This is a floor cleaning robot built to demonstrate SRI's electrostatic adhesion technology. In use, this robot wanders the floor and gathers debris by using the electrified adhesion roller and subsequently, deposits the particles in its tray.
For this project, I was tasked with developing a reliable, attractive, demonstration device that SRI could bring to conferences to display its technology in a familiar form factor. This necessitated a modular design that could be serviced without any tools. What I developed was a robot in which each subsystem connected to the main body like a cartridge, released by a quick release lever. Most parts were constructed with folded sheet metal to keep the cost low. I encapsulated it all in a 3D printed shell with features inspired by sports car bodies.
This is an attachment to a popular wheelchair model that enables mobility to those unable to propel a wheelchair. It is important for stroke victims to exercise frequently in the first few months of recovery. The complex motion required to propel a wheelchair is initially too difficult for people who have recently suffered from a stroke. This device enables the user to propel a wheelchair with a much simpler push-pull motion.
I designed the tubular frame and spec'd the components that interfaced with this device. The frame was shaped to match the motion paths of the hands of test subjects when asked to push and pull a handle in a comfortable manner. The vision of this project was to deploy a vast number of these into third world countries; therefore, a low cost prototype was critical to its success. I designed this frame to be made with just a couple manufacturing steps and maximized the utilization of as many off-the-shelf parts as possible.
This is a dishware scrubbing device whose contact surface dynamically conforms to the profile of a spectrum of bowls and plates, effectively cleaning each surface. A need was realized to develop a single end effector that could clean the surface of any ware that the robot might encounter (as opposed to a different cleaning head for each ware type). This greatly reduced the complexity of Dishcraft’s solution.
I was handed down the concept of continuous conformation via leaf springs and I took the design from there. Many months were spent testing different forms of this subsystem, comparing the prototypes side-by-side to determine the most effective form of the design. The final form of this device features a cast flexible urethane pad on top of a linkage, achieving three levels of compliance. First level of compliance is a spring loaded mechanical linkage, then an array of leaf springs, and lastly, a series of cantilevered wipers. After fully vetting the design, I brought it through low-volume production.
This is a multi-functional gripper made to handle an array of drinking glasses, plates, bowls, silverware, and napkins. Commercial robotic grippers proved ineffective at reliably grasping all items that could be found in a typical dishroom (especially when slippery). This device enables a robotic arm to achieve a more stable, reliable, grip around objects compared to a traditional pinch style gripper.
My contribution was the design of the flexible blue parts shown here. The biggest challenge here was creating a single continuous shape that could grasp a set of objects with a wide variety of shapes, weights, and surface properties. I converged on a design that enabled three different modes of grasping within the same package. First, is a notch cutout of the main body that runs most of the length of the finger. The gripper can nest these notches around the rim of plates and bowls for transport. Second, is a continuous grippy surface backed by a row of leaf springs for closing around cups of varying diameters. Lastly, there are three delicate fingers for pinching napkins and silverware.
This is a fluid system composed of a manifold and six nozzles. In order to ensure a consistent clean quality across all wares exiting Dishcraft’s robot, water must be delivered in controlled amounts to different areas around the cleaning volume during scrubbing. This necessitated a custom manifold to divert a single flow of water to six different channels with specific flow percentages between each channel.
I designed the manifold with the aid of CFD software. The tricky part here was crafting the interior of the manifold to partition the flow in specific amounts in a way that minimized head loss. The flowrate out the pump was very sensitive to changes in head, so dialing in the relative flow rates between the channels by varying orifice sizes was not an option. The solution came byway of branching the flow in multiple places sequentially down the manifold.
F.I.R.S.T. Robotics Competition is an international high school robotics competition. Each year teams of high school students work during a six week period to build robots to compete in a game that changes annually.
I was a member of team 971 - Spartan Robotics for four years, two of which I led as team captain. The foundation of my skills in engineering were built during my years in this organization. I was introduced to CAD, mechanical design, metal machining, and electronics. My team was consistently in the top three teams in California, and won most tournaments that we enrolled in.
WINNING TITLES:
2009 World Championship
2009 Silicon Valley Regional
2010 Silicon Valley Regional
2012 Davis Regional
2012 Silicon Valley Regional
