Urban accessibility remains a major challenge for people with disabilities due to inconsistent infrastructure, lack of reliable accessibility information, and limited awareness. This paper proposes NavAble, an AI-driven accessibility mapping platform designed to assess, rate, and visualize accessibility features within urban environments. The system integrates community- sourced data, machine-learning- based accessibility inference, and personalized route planning to support independent mobility for disabled citizens. NavAble utilizes a modern full-stack architecture incorporating React.js, Node.js, and Firebase, TensorFlow.js, and Leaflet maps to deliver scalable, real-time accessibility insights. The platform enhances civic inclusivity, supports governmental compliance monitoring, and empowers users through data-driven navigation.