PDCA-Guided Development of Fuzzy Logic Controller for Autonomous Vehicle Cruise Control and Collision Avoidance

<div class="section abstract"><div class="htmlview paragraph">The advancement of the automotive industry towards automation has fostered a growing integration between this field and automation. Future projects aim for the complete automation of the act of driving, enabling the vehicle to operate independently after the driver inputs the desired destination. In this context, the use of simulation systems becomes essential for the development and testing of control systems. This work proposes the control of an autonomous vehicle through fuzzy logic. Fuzzy logic allows for the development of sophisticated control systems in simple, easily maintainable, and low-cost controllers, proving particularly useful when the mathematical model is subject to uncertainties. To achieve this goal, the PDCA method was adopted to guide the stages of defining the problem, implementation, and evaluation of the proposed model. The code implementation was done in Python and validated using different looping scenarios. Three linguistic variables were used, one with three fuzzy sets. As a result, nine rules were implemented in order to evaluate the vehicle’s response. An iterative loop was proposed to model different acceleration, deceleration or speed maintenance scenarios. The implementation of a system controlled by fuzzy logic was performed using the Python programming language. The simulations validated the speed adjustment, proving to be efficient for applications in autonomous vehicles as a simple and low computational cost approach.</div></div>