When wired correctly and powered using the recommended battery pack, the Pi Cam Car should be able to drive by itself. While driving, it can detect obstructions in front of it. This is done using the ultrasonic sensor placed at the top front of the car. When the obstruction is close enough, the car will avoid the obstruction by turning left or right, until there is no more obstruction in its way. The Pi Cam Car is able to detect faces within its viewing perspective, which is the perspective of the attached Pi Camera. This perspective can be observed on another computer or laptop through the Remote Desktop function installed on the Raspberry Pi. As such, we can have kind of a live stream of what the car actually “sees” and check whether it has correctly recognised faces. Once the car has detected a face at a sufficiently close distance, it will drive by itself towards the person and interact with the person through audio output. Pi Cam Car is able to play supported audio file through its attached loudspeaker in response to certain triggers. For example, after approaching the person whose face has been recognised, it will say hello or introduce itself. It can also play warning sound when facing with obstruction, and notification sound when obstruction is successfully avoided.
Below is the list of all documents and files needed to build the project, all can be found under the profect_files folder:
car_frame_design.pdf
code_package.zip
materials&tools_list.pdf
project_description.pdf
setup_instructions.pdf
Pi Cam Car by Thi Yen Thu Nguyen is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
For further information, please contact us via email: andreas.markwart@hsrw.org or thi.yen-thu.nguyen@hsrw.org