Artemisa: A Personal Driving Assistant for Fuel Saving

In this paper, we propose a driving assistant that makes recommendations in order to reduce the fuel consumption. The solution only requires a smartphone and an OBD/Bluetooth device. Eco-driving advices try to avoid situations that cause an increase in the fuel consumption such as inappropriate speed or slow reaction to the detection of traffic signs and traffic incidents. The main contribution of this paper is the use of artificial intelligence techniques in order to issue the eco-driving tips that are best adapted to the user profile, the characteristics of the vehicle, and the road state conditions. This is very important because the driver may lose the interest due to the high requirements that tend to be provided by general use eco-driving assistants. In order to properly assess and validate the proposed solution, it has been implemented on several Android mobile devices and has been validated using a dataset of 2,250 driving tests using three different models of vehicles with 25 different drivers on three distinct routes. The results show that the system reduces the fuel consumption by 11.04 percent on average and even, in certain cases, the fuel saving is greater than 15 percent.

1 Introduction

The vehicle is one of the most significant innovations of the past century. Its invention has had a huge impact at the social, ecological, and economic levels. The expansion of the automobile has contributed to globalization and economic growth in Europe, United States, Japan and many other parts of the world. It has also helped the expansion of suburban areas. However, it has caused many negative effects both at the social [1] and at the environmental levels [2]. This work is mainly focused on the environmental level.