<p>Assistive technologies play a significant role in improving the quality of life for visually impaired individuals by enabling autonomous navigation and access to information. However, adoption is often limited by high device costs, frequent grid charging requirements, and constraints in portability, highlighting the need for affordable and self-sustained solutions. This study presents a compact solar-powered assistive device that integrates real-time object detection and multilingual reading assistance to enhance spatial awareness and reading capability. The system employs an edge-optimized deep learning model for efficient object detection and classification, while object proximity is measured using ultrasonic sensing. Text captured by the reading module is converted to speech to support seamless access to printed and digital content. The device offers a lightweight form factor and unifies navigation and reading functionalities within a single platform. The system consumes 10 to 15 Wh per hour and uses a solar subsystem with rechargeable storage, yielding up to 28 kWh of annual energy savings and supporting reliable operation in low-resource settings. Experimental evaluation demonstrates improved performance, with an inference rate of 4 FPS that surpasses typical Raspberry Pi-based implementations and a mean average precision mAP⁵⁰ of 0.73 on the COCO25k Kaggle dataset. The reading module accurately recognizes text across a range of fonts, sizes, orientations, and colors under bright and moderate illumination. Indoor trials along a circular path with randomly placed obstacles and participants walking at an average speed of 0.35&#xa0;m/s verified effective navigation. Field evaluations in diverse Indian climatic conditions confirmed system reliability and ergonomic suitability.</p>