Settling on a project

We settled on this idea after a lot of group brainstorming. We wanted to work on something with significant mechanical, software, and electrical complexity to fulfill the learning goals of each team member. Collectively, we also wanted to create something that looked cool and/or involved complex biomechanical-style motions. This led to us exploring ideas at the intersection of art and robotics, as well as the intersection of biomechanics and robotics. One idea which generated a lot of excitement was creating a table tennis robot, but we quickly realized that this might be over-scoped for a six-week project. After some more brainstorming, we realized that an air hockey robot might be a reasonable simplification of a table tennis robot that we would still be excited to work on. While a table tennis player should be able to move their paddle in six degrees of freedom, air hockey only requires two degrees of freedom—translation in the width and length dimensions of the table. It is also much easier to predict the movement of an air hockey puck than a table tennis ball since the motion is not affected by spin. After discussing this with the teaching team and receiving positive feedback, we decided to settle on creating an air hockey robot for our project.

Scoping and Goals

Stretch Goal

Our goal for our air hockey robot was that it would be able to successfully play offense and defense against a beginner human player and beat them in a game of air hockey.

Minimum Viable Product (MVP)

Though our goal was to create a robot that could play air hockey both offensively and defensively, we needed to have a fall back option in case we ran into challenges, as almost always happens in engineering projects. We decided that having a robot that could play defense well in one axis would be a good milestone and would still represent a completed project.

Middle Ground (MVP+)

To bridge the gap between our minimum viable product and our stretch goal, we determined that we would add limited two axis motion to our minimum viable product, such that the robot could play defense well but also move in a second axis to hit a puck that had gotten stuck on the robot’s side of the board. This would not require the same speeds, control, and puck motion prediction in the second axis of motion that full offense would require.