Robotic Grip

[Add text here]

Ever since I was younger, I thought it was cool how robotics and mechanical machines could replicate how natural animals moved. Robotic hands with a glove could be almost indistinguishable from real hands, and robotic spiders could navigate terrain and calculate routes faster than me. I never really made anything robotic before this science fair, only small circuits with transistors and various micro controllers. I always wanted to do something like a robotic hand for a competition, and this was a perfect chance. I was skeptical at first but I still gave it a try.

I had one main goal in making this hand: to keep it under $50. Before this, I noticed how expensive robotic hands were, and I thought to myself that it didn’t have to be that expensive. If it’s under $50, this could really benefit the price of prosthetic and robotic hands. It came to $44 in total, just under the limit.

When making a prosthetic arm, this hand could be a great starting point because it shows basic finger movement and control while reducing the price by hundreds. Since it’s cost effective and uses fewer materials, by switching to biodegradable materials it will help the environment a lot. Using lithium ion batteries instead of disposable ones could reduce waste, making it better for the environment as well.

When I started this project, I researched how robotic hands worked. I watched YouTube videos, read Reddit posts, and used the Arduino wiki to learn about programming motors. I looked on content that was popular, easy to follow, and credible.

After research I planned all my materials: 5 SG90 servo motors, an Arduino board, wires, 3D printed parts made with PLA plastic on an Ender 3 V3 SE printer, screws, a breadboard, batteries, and two pieces of wood to hold the hand and electronics together. I designed the hand in Onshape including the fingers, mounts, and the palm. To print the parts I exported the models as STL files, and placed them on an SD card which I then transferred to the 3D printer.

Each finger was assembled with rotating joints and screws, as well as mounts for the servo motors.

I programmed the servos in Arduino IDE to move individual fingers or all fingers at once. To test the hand I first tried moving the fingers to check the speed strength. There was occasional jittering if moved a lot in a short amount of time quickly. I grabbed objects like water bottles, mugs, containers, toys, gaming controllers, and other every day objects.

When fingers didn’t move correctly or jittered, I adjusted the calibration or code.

Through this project, I learned how mechanics, electronics, and coding work together. I also had issues like limited motion, servo jitter, and servo heating under load which I had to solve. The hand could lift objects up to about 3 kg in total and grip many household items. The project shows that even a low cost, simple robotic hand can be educational, and be used in many applications.

During testing I recorded many forces in the fingers. When accounting for the tip of the fingers, each of them gave ~150g of force, so when all combined the fingers did ~1.2kg of force. The closer to the center of the hand, the finger will do more force while less force when further from the hand. Because of the thumb and its placement, it helps a lot in holding objects, allowing for objects up to 3 kilograms in weight. I tested on stress balls, plastic cups, water bottles, controllers and other at home items.

The hand was modeled using Onshape and 3D printed using PLA plastic and an Ender 3 V3 SE printer. The fingers were mechanically assembled so that they could bend at the joints, and an Arduino was programmed to rotate motors attached to the end of each finger to give a bend. All the fingers were put on a mount which would then be placed on the palm. The thumb didn't require mounting as it was already attached to the palm. The hand sits on a wooden stand to be more portable and less prone to having its wires pulled out. When moving fast the fingers did well and stayed in position, when moving slowly, due to its flexibility it would shake slightly.

To test their range of motion I rotated each finger to its maximum and minimum point, this would help me to manually align them to do basic calibration. Calibration is important and if not done correctly can lead to specific motors doing more work than others, making it possible that they break faster because of the weight. The load can also cause a lot of current through the motors, leading to them heating up. To fix this I placed resistors to limit the voltage and strength to keep it cooler while keeping decent strength.

I made a chart of the current in different states of the fingers by reading the voltage drop of the resistors that kept the motors from heating up. For example when they are still, they can draw barely any mill amps of current, when they are in the state of moving, they can show double the mill amps, when holding objects the current goes up much higher to 8* times more of mill amps of current, but the current peaks to its maximum when getting actively pushed back nearing an amp. Even with resistors the motors can still heat up. If holding an object, after ~30 minutes they can burn out, to maximize the time you would need software changes.

In conclusion, with good calibration the robotic hand can do really well, being able to hold lightweight house hold objects with good strength, around 3kg objects at maximum. There aren't many issues but there are problems where the finger would slightly shake if moving slowly, or if used for an extended period of time the current could build up and heat the motors.

The results from this project showed that low cost robotic hand can still be practical and useful even though their not as expensive. The design was made using basic 3D printed parts, small servo motors, and simple electronics, and was able to grab and move many everyday objects. This makes it viable for prosthetic hands where on average, dont need to lift very heavy objects.

One important conclusion from these results is that the design works better when gripping medium or larger objects. Smaller objects are harder to grip because the fingers did not have enough range of motion for more precise movements. Currently the hand has 5DOF but it would be much better to have around 11DOF to fix these issues. This means that while the hand works well, it could still be improved to make it more accurate.

From this project, I learned that robotic hands depend not only on strength, but more on finger movement, control, and overall design. I learnt a lot about 3D modeling, electronics, testing, and prototypes while making this project. Some parts of the project worked well, while I couldve done better and saved time on others. Overall, this project proves that a low cost robotic hand can be effective even with a low price, this could save thousands on prosthetics and help save money for new companys that are mass producing.

If I kept working on this project, I would try and make the hand much more robust. I would use stronger motors so each finger could produce at least 5 kg of grip force, change the material to improve grip, and make it better at holding smaller shaped objects by compacting the hand. I would use gears running through the palm so all the motors could be placed in the wrist eliminating the big gaps in between the fingers. My next steps would be redesigning the hand, testing stronger parts, and improving it for a more realistic prosthetic prototype.

While I did all the modeling, printing, and work my self, I would like to thank my parents for helping me with different parts of this project. My dad gave me advice when I was solving problems with servo heating and current, which helped me make the robotic hand work more reliably. He provided wood and glue that I used to build the stand and support structure to hold the hand and electronics in place. My mom helped with buying some of the important components for the project, including the Ender 3 V3 SE 3D printer which without I could not have completed this project. Their support made it possible for me to build, test, and improve my robotic hand more successfully.

Arduino. (n.d.). *Arduino Servo library*. https://www.arduino.cc/reference/en/libraries/servo/

Arduino. (n.d.). *Software*. https://www.arduino.cc/en/software/

Autodesk. (n.d.). *Fusion 360 for personal use*. https://www.autodesk.com/products/fusion-360/personal

Aviv Makes Robots. (2023, October 17). *Onshape tutorial for absolute beginners!* [Video]. YouTube. https://www.youtube.com/watch?v=2utLjjkXpIg

Castro, M. C. F., Pinheiro, W. C., & Rigolin, G. (2022). *A hybrid 3D printed hand prosthesis prototype based on sEMG and a fully embedded computer vision system*. *Frontiers in Neurorobotics, 15*. https://www.frontiersin.org/articles/10.3389/fnbot.2021.751282/full

Cogley, W. (2025, January 29). *Designing a new bionic hand (which prints pre-assembled)* [Video]. YouTube. https://www.youtube.com/watch?v=ILK5vl8Apew

Cuellar, J. S., Plettenburg, D., Zadpoor, A. A., Breedveld, P., & Smit, G. (2021). *Design of a 3D-printed hand prosthesis featuring articulated bio-inspired fingers*. *Proceedings of the Institution of Mechanical Engineers, Part H: Journal of Engineering in Medicine, 235*(3), 336–345. https://doi.org/10.1177/0954411920980889

Donaldson, W. (2017, July 22). *DIY robotic arm 3D printed (an initial prosthetic prototype)* [Video]. YouTube. https://www.youtube.com/watch?v=s0LA48Sw62k

HowToMechatronics. (2018, February 20). *Arduino robotic arm and Mecanum wheels robot* [Video]. YouTube. https://www.youtube.com/watch?v=VsYniCv5gDA

HowToMechatronics. (n.d.). *Arduino and servo motor tutorial*. https://howtomechatronics.com/tutorials/arduino/arduino-servo-motor-control-tutorial/

Onshape. (2014). *Onshape | Product development platform*. https://www.onshape.com/en/

Open Bionics. (n.d.). *Hero Arm: The world’s first medically certified 3D-printed bionic arm*. https://openbionics.com/hero-arm/

Price, T. (2019, November 23). *EXTREMELY simple way to sense servo current draw* [Video]. YouTube. https://www.youtube.com/watch?v=LKJLCJvyVdk

Product Design Online. (2022, July 2). *Day 1 of learn Fusion 360 in 30 days for complete beginners! - 2022 edition* [Video]. YouTube. https://www.youtube.com/watch?v=d3qGQ2utl2A

ResearchGate. (n.d.). *Figure 2: Hand measurements as a diagram*. https://www.researchgate.net/figure/Hand-measurements-as-a-diagram_fig1_264663607

Slant 3D. (2024, March 19). *Connect 3D printed parts | Design for mass production 3D printing* [Video]. YouTube. https://www.youtube.com/watch?v=djm5tCFn9S0

TooTallToby. (2024, April 5). *Onshape tutorial: Go from a 2D print to a 3D model using Onshape* [Video]. YouTube. https://www.youtube.com/watch?v=Y8t_1mkpuFo

TooTallToby. (2024, August 2). *Onshape tutorial - STEP by STEP - Latch hinge* [Video]. YouTube. https://www.youtube.com/watch?v=uH8UAAe98PA

TooTallToby. (2025, June 16). *Step-by-step Onshape tutorial (beginners/intermediate)* [Video]. YouTube. https://www.youtube.com/watch?v=gp7bK1fJdBQ

UltiMaker. (n.d.). *UltiMaker Cura*. https://ultimaker.com/software/ultimaker-cura/

Python Software Foundation. (2024). Python (Version 3.12.3) [Computer software]. https://www.python.org/

Google. (2019). MediaPipe (Version 0.10.x) [Computer software]. https://github.com/google/mediapipe

Microsoft. (2024). Visual Studio Code (Version 1.88) [Computer software]. https://code.visualstudio.com/

All other parts of this projects were made by Maxim Denisov