Traffic Flow with Autonomous Vehicles in Real Life Traffic Situations
Nominated for The Neil Mansfield Award
Eirin Ryeng, NTNU, Eivind Myklebust Lindseth, NTNU, Torbjorn Haugen, NTNU
Autonomous vehicles will probably be a part of normal traffic within few years. In this field study traffic flow effects will be registered for different headway settings in these vehicles, and a signalised intersection will be observed.
Autonomous vehicles will probably be part of normal traffic in most countries within few years. Will their introduction contribute to better traffic flow when mixed into traffic with conventional vehicles? Previous research, mostly based on microsimulation studies, suggest that traffic solely composed of autonomous vehicles will flow better, and that traffic volumes will be higher than today. However, in the near future, a more realistic scenario is that there will be a mix between autonomous and traditional vehicles. Autonomous vehicles may use technologies similar or identical to adaptive cruise control (ACC) to adjust headway. In cars with ACC the driver can choose from different safety settings for headway. In this field study, performed in real life traffic situations, five cars utilizing ACC served as substitutes for autonomous vehicles. Saturation flow rates in a signalized intersection were examined, comparing normal traffic flows to traffic flows with varying shares of cars with activated ACC, for varying headway settings. Compared to normal traffic, saturation flow rates were found to be significantly lower for traffic flows including cars with maximum ACC headway settings, while minimum ACC headway settings gave no significantly different saturation flow rates. Given that cars equipped with and activated ACC are realistic substitutes for future autonomous vehicles, these findings indicate that the traffic flows will not become more effective in a future transition period with mixed traffic flows. On the contrary, if passengers of the autonomous vehicles are to choose the headway settings by themselves, they might choose maximum settings, leading to less effective traffic flows compared to today’s traffic.
Association for European Transport