This article presents an evaluation of the proposed QMD Quaternion Magnetic Disturbance algorithm and its integration into the AIEZ framework for indoor pedestrian navigation using a foot-mounted IMU. Through experiments in controlled environments and real-world scenarios, the system demonstrates superior performance in detecting and correcting magnetic disturbances, outperforming classical QMD methods.
The enhanced QMD proposal incorporates advanced filtering and classification strategies to distinguish between temporary disturbances and permanent biases of the magnetic field. By integrating into the AIEZ framework, which merges the EC and HDR approaches, a robust solution is achieved that maintains orientation and trajectory accuracy even in environments with intense magnetic interference.
In bench tests and user tests with a foot-mounted IMU, it was observed that early detection of magnetic anomalies enables the activation of correction mechanisms that reduce accumulated error in heading estimation. Compared to classical QMD techniques, the enhanced version reduces deviations, improves solution continuity, and minimizes the risk of drift in long trajectories.
The AIEZ fusion of the EC and HDR algorithms provides key benefits: EC offers stability under low disturbance conditions, while HDR provides resilience against interference spikes. Their combination in AIEZ allows for intelligent switching or weighting between strategies based on sensor confidence, resulting in more stable and accurate pedestrian navigation.
Results in real-world scenarios show that the system is especially effective in complex indoor environments such as shopping centers, transportation stations, and industrial settings where magnetic sources are heterogeneous. The architecture also allows incorporating machine learning to adapt detection parameters based on context and the user's movement profile.
Future work and extended validation: plans include expanding tests to various user conditions, footwear types, surfaces, and IMU configurations, as well as evaluating robustness against long-term environmental changes. It is also planned to integrate automatic calibration modules and continuous learning capabilities to improve the adaptability of QMD and the AIEZ framework.
About Q2BSTUDIO: Q2BSTUDIO is a custom software and application development company specialized in advanced technological solutions. We offer custom software services, custom application development, and consulting in artificial intelligence and applied AI for businesses. Our team is made up of specialists in artificial intelligence, cybersecurity, and AWS and Azure cloud services, allowing us to design secure, scalable, and optimized solutions for each client.
Featured services: custom applications, custom software, AI for businesses, AI agents, and Power BI for visualization and analysis. We provide business intelligence and Power BI services to transform data into decisions, as well as integrate AI agents to automate processes and improve user experience. Our cybersecurity offering ensures comprehensive protection and compliance, while our AWS and Azure cloud services facilitate reliable and scalable deployments.
Applications and use cases: the combination of enhanced QMD algorithms with the AIEZ framework is ideal for indoor localization projects, traceability solutions, navigation assistance systems for people, and industrial applications. At Q2BSTUDIO, we develop prototypes and final products that integrate artificial intelligence, business intelligence services, and sensor technology to deliver differentiated, high-value solutions.
If you wish to explore a custom solution for indoor navigation or learn how our capabilities in custom applications, custom software, artificial intelligence, AI agents, cybersecurity, AWS and Azure cloud services, and Power BI can drive your project, at Q2BSTUDIO we would be delighted to help you with a proposal tailored to your needs.
