Optimized TOPSIS technique for trajectory selection of self-driving vehicles on highways

Article type: Research Article

Authors: Arenas Muñiz, Andrés Antonio | Mújica-Vargas, Dante | Rendón Castro, Arturo | Luna-Álvarez, Antonio | Vela-Rincón, Virna V.

Affiliations: Tecnológico Nacional de México/CENIDET, Cuernavaca, Morelos, México

Correspondence: [*] Corresponding author: Dante Mújica-Vargas, Tecnológico Nacional de México/CENIDET, Cuernavaca, Morelos, México. E-mail: [email protected].

Abstract:  The selection of an appropriate trajectory for self-driving vehicles involves the analysis of several criteria that describe the generated trajectories. This problem evolves into an optimization problem when it is desired to increase or decrease the values for a specific criterion. The contribution of this thesis is to explore the use and optimization of another technique for decision-making, such as TOPSIS, with a sufficiently robust method that allows the inclusion of multiple parameters and their proper optimization, incorporating human experience. The proposed approach showed significantly higher safety and comfort performance, with about 20% better efficiency and 80% fewer safety violations compared to other state-of-the-art methods, and in some cases outperforming in comfort by about 30.43%.

Keywords: Decision-making, human experience, trajectory selection, self-driving

DOI: 10.3233/JIFS-219365

Journal: Journal of Intelligent & Fuzzy Systems, vol. Pre-press, no. Pre-press, pp. 1-10, 2024