Longitudinal Vehicle Dynamics Parameter Identification for Effective Control System Design

Accurate parameter identification in longitudinal vehicle dynamics is essential for enhancing a road vehicle's performance, safety, and efficiency. In a mathematical model, parameters are constants that scale units or terms, representing the behavior of a physical system. However, identifying and estimating these parameters is challenging, requiring specialized methods and extensive testing. This paper presents a comparative study of metaheuristic optimization methods, including Genetic Algorithms (GA) and Particle Swarm Optimization (PSO), for estimating key parameters of the longitudinal vehicle model. The model uses the Newton-Euler approach, incorporating key dynamic forces such as aerodynamic drag, rolling resistance, and throttle input effects. In addition, the parameter identification process minimizes the error between simulated and experimental vehicle data, improving the accuracy of the control system. The experimental results substantiate the efficacy of diverse metaheuristic techniques in estimating the dynamic parameters of a mathematical model. These techniques are then applied to the design of cruise control and speed limiter systems.