Model-free control of automotive engine and brake for Stop-and-Go scenarios

Abstract

In this paper we propose a complete strategy for the longitudinal control of automotive vehicles in Stop-and-Go situations. Firstly, a upper level grey-box torque control is proposed to compensate for neglected dynamics at chassis level (due for example to road slopes, aerodynamic forces, rolling resistance forces, etc.). Secondly, to obtain the desired torque, we have considered a model-free approach to elaborate the suitable low level engine or braking torque. Convincing simulation results are presented to validate our method.

Low-Speed Longitudinal Controllers for Mass-Produced Cars: A Comparative Study

“…Finally, the dynamics between the braking control variable t¿5 and braking torque 75 can be modeled as a second-order linear system [13] n(s)…”

Data-driven optimal model-free control of twin rotor aerodynamic systems

“…Some current design approaches related to (5) include Intelligent-PI (i-PI) controllers [11]- [19], pole placementbased design [20] and sliding mode-based design [18]. They can also be viewed in the context of data-driven control with attractive applications reported in [27]- [30].…”

“…It is called model-free not because the first principle modeling is not used for control design but rather a simple phenomenological model is enforced on the process and the developed control strategy aims at both compensating the mismatch between the true process and the enforced model and solving for both a reference trajectory tracking problem and disturbance rejection. MFC is applied in automotive processes [11], [12], hydroelectric power plants [13], active magnetic bearings [14], laboratory manipulators [15], electro-hydraulic systems [16], DC motor servo systems [17], [18], shape memory alloy active springs [19] and aerodynamic systems [20]. Another similar adaptive MFC approach is proposed in [21] and [22] and successfully applied in [21]- [24].…”

Model-free control techniques for Stop & Go systems

“…Finally, the dynamics between the braking control variable u b and braking torque τ b can be modeled as a second-order linear system [19] …”