Energy-efficient automated driving: effect of a naturalistic eco-ACC on a following vehicle

The optimal control problem of eco-driving, driving a vehicle using a minimal amount of fuel or electrical energy, has received much attention in the intelligent vehicles literature with many recent proposals for eco-driving adaptive cruise control systems (eco-ACC). In this paper, we consider a recently-introduced 'naturalistic' eco-ACC approach, which was designed to give human-like behaviour in vehicle following. For this eco-ACC, we show that in car following and start-stop traffic scenarios, the eco-ACC benefits not only the ego vehicle but also a further following vehicle. To see if further reductions to total energy consumption are possible, we extend the eco-ACC system with an optimal control formulation that also minimises energy losses of the following vehicle assuming it behaves according to the intelligent driver model (IDM). This gives some minor reductions in energy usage but surprisingly, for a follower that behaves according to the IDM, the naturalistic eco-ACC appears to be nearly optimal for the problem of minimising total energy loss of both the ego vehicle and its follower.