Roosevelt Island Tree Care Robot

Summary

The Roosevelt Island Tree Care Robot is a proposal for a low-speed autonomous robotot that monitors tree and plant health along Roosevelt Island parks and waterfronts. The robot will use cameras, sensors, and thermal imaging to detect plant disease, storm damage, stress from droughts, soil moisture, and heat-effects on vegetation. This project turns routine landscape inspection into a mobile data service and sllows for healthier greenery on Roosevelt Island.

AV Use Case

What AVs are involved?

The AV is a small level four autonomous service robot twuth a geofenced park/corridor. There will be about 2-3 robots that road around parks and around the island at walking speed, meaning the robot will not travel faster than 3-4 miles per hour. There will be a specific route - either on a park edge or green space - that the robot operates around only during early morning and late evening hours in order to avoid pedestrians and traffic during peak hours. The robot will use cameras, thermal imaging, LiDAR, soil moisture probes, air temperature and humidity theremometers, and particulate sensors. With these sensors, the robot will be able to inspect trees, soil considitions, planting beds, storm damage, and heat stress. The output will be reported to the New York State Department of Parks, Recreation & Historic Preservation, which will include tree-health maps, watering priorities, storm recovery reports, and maintenance alerts.

What are they doing?

Tree Care Robots run a scheduled, pre-planned route where they stop and monitor different trees, planting beds, lawn edges, bioswales, and waterfront vegetation zones. Within their scheduled route, the robot will scan for branch damage, leaf lost, pest indicators, and storm-related breakage. It checks for soil moisture and heat stress consditions nearby planting beds and sends maintenence alerts for trees or lanascaped areas that require attention. The robots will return to small solar charging dock when not on route.

The robot runs these pre-planned routes during low-conflict periods: early morning inspections during 6AM to 9AM and 7PM to 10PM for evening follow-up scans. These robots will be able to run inspections during events such as storms and heat waves in order to identify damage from flooding, erosion, stress, and more.

Why here?

Roosevelt Island and the Four Freedoms Park is a strong site as it combines waterfront vegetation exposure, planted landscapes, and trees in one area. The area is relatively flat and does not get extremely busy, making it suitable for a service robot to monitor. The waterfront and island setting also creates a real need to more frequent environmental monitoring. Furthermore, a robot that maintains and promotes a healthy neture landscape on Roosevelt Island will imrpove everyday experience for residents, visitors, tourists, Cornell Tech students, and more. This robot can become the perfect demonstration for how autonomous vehicles can support urban landscapes without overwheling pedestrian space.

Stakeholders

Who participates?

The Roosevelt Island Operating Corporation (RIOC) will serve as the project sponsor and will need to assist in coordinating sign off’s and access to park paths, maintenance areas, and space for charging infrastructure. New York State Department of Parks, Recreation & Historic Preservation or Roosevelt Island landscape maintenance teams will be able to use the robot’s data to guide tree care, watering, pruning, and storm response. Cornell Tech can support sensor evaluation and installation as well as data analysis and public-facing research on urban technology and service AVs. A robotics provider will be required to purchase supplies such as the vehicle fleet, remote monitoring system, and maintenance software. Additional participants include environmental consultants and community organizations that care about public space quality. Residents and park users also participate indirectly by giving feedback on whether the robot feels safe, helpful, or intrusive in the landscape.

Who is impacted?

Those that live, work, and visit Roosevelt Island will benefit from healthier trees, well-maintained walking paths, and faster response after storms. In addition, people with disabilities, children, and edlerly folks will benefit as the robot will hlep assure clear paths after post-storm fallen branches, damaged pavement, flooding, or overgrown vegetation. Maintenance workers are directly affected because the robot allows for a more efficient and faster method for how landscape issues are detected and fixed. The robot does not replace human workers, but rather provides them with a larger breadth of knowledge and range of information before problems beocme emergencies.

How does the solution use their capabilities?

