CMU Collaborates on Project To Develop Future Robotic Wheelchair Carnegie Mellon Robotics Institute Team Will Integrate Robotic Arm Into Design

Aaron AupperleeTuesday, November 4, 2025Print this page.

The Robotic Caregiving and Human Interaction Lab inside the School of Computer Science’s Robotics Institute will spearhead the software integration of a robotic arm into the wheelchair design as part of an Advanced Research Projects Agency for Health project.

Carnegie Mellon researchers will collaborate on a federally funded project to wholly rethink and redesign wheelchairs to incorporate new technologies and offer greater mobility.

The Robotic Caregiving and Human Interaction Lab inside the School of Computer Science’s Robotics Institute (RI) will spearhead the software integration of a robotic arm into the wheelchair design to assist people with grasping objects, opening doors and other chores and tasks both in and outside the home.

“The central idea is to make the future of wheelchairs,” said Zackory Erickson, an assistant professor in the RI and head of the Robotic Caregiving and Human Interaction Lab. “The wheelchair will be built around the robotic arm. If people are going shopping, they can use the arm to pick items off the shelf. If they are in their homes, they can clean their dishes, do laundry, water plants and more with their robotic wheelchair.”

The University of Pittsburgh will lead the project, which is funded up to $41.5 million by the Advanced Research Projects Agency for Health (ARPA-H) project, to develop the Robotic Assisted Mobility and Manipulation Platform (RAMMP) system. The platform will use next-generation robotics and new assistive technology to reimagine a wheelchair and assistive robotic arm that will improve the independence, safety and quality of life for people with disabilities, including veterans. In addition to Pitt and CMU, the project also includes Northeastern University, Cornell University, Purdue University and companies ATDev, Kinova Robotics and LUCI Mobility.

The new wheelchair will integrate advanced robotics, artificial intelligence, a novel operating system and digital twin technology. The RAMMP system will advance the design of powered mobility and manipulation devices by improving their function, obstacle detection and negotiation abilities. It will also seamlessly integrate with robotic arms for more effective object interaction. Its real-time, 360° environmental awareness and adaptive control features will allow users to navigate complex environments with enhanced capabilities, confidence and safety.

“Most powered wheelchairs aren’t designed to overcome many of the common challenges in the real world — and changing the environment to accommodate them is nearly impossible,” said Rory Cooper, founding and current director of Pitt’s Human Engineering Research Laboratories, and a co-lead of the project. “We need smarter technology that prevents tipping and falling, improves mobility, and adds more function such as coordinated mobility and robotic arm manipulation of objects so people with disabilities can fully participate in everyday life.”

CMU researchers will focus on the robotic arm. While some wheelchairs already have robotic arms, they are often bolted on after the fact and not integrated into the overall design. This robotic arm will be part of the wheelchair design from the beginning, and the CMU team will apply advances in robot learning, simulation, autonomy and advanced sensing to make it easier to use.

The CMU team will develop and refine algorithms to control the robotic arm mounted on the wheelchair. They’ll also create a digital twin of the wheelchair to test its functionality and train it in a simulated environment.

Robotic arms and grippers can be tricky and tedious to control manually. Using a robotic arm with a joystick to grab an item off a shelf in a grocery store or turn a doorknob in the home could take as long as three minutes, and being even a centimeter off could mean pushing objects around or having an unsteady grip. The automated assistance developed by CMU researchers could help align the gripper or perform other precise movements.

“Precision is really important for a lot of these tasks,” Erickson said. “We want to enable some underlying intelligence in the system with shared control between the arm and the user. If we can automate some level of manipulation assistance using robot learning methods and shared control techniques, we can make it easier to operate — not tedious or slow.”

The CMU researchers will also develop and integrate new sensors into the arm to further enhance the capabilities of the robotic arm. While cameras will be essential for allowing the arm to locate and align with objects, other tactile and haptic sensors could be used to perceive force, temperature and vibration or to use a touchscreen and other interfaces.

Finally, the CMU team will test the new wheelchairs with individuals who have disabilities. The studies will test the new technologies, specifically the robotic arm, sensors and control algorithms, in scenarios such as shopping and performing household chores. The CMU team will meet regularly with users and stakeholders and gather critical feedback to better understand their needs and provide data to the broader project team.

The Robotic Caregiving and Human Interaction Lab is no stranger to developing assistive technologies. Past projects in the lab have sought to create robotic systems to assist with getting dressed, hanging up laundry and eating. The lab even designed a head-worn assistive teleoperation (HAT) device — an experimental interface to control a mobile robot — and spent a week testing it in the home of a person with quadriplegia.

This research was funded, in part, by ARPA-H. The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies, either expressed or implied, of the U.S. government.