Smart Assistant Robot using Image Processing and Computer Vision
This Rover can use Computer Vision and Machine Learning models to dynamically follow humans around or find any 90 unique objects, pick them up and bring them to you. The rover truly is a personal assistant and its powered by the RoboClaw motor controller, the Raspberry Pi & Google Coral.
2025-06-28 16:35:05 - Adil Khan
Smart Assistant Robot using Image Processing and Computer Vision
Project Area of Specialization Artificial IntelligenceProject SummaryThis Rover can use Computer Vision and Machine Learning models to dynamically follow humans around or find any 90 unique objects, pick them up and bring them to you. The rover truly is a personal assistant and its powered by the RoboClaw motor controller, the Raspberry Pi & Google Coral.
Project Objectives- You can ask the robot to follow you around. (Done via the TensorFlow face detection computer vision algorithm)
- You can ask the robot to find you and then follow you. (Again this will use computer vision + ML)
- You can ask the robot to find any object (from this list) for you, pick it up using its robotic sled and bring it to where you are.
- You can give the robot a voice command to carry the collected item and follow you around like an assistant robot.
- Best part of all this is you can issue all these commands via speech as the robot has voice recognition.
This robot is truly special because it can use Machine Learning models to 'see' the world via a generic camera and perform tasks depending on how the detected object's position is changing in the camera.
This robot is built around the ever popular Raspberry pi, the incredibly powerful RoboClaw motor controller, and the common Rover 5 robot platform. Furthermore, all the additional physical parts are 3D printed. This robot also uses the Tensorflow USB Coral accelerator to speed up the Raspberry Pi's slow object detection. More on all this in the next few steps
The first part of this Instructables guide will cover some the theory behind how the algorithms and code works. The second half of this Instructables guide will cover how you can build this robot yourself including the physical building + installing libraries etc.
Benefits of the ProjectThis Rover can use Computer Vision and Machine Learning models to dynamically follow humans around or find any 90 unique objects, pick them up and bring them to you. The rover truly is a personal assistant and its powered by the RoboClaw motor controller, the Raspberry Pi & Google Coral.
Technical Details of Final DeliverableThis robot is built around the ever popular Raspberry pi, the incredibly powerful RoboClaw motor controller, and the common Rover 5 robot platform. Furthermore, all the additional physical parts are 3D printed. This robot also uses the Tensorflow USB Coral accelerator to speed up the Raspberry Pi's slow object detection. More on all this in the next few steps
Final Deliverable of the Project HW/SW integrated systemCore Industry SecurityOther Industries Education, IT, OthersCore Technology Artificial Intelligence(AI)Other Technologies RoboticsSustainable Development Goals Partnerships to achieve the GoalRequired Resources| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 78650 | |||
| raspberry pi | Equipment | 1 | 5000 | 5000 |
| Roboclaw 2x7A Motor Controller | Equipment | 2 | 12000 | 24000 |
| Google Coral USB accelerator | Equipment | 1 | 19500 | 19500 |
| USB Webcam | Equipment | 1 | 5000 | 5000 |
| Rover 5 platform | Equipment | 1 | 10000 | 10000 |
| 7A-12V battery | Equipment | 1 | 3500 | 3500 |
| Geared brushed DC motor | Equipment | 1 | 1000 | 1000 |
| Ultrasonic Distance sensor | Equipment | 1 | 150 | 150 |
| Servo SG90 | Equipment | 1 | 500 | 500 |
| 