Analysis: Could a smart parking recommendation system bring an end to streets and footpaths cluttered with discarded e-bikes and scooters?
Shared e-bikes and scooters have been widely accepted around the world as an eco-friendly alternative to traditional transportation methods. But finding a suitable parking space is a pressing problem for every user, especially in crowded urban areas, which will lead to wasted time and the potential for improper parking. Therefore, an effective and convenient assistance system is essential for users to help them find parking spaces, especially to meet the changing needs of users during their trips.
Recent media coverage has highlighted the increasing problem of inconsiderate parking in cities, where e-bikes and scooters left carelessly on pavements can create significant obstacles for pedestrians, especially those with disabilities. This situation underscores the urgent need for effective parking management solutions in our cities.
We need your consent to load this rte-player contentWe use rte-player to manage extra content that can set cookies on your device and collect data about your activity. Please review their details and accept them to load the content.Manage Preferences
From RTÉ Radio 1's Drivetime, the Road Safety Authority answer listeners' questions about new laws around e-scooters
In Dublin, the inconsiderate parking issue is also present. Our recent study analysed the parking behaviour of shared e-bike users in Dublin using real-world data from MOBY Bikes. Our research found that up to 12.9% of shared e-bike users did not park their shared e-bikes properly at designated stands, which inevitably reduced the overall operational efficacy of these tools.
To address these challenges, our research team has developed U-Park, a user-centric smart parking recommendation system for e-bikes and scooters. Using advanced AI technologies and real-world datasets from MOBY Bikes and Dublin Bikes, U-Park aims to enhance the user experience by providing accurate and personalised parking suggestions based on historical journey data and real-time trip trajectories.
U-Park combines historical journey data and real-time trip trajectories to predict both the user's destination and the availability of nearby parking spots accurately and proactively. Based on the prediction results, personalized parking suggestions are provided to enable users to efficiently locate parking spaces.
From RTÉ Prime Time, concerns around e-scooters and collisions
Imagine a commuter near Dublin City University in Glasnevin who needs to get to their office in the city centre in the afternoon and is thinking about parking for their shared e-bike. U-Park can help by identifying the best parking spot near their destination before they even set off.
As illustrated here, U-Park provides pre-journey recommendations based on current availability and traffic flow. Users can choose from predicted destinations or manually input their preferred location. During the trip, U-Park offers real-time updates on the available parking spots. This proactive approach not only saves users time searching for parking but also helps them avoid fines for improper parking, all in a user-friendly manner.
The advantages of this system are significant. On the one hand, experiments have shown that this system can increase the probability of finding a parking spot by about 25% on average, and even more when parking is in high demand. These improvements are brought by the system's ability to adapt and learn the trip behaviours of each user. On the other hand, by enhancing the convenience and appeal of shared e-bikes and scooters, U-Park encourages people to switch to more environmentally friendly modes of transportation and can help reduce traffic congestion and pollution, promoting a greener and more sustainable environment.
U-Park employs advanced machine learning algorithms, including graph neural networks and transformers, to analyse and extract complex data patterns from the collected datasets. This analysis allows the system to predict parking availability with high accuracy. By factoring in variables such as time of day, day of the week, weather conditions and local events, U-Park can provide dynamic and context-aware parking recommendations. Importantly, all data used in U-Park's analysis is anonymized and fully compliant with GDPR regulations, ensuring user privacy and data security.
We need your consent to load this rte-player contentWe use rte-player to manage extra content that can set cookies on your device and collect data about your activity. Please review their details and accept them to load the content.Manage Preferences
From RTÉ Radio 1's News at One, Ireland's first e-scooter rental scheme launches in Wexford
Another critical aspect of U-Park is its planned integration with mobile applications. Users will receive real-time notifications about parking availability and recommendations directly on their smartphones. The app interface is intended to be user-friendly, with clear visual cues and interactive maps to guide users to the nearest available parking spots. The planned integration also includes features such as turn-by-turn navigation and estimated walking time from the parking spot to the final destination, enhancing the overall user experience.
In summary, we believe that U-Park represents a significant advancement in the field of smart urban shared e-mobility. By leveraging AI and real-world data, it enhances the efficiency and convenience of shared e-bike services, promotes sustainable transportation, and contributes to the development of smarter, more connected cities.
More details of this project can be found here.
Follow RTÉ Brainstorm on WhatsApp and Instagram for more stories and updates
The views expressed here are those of the author and do not represent or reflect the views of RTÉ