Design and development of an Affordable Sinhala Language-Enabled Voice-Assisted Smart Wheelchair

Assistive mobility technologies have undergone significant advancements with the integration of embedded systems, speech recognition, and intelligent control mechanisms. However, commercially available voice-assisted smart wheelchairs predominantly support English and a limited set of widely spoken international languages, thereby restricting accessibility for native language speakers and remaining financially inaccessible to many users in developing countries. In Sri Lanka, where Sinhala is the primary language of communication for the majority of the population, this linguistic limitation presents a significant barrier to independent assistive mobility. This study presents the design and development of a Sinhala Language-Enabled Voice-Assisted Smart Wheelchair (VASW), tailored specifically to the linguistic, economic, and technological context of Sri Lanka. The proposed system integrates an Arduino Uno-based embedded control architecture, dual 12 V DC motors for propulsion, ultrasonic sensors for obstacle detection, a microphone module for voice acquisition, and a micro-SD storage module for command processing. Direction-specific Sinhala commands corresponding to forward, backward, left, right, and stop are trained and processed locally without reliance on external smartphone applications or cloud-based platforms. The voice recognition module provides an approximate response latency of 10 ms, enabling near real-time operation. The architecture prioritizes affordability, modularity, and reduced system complexity while ensuring functional reliability. Experimental validation conducted in controlled indoor environments demonstrates accurate command recognition and safe maneuverability within predefined operational constraints. The proposed system offers a linguistically localized, economically feasible assistive mobility solution tailored to the Sri Lankan context. Furthermore, a scalable framework for future integration of adaptive Artificial Intelligence (AI)-based speech processing is proposed to accommodate regional Sinhala dialect variations. The novelty of the system lies in its linguistic localization, economic accessibility, and modular design architecture. The proposed VASW contributes toward equitable assistive technology deployment in resource-constrained environments and establishes a foundation for language-inclusive mobility solutions.