Voice Controlled Car
UC Berkeley, 2020, in collaboration with Matthew Ye and Tony Kim
Description.
For EE 16B class, our semester-long project employed linear algebra and circuit principles to build a robot car that could respond to a constrained set of voice commands and perform actions based on what was spoken. A microphone collected the audio, which was processed by the on-car low-power Ti Launchpad microcontroller board. Once the audio signal was filtered, it was then classified as one of 4 words; the selected word determined the car's action. Certain control principles, such as close-loop feedback, SVD, and PCA were employed to ensure steady-state error correction and smooth performance.