error measures are evaluated for each model corresponding to the training dataset.…”

“…The minimum lateral clearance data was considered to obtain the width of the dynamic lane. In parallel, different researchers incorporated speeds in the development of lateral clearance models of MTWs ( 5, 20–22 ) and accordingly, a linear relationship was obtained in all the studies. Conversely, Mallikarjuna et al ( 23 ) indicated non-linear variation in lateral clearances with speeds for MTW interactions, but such non-linear dependent relationship has not yet been developed.…”

Macroscopic Traffic Flow Characterization for Stimuli Based on Driver Reaction

“…Lateral distance headway is the safe distance between two adjacent parallel vehicles. This can be used to characterize traffic heterogeneity [35]. As the lane width decreases, this headway decreases.…”