Implementation and design of ultra-local model-based control strategy for autonomous vehicles

Abstract

The paper presents a novel control strategy for autonomous vehicles, which can effectively handle the nonlinear effects and uncertainties of the controlled system. This solution is based on a new formulation of the ultra-local model-based control approach. This method can deal with the unknown dynamics and other uncertainties of the vehicle without any further measurement or complex modelling process. The control methods require a reduced-complexity nominal linear model of the system, and the unknown part is continuously estimated by the proposed ultra-local model. The proposed control-oriented vehicle modelling method leads to an increased performance level of vehicle motion. The effectiveness of the proposed method through a high-fidelity vehicle simulator (CarMaker) is illustrated. Moreover, the operation of the control through its implementation on a test vehicle of ThyssenKrupp is also demonstrated.

Keywords:

• Ultra-local model
• model free control
• lateral control
• trajectory tracking

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The research was supported by the European Union within the framework of the National Laboratory for Autonomous Systems (RRF-2.3.1-21-2022-00002). The paper was partially funded by the National Research, Development and Innovation Office [OTKA grant agreement number K 143599]. This paper was partially supported by the Janos Bolyai Research Scholarship of the Hungarian Academy of Sciences.