Development of a Cost-Effective On-Device Natural Language Command Navigation System for Mobile Robots in Challenging Indoor Scenarios

The increasing demand for mobile robots in in- door environments such as hospitals, offices, and residential buildings has highlighted the need for affordable, privacy- preserving navigation and interaction capabilities. This study introduces a cost-effective, on-device BERT-based natural lan- guage navigation system that enables robots to interpret human commands into goals. The system is designed for deployment on lightweight embedded computers and updates without requir- ing model retraining, ensuring scalability and flexibility. We also propose an AprilTAg-augmented SLAM system to reduce navigation errors in common indoor challenges like ramps and transparent obstacles. Experiments in real-world settings statistically demonstrate that our solution significantly reduces errors in these scenarios, offering a more reliable approach to indoor robot navigation.