Wireless communication through automated and connected vehicles is an evolving technology. This ameliorates the driving conditions, reduces time spent in traffic and curtails the crash occurrences. One of the most challenging areas, where these interactions can be most useful, are freeway merge ramps. Both the drivers on mainline and the drivers merging would be skeptical about their decisions at this location. The drivers who want to merge to the freeway mainline would seek to find an appropriate gap to enter the mainline of the freeway. While the technology of connected and automated vehicles is being promoted, the reality now is that for the foreseeable future, the traffic would not comprise 100% of such connected and automated vehicles. In other words, there will be a mixed traffic of manually-driven and connected/automated vehicles, with various levels of automation in the latter types of vehicles. Capturing the driver behavior at the merge locations into a freeway with such mixed traffic, will be useful in learning and improving safety on the roadways. The Driving Simulator is a useful device in capturing driver behaviors. In this study scenarios are developed in the Driving Simulator which allows mixed traffic on mainline and also observe the driver behaviors from the ramp onto the merge. Overall there were three variations in the mixed traffic flow for the mainline freeway: 0%, 50% and 75% penetration rates. The freeway traffic was generated for the mixed traffic by first developing a mixed probability distribution which assumes exponential distributions for the inter-arrival times of manually-driven vehicles and a constant headway (uniform distribution) is assumed between connected vehicles. The mixed distribution was then used to randomly generate vehicles through Monte Carlo simulation, with assigned headways in the Driving Simulator for the various connected vehicle penetration rates. The subject driver’s speed along the ramp is monitored, as well as the speeds of those vehicles on the freeway. The gaps between freeway vehicles, which were accepted by the subject driver, were recorded for the various situations and scenarios. There were a total of 41 participants, with 29 young drivers (younger than 65 years) and 12 elderly drivers (65 years and older, amongst which 2 were between 55 and 65 years old). Three scenarios were presented to the drivers. The first driving task was to determine headway gap acceptance for the three penetration rates, based on the perception of the subject drivers (without driving). The second test involved the subjects actually driving on the ramp and implementing a suitable gap to merge on the freeway traffic at each ramp. From the data collected, the critical gaps were estimated based on perception. The gaps accepted while driving were also tabulated analyzed. It was observed that the critical gap for the young drivers in 0%, 50%, 75% penetrations rate are 2.9 sec, 1.8 sec, and 1.7 sec respectively. The critical gaps observed for elderly drivers aged over 65 are 3.5 sec, 2.0 sec, and 1.9 sec respectively. Based on an Analysis of Variance (ANOVA), there is no evidence to prove the equality of means for different groups classified by age, gender and driving experience in both perception and actual driving conditions for 0% and 50% penetration rates. It was observed that the headway gaps accepted by young and drivers, both by perception and driving in 0% penetration rate were 2.39 sec and 2.35 sec respectively. The headway gaps accepted by elderly drivers both by perception and driving in 0% penetration rate were 2.4 sec and 2.72 sec respectively. When the ANOVA was performed between the 0% and 50% penetration rates of driving conditions, it was observed that there is a lot of variation in the mean headway gaps accepted. The values of average headway gaps accepted for young drivers were estimated as 2.36 sec and 1.53 sec respectively, in the 0% and 50% penetration rates. For the elderly drivers the average headway gap values observed were 2.72 sec and 1.55 sec respectively, in the 0% and 50% penetration rates traffic. The results also indicated the subject driver acceleration and deceleration behavior at the merge ramp. The results also showed that when the (aggressive) drivers accelerated to match the velocity of mainline traffic and merged in between connected-automated vehicles with the shortest gap, effects were noticed on the mainline traffic, where the main line traffic had to decelerate rapidly. Overall, it was observed that the subject drivers accepted shorter headway gaps as the penetration rates increases.