Capabilities such as site access and operational authority from RIOC as well as research capacity from Cornell Tech. These capabilities allow the solution to become a part of a larger workflow of maintainence on Roosevelt Island. The robot collects data from the field, translates its observations into a readable dashboard of maintenence priorities, and helps optimize human crew deployment.

How does it address their concerns?

To address concerns such as pedestrian safety, the robot runs slowly, stays contained within geofences routes, had emergency stop buttons, and automatic yielding for people and other barriers. To address privacy concerns people may have about sensors and cameras, the solution limiting camera use to only navitation and analysis rather than using it does any facial recognition or other uneccessary imaging. Labor concerns are addressed by using the robot as a tool within the workflow for maintainence and decision-making rather than as a full replacement for human crews.

Relevant Blueprints for Autonomous Urbanism

The following urban design strategies are drawn from the NACTO Blueprint for Autonomous Urbanism, 2nd Edition.

Slow Zones & Shared Space Design

According to the NACTO Blueprint for Autonomous Urbanism, autonomous vehicles that are intended to operate near or around pedestrians should be managed through slow speeds, clear spatial cues, and pedestrian-priority design. At Four Freedoms Park, the robot’s pre-planned route will be a slow mixed-used space rather than a sole vehicle lane. The robot will always yield to pedestrians and will operate at walking speed. In order to design around this shared space, it will be important to add signage about robot yield zones at entrances/ecits and small pavement markings along the robot’s route. There should also be areas where the robot can pull out onto near narrow path segments and benches in order to properly yield to people travelling.

Dedicated AV Service and Maintenance Zones

The NACTO Blueprint also supports clearly defined zones where autonomous vehicles can stop, load, charge, or perform service functions without blocking pedestrian movement. In order to keep the robot’s operational duties and needs separate from that of humans, Four Freedoms Park should recieved two small AV service areas where one is a solar-assisted charging dock near a maintenance access point and acts as a mid-route pull-off zone where the robot can pause for longer scans. By using permeable pavement, low bollards, curb-level markings, and planted buffersm these service zones can help to distinguish robot activity from normal pedestrian circulation. The charging dock is placed away from the most crowded areas of the park and the scan zone is located near a cluster of trees or planting beds that require frequent monitoring.

Green Infrastructure Integration

Monitoring points are aligned with tree pits, planting beds, waterfront vegetation, bioswales, and heat-exposed paved areas. Where the robot identifies repeated drought stress or flooding, it will be able to propose solutions to improve the health of the green infrastructure it monitors, potentially turning this project into part of a larger climate-resilience strategy.

Methods

Step 1

Tool: ChatGPT image generation
Transformation: Generated the featured rendering using the prompt: “Autonomous tree care robot scanning trees and planting beds along a Roosevelt Island green corridor.” This prompt was intended to design a small low-speed service robot that operates in waterfront setting with trees, planting beds, and pedestrians.
Result: Produced hero-image.jpg, which shows the robot’s purpose and relationship to green corridors on Roosevelt Island.

Step 2

Tool: ChatGPT image generation
Transformation: Generated a street-level rendering using the prompt: “Street-level rendering of the robot operating at walking speed while yielding to pedestrians and park users.” This image was intended to focus on pedestrian safety and hihglifht how the robot will be able to share streets with other park goers.
Result: Produced street-view.jpg, showing the robot operating cautiously in a human-centered public-space environment.

Step 3

Tool: ChatGPT image generation
Transformation: Generated a site-plan-esque image using the prompt: “Site plan showing the robot route, tree-monitoring zones, charging dock, pedestrian-priority areas, and maintenance access points.” The intention here was to show how the robot moves through the site and where it pauses, scans, charges, and yields.
Result: Produced site-plan.jpg, a visual diagram clarifying the robot’s operating route, monitoring zones, charging location, and pedestrian-priority areas.

Step 4

Tool: geojson.io
Transformation: Drew the project boundary around the selected Roosevelt Island green corridor where the autonomous tree care robot operates. The boundary was used to define the robot’s geofenced service area and connect the proposal to a real site.
Result: Created boundary.geojson, which maps the project area for the Robot Island case study.