This BattleBots-Inspired Spinning Robot Claw Swaps a Trophy for a Coffee Cup

Pseudonymous maker "drpykachu" has created a do-it-yourself version of the famous BattleBots claw-shaped trophy stand — to slowly spin a coffee cup on a desk.

"Growing up, my father and I watched BattleBots quite often. During the fights where robots were being shredded to pieces, I noticed there was a giant, robotic, claw trophy stand in the background that I thought was more impressive than the bots themselves," drpykachu recalls. "I wanted one for myself, but after looking into it some more, I realized the stand is massive, heavy, and expensive. I decided that I could attempt to make one myself, but a smaller, cheaper desktop version to rotate my coffee mug (because why not!)."

The trophy wasn't anything particularly special — an oversized, polished nut engraved with the show's logo and the word "champion" — but the way it was displayed was undeniably eye-catching: the trophy stood on a plastic circular plate that was held aloft by a five-clawed robot arm. The arm's claws would slowly spin the plate to rotate the trophy, "walking" along the bottom of the plate in a fluid motion while keeping the trophy perfectly balanced.

The replica designed by drpykachu is considerably more compact, and designed for desktop rather than stadium use. Like its full-size inspiration, it has five jointed "fingers" that end in claws designed to hold a flat, circular plate aloft. Where the original trophy would have stood, though, is something a little smaller and at the same time considerably more practical: a coffee cup, rotated via a controlling Raspberry Pi running Python — with a camera-based vision system to ensure the mug remains balanced.

"Each finger in the claw contains three motors," the maker explains. "We need to perform inverse kinematics to find the angles needed to position the finger tip at a given 3D coordinate. There are several classes of motor that were considered, but servo motors are chosen as the range of movement (<180°), built-in rotary encoder (for position tracking and tuning with PID), and step accuracy (the smaller the better) are ideal. The translation from theoretical model, to CAD model, to 3D printing and fine-tuning took a considerable amount of time and effort."

The project is documented in considerable detail on GitHub, where you can also find the source code and 3D print files under an unspecified license; additional discussion can be found on Reddit.