Abstract
"Herd Routes" is a preventative IoT-based system designed to improve female pedestrian safety by incentivizing people to walk along specific "busy" routes. By combining a dynamic pricing controller (PID) with a zero-fee distributed ledger (IOTA Tangle), the system creates a "safety in numbers" effect through societal incentivization rather than reactive emergency alerts. This paper presents the architecture, simulation results, and hardware-in-the-loop testing of the proposed system.
Introduction
Despite the prevalence of location-based safety apps, most existing solutions are reactive, relying on "panic buttons" or triggers that activate only after a threat is perceived. Furthermore, these tools often place the burden of safety entirely on the potential victim. Research indicates that pedestrians, particularly women, feel significantly safer on busier, well-lit streets. This paper addresses the challenge of "preventative safety" by proposing a system to proactively orchestrate urban pedestrian flow, ensuring individuals do not have to walk alone on quiet, high-risk streets.
System Architecture and Methods
The proposed system utilizes a three-layer architecture: User, Application, and Ledger. This structure facilitates a seamless transition from user onboarding to routine use, where the app calculates routes and manages token incentives.
Dynamic Incentivization
The core of the system is a Proportional-Integral-Derivative (PID) controller used to manage pedestrian density. If a specific route is under-populated, the "token price" (reward) for walking that route increases to attract more users. As the route reaches a target density, the reward decreases to maintain stability.
Distributed Ledger Technology
To handle micro-payments and maintain a secure, immutable record of user locations, the authors selected the IOTA Tangle. Unlike traditional blockchains, IOTA is designed for IoT; its fee-less structure is essential for high-frequency messaging and micro-incentives where transaction fees would otherwise exceed the reward value.
Hardware-in-the-Loop (HIL) Testing
The system was validated through an Android app integrated with the Simulation of Urban Mobility (SUMO). This allowed real-world users to interact with a simulated crowd of 750 agents, testing technical feasibility and user desirability within a map of the Imperial College London campus derived from OpenStreetMap (OSM).
Results
Experimental results demonstrate the effectiveness of the PID controller in regulating pedestrian density. The error signal—the difference between actual and target density—consistently trended toward zero across 2,000-step simulations. The system successfully achieved a "critical mass" benchmark (roughly 5 users per street) in the majority of test cases. On average, it took 174.64 simulation steps for the first agent to navigate onto an incentivized route once the system was active. User validation through HIL tests indicated that participants felt more aware of others and perceived themselves as part of a community.
Discussion and Future Work
While the system shows promise, several challenges remain. Data privacy is a primary concern, though 63% of study participants felt location sharing was justified for safety. Potential "honey pot" risks could be mitigated through a referral-based trust system. Future work will explore sustainable financing models (e.g., local business or city council sponsorship) and integrate environmental data such as lighting and CCTV coverage into the routing intelligence.
Summary updated