On Board Eco-Driving System for Varying Road-Traffic Environments Using Model Predictive Control M.A.S. Kamal, M. Mukai, J. Murata and T. Kawabe Abstract— This paper presents model predictive control of a vehicle in a varying road-traffic environment for ecological (eco) driving. The vehicle control input is derived by rigorous reasoning approach of model based anticipation of road, traffic and fuel consumption in a crowded road network regulated by traffic signals. Model predictive control with Continuation and generalized minimum residual method for optimization is used to calculate the sequence of control inputs aiming at long run fuel economy maintaining a safe driving. Performance of the proposed eco-driving system is evaluated through simulations in AIMSUN microscopic transport simulator. In spite of non- linearity and discontinuous movement of other traffic and signals, the proposed system is robust enough to control the vehicle safely. The driving behavior with fuel saving aspects is graphically illustrated, compared and analyzed to signify the prospect of the proposed eco-driving of a vehicle. I. I NTRODUCTION Eco-driving aims at cruising or safely following a car in the traffic stream in such a way that the fuel consump- tion in long run is minimum. Maneuvering a vehicle is a complex task due to changing and discontinuous nature of the road-traffic environment. Human drivers cope with such environment utilizing their complex and experience-based reasoning, and driving style among drivers varies widely [1]. Besides many physical factors, driving style has a great influence on vehicle emissions and energy consumptions [2]. Proper driving styles can improve the driving efficiency considerably. A recent experiment that was conducted in the form of driving contest on urban roadways showed that reduction in fuel consumption can be as high as 25% [3]. Generally, fuel economy is maximized when acceleration and braking are minimized. Therefore, a fuel-efficient or eco- strategy is to anticipate what is happening ahead, and drive in such a way so as to minimize acceleration and braking, cruising at the optimal speed and braking slowly at stops. Driver assistance in various ways for eco-driving has been emerged recently. Speculative features of eco-driving are available in the form of driving tips [4]. Some recently manufactured cars have an indicator that shows green ‘ECO’ marks to a driver when it consumes a little or no fuel. A driver may find his driving as ecological only when he maintains a steady velocity at a reasonable velocity or brakes the car. Nissan launched an off-board eco-driving support M. A. S. Kamal is with Fukuoka Industry, Science, and Tech- nology Foundation, 3-8-33 Momochihama, Sawaraku, Fukuoka, Japan. Email: maskamal@ieee.org M. Mukai, J. Murata and T. Kawabe are with the Faculty of Information Science and Electrical Engineering, Kyushu University, 744 Motoka, Nishiku, Fukuoka, Japan. Email:{kawabe, murata, mukai}@ees.kyushu-u.ac.jp service for some users in which, after driving record is sent to a telemetric data center for off-line analysis, advice is sent to the driver for improving his driving style in the next time. Based on past performance they have proposed an on- board assist system to motivate the driver for ecological driving by showing his comparative driving efficiency, his position in fuel composition ranking, etc [5]. A recent work in determining ecological strategy uses an optimal control approach in which only the model of the engine in terms of velocity, gear ratio and load is considered [6]. A more realistic approach to assisting a driver uses information of traffic signal, jams, road gradient and distance between cars, and the advice is given in a very rough form, such as ‘keep driving’ or ‘reduce pressure on pedal’, depending on motivation of the driver [7]. However, existing approaches to eco-driving assistance are very superficial, they do not provide concrete information such as the level of velocity or acceleration required for long term fuel efficient driving by analyzing current vehicle-road-traffic situation and its trend. Earlier a simplified form of the eco-driving assist system using model predictive control (MPC) was presented, where eco-driving was mostly based on driving at steady rated speed with slow braking at stops [8], [9]. In this paper, a completely revised and enhanced approach of eco-driving is presented. The effects of resistances and traction forces are included in the vehicle dynamics and fuel consumption is estimated using engine characteristic map. In addition the performance index is revised to emphasize mostly on the driving efficiency instead of given rated speed. The proposed on board eco-driving system senses the status of the host vehicle and the preceding vehicle, and the traffic signal ahead through information technology. It anticipates the possible state of the preceding vehicle in near future using some sort of simplified model. Based on this information and fuel consumption model of the vehicle, it generates an immediate control input using optimization process that may lead to a long-run economy travel. This rigorous approach of deriving optimal control input would make the proposed system more efficient, reliable and trustworthy for eco-driving. It is expected that new sensing and communication technologies will be incorporated for the development of road-traffic infrastructure in coming future. Through the advancement in intelligent transportation system, the proposed eco-driving can be realized. Performance of the proposed on board eco-driving system is evaluated through various observations in AIMSUN, a microscopic transport simulator with an integrated traffic en- vironment, using its extension developed through application Cite original Paper as M.A.S. Kamal, M. Mukai, J. Murata and T. Kawabe, "On Board Eco-Driving System for Varying Road-Traffic Environments Using Model Predictive Control," in Proc. of IEEE MSC2010, Yokohama September 8-10, 2010, pp.1636-1641.