14.5 Conclusions & Future Work
Conclusion
Our final assembly of The Thing, successfully fulfilled our project objective: to replicate the organic motion of an articulated robotic hand from the show, The Addams Family’s. By optimizing gear placement and using a single degree of freedom, we achieved coordinated motion across three fingers that allowed the hand to crawl forward. Initially, conceptual finger assemblies were analyzed in motion-gen and Solidworks, with a emphasis on the displacement and velocity profiles of the fingertips. When prototypes were first fabricated, friction and other real-world constraints deviated the performance from simulation. Critically, we noted that the links length and gear size were extremely sensitive to Grashof conditions. After simplifying the design, speccing our motor, and creating 3D fingers that replicated a real hand, we began our final assembly. Ball bearings were embedded to the thumb and pinky to reduce friction and half the hand was fabricated with acrylic to allow observations into the mechanism. Despite challenges related to center of gravity and consistent friction at the fingertips, the robot was able to move across a flat surface, validating our design.
Lessons Learned
Early iterations of our prototype taught us that reducing part count not only simplified assembly but also improved reliability and minimized mechanical losses.
The transition from CAD to a physical prototype revealed tight tolerances and misalignments that required design reconsideration.
The importance of speccing a motor cannot be understated as this dictated the force analysis. A motor that is too powerful could risk damaging the mechanism, while a motor that is too weak could risk no linear movement across the assembly.
To increase prototyping efficiency, reprints should be done on parts that are the quickest to produce. It is much faster to print a large part once and then iterate through smaller parts until the fit is satisfied.
The use of a test piece to tolerance fits greatly increased efficiency. These were produced on the laser cutter.
Future Work
Although our final design successfully demonstrated linear crawling using a single DOF, it opened the door to numerous directions for improvement and functional enhancement:
Currently, all finger joints operate within a fixed plane. A future design could integrate rotational motion by introducing ball joints or additional degrees of freedom, enabling the hand to climb or gesture.
Adding modular joints or variable-length links could allow the hand to adapt its to new surfaces, slopes, or even curved paths, increasing its versatility.
Implementing closed-loop control would add to the robustness of the design. This would allow the hand to adjust motion in real time based on terrain, slippage, or obstacle feedback.
By actuating the thumb and pinky fingers independently using a servo motor, the hand could steer itself to the right or left by shifting its center of gravity.
As we are trying to mimic the animatronic, future prototypes could utilize silicone molds to improve its aesthetics.
Tips for Future Teams
Prototype as quickly as possible. MotionGen and CAD may lead to a false sense of confidence as issues only reveal themselves in physical builds.
Plan wiring and electronics layout early, especially if the build involves components that need to be accessible such as batteries. This should ideally be integrated in the Design for Assembly considerations.
Assign sub-team roles based on strengths. Allocate tasks by design, manufacturing, analysis, and documentation. This will allow for better planning and understanding which tasks require more lead times or codependencies.
Revisit the problem statement regularly and revise if necessary. Check if each new idea truly serves the project goals or adds unnecessary complexity.
Make documentation a live process. Capture test results, failed iterations, and lessons learned as you go.
Acknowledgments
We would like to sincerely thank Dr. Meredith Symmank for providing the foundational knowledge and feedback that shaped every iteration of our design. A huge thank you to our TA, Connor Hennig, for helping us refine our mechanism and pave the path for the aesthetics. We also extend our gratitude to the staff at Texas Inventionworks, whose resources and guidance made this project possible.
Link