Ronan Furlong wears many hats. An architect by training, he understands urbanism and cities. As the executive director of Alpha, Dublin City University’s Innovation Campus and a centre of Internet of Things (IoT) innovation, he explores the ways technology advances how people and things move in urban environments. A big believer in e-scooters as the future of transportation, he is also co-founder of Luna, a precise positioning and computer vision technology company for micro-mobility. Through his work, Furlong realised that changing the public perception and safety of e-scooters is essential to securing their place in the transportation network.
“I was bitten by the micro-mobility bug, and I see the challenges and possibilities inherent in it,” says Furlong of his interest in the industry. “I also see a passion and clarity of mission to make transportation better and to make the world a better place. The current trajectory of car traffic is completely unsustainable. I’m looking forward to the opportunity to help change that through the advancement of micro-mobility technology.”
That message perfectly encapsulates Voi’s mission to create cities for living, so it’s only natural that we, along with the city of Northampton, would join forces to pilot a new program combining the latest in computer vision and GPS technology with our e-scooters to create safer riding experiences. This new program, which begins today in Northampton, will run for the remainder of the city’s e-scooter trial.
“It’s a case of a micro-mobility safety tech company — Luna — getting involved with one of the most innovative and safety-focused operators out there — Voi,” he explains. “It was inevitable that we would find each other.”
Why should pedestrians have to detect and avoid e-scooters? Scooters should be smart enough to flip that on its head and detect and avoid pedestrians.
The project aims to improve e-scooter safety by tackling two key e-scooter issues: pavement riding and pedestrian detection. Street clutter and fear of scooters mowing down pedestrians, especially vulnerable road users (including people with compromised vision), are major public concerns about the introduction of e-scooters to city streets. The solution to these issues lies in improving GPS accuracy and the scooter’s understanding of its immediate vicinity, so where scooters park and ride can be controlled — and these are the capabilities that Luna enables.
“My view is: why should pedestrians have to detect and avoid e-scooters? Scooters should be smart enough to flip that on its head and detect and avoid pedestrians?” says Furlong.
E-scooters, as with all modern connected devices, rely on the global navigation satellite system (GNSS) to identify their location and get where they are going, from Point A to Point B. (GNSS is comprised of several satellite clusters, including North America’s GPS, China’s BeiDou, Russia’s Glonass, and Europe’s Galileo.) However, GNSS has its limitations. The current standard GPS in all e-scooters is the same as the GPS in a smartphone. There is only a five-to-10-metre radius of location accuracy to any particular device. Even in good conditions, GPS accuracy in a phone is only this accurate some of the time, due to various factors; multipath interferences (for example, “urban canyons”, which cause signals to bounce off multiple surfaces of tall buildings) and ionospheric interferences (which degrade signals as they enter the Earth’s atmosphere) make it difficult to get an exact fix on location.
Technology that improves location accuracy isn’t new. We see it in high-end cars and autonomous agricultural machinery, such as combines equipped with precision hardware that can harvest a field a corn to 20-centimeter accuracy. But this technology doesn’t come cheap — the kit in that combine harvester could cost upwards of US$20,000. Luna has managed to reduce the form factor and building materials without reducing any of the accuracy and create the same level of precision for e-scooters.
So, how will it work?
Luna is the result of innovations in fleet telematics (in layman’s terms, the information exchange between a vehicle fleet and its central authority) and IoT. Luna’s technology stack consists of two key facets: centimetre-level precision (essentially next-generation GPS) and computer vision technology. Luna is able to offset the inaccuracies of legacy GNSS by overlaying a real-time kinematic (RTK) correction service on top of the “flicker” (error) in GPS to bring location accuracy to centimetre level.
“The secret sauce here is the correction service, purely on the RTK side of the tech stack, that brings the traditional GPS inaccuracy of five to 10 metres to a hyper-accuracy of five to 10 centimetres,” explains Furlong.
Luna is also developing computer vision technology that uses onboard smart cameras as sensors to govern and control where and how scooters are ridden. They can detect and count the presence of pedestrians, and with the help of a lane-segmentation algorithm, they can tell what surface or lane the scooter is being ridden on — footpath, cycle lane, or road carriageway. These cameras are outfitted on a select number of the Voi fleet in Northampton. Riders won’t know the difference between a Luna-enabled Voi e-scooter and a regular Voi e-scooter, because the technology is incorporated into the e-scooter’s existing IoT box. “Only under close inspection will they see a camera lens on the front of the scooter,” says Furlong.
The cameras, enabled by Edge AI, will detect and process instances of pedestrian recognition and count the number of pedestrians in its field of view. The cameras do not record faces or take images, so they are compliant with GDPR and other privacy regulations.
“All of this information can be processed ‘at the edge’,” explains Furlong. “It doesn’t have to be sent via the Cloud to a data centre to be processed and then relayed back to the scooter 20 seconds later, when it’s too late to do anything about the infraction. It’s all happening on the edge, in real time, and that makes the scooter far more responsive to its environment.”
This new information will enable Voi and the Northamptonshire City Council to make more informed decisions about e-scooter usage. “The AI doesn’t know what a pedestrian is until it’s trained to recognize one,” says Furlong. “Once we have a count of the number of pedestrians, then Voi and the city council can start to make decisions on the appropriate course of action. If there are 20 pedestrians immediately in front of an e-scooter, Voi can alert the rider, slow the scooter down, bring it to a stop, or warn the rider that they are on a footpath, as the case may be.”
Although Voi has yet to determine exactly how they’ll use these insights in Northampton (and, eventually across all of its 50+ European cities), the technology not only unlocks potential in accuracy of positioning and riding, but also from an operational and regulatory standpoint. For example, if the fleet management and rebalancing teams can pinpoint exactly where a scooter is, then they will save time and money retrieving scooters. The Luna mobile camera system can also be trained on any number of algorithms for a variety of applications. In the future, the cameras could recognize potholes on the roadway or recognize delivery vans blocking bike lanes outside of delivery hours, and then report this information back to the local authority.
“It’s an order of magnitude change in the resolution of GPS and a step change in the business applications that are possible with that new level of accuracy,” says Furlong.
Together, we are building the future of micro-mobility.
“Aside from all the well-documented first- and last-mile and environmental benefits, in the post-COVID recovery era, when public transport capacity is down, there just isn’t enough room for socially distanced one-and-a-half tonne metal boxes called cars on our streets,” says Furlong.
“Our partnership will complement public transport and allow people to travel in a safe, controlled, socially distanced fashion that bolsters the overall value proposition of micro-mobility and gets rid of some of the negative public perceptions associated with it.”