A model predictive approach to control vehicle skidding by sideslip angle estimation

• Main Authors: Simone Graffione (Author), Roberto Sacile (Author), Enrico Zero (Author), Chiara Bersani (Author)
• Format: Book
• Published: Annals of Robotics and Automation - Peertechz Publications, 2021-08-19.
• Subjects: Research Article
• Online Access: Connect to this object online.

<p>Nowadays, many cars are equipped with Advanced Driving Assistance Systems (ADAS), from cruise control to collision detection and speed limit signalling. In recent years, the scientific community has carried out a lot of research in the field of autonomous driving system with the aim of improving fuel consumption, traffic and safety, but it is still a developing field that will require a lot of work despite the great progress. This paper proposes a new approach to road safety, introducing an autonomous driving system that intervenes when dangerous situations occur or when the driver is not able to drive, keeping the passengers and the other vehicles safe. </p><p>A Linear Time Varying (LTV) Model Predictive Control (MPC) is proposed to control the vehicle minimising the lateral error with respect to lane centre while maintaining a constant speed. The lane centre is estimated by the lane boundaries that are supposed to be given by a vision detection system.</p><p>The risk situation is evaluated by the slip angle and rate.</p>