Enhancing Crowd Monitoring System Functionality through Data Fusion: Estimating Flow Rate from Wi-Fi Traces and Automated Counting System Data

Abstract

Crowd monitoring systems (CMSs) provide a state-of-the-art solution to manage crowds objectively. Most crowd monitoring systems feature one type of sensor, which severely limits the insights one can simultaneously gather regarding the crowd’s traffic state. Incorporating multiple functionally complementary sensor types is expensive. CMSs are needed that exploit data fusion opportunities to limit the number of (more expensive) sensors. This research estimates a data fusion algorithm to enhance the functionality of a CMS featuring Wi-Fi sensors by means of a small number of automated counting systems. Here, the goal is to estimate the pedestrian flow rate accurately based on real-time Wi-Fi traces at one sensor location, and historic flow rate and Wi-Fi trace information gathered at other sensor locations. Several data fusion models are estimated, amongst others, linear regression, shallow and recurrent neural networks, and Auto Regressive Moving Average (ARMAX) models. The data from the CMS of a large four-day music event was used to calibrate and validate the models. This study establishes that the RNN model best predicts the flow rate for this particular purpose. In addition, this research shows that model structures that incorporate information regarding the average current state of the area and the temporal variation in the Wi-Fi/count ratio perform best.

Data Assimilation for Agent-Based Models

“…Most traditional methods for online pedestrian monitoring and behavior prediction do not take advantage of any simulation engine [23], and directly work on data and are often coined with the term "Crowd Monitoring Systems (CMS)" in the literature [21]. CMS usually focuses on real-time data collection, feature extraction, and pedestrian tracking.…”

“…We will start with traditional crowd monitoring systems, which are purely data-based (not incorporating models), and compare them with DA methods so that the reader will better understand the challenges that data assimilation will address. Many of the current real-time pedestrian/passenger monitoring systems do not use simulation engines for crowd monitoring and prediction and are purely data-based [21][22][23].…”

Sensing and Forecasting Crowd Distribution in Smart Cities: Potentials and Approaches

“…However, using Wi-Fi data only to monitor and forecast the crowd size and dynamics may yield inaccurate results. For this reason, some works in the literature propose to integrate Wi-Fi data with other data sources, such as data coming from stereoscopic cameras [48] in order to estimate the crowd size, or from an automated people counting system [49] in order to better approximate crowd sizes.…”

Pictorial depiction on controlling crowd in smart conurbations using Internet of Things with switching algorithms

“…The traces of crowd in public management system is monitored using flow rate of wireless modules 7 where all the system data is counted. In this counting stage it is not possible to predict the number of human interference as no explanation on such procedures is provided.…”