AR BASED INDOOR NAVIGATION: A REVIEW

Abhinav ShrivastavaInternational Journal of Advanced Research in Computer Science 2025 · DOI

TL;DR

This review paper explores the evolving landscape of indoor navigation, highlighting the shift from traditional 2D maps to Augmented Reality (AR) for a more immersive and intuitive user experience. It critically examines various positioning and navigation technologies, comparing their strengths, weaknesses, and future potential in creating robust AR-enhanced indoor navigation systems.

What Problem Does This Paper Address?

Global Positioning Systems (GPS) are notoriously unreliable indoors due to signal obstruction. This limitation creates a significant challenge for indoor navigation in environments like malls, airports, and hospitals, where precise location tracking is crucial. Traditional methods, relying on 2D maps displayed on screens, often lack spatial depth and can be less intuitive. The paper addresses the need for effective indoor navigation solutions and investigates how Augmented Reality (AR) can enhance these systems by overlaying digital information onto the real world, offering a more engaging and spatially aware experience.

Methods

This paper is a review, so it doesn't introduce new methods but rather synthesizes existing research. It categorizes indoor navigation approaches into two main areas:

  • Positioning:
    • RF-based systems: This includes Wi-Fi, Bluetooth Low Energy (BLE), RFID, and Ultra-Wideband (UWB). BLE is highlighted for its cost-effectiveness and efficiency, often using beacons for localization. Techniques like RSSI fingerprinting are discussed for accuracy.
  • Other positioning methods: Satellite-based (limited indoors), Magnetic-based, Inertial Sensor-based (accelerometers, gyroscopes), Sound-based (audible, ultrasonic, acoustic), and Optical-based (Infrared, Visible Light Communications).
  • Navigation:
    • Artificial Intelligence (AI): Machine learning models (KNN, DNN, CNN) are discussed for improving accuracy, energy efficiency, and handling environmental noise. Frameworks like LearnLoc, CNN-LOC, and QuickLoc are mentioned.
  • Simultaneous Localization and Mapping (SLAM): This approach incrementally builds a map of an environment while simultaneously tracking the user's location, making it suitable for unknown spaces. It's often used with pathfinding algorithms like A*.
  • Traditional Pathfinding Algorithms: A* (A-star) is mentioned as a common algorithm used in conjunction with AI and SLAM.

The paper also discusses the role of AR Interfaces in presenting navigational information, moving beyond 2D screens to overlaying digital cues onto the real world.

Datasets

Not discussed in the paper.

Key Insights and Ideas

  • The AR Shift: The move from 2D maps to AR for indoor navigation is driven by the desire for a more immersive experience, improved spatial awareness, and greater ease of navigation by overlaying digital cues onto the real world.
  • BLE's Promise and Pitfalls: BLE beacons offer a cost-effective and scalable solution for indoor positioning, but challenges remain with beacon placement, signal interference, and database management.
  • AI and SLAM as Navigation Powerhouses: AI and SLAM technologies have significantly advanced indoor navigation, offering improved accuracy and adaptability, especially in dynamic environments.
  • The Cost-Complexity-Accuracy Trade-off: There's an inverse relationship between cost, complexity, and accuracy. Simple solutions like QR codes are cheap but less flexible, while AI/SLAM offer higher accuracy and scalability but come with greater complexity and computational demands.
  • The Need for Hybrid Approaches: Given the limitations of individual technologies, future research should focus on hybrid systems that combine the strengths of cost-efficient methods with the accuracy and scalability of advanced techniques.

Key Results

The paper is a review, so it doesn't present new quantitative results. However, it highlights the following qualitative findings:

  • BLE positioning accuracy: Can achieve accuracy within one to three meters.
  • AI-based navigation: Frameworks like CNN-LOC have shown improved accuracy and reduced energy consumption compared to earlier DNN-based solutions.
  • SLAM: Enables efficient navigation in unknown environments without requiring pre-made maps.
  • AR Interfaces: Offer enhanced user satisfaction and reduced cognitive load by providing spatial context.

Limitations and Future Work

Limitations:

  • Computational Demands: AI and SLAM systems, while powerful, require significant computational resources, leading to potential delays and errors, especially in high-traffic zones or large areas.
  • Environmental Sensitivity: Signal interference, physical barriers, and environmental changes can degrade the accuracy and reliability of many positioning and navigation systems.
  • Scalability and Adaptability: While AI and SLAM have improved, challenges remain in scaling these systems to larger areas and adapting them to constantly changing environments.
  • Hardware Limitations: AR interfaces can be constrained by hardware capabilities, and there's a risk of cognitive overload for users.
  • Standardization: A lack of standardized frameworks makes evaluating and comparing AR systems difficult.
  • User-Centric Challenges: Engaging representative user groups and effectively measuring unique AR experiences like immersion and collaboration remain challenging.

Future Work:

  • Hybrid Approaches: Developing systems that combine the cost-efficiency of simpler technologies (like BLE or QR codes) with the accuracy and scalability of AI and SLAM.
  • Edge Computing: Leveraging edge computing for instant processing to address latency issues.
  • User-Centric Design: Focusing on hardware compatibility and user experience to ensure widespread adoption.
  • Novel Solutions: Addressing persistent issues like computational efficiency, environmental adaptability, signal reliability, user retention, and computational efficiency.

Summary updated

Abstract

: Indoor navigation systems have gained traction recently since GPS reception is extremely low inside buildings and enclosed spaces. These systems involve using technologies such as Wi-Fi, Bluetooth, Artificial Intelligence, or even simple hard-coded maps with pathfinding algorithms between two marked places. Furthermore, the method to present this system is also slowly moving away from traditional approaches such as using a map on the mobile screen, to methods like using AR. This review paper intends to provide a comprehensive understanding and explore the potential of Augmented Reality enhanced Indoor Navigation Systems by comparing and contrasting various technologies, methodologies, and approaches. The paper also aims to provide a deeper insight into the capabilities, limitations, and future directions.
BibTeX
@article{shrivastava2025ar,
  author = {Abhinav Shrivastava},
  title = {{AR BASED INDOOR NAVIGATION: A REVIEW}},
  journal = {International Journal of Advanced Research in Computer Science},
  year = {2025},
  doi = {10.26483/ijarcs.v16i1.7167},
}
Figure