MirrorBallBot Uses Mirror to Expand Vision

Hackster News reports Andrea Favero created MirrorBallBot, a compact balance-bot that uses computer vision and an angled mirror to balance and actively chase a finger-mounted target. Hackster News describes the mirror as increasing the camera's effective viewing distance and expanding the field of view by roughly 60%, enabling a larger apparent platform without increasing the robot's physical footprint.

Per Hackster News, the build uses a Raspberry Pi Camera Module 3 Wide in the optical path and a Raspberry Pi 4 Model B as the central controller. The Pi handles vision, inverse kinematics calculations, and PID control while a 7-inch touchscreen provides the user interface. Motion control is split across three RP2040-Zero boards, each driving a stepper motor through TMC2209 drivers. The project implements a custom I2C packet protocol and a dedicated synchronization signal so the three motors start in lockstep, according to the article.

Using an angled mirror to create a virtual increase in camera distance is a low-cost optical trick that trades optical complexity for mechanical or enclosure size. Projects that need wider scene coverage but must remain compact often face trade-offs among sensor selection, optics cost, and system size. Optical folding with mirrors can be especially attractive for prototyping where distortion and alignment are manageable.

For practitioners, MirrorBallBot is a clear example of pragmatic system integration: commodity vision hardware, microcontroller-based motor control, and simple control algorithms (PID, inverse kinematics) are combined with a physical optics trick to solve a spatial-coverage problem. The build documents practical issues relevant to robotics hobbyists and educators, including synchronization of multiple motor drivers and a lightweight inter-board communications approach.

For builders, replication questions include mirror alignment tolerances, latency introduced by the vision pipeline, and how the custom I2C protocol handles error recovery and step synchronization under load. Observers interested in compact robot perception should watch for follow-up documentation or code that details calibration and timing choices.