Ensuring reliable and safe braking in autonomous vehicles is particularly challenging in urban environments, where dynamic signal processing, such as traffic lights, requires rapid and accurate decision-making. This study introduces an advanced braking system that employs robust control techniques specifically designed for emergency response involving traffic signal interactions. The system integrates a fuzzy-logic-based decision-making algorithm that prioritizes traffic light status and detected distance within the braking framework, enabling context-sensitive, adaptive emergency responses. A mechanically modified braking mechanism is controlled using a super-twisting strategy, ensuring robustness against unmodeled dynamics, nonlinearities, and external disturbances. Online traffic light detection is integrated with adaptive braking profiles for green, yellow, and red traffic light states, achieving a minimum response time of just 0.25 seconds. The experimental results indicated an angle deviation within the range of 0.06¿ to 0.35¿, which corresponds to a variation of less than 0.2%. These results confirm the feasibility of integrating low-cost hardware with advanced control and decision-making logic to improve safety in autonomous platforms.
