The Smart Stick is an advanced assistive device designed to enhance mobility for visually impaired individuals by integrating real-time obstacle detection and surface classification. This study proposes a CNN-GWO optimized model, significantly improving classification accuracy (98.8%) compared to traditional CNN (95.3%). The system ensures sustainable operation using solar-powered energy and enhances navigation through sensor fusion. Experimental evaluation demonstrated high detection accuracy, reducing false positives and improving obstacle classification. Compared to conventional mobility aids, the proposed provides an accurate, reliable, and adaptive solution, improving safety and independence for visually impaired users in diverse environments.