What do Insects and Bycicles have in common? Find out in Eduardo Castello’s Paper about Urban Mobility Swarms

February 26, 2024

This week opens with great news: we are celebrating a new publication in the field of robotics! Eduardo Castello, member of the IE Research Datalab, has recently showcased his article “Urban Mobility Swarms: Towards a Decentralized Autonomous Bicycle-Sharing System” in the IEEE 26th International Conference on Intelligent Transportation Systems. Professor Castello’s paper is his first publication under his new affiliation with IE, and it showcases his innovative research on urban mobility and self-organizing systems.

Urban mobility can often be categorized as a complex system – e.g., a nonlinear system composed of many interacting components with interdependent relationships. The growing trend towards shared, lightweight, and autonomous vehicles requires planning solutions that are less centralized and can manage the increasing complexities of new mobility. Professor Castello’s research investigates planning strategies for shared micro-mobility systems, focusing on shared autonomous bicycles.

Vehicle rebalancing within such systems poses a critical technical challenge and has substantial environmental and economic implications. To tackle this challenge, they propose a fully decentralized approach that allows autonomous bicycles to rebalance in a self-organizing manner via stigmergy, a bio-inspired mechanism for indirect communication. While the bicycles autonomously navigate their urban environment, they locally update RFID tags at intersections, leaving virtual pheromone trails that collectively guide each other toward high-demand areas.

The efficacy of professor Castello’s approach is assessed through a realistic agent-based model of Cambridge, MA (USA). Results highlight the capacity of autonomous bicycles to rebalance in a self-organized manner, using strictly decentralized local communication, while significantly reducing the average user wait time compared to no rebalancing and random rebalancing. These findings emphasize the feasibility and potential of decentralized planning strategies in handling complexity within new mobility systems.

Readers can find the article in the following link: https://ieeexplore.ieee.org/abstract/document/10421869