Science, Technology, Engineering and Mathematics with Science and Technical Writing (STEM) is a course taught by Dr. Crowthers. The coursework includes completing an updated logbook, understanding and analyzing previous research, writing detailed article notes, collecting and analyzing data, completing a Grant Proposal, and writing a STEM thesis. At the same time, students work on their independent Research Projects, where they can choose a topic of interest and work on it. This class allocates a lot of in-class work time for the project and has taught me a lot about independence.
The aim of the project is to design a modular robotic system capable of autonomous self-assembly with a novel mechanical latching connector. The key innovation is the articulated, multi-axis latching mechanism, allowing individual modules to tolerate angular misalignment during docking and have mechanical deformation even during assembly. In this way, it enables stronger connecting and adaptability between robots during self-assembly. The latch is used in a traditional rigid robot system of cube-like modules to demonstrate functionality in both chain and lattice configurations. Later, they will be applied to a novel system of origami modules integrating soft components into modular robotics. These applications will demonstrate a proof-of-concept design of how small adaptable modules can autonomously form larger, and more complex structures, extending the application of self-assembling robots to the real-world.
Self-assembling robots have shown promise in many laboratory environments, yet their performance and application remain limited to those environments due to their rigid units, fragile docking, and poor tolerance. This work introduces a novel type of soft-rigid hybrid modular robot inspired by origami structures and includes a new mechanical coupling mechanism designed to accommodate errors and misalignment. Each module contains a soft deformable body with a novel connector combining previous magnetic and mechanical methods, allowing passive alignment and locking even under imperfect scenarios. Mechanical testing shows that the connector withstands significant force loads while maintaining reversibility and success rate. Connector prototypes were assessed based on six criteria: alignment tolerance, maximum strength, reversibility, simplicity, scalability, and cost. Each prototype was also mounted onto a custom ball joint mechanism with yaw and roll rotational abilities to allow for more advanced functionality of robot assembly. After the components were integrated, the module was tested for critical positions the robot could reach to evaluate the effectiveness of the utilization of the origami deformation; the connecting mechanism was tested for maximum withstand-able force; and the self-assembly was tested for alignment reliability from different start orientations. This approach advances the practicality of modular swarms and offers insights into the effectiveness of origami deformation for self-assembly.
Autonomous modular robots are limited in individual capability and require connector mechanisms tolerating positional and angular misalignment.
Design and evaluate novel connector mechanisms that facilitates reliable autonomous self-assembly, targeting a soft robotic application.
Current research in self-assembling robots often has limitations in physical docking because existing connectors are purely magnetic or purely mechanical. Magnetic connections are weak, while mechanical connectors are often too rigid to tolerate error or deformation. As a result, modules often fail to assemble correctly outside of controlled laboratory conditions, reducing the real-world application in limited. There is a need for a connection mechanism that addresses these issues to allow modules to form both stable chains and lattice structures. Purely magnetic or purely mechanical connectors offer unique strengths while retaining limitations that could be overcome by cross-integration. The goal of this engineering project is to bridge those gaps by developing a latch that can form chain-like structures when applied to a soft worm robot system and prove feasibility in lattice configurations as well using more traditional modular robot models.
Initially, the approach is to design and try prototypes of the Yoshimura design using CAD software and those designs will be prototyped to better understand the folding technique. Of all the prototypes, the most effective design was saved for the robot module. Next, the front and back plates that drive the robot and control its cables were designed using CAD software, and the iterations were finalized. The individual module was then completely assembled, and the electrical units were all connected to understand the performance of a singular module. After the basic module was created, three prototypes of the coupling mechanism were designed using CAD, and they were prototyped with PLA. The coupling accuracy and strength of each of the prototypes were recorded and compared to existing designs. The best coupling design was then attached onto a module, and multiple robots were assembled to test the capabilities of assembling and the performance of the assembled system. Lastly, the capabilities and performance of the modular system was lastly compared to existing models of self-assembling robots.
Robustness (25%) and alignment tolerance (20%) were prioritized, while reversibility, simplicity, and scalability were each weighted at 15%, and cost was weighted at 10%. Prototype 3 achieved the highest overall score (7.85) due to its strong alignment tolerance, high mechanical strength, and low projected cost. SMORES-EP ranked second (6.75), followed by the Soft Magnetic Connector (6.30).
Prototype 3 achieved the strongest overall performance balance, driven by its high load-bearing capacity, strong reversibility, and low projected manufacturing cost. This performance highlights the effectiveness of integrating mechanical strength with compliant design features in modular connectors. The results indicate that hybrid connector designs can outperform purely magnetic or purely rigid approaches by simultaneously providing misalignment tolerance and robust structural strength. Overall, the proposed designs demonstrate improved scalability and projected cost efficiency, supporting their feasibility for larger multi-module robotic assemblies.
Having Trouble Viewing? Try Here!