MPC trajectory planner for autonomous driving solved by genetic algorithm technique. Issue 12 (2nd December 2022)

• Record Type: Journal Article
• Authors: Arrigoni, S.; Braghin, F.; Cheli, F.
• Is Part Of: Vehicle system dynamics. Volume 60:Issue 12(2022)
• Page Start: 4118
• Page End: 4143
• Publication Date: 2022-12-02
• Subjects: MPC -- GA methods -- motion planning -- autonomous driving -- obstacle avoidance; Motor vehicles -- Dynamics -- Periodicals; Electronic journals; 629.231
• DOI: 10.1080/00423114.2021.1999991
• Languages: English
• ISSNs: 0042-3114

Abstract

Abstract : Focusing on autonomous driving algorithm development, this paper proposes a novel real-time trajectory planner formulated as a Nonlinear Model Predictive Control (NMPC) algorithm. The mathematical formulation of the problem is deeply reported and discussed. The numerical solution of the NMPC problem is the result of a novel genetic algorithm strategy that represents the innovative aspect of the work proposed. The aim of this paper is also to show how genetic algorithm can be a valid approach for motion planning strategies. Numerical results are discussed through simulations that show a reasonable behaviour of the proposed strategy in the presence of moving obstacles as well as in a wide range of road friction conditions. Moreover, a real-time implementation for research purposes is assumed as possible by considering computational time analysis reported.