This is the "Stobot" - a robot designed and built by my team and me for the subject Mechatronics 2 at the University of Technology Sydney.
The Stobot is a mobile differential drive robot, with a Dobot Magician robot arm mounted on top. It has sensors that allow it to navigate around a maze/map using Monte Carlo Localisation - a particles filter algorithm, here implemented on MATLAB. It also has a camera to find the pre-determined object, using a trained neural network based on LeNet. Once the object has been found, a pose-estimation algorithm using computer vision will guide the robot arm to pick up the object.
Please leave a comment below if you have any questions or things to add! I am always looking forward to getting feedback!
- 12V 30RPM DC Motor with wheel, mount and bracket x2
- 5V DC Servo Motor
- Dual H Bridge L298N DC Stepper Motor Driver
- HC-05 Wireless Bluetooth RF Serial Transceiver (used during the initial testing phase)
- Time of Flight sensor VL53L0X
- Voltage Regulator
- Arduino Mega 2560
- Raspberry Pi Model 3 B+
- 11.1V RC LiPo Battery Pack x2
You can find it on my GitHub
Link to the documentation can be found here.
If you found the project interesting and want to know more about the details, especially in our implementation of Monte Carlo Localisation and Computer-Vision based Pose Estimation, you should read the full documentation linked above.
Thank you for reading, if you have any questions, feel free to leave it below!