[Paper Review] Eco-Routing based on a Data Driven Fuel Consumption Model

Xianan Huang, Huei Peng|arXiv (Cornell University)|Jan 18, 2018

Vehicle emissions and performance12 references22 citations

TL;DR

This paper proposes a data-driven nonparametric fuel consumption model using six months of real-world driving data from 2,000 vehicles in Ann Arbor, combined with road grade data and simulations via Autonomie, to enable eco-routing. The study demonstrates that eco-routing reduces fuel consumption significantly, and a travel-time-constrained variant maintains most fuel savings with minimal travel time increase.

ABSTRACT

A nonparametric fuel consumption model is developed and used for eco-routing algorithm development in this paper. Six months of driving information from the city of Ann Arbor is collected from 2,000 vehicles. The road grade information from more than 1,100 km of road network is modeled and the software Autonomie is used to calculate fuel consumption for all trips on the road network. Four different routing strategies including shortest distance, shortest time, eco-routing, and travel-time-constrained eco-routing are compared. The results show that eco-routing can reduce fuel consumption, but may increase travel time. A travel-time-constrained eco-routing algorithm is developed to keep most the fuel saving benefit while incurring very little increase in travel time.

Motivation & Objective

• To develop a nonparametric fuel consumption model based on real-world driving data for eco-routing applications.
• To evaluate the trade-off between fuel savings and travel time in different routing strategies.
• To design a travel-time-constrained eco-routing algorithm that preserves most fuel savings while limiting time increases.
• To compare routing strategies including shortest distance, shortest time, standard eco-routing, and time-constrained eco-routing.
• To validate the model using real vehicle data and high-resolution road network information.

Proposed method

• A nonparametric fuel consumption model is trained on six months of real driving data collected from 2,000 vehicles in Ann Arbor.
• Road grade information is extracted for over 1,100 km of road network to improve fuel consumption accuracy.
• The vehicle simulation software Autonomie is used to compute fuel consumption for all trips across the road network.
• Four routing strategies are evaluated: shortest distance, shortest time, eco-routing, and travel-time-constrained eco-routing.
• The travel-time-constrained eco-routing algorithm limits detours by setting a maximum allowable time increase.
• Performance is assessed by comparing fuel consumption and travel time across the different strategies.

Experimental results

Research questions

• RQ1 How effective is a data-driven, nonparametric fuel consumption model in enabling accurate eco-routing?
• RQ2 To what extent does eco-routing reduce fuel consumption compared to conventional routing strategies?
• RQ3 What is the trade-off between fuel savings and increased travel time in eco-routing?
• RQ4 Can a travel-time-constrained eco-routing algorithm maintain significant fuel savings while minimizing time penalties?
• RQ5 How do different routing strategies perform in terms of fuel efficiency and travel time on real urban road networks?

Key findings

• Eco-routing reduces fuel consumption compared to shortest-distance and shortest-time routing strategies.
• Standard eco-routing leads to notable fuel savings but increases travel time due to route detours.
• The travel-time-constrained eco-routing algorithm achieves most of the fuel savings of unconstrained eco-routing with only a minimal increase in travel time.
• The nonparametric fuel consumption model, trained on real-world data, effectively captures complex driving behavior and road conditions.
• The integration of detailed road grade data and vehicle simulation (Autonomie) improves the accuracy of fuel consumption estimation.
• The results demonstrate the feasibility and effectiveness of data-driven eco-routing in real urban environments.