Robust Model Predictive Control-Based Autonomous Steering System for Collision Avoidance

Harsh road conditions and sudden obstacles often arise when autonomous vehicles (AVs) operate on real roads. Therefore, designing control for autonomous steering systems under these conditions is challenging. To overcome this challenge, we introduce a robust model predictive control-based (RMPC) aut...

Full description

Saved in:
Bibliographic Details
Main Authors: Nam, Nguyen Ngoc, Nguyen, Hung Duy, Han, Kyoungseok
Format: Conference Proceeding
Language:English
Subjects:
artificial potential field
autonomous vehicles
collision avoidance
Linear matrix inequalities
linear matrix inequalities (LMIs)
Prediction algorithms
Predictive models
Roads
Robust model predictive control (RMPC)
Simulation
Steering systems
Trajectory
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Harsh road conditions and sudden obstacles often arise when autonomous vehicles (AVs) operate on real roads. Therefore, designing control for autonomous steering systems under these conditions is challenging. To overcome this challenge, we introduce a robust model predictive control-based (RMPC) autonomous steering system. To deal with a sudden obstacle that appears on real roads, an artificial potential field (APF) approach is introduced. It can generate an optimal-safe trajectory based on the receding horizon control (RHC) algorithm. In addition, to ensure the AV system operates well in slippery road conditions with highly varied coefficients, the RMPC is proposed, considering uncertain parameters and the varying velocity of the AV system. Specifically, the optimal control is designed by solving an optimization problem based on linear matrix inequalities to ensure a fast convergence rate. Furthermore, the input and output constraints are considered to guarantee the AV system works safely in complex and dynamic environments. Finally, various simulation results are provided to verify the effectiveness of the proposed control method.
ISSN:2642-3901
DOI:10.23919/ICCAS59377.2023.10316984