Methodology

Tractive energy consumption serves as the energy impact indicator to exclude the effect of propulsion system. Two techniques were used to further isolate driving behavior and allow comparison of ACC driving behavior with human driving. Tractive energy consumption was calculated by integrating the tractive power requirements at the wheels over time, not considering regenerative braking.  

Five test passenger cars of different manufacturers and models were driven by human drivers on one direction of the 124 km round trip route. On the return trip, three of the cars used ACC to regulate speed and distance, following a human-driven lead vehicle. Multi-constellation Global Navigation Satellite System (GNSS) receivers were installed on each vehicle to collect detailed trajectory data at 10 Hz.

Findings

• The results suggest that ACC results in string instability, i.e., amplifying downstream speed variations, in followers. The inability of the ACC systems to absorb speed overshoots may be partly explained by high responsiveness. Followers in the automated or mixed traffic flow generally perform worse in reproducing the driving style of the preceding vehicle.
• Due to the challenges from different road and traffic condition in assessing tractive energy consumption on an individual level, the researchers compared energy consumption with the lead vehicle on the same trip. On an individual level, ACC followers have tractive energy consumption 2.7 – 20.5 percent higher than human counterparts. This was attributed to string instability and high responsiveness of the ACC systems. On a platoon level, tractive energy values of ACC followers tend to consecutively increase (11.2-17.3 percent) as the speed perturbations propagate upstream, relating to the string instability characteristic.
• The research provides a feasible path for evaluating ACC in real-world applications and have significant implications for ACC safety design when handling the stability-responsiveness trade-off.