Abstract
Individual tracking of museum visitors based on portable radio beacons, an asset for behavioural analyses and comfort/performance improvements, is seeing increasing diffusion. Conceptually, this approach enables room-level localisation based on a network of small antennas (thus, without invasive modification of the existent structures). The antennas measure the intensity (RSSi) of self-advertising signals broadcasted by beacons individually assigned to the visitors. The signal intensity provides a proxy for the distance to the antennas and thus indicative positioning. However, RSSi signals are well-known to be noisy, even in ideal conditions (high antenna density, absence of obstacles, absence of crowd, ...). In this contribution, we present a method to perform accurate RSSi-based visitor tracking when the density of antennas is relatively low, e.g. due to technical constraints imposed by historic buildings. We combine an ensemble of "simple" localisers, trained based on ground-truth, with an encoding of the museum topology in terms of a total-coloured graph. This turns the localisation problem into a cascade process, from large to small scales, in space and in time. Our use case is visitors tracking in Galleria Borghese, Rome (Italy), for which our method manages >96% localisation accuracy, significantly improving on our previous work (J. Comput. Sci. 101357, 2021).
Highlights
The behavioural analysis of museums’ visitors has a long-standing multidisciplinary tradition [1, 2], and underlies the capacity of profiling exhibitions, increase visitors’ comfort and safety, improve public reception, increase the number of sold tickets, and enhance artworks preservation [3]
We present a method to perform accurate Received Signal Strength intensity (RSSi)-based visitor tracking when the density of antennas is relatively low, e.g. due to technical constraints imposed by historic buildings
These yield even noisier RSSi signals with a quick decay, causing ambiguous or even void positioning readings. These constraints likewise jeopardise the success of any approach based on instantaneously “maximum RSSi” readings, even when the ambition is the sole room-level localisation. To deal with these limitations, we proposed in [4] a method based on an end-to-end Multi-Layer Perception (MLP) neural network
Summary
The behavioural analysis of museums’ visitors has a long-standing multidisciplinary tradition [1, 2], and underlies the capacity of profiling exhibitions, increase visitors’ comfort and safety, improve public reception, increase the number of sold tickets, and enhance artworks preservation [3]. These yield even noisier RSSi signals with a quick decay, causing ambiguous or even void positioning readings These constraints likewise jeopardise the success of any approach based on instantaneously “maximum RSSi” readings (argMax), even when the ambition is the sole room-level localisation. To deal with these limitations, we proposed in [4] a method based on an end-to-end Multi-Layer Perception (MLP) neural network. Typical museum geometries are an obstacle to end-to-end learning, including physically disconnected areas, geometrically close, that can be ambiguously represented by RSSi readings It is reasonable that a methodology as an MLP trained on short RSSi windows is incapable of resolving these aspects: no “expert” knowledge, such as constraints dictated by museum topology (preventing, e.g., “nonphysical jumps” amongst rooms) is injected.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
