Sidewalk Labs’ mission is to radically improve quality of life in cities. The ability to confidently and comfortably ride a bike or meander down the street is critical to that mission. So is the ability to get where you need to go as efficiently as possible, which often involves traveling in a vehicle. These two needs can often be at odds with each other, but while the vehicle usually wins today, the balance is starting to shift.
Many cities, like Boston and Toronto, have published Complete Streets Guidelines to promote design standards for pedestrians, bicycles, transit, and public space. In 2017, NACTO released its Blueprint for Autonomous Urbanism to “proactively guide the [self-driving vehicle] technology to prioritize people-first design.” Sidewalk Labs aims to build on these ideas by asking: “Instead of teaching self-driving vehicles to operate on today’s streets, can we take advantage of new technologies to fundamentally redesign the street?”
This living document proposes design principles that strive to harness these advances to create safer and more flexible streets. These principles will be updated periodically based on collaboration with city planners, engineers, mobility providers, and technology companies — and by Sidewalk Labs itself, as we test designs in prototype and pilot environments.
Cities often have the worst of both worlds when it comes to street design: top speeds that create safety risks, but average speeds that frustrate everyone. The solution is often to make streets wider by adding lanes and buffers, but that approach can do more harm than good.
Cities often design streets to be safe by making them wide, but wide streets cause speeding.
Streets today are designed to allow vehicles to move quickly. But this decision requires streets to be designed defensively as well — because speed kills. As a result, engineers design wider lanes to account for drivers who drift or veer, and they design buffer spaces like shoulders, medians, and street-parking areas to try to improve pedestrian and cyclist safety. But they are not safe; more than 6,700 pedestrians and cyclists died on streets in the United States in 2017 due to automobile crashes.(1) Neither pavement markings nor bollards are enough to protect vulnerable bicycles and pedestrians — and certainly not enough to make them feel comfortable.
This approach doesn’t help move people, either.
Despite being engineered for speed, today's streets are often congested — and frustratingly slow. Congestion caused by double-parking and uneven distribution of traffic volume across the day leads to lower average speeds overall. In 2018, nearly every major U.S. city recorded a downtown last-mile travel speed below 20 mph.(2) In downtown Toronto, the speed limit is 40 km/h (~25 mph), but most vehicles travel at an average speed of 24 km/h (~15 mph) — and some much, much slower than that.(3) As a result, drivers and passengers are still frustrated with long, stop-and-go commutes.
One common solution is to add even more lanes, but this leads to streets that can feel empty, because they’ve been designed for the worst-case traffic scenario.
In an effort to accommodate more vehicles, engineers have defaulted to calculating the space needed to handle peak, rush-hour demand. The result is acres of pavement that are empty at all other times and are neither pleasant to walk around nor conducive to the types of welcoming urban spaces that encourage street life. Part of the reason engineers feel the need to plan for worst-case traffic scenarios is because curbs and pavement markings are set rigidly into place and unable to adapt to changing needs.
Technology is not a cure-all solution to mobility challenges. But it offers the chance to fundamentally redesign our street system with narrower, safer streets that still get people where they need to go.
Connected and autonomous vehicles (CAVs) can be required to follow speed limits and can operate in narrow streets where lanes may appear, disappear, or change direction.
Connected vehicles are vehicles driven by people that receive warnings on speed limits, potential conflicts, hazardous conditions, and other detailed information to improve safety. Autonomous or self-driving vehicles are able to ingest this information and have the vehicle itself respond, without a person driving.
Together, CAVs can be expected to follow speed limits, stay out of areas that are restricted, and obey rules of interaction with cyclists and pedestrians. These advances also apply to e-bikes and e-scooters that could be hard-coded to remain in vehicle or bike lanes. Similarly, CAVs could safely travel on narrower streets that are prioritized for transit, bicycles, and pedestrians, including pedestrians using wheelchairs or other assistive devices.
Dynamic (LED-embedded) pavement and moveable street furniture can help adapt the number of lanes, the width of the sidewalk, and even the direction of the street, meaning that a narrower street can serve multiple uses based on demand.
The operation and character of a street can change daily when raised concrete curbs can be removed in favor of dynamic pavement and moveable street furniture. Several companies have started to experiment with dynamic pavement, which embeds LEDs into the surface to change the color and shapes of markings. These design features can be used to create travel lanes, bike lanes, transit lanes, or pick-up/drop-off zones. They can also be used to change a lane’s travel direction, providing more flexibility than a fixed, grade-separated curb ever could.
Such a dynamic allocation of space allows for a potential reduction of vehicle space, creating safer crossing distances for pedestrians; providing a more pleasant walking and cycling environment; improving the travel experience for pedestrians using strollers, wheelchairs, or other types of wheels; and naturally slowing down vehicles that are used to wide lanes.
