Subject Area
Mechanical Engineering
Abstract
Conventional robot actuation systems provide significant flexibility in powering robotic mechanisms. However, they inherently increase system complexity by occupying more space, adding weight, and requiring onboard power sources. In addition, these systems are difficult to miniaturize beyond a certain scale. To address these limitations, alternative actuation mechanisms such as magnetic actuation have been proposed.
Magnetic actuation uses the interaction between a controlled external magnetic field and magnets embedded within the body of robots. As a result, it avoids the limitations of conventional actuation methods. Since magnetic robots lack internal actuators and power sources,they feature simpler designs and are easier to manufacture, operate, and miniaturize. Due to these advantages, they have been used in biomedical and micromanipulation applications, where their small scale and simplicity offer significant benefits.
Despite their advantages, magnetic robots have a major drawback. Independently operating multiple magnetic robots is highly challenging, if not generally impossible. One proposed solution to control multiple magnetic robots independently is to design robots with heterogeneous responses to the magnetic field by altering their geometry or magnetic properties nonuniformly.
In this thesis, we investigate this solution by studying systems of multiple heterogeneous magnetic robots that move in parallel directions with various selectable fixed velocity ratio sets under a global magnetic field. We develop a simple hybrid linear model with multiple selectable modes to describe the evolution their motion. Then, we perform controllability analysis and prove that if the velocity ratio sets are linearly independent and exceed the number of robots, the multi-robot system is controllable.
Subsequently, we develop a simple motion planning algorithm that navigates robots between arbitrary feasible positions. We incorporate the collision avoidance between robots as a nonlinear constraint, and formulate the motion planning problem as a nonconvex nonlinear optimization problem.
Finally, we introduce a sampling-based motion planning approach by adapting the asymptotically optimal Rapidly-exploring Random Trees algorithm to the hybrid multi-robot system. The adapted algorithm enables navigation of the robots through obstacles in constrained workspaces.
Degree Date
Fall 12-20-2025
Document Type
Dissertation
Degree Name
Ph.D.
Department
Mechanical Engineering
Advisor
Yildirim Hurmuzlu
Number of Pages
93
Format
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-Share Alike 4.0 License.
Recommended Citation
Asadi, Farshid, "Motion Planning for Multiple Magnetic Robots" (2025). Mechanical Engineering Research Theses and Dissertations. 62.
https://scholar.smu.edu/engineering_mechanical_etds/62
Included in
Acoustics, Dynamics, and Controls Commons, Controls and Control Theory Commons, Robotics Commons
