Having a pet can really make a difference to your happiness at the end of the day, but they’re also a lot of work. This project by [Ioannis Stoltidis] does something similar — minus all the responsibility. The Smart Car Follower Project is designed to track people using Bluetooth and IR and follow them around from room to room.
Submitted as part of a Master’s thesis, this project hacks a toy car and uses a key chain transmitter that sends the tracking signals. A Raspberry Pi 3 combines the Bluetooth RSSI and IR signals to make create an estimate of the position of the beacon. Arduinos facilitate the IR signaling as well as the motor control allowing the robot to chase the user around like a puppy. The whole thing also comes with obstacle avoidance using ultrasonic sensors on all sides which are good if you have a lot of furniture in the house.
You can also choose to go manual-mode and drive it around the block using a PC and gamepad. A webcam connected to the onboard computer allows a first person view of the vehicle by sending the video feed over wifi to a PC application. OpenCV is used to create the final GUI which allows you to see and control the project remotely. The source code is available for download for anyone who wants to replicate the project. Check out the video of it in action below.
The use of OpenCV means that the design can be reused to make a robot with vision later on. The Raspberry Pi has been proven to be an able platform for image processing however for instances where you need that extra muscle, this project may be a good starting point. For more ambitious uses of OpenCV, check out this Laser Firing Robot.