Sensors, digital signage, and integrated navigation apps and fleets can communicate real-time information on speed limits and lane closures.
Spatial occupancy sensors can give cities a better understanding of street conditions by generating real-time feedback like curb space availability or congestion on a given road. That information can be communicated directly to travelers through digital signage or via integration with vehicles and navigation apps.
It can also identify patterns that emerge over time, information that is critical to urban planners and traffic engineers. For example, BriskLUMINA sensor applications have helped planners in Atlanta and Pittsburgh identify intersections with higher than normal risk of pedestrian injury. Other cities have used sensors to help optimize traffic light timing.(4)
Traffic management tools can recommend changes to lanes, speed limits, and pricing to maintain person-throughput or meet policy goals, such as Vision Zero.
Traffic management tools can make the most of roadway space and increase “person throughput,” or the total amount of people traveling through an intersection, across all modes (not just vehicles). These tools include low-cost sensors, edge computing capabilities, machine-learning simulation models, and adaptive traffic signals that can adjust green times to optimize flow or prioritize certain modes. Together, these tools can form a mobility management system that can adapt to real-time street conditions by reallocating lanes and adjusting signal timings to keep all modes moving — and safe.
One promising management advance is the bicycle “green wave,” which works with adaptive traffic signals to give cyclists a premium experience. LED indicators embedded at the edge of a bicycle lane can light up in front of cyclists to form a moving green segment. The segment sets the ideal travel speed for cyclists, so they arrive at intersections when the traffic signal is green. Information on speed and green times can be communicated by fleets and navigation apps.
Principle 1: Tailor streets for different modes
New capabilities make it possible to design streets that prioritize certain modes, instead of aiming to accommodate all uses at all times of day. Laneways prioritize pedestrians while Accessways prioritize cyclists. Transitways give priority to public transit through dedicated lanes and signal priority. Boulevards are intended for all modes but primarily for vehicles.
Principle 3: Incorporate flexibility into street space
Adaptable infrastructure and real-time traffic insight make it easy for lanes to become “dynamic,” serving different purposes across the day. Sidewalk Labs is exploring a concept we call the "dynamic curb" which could be reserved for vehicles or converted into public space, depending on priorities. Optimizing this space requires a management system to understand demand and congestion patterns at various times and can vary depending on local policy objectives.
Principle 4: Recapture street space for the public realm, transit, bikes, and pedestrians
CAVs, adaptable infrastructure like dynamic pavement, and moveable street furniture enable cities to recapture space once devoted to parking and vehicles. This space can be reallocated to the public realm and high person-throughput modes, such as transit, while still enabling all travelers to get where they need to go.
Streets for an Integrated Mobility System
The Street Design Principles should be considered just one part of an overall mobility strategy. Even the best-designed street network can only realize its full potential as part of an integrated transportation system with many trip options.
For our Sidewalk Toronto project, this overall mobility strategy is anchored in the extension of a high-capacity light rail transit network — knowing that public transit is by far the most efficient way to connect people and jobs across dense urban areas.
This strategy continues with expanded walking and cycling infrastructure to encourage the use of active transportation modes, with, bike-share, scooter-share, and other low-speed vehicle options playing an increasing role.
Finally, new mobility options — such as carshare, taxi, and ride-hail services — can help reduce the need for residents or workers to own a car while still facilitating vehicle trips.
Sidewalk Labs recognizes that a manager (often a city department or agency) is required for all these types of streets and trip options to work in concert. It is important that this manager be empowered to use tools like regulation changes, pricing, and adaptive traffic signal management to achieve the policy goals and performance targets that are set.
The Street Design Principles are the foundation for this integrated mobility system, providing the infrastructure and framework for cities to balance the need to move people with the re-emergence of streets as vital community space.
The goal of these prototypes will be to gauge how drivers, pedestrians, and cyclists react to these designs and, in particular, the dynamic elements.
Over the course of 2018, Sidewalk Labs hosted a series of co-design sessions, events, and workshops in order to engage with the accessibility community and co-create our accessibility principles with them. We remain committed to these principles, which will evolve as we receive more feedback, and we will continue to work with the accessibility community to ensure our street designs work for all people with lived experience of disability.
We’ll have a better understanding of how dynamic pavement, bicycle LEDs, and sensor hardware work — and begin to test operational, maintenance, and life-cycle costs.
We’ll bring these elements together in order to test for safety, operability, and throughput.
Most importantly, we’d like to hear from you — the mobility engineers, planners, advocates, providers, disrupters, and enthusiasts. Let us know what you think, and help us drive towards the next version of these designs.