Control of a 7-DOF MRI Compatible Robot Arm

Naichen Zhao

EECS Department
University of California, Berkeley
Technical Report No. UCB/EECS-2025-100
May 16, 2025

http://www2.eecs.berkeley.edu/Pubs/TechRpts/2025/EECS-2025-100.pdf

This work describes the structure and testing of the software, firmware, and electronics used to control a custom built MRI compatible 7DOF robot arm. The arm itself is designed to maneuver a Transcranial Magnetic Stimulation (TMS) coil around a patient's head within a MRI bore, allowing for precise positioning of the TMS coil.

The Arm is constructed using MRI compatible materials, driven by 7 Ultrasonic motors, each with its own Series Elastic Actuator (SEA), allowing for torque sensing on each joint. A microcontroller oversees the arm's low-level functions, running firmware written using the Lingua Franca (LF) framework, while the higher level planning is performed on a Linux computer running ROS2. The system is capable of tacking the force experienced on each joint while performing gravity compensation to filter out any gravitational load. The controller is also shown using position-based PID control to maneuver the arm to various positions while operating with a variety of end-effector weights.

This project's documents (including PCB schematics/layouts, microcontroller firmware, and ROS packages) can be found at:[https://github.com/biomimetics/MRIRobotProject].

Advisor: Ronald S. Fearing

\"Edit"; ?>

BibTeX citation:

@mastersthesis{Zhao:EECS-2025-100,
    Author = {Zhao, Naichen},
    Title = {Control of a 7-DOF MRI Compatible Robot Arm},
    School = {EECS Department, University of California, Berkeley},
    Year = {2025},
    Month = {May},
    URL = {http://www2.eecs.berkeley.edu/Pubs/TechRpts/2025/EECS-2025-100.html},
    Number = {UCB/EECS-2025-100},
    Abstract = {This work describes the structure and testing of the software, firmware, and electronics used to control a custom built MRI compatible 7DOF robot arm. The arm itself is designed to maneuver a Transcranial Magnetic Stimulation (TMS) coil around a patient's head within a MRI bore, allowing for precise positioning of the TMS coil. 

The Arm is constructed using MRI compatible materials, driven by 7 Ultrasonic motors, each with its own Series Elastic Actuator (SEA), allowing for torque sensing on each joint. A microcontroller oversees the arm's low-level functions, running firmware written using the Lingua Franca (LF) framework, while the higher level planning is performed on a Linux computer running ROS2. The system is capable of tacking the force experienced on each joint while performing gravity compensation to filter out any gravitational load. The controller is also shown using position-based PID control to maneuver the arm to various positions while operating with a variety of end-effector weights. 

This project's documents (including PCB schematics/layouts, microcontroller firmware, and ROS packages) can be found at:[https://github.com/biomimetics/MRIRobotProject].}
}

EndNote citation:

%0 Thesis
%A Zhao, Naichen
%T Control of a 7-DOF MRI Compatible Robot Arm
%I EECS Department, University of California, Berkeley
%D 2025
%8 May 16
%@ UCB/EECS-2025-100
%U http://www2.eecs.berkeley.edu/Pubs/TechRpts/2025/EECS-2025-100.html
%F Zhao:EECS-2025-100