Articles
Making Cobots Easier to Deploy: Lessons From a Mayo Clinic Study
What a Mayo Clinic feasibility study taught us about cobot training, remote support, and making automation more practical for smaller manufacturers.
Cobot Team Published: September 3, 2026 7 min read
During the COVID-19 pandemic, Cobot Team had the opportunity to work with a multidisciplinary team at the Mayo Clinic Medical Simulation Center in Jacksonville, Florida on a unique challenge: could a collaborative robot perform physical tasks in a simulated ICU room that would otherwise require a health care worker to enter the room?
Travel restrictions meant our team could not simply fly across the country and install the system in person. Instead, Cobot Team shipped a Sawyer collaborative robot from Portland, Oregon to Jacksonville and supported the Mayo team remotely as they learned to operate and program the robot for a series of simulated tasks.
That collaboration ultimately became a published feasibility study in Mayo Clinic Proceedings: Innovations, Quality & Outcomes. Cobot Team team members Tom Szambelan, Justin Doty, and Brendan Ball were among the study's authors.
The study focused on healthcare and using robotics to interact with medical equipment, but looking back at the experience, some of the most useful lessons apply well beyond a hospital room. They speak directly to questions small and midsize manufacturers still ask when considering collaborative robots. How difficult are they to learn? Do we need a robotics specialist on staff? What happens when we need support? And how much of that support can happen remotely?
A Different Kind of Robotics Challenge
During the height of the COVID-19 pandemic, even a routine trip into an intensive care room could require the time consuming process of putting on and removing personal protective equipment, while also increasing a health care worker's potential exposure to the virus.
The Mayo team wanted to explore whether a collaborative robot could perform certain physical tasks inside a simulated COVID-19 ICU room. The goal was not to replace doctors or nurses. It was to test whether a robot could handle specific interactions with medical equipment and controls that might otherwise require someone to enter the room.
The resulting project was designed as a proof of concept feasibility study in Mayo Clinic's Medical Simulation Center, testing whether these tasks could be performed robotically in a simulated ICU environment.
Five Tasks Tested With a Collaborative Robot
The team selected five common physical tasks for the Sawyer cobot to attempt:
-
Push a button on an IV pump after an alert
-
Adjust a ventilator control
-
Silence an ICU monitor alarm
-
Adjust oxygen flow at the wall
-
Press a nurse call button
All five tasks were successfully completed robotically during the simulated experiments. The research team completed the five experiments in less than 30 days.
The individual tasks were clearly defined. The study tested whether those specific physical tasks could be taught to a collaborative robot and carried out successfully.
That same principle matters in manufacturing. Successful automation often starts with a well defined task rather than trying to automate an entire operation at once.
Supporting the Team From Across the Country
One of the most interesting parts of the project for us was how the robot was supported.
Cobot Team was in Portland. The Mayo team was in Jacksonville. With travel restricted, Cobot Team personnel provided real time video conferencing support to guide the medical team through robot operation, programming, and execution of the experimental tasks. The team was able to learn and work with the robot without Cobot Team being physically onsite.
The learning curve was also relatively short for the specific tasks being tested. The researchers reported approximately 30 to 40 minutes of task specific learning during each session, while programming a task from scratch generally required about 45 minutes to one hour of training.
The time required to learn and program a cobot will vary with the complexity of the application, but the experience demonstrated how quickly a new team could begin working with the technology.
The experience demonstrated something important: the people using the robot did not need Cobot Team physically standing beside them to begin learning the system and accomplishing useful tasks.
What That Experience Means for Manufacturers Today
The setting was very different from a manufacturing floor, but the experience highlighted several lessons that still apply to manufacturers considering collaborative robots today.
You May Not Need a Robotics Department
One of the barriers to automation for smaller manufacturers is the assumption that adding a robot means adding a robotics engineer.
That is not necessarily the case.
Collaborative robots were developed in part to make automation more accessible. The Mayo project demonstrated what that can look like in practice. The team learned to operate the system, program specific tasks, and carry out the planned experiments with Cobot Team providing guidance remotely from across the country.
Manufacturing applications still require proper engineering, integration, and training. What this experience showed is that the people who already understand the process can play an active role in operating and adapting the automation, without needing to become robotics specialists themselves.
Remote Support Can Make Automation More Accessible
Not every question or programming issue requires an onsite service visit. During the Mayo collaboration, Cobot Team helped a team roughly 3,000 miles away learn to operate and program the robot through remote support.
That experience demonstrated how video conferencing and remote technical guidance could provide practical support even when Cobot Team couldn't be physically onsite. Some issues will always require hands-on service, but many programming questions, troubleshooting needs, and adjustments can be addressed remotely.
For small and midsize manufacturers, that can make automation easier and more cost effective to support over time, especially for facilities located farther from their integrator.
Ease of Use Matters More for Smaller Manufacturers
Large manufacturers may have dedicated engineers, automation specialists, and maintenance teams. Smaller operations often have fewer internal resources to dedicate specifically to robotics and automation.
That makes ease of programming, training, and access to technical support especially important. Robot specifications like payload and reach matter, but they are only part of what determines whether an automation system will work well for a manufacturer.
The system also has to fit the people who will actually use and support it.
That was part of the early promise of collaborative robotics, and it remains important today. Cobot Team focuses on systems that manufacturers can operate and support with their existing teams, backed by the training and technical support they need.
Collaborative Robots Have Continued to Evolve
The Mayo experiments took place in 2020 using a Sawyer collaborative robot from Rethink Robotics. Sawyer was designed around many of the qualities that helped make collaborative robots more accessible, including intuitive programming and integrated sensing. While Rethink Robotics is no longer in operation, Cobot Team continues to support existing Sawyer systems today.
Collaborative robot technology has continued to advance since the study. Manufacturers now have access to a wider range of robot sizes, payloads, tooling, vision systems, and other capabilities for applications ranging from machine tending and palletizing to welding, inspection, assembly, and material handling.
The technology has evolved, but the goal remains much the same: make automation capable enough to solve the application while keeping it practical for the manufacturer to operate and support.
Start With the Right Application
For manufacturers considering their first collaborative robot, one of the most important lessons from this project is the value of starting with a well defined application. Automation becomes much more approachable when the task is understood, the robot is well matched to the application, and the people using the system have the right training and support.
A good first project often starts with a repetitive task that can be evaluated against clear requirements. From there, the robot, tooling, controls, safety requirements, and integration approach can be designed around the actual process.
Cobot Team takes a phased approach to automation: start with an application that makes sense, demonstrate the value, build confidence within the team, and expand from there.
The Mayo Clinic collaboration took place in a very different environment from a manufacturing floor, but it gave us an early example of something we continue to see today: collaborative robots become far more useful when the technology is approachable and the support behind it is accessible.
Read the Published Study
The full study, "Robotics in Simulated COVID-19 Patient Room for Health Care Worker Effector Tasks: Preliminary, Feasibility Experiments," was published in Mayo Clinic Proceedings: Innovations, Quality & Outcomes.
