Informative Path Planning for Location Fingerprint Collection

Fingerprint-based indoor localization methods are promising due to the high availability of deployed access points and compatibility with commercial off-the-shelf user devices. However, to train regression models for localization, an extensive site survey is required, which collects fingerprint data from the target areas. In this paper, we consider the problem of informative path planning (IPP) to find the optimal walk for a site survey subject to a budget constraint. IPP for location fingerprint collection is related to the well-known orienteering problem (OP) but is more challenging due to its edge-based non-additive rewards and revisits. Given the NP-hardness of IPP, we propose two heuristic approaches: a Greedy algorithm and a Genetic algorithm. Through experimental data collected from two indoor environments with different characteristics, we show that the two algorithms have low computation complexity, and can generally achieve a higher utility, as well as lower localization errors compared to the extension of two state-of-the-art approaches to OP.