Platooning – grouping vehicles in a single file on roadways – may suggest to some drivers a nightmarish delay and the impossibility of passing for miles on end. But the growing concern about greenhouse gas emissions, coupled with technological advances in smart controllers and wireless networks, has brought the topic of platooning to the forefront.
Platooning, particularly platooning of heavy-duty vehicles (HDVs), offers several benefits. By maintaining short distances between vehicles, aerodynamic drag can be significantly reduced. This reduction in turn means a reduction in the amount of fuel consumed and hence can result in significant savings in fuel costs.
Numerous studies have been done on platooning, but most of these have focused on single platoons or individual vehicles within a platoon. Recently, a team of researchers from Argonne National Laboratory and the KTH Royal Institute of Technology in Sweden has been tackling the problem from a different angle.
“Unlike previous studies, our concern is how to coordinate HDVs throughout a large-scale real-world road network,” said Jeff Larson, a postdoctoral appointee at Argonne.
To achieve this goal, the researchers simulate a distributed network of virtual controllers placed at major intersections in a road network. By knowing the destination of the approaching vehicles, the controllers determine the possible fuel savings incurred from platooning. The controllers then recommend slight adjustments to the HDVs’ speeds as the vehicles approach an intersection, so that they can form a platoon. Control is initiated only if the cost of forming the platoon is smaller than the savings incurred from platooning.
“Our approach avoids having to plan routes days beforehand,” said Larson. “We also don’t need to find every HDV’s current location and eventual destination. While a more global approach might seem desirable, no single global coordinator exists with the authority to route vehicles into platoons; and even if it did, the problem is known to be computationally intractable. ”
Distributing controllers at junctions in the road network simplifies the problem considerably. Nevertheless, the researchers still faced several challenges in developing their platooning control methodology. For example, they had to ensure that the HDV drivers did not have to alter significantly their routes in order to form a platoon. Their calculations showed that allowing one vehicle to travel 10 or 20 km extra can help improve the fuel use because more options for platooning are available. On the other hand, allowing every vehicle to possibly detour 10 km or more is not cost effective.
Another challenge was the density of HDVs in the network. As one might expect, when the density is low, few opportunities for platooning exist. But the researchers also determined that when the road network becomes “saturated” with more than 2,000 vehicles, the benefits of platooning drop sharply.
“Our calculations indicated a savings of more than 5% with only a few thousand vehicles participating in the system,” said Larson.
To test the feasibility of platooning on a real-world problem, the researchers conducted a large-scale simulation of the German Autobahn network. The probe data came from HDVs over a 24-hour period in 2013; each of the 875 vehicles considered traveled over 150 km from start to finish. For the simulation, the researchers assumed that each HDV could speed up to catch another vehicle only if it was trailing by at most one edge length. Vehicles were placed at random points throughout the network and started toward their random destination (following their shortest path). Whenever a vehicle left an intersection, it broadcast its route to the next intersection on its path. At every time step, the controller attempted to find an optimal routing. If platooning savings were larger than the cost of adjusting speeds in order for the relevant vehicles to arrive at the intersection simultaneously, the vehicles’ speeds were adjusted. The results showed that when the distance between the vehicles was 5 km, nearly 90% of the 875 vehicles had an opportunity for platooning.
“If these vehicles used our local controller to adjust their speeds by a small amount in order to synchronize travel with other HDVs, at least a 1.2% fuel savings was possible,” said Larson. “We believe similar savings can be achieved on a road network with a similar geography, for example, the U.S. Interstate Highway System.“
For further information, see the paper “A Distributed Framework for Coordinated Heavy-Duty Vehicle Platooning,” by J. Larson, K.-Y. Liang, and K. H. Johansson, special issue of IEEE Trans. on Intelligent Transportation Systems, 2014.