A Hybrid Metaheuristic Embedded System for Intelligent Vehicles Using Hypermutated Firefly Algorithm Optimized Radial Basis Function Neural Network

Abstract

This paper presents a hybrid metaheuristic embedded system for intelligent vehicles using hypermutated firefly algorithm (FA)-optimized radial basis function neural network (RBFNN), called FA-RBFNN. With the Mecanum vehicle's dynamic model, the FA with hypermutation is fused with RBFNN to develop a real-time optimal controller of the four-wheeled Mecanum vehicles in a field-programmable gate array (FPGA) chip. This hybrid metaheuristics takes the benefits of neural network, FA, real-time control, and FPGA realization. All the FA-RBFNN, dynamic controller, and hardware circuits are implemented in one FPGA chip using System-on-a-Programmable Chip methodology. Comparative works and experimental results clearly illustrate that the proposed FPGA-based FA-RBFNN optimal controller outperforms the conventional control methods.