Comparision of pathfinding algorithms for visually impaired people in IoT based smart buildings

Indoor navigation is highly challenging for visually impaired, particularly when visiting an unknown environment with complex design. In addition, a person at the entrance of the building might not be aware of distant changes/disruption in the path to the destination. Internet of Things devices can become the foundation infrastructure for scanning the dynamic changes in such an environment. With the sensory data of the scanned nodes, a dynamic pathfinding algorithm can provide guided route considering the changes to the destination. There are various pathfinding algorithms proposed for indoor environment including A*, Dijkstra's, probabilistic roadmap, recursive tree and orthogonal jump point search. However, there is no study done to find if these algorithms are suited to the special requirements of low vision people. We have carried out simulations in MATLAB to evaluate the performance of these algorithms based on parameters such as distance and nodes travelled execution time and safety. The results provide strong conclusion to implement most suitable orthogonal jump point search to achieve optimal and safe path for low vision people in complex buildings.