CHECK IT OUT: https://a360.co/30ipv3s
This is a 3 finger gripper based off of 4ndreas's Three Finger Gripper on Thingiverse. He based his off of Robotiq's gripper design.
Printing this in ABS with 25% infill should be strong enough. I am having mine printed on Treatstock right now for about $60.
I have redesigned the base of the gripper for use with stronger motors than the original. I scaled 4ndreas's original fingers to allow them to better fit the redesigned base. I also made the rotation easier, with 15 degrees of motion in either direction for the left and right fingers.
Each finger can move independently. The left and right fingers can also rotate inwards, creating a tweezer effect to pick up small objects. The original robot hand was quite strong, capable of lifting a cup of water, and crushing a standard plastic soda bottle if the cap was off. Mine will be even stronger since I'm using 12v dc motors (approx 9 kg/cm) for each finger and two 25 kg/cm servos for the tweezer effect.
This is part of a larger robot arm I'm designing. I will post those files as I finish them. In some of the pictures, you can see the (unfinished) wrist. The wrist is a parallel spherical manipulator modeled after the parallel spherical manipulator with coaxial inputs from Alaris NU. See: https://www.youtube.com/watch?v=soWiX1eqZq0
I designed the motor holders and base components in the design, but I copied and scaled the finger components from 4ndreas. Then I hollowed them out and made room for embedded electronics.
I am using capacitive touch sensors in conjunction with pressure-sensitive resistors as well as the encoders in the motors. With a capacitive touch sensor, you can sense through materials and gauge change in distance for a nearby object. The measurement depends on the distance of the material as well as the material's dielectric properties. Unless you know which material you are trying to interact with, the use of this sensor is somewhat limited...
That's where the pressure sensor comes into play. I will record the readings from the capacitive touch sensor. Then, as soon as I have confirmation from the pressure sensor, I know that the object is touching/ at the minimum distance possible. I also have the amount of time that it took to finally touch as well as the angular velocity of the finger. With this information, I can interpolate and know the distance as a function of time before the object made contact with the pressure sensor.
In short, I can also determine what dielectric property was necessary to produce those results and infer the material composition. The gripper can detect what material it is gripping and how hard it is gripping it.
Not only that but the same information from which I deduced the dielectric constant can be used to produce a bounding polygon for the object being gripped.
The gripper will learn to recognize the objects it grips by size, dielectric constant, and appearance**.
This will be a machine learning project with the brand new teensy 4 from PJRC.
**There will also be a Jevois camera mounted to the wrist, which is able to view the object being held by the gripper as well as the surrounding environment
Motor: Lheng 25GA DC Geared Motor with Encoder Speed Measuring Code High Power Large Torque Smart Car Dedicated 12V 500RPM
ZOSKAY Waterproof High Torque Metal Gear Standard Digital Servo 25KG/0.13S 6.8V for 1/8 1/10 RC Cars (Control Angle 270)
All screws needed will be M3's. The longest screws will be 30 mm and there are approximately 10 of those. There are probably another ten 20 mm screws and perhaps twenty smaller screws 5 - 10 mm long.
I will update the specifics later.
I'm still 3d printing the design. I'll update this once I build it.
I suggest using 3mm and 2 mm carbon fiber rods as the joints in the fingers.
USAQ 12 Piece Pure Carbon Fiber Rod 3 Pieces Each 2mm 3mm 4mm 5mm 500mm Length Lightweight Spar Support Assortment
DF9-40 Thin Film Resistance Force Pressure Sensor High Precise Sensitive Be D8C3
WMYCONGCONG 20 PCS TTP223B Digital Touch Capacitive Sensor Switch Module DIY for Arduino (20PCS)
Digital Touch Sensor Capacitive Switch Module for Arduino PIC DIY TTP223B US
total Quantity needed: 11
SlipToGrip Prescored Furniture Grippers
-Cut the foam out, cut the rubber to size, and screw it on the the gripper
JEVOIS-A33 SMART MACHINE VISION CAMERA - BEGINNER TURNKEY KIT
Teensy 4.0 Development Board