'Robotability Score' is based on the streets of the NYC for future robot operations

Not all sidewalks are created for delivery robots – some are uneven or clogged with people and bus stops – i.e. developed researchers from Cornell Tech “”Robotic abilityAnd rated every street in New York City how hospitable it would be for robots.

Its evaluation system is the first of its kind, according to the researchers, and could help urban planners and robotics companies to plan future robot deployments that do not disturb existing sidewalks.

“I don't know that every robot wants to have in his neighborhood, but if this is the case, the robotability assessment can help you think about what functions we have installed to welcome robots?” said senior author Wendy JuAssociate Professor of Information Science at Cornell Tech, the Cornell Ann S. Bowers College for Computer and Information Science and the Jacobs Technion-Cornell Institute and the Multicolege Department of Design Tech. “We just try to make these things a little more visible.”

Matt Franchi, doctoral student in the field of computer science, and Maria Teresa Parreira, a doctoral student in the field of information science, will present the study.The robotability assessment: enable harmonious robot navigation on urban streets”At the Computing Machinery conference on human factors in computer systems (CHI), April 28 in Yokohama, Japan.

Delivery robots have already driven on the street in Los Angeles and roam some college campus and airports. But most communities are currently undesirable places for robots.

Inspired by the reviews of walkability and accessibility in the neighborhood, the researchers developed the robotability value to create several characteristics that influence robot navigation into a single number.

“People love convenience,” said Franchi. “We believe that this tool will be easy to use when people imagine working in urban rooms.”

In order to develop the robotability value, the researchers first interviewed 10 experts in urban planning, robot navigation and accessibility of science and industry to decide which functions are included. The team then used an online survey among additional experts to determine how the meaning of each function for the end result can be weighted.

The robotability value comprises 24 characteristics, although the researchers 19 were used in their analysis of New York City. Six characteristics – pedestrian density, crowd dynamics, pedestrian flow, sidewalk quality, road width and density of the street furniture – formed almost half of the score.

New York City is the perfect place to develop such a score, according to the researchers due to the urban data available via the NYC Opendata website. This database contains information such as the width and the condition of the sidewalk and the locations of bus stops, bicycles and newspaper kiosks. The researchers also used around 8 million dashcam pictures that were collected in the city at the end of 2023 to estimate the vehicle, bicycle and pedestrian traffic.

In the entire city, the areas with the highest robotability were 4.3 times “robot” as areas with the lowest score. Franchi developed one Interactive mapWhere users can see how different locations compare.

The team tested the accuracy of their results by operating A Garbage – A garbage bucket on recycled hoverboard parts that is operated with a joystick – in two places with low and high robotability in Queens and Manhattan.

“Rush Hour enlarged the scores,” said co-author Frank Bu, a doctoral student in the field of computer science. Even at peak times, the trash can no problem rolling on empty sidewalks in areas with high robot ability. “Other places with containers, high pedestrian flow, flow of traffic and even with food trucks or shops are on the side of the sidewalks -that intensified the scene,” said Bu.

There were additional features that the researchers wanted to include, but the data were not yet available, such as: B. Local settings for robots. Future versions can include these functions and possibly dynamic functions such as real -time weather reports and pedestrian traffic.

Just as the existence of evaluations of walk -in and accessibility has influenced developers and urban planners to improve these qualities in districts, the team hopes that the robotability evaluation will do the same – but only in communities who want robots.

“We hardly start building cities for people – we are not trying to say that cities should be built for robots,” said Parreira. “Only to see the urban environment through the lens of a robot was our goal.”

This work was carried out by financing the National Science Foundation and the Digital Life initiative At Cornell Tech.

Patricia Waldron is the author of Cornell Ann S. Bowers College for Computer and Information Science.