776 IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, VOL. 16, NO. 2, APRIL 2015 Longitudinal Model Identification and Velocity Control of an Autonomous Car Jullierme Emiliano Alves Dias, Guilherme Augusto Silva Pereira, Senior Member, IEEE, and Reinaldo Martinez Palhares Abstract—This paper presents the model identification and the velocity control of an autonomous car. The control system was designed so that the car is controlled at low speeds, where the main applications for the vehicle’s autonomous operations include parking and urban adaptive cruise control. A longitudinal model of the car was used in the control loop to compensate the nonlinear behavior of its dynamics. Since the determination of the vehicle’s model is a difficult step in the design of model-based controllers, the main contribution of this paper is the use of an empirically determined model to this end. In this paper, the structure of the model was conceived from the car’s physics equations, but its parameters were estimated using data-based identification tech- niques. An important contribution of this paper is the fact that, although the model is strictly linear, we can change its parameters as a function of the operation point of the vehicle to represent the engine’s and the transmission’s nonlinear behaviors. Moreover, in this paper, we propose a way to include changes in the longitudinal dynamics caused by the automatic gear shifting. The validation of the proposed controller was conducted by computer simulations and real-world experiments. Index Terms—Intelligent vehicle, mathematical model, system identification, velocity control. I. I NTRODUCTION A UTONOMOUS vehicles are likely to play a major role in future transportation systems, since they provide many potential benefits, such as increased safety and higher utiliza- tion of roads. Motivated by this, since 2007, the Group for Research and Development of Autonomous Vehicles (PDVA) at the Federal University of Minas Gerais (UFMG/Brazil) has been working on the design and development of the Au- tonomous Car Developed at UFMG (CADU, a Portuguese acronym), which is a commercial automobile (Chevrolet Astra Sedan 2003) equipped with automatic transmission, ABS, and drive-by-wire throttle (see Fig. 1). Commands such as throttle, Manuscript received February 5, 2014; revised June 4, 2014; accepted July 14, 2014. Date of publication August 28, 2014; date of current version March 27, 2015. This work was supported in part by the Fundação de Amparo à Pesquisa do Estado de Minas Gerais (FAPEMIG/Brazil), Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq/Brazil) and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES/Brazil). The work of J. E. A. Dias, G. A. S. Pereira, and R. M. Palhares was supported by a CNPq Scholarship. The Associate Editor for this paper was M. Da Lio. The authors are with the School of Engineering, Universidade Federal de Minas Gerais, Belo Horizonte, MG 31270-901, Brazil (e-mail: julliermeead@ ufmg.br; gpereira@ufmg.br; palhares@cpdee.ufmg.br). Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org. Digital Object Identifier 10.1109/TITS.2014.2341491 Fig. 1. Autonomous Car Developed at UFMG (CADU). brake, steering, and gear shifting have been automated and controlled by a computer through a real-time network. Several developments have been made in this project, which allowed the car to drive itself in scenarios with obstacles using cameras and other sensors [1]. In this paper, we present the longitudinal velocity controller of this vehicle. Commercially, the longitudinal velocity controllers are in- cluded in the cruise control (CC) systems. In general, the CC is functional only in medium and high speeds (above 30 or 40 km/h); for this reason, it cannot be directly used in au- tonomous cars that operate at urban traffic speeds. The main difference between the standard CC systems and the controllers needed by autonomous vehicles is the fact that the former only controls the throttle to maintain a constant desired speed. It is the driver’s responsibility to brake the car if the current speed is no longer safe for travel. Longitudinal controllers for au- tonomous cars need to control both throttle and brake to ensure that a velocity profile (specified by a higher level controller) is followed by the car. Only recently some commercial cars have been equipped with systems, such as the adaptive CC (ACC) systems, which also brake the vehicle and allow it to operate in traffic conditions, in which the vehicle’s speed can range from 0 to more than 100 km/h. In this paper we present a practical way to implement a model-based velocity controller that operates in longitudinal velocities that range from 0 to 40 km/h. The system controls the vehicle using throttle and brake and, thus, can be used in autonomous operations. Provided that some specific tests can be executed with the vehicle to construct a mathematical model, this controller is a candidate to control any autonomous car. Since it operates at low speeds, it can be also useful, for ex- ample to compose autonomous parking or ACC [2] systems of 1524-9050 © 2014 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.