Starling Crossing is an interactive pedestrian crossing that responds dynamically in real-time to make pedestrians, cyclists & drivers safer and more aware of each other.
Starling was spunout from Umbrellium in 2020. For more information on the product, please visit Starling Technologies.
The pedestrian crossing is one of the most complex points of urban negotiation that almost everybody experiences every day. However, crossing designs have not been updated for the ways that we use, or need to use, our streets in the 21st century. Most discussion about road technology focuses on vehicles, but with the Starling Crossing (STigmergic Adaptive Responsive LearnING Crossing) we have created a responsive road surface that puts people first by updating the design of pedestrian crossings (first introduced in the 1940s) to account for streets with more cars, pedestrians and technology, and a different societal relationship to urban transport infrastructure.
While it uses familiar and understandable road markings and colours, the Starling Crossing reacts dynamically in real-time to different conditions and is able to modify the patterns, layout, configuration, size and orientation of pedestrian crossings in order to prioritise pedestrian safety. The entire road surface at the crossing area is monitored by cameras and embedded with computer-controlled LEDs that can be seen from all angles, during both day and night.
Drawing on research by the Transport Research Laboratory, our full-scale prototype, installed temporarily in South London, is designed to support the weight of vehicles, remain slip-free in pouring rain and to display markings bright enough to be seen during daytime.
Using a neural network framework, cameras track objects that are moving across the road surface, distinguishing between pedestrians, cyclists and vehicles, calculating their precise locations, trajectories and velocities and anticipating where they may move to in the next moment.
At different times of day, and in different situations, the road can alter its configuration in real-time.
Early in the morning when there are few pedestrians, the Starling Crossing may only appear when someone approaches, guiding them to the crossing location that it has learned over time is the safest, leaving the road otherwise free for vehicular traffic.
Later in the day, when pubs close or a film ends and many people need to cross the road at the same time, the Starling Crossing automatically expands in width to accommodate increased pedestrian traffic.
If a person is distracted, looking down at their mobile, and veers too close to the road surface when a car is nearby, a warning pattern lights around them to fill their field of vision. If a child runs into the road unexpectedly, a large buffer zone is created around them to make their trajectory clear to any nearby drivers or cyclists.
In a particularly dangerous situation, when a pedestrian is rushing across the street but is in a cyclist's or driver's blindspot, the Starling Crossing adapts in the real-time to draw their attention directly to the hidden pedestrian’s location and trajectory.
The Starling Crossing can also adapt to different specific environmental conditions (e.g. creating larger pedestrian buffer zones in wet weather or to keep particularly polluting vehicles further from a school crossing).
Using the principles of stigmergy (c.f. the pheromone traces that ants leave, attracting other ants to the best paths toward food sources) the Starling Crossing is able to monitor and adapt to pedestrian desire lines over long term use so that, for example, if most people exiting a tube station end up walking diagonally across the road towards a park entrance, the crossing is able to reconfigure as a diagonal or even trapezoidal crossing, with corresponding safety buffer zones.
The Starling Crossing is a pedestrian crossing, built on today's technology, that puts people first, enabling them to cross safely the way they want to cross, rather than one that tells them they can only cross in one place or a fixed way. Key design principles include aiming to enhance people's perceptual awareness without distracting them, and highlighting safety relationships between people and cars so they can make their own decisions, rather than telling them what to do.
With particular thanks to Dan Sloane, Vincent Rebers, Immersive AV, Set Works, Professor Nick Tyler and Christopher Burman.