In the event you’ve put in all the required follow to study bike methods, you’d most likely like an appropriately dramatic soundtrack to accompany your stunts. A workforce of scholars engaged on a capstone mission on the College of Washington took this pure need a step additional with the Music Bike, a system that generates adaptive music in response to the bike’s movement.
The Music Bike has a set of sensors managed by an ESP32-S3 mounted beneath the bike seat. The ESP32 transmits the info it collects over BLE to an Android app, which in flip makes use of the FMOD Studio adaptive sound engine to generate the music performed. An MPU9250 IMU collects most place and movement knowledge, supplemented by a corridor impact sensor which tracks wheel velocity and path of rotation.
When the Android app receives sensor knowledge, it performs some processing to detect the bike’s actions, then makes use of these to regulate FMOD’s output. The scholars tried utilizing machine studying to detect bike methods, however had hassle with latency and accuracy, in order that they switched to a threshold classifier. They have been ultimately in a position to detect jumps, 180-degree spins, ahead and reverse movement, and wheelies. FMOD makes use of this data to switch music pitch, alter instrument layering, and alter the observe. The scholars gave a formidable in-class demonstration of the system within the video under (the demonstration begins at 4:30).
Surprisingly sufficient, this isn’t the primary music-producing bike we’ve featured right here. We’ve additionally seen a music-reactive bike lighting system.
Because of [Blake Hannaford] for the tip!
