Constrained Time-Dependent Adaptive Eco-Routing Navigation System

Matej Kubicka

2017-11-16

Abstract:Eco-routing is a vehicle navigation method that selects those paths to a destination that minimize fuel consumption, energy consumption or pollutant emissions. It is one of the techniques that attempt to lower vehicle's operational cost and environmental footprint. This work reviews the current eco-routing methods and proposes a new method designed to overcome their shortcomings. Most current methods assign every road in the road network some constant cost that represents either vehicle's consumption there or the amount of emitted pollutants. An optimal routing algorithm is then used to find the path that minimizes the sum of these costs. Various extensions are considered in the literature. Constrained eco-routing allows imposing limits on travel time, energy consumption, and pollutant emissions. Time-dependent eco-routing allows routing on a graph with costs that are functions of time. Adaptive eco-routing allows updating the eco-routing solution in case it becomes invalid due to some unexpected development on the road. There exist published optimal eco-routing methods that solve either the time-dependent eco-routing, or constrained eco-routing, or adaptive eco-routing. Each comes with considerably higher computational overhead with respect to the standard eco-routing and, to author's best knowledge, no published method supports the combination of all three: constrained time-dependent adaptive eco-routing. It is argued in this work that the routing costs are uncertain because of their dependence on immediate traffic around the vehicle, on driver's behavior, and other perturbations. It is further argued that since these costs are uncertain, there is little benefit in using optimal routing because the optimality of the solution holds only as long as the routing costs are correct. Instead, an approximation method is proposed in this work. The computational overhead is lower since the solution is not required to be optimal. This enables the constrained time-dependent adaptive eco-routing.

Environmental Science,Engineering,Computer Science