Redesigning Robotic Gripper for Teleoperation
Though promising, the world of Robot-assisted minimally invasive surgery (RMIS) has much room for improvements in accuracy and dexterity of a surgeon in order to minimize trauma to patients. Significant clinical success with RMIS has been limited. The lack of haptic (force and tactile) feedback presented to the surgeon is a limiting factor.
My focus in the Stanford's Collaborative Haptics and Robotics in Medicine (CHARM) lab was to redesign the existing Phantom Omni Master Grip interface--a non-intuitve two-button open and close interface.
I explored existing commercial and academic RMIS systems such as the various versions of Intuitive Surgical's DaVinci (and seeing the latest design), various robotic designs at UCSC to understand existing solutions and device designs. Beyond basic functionality in design and housing the electronics, I aimed to achieve an efficient, lightweight, frictionless design that appeared more streamlined and ergonomic to its user than most other robotic grippers currently existing in the market. Pictured above are a few of my concept sketches.
Mechanical Design Iteration
In my first iteration, efficiency took precedence, thus my initial design would involve one 3D printed capstan press-fit into a Maxon backdrivable motor with two levers attached on top and bottom of the capstan. Eventually, the nine iterations thereafter involved two 3D-printed capstans and six flush pieces.
In order to decrease user fatigue, I designed a two finger platform instead of one for one lever in addition to extending a handle unlike many existing robotic grippers, which neglect handles or a platform for both the index and middle finger
I was specifically responsible for the CAD drawings seen below. This is the 14th and final iteration.
My team decided to provide open-source content of the hardware designs: you can see the SolidWorks files here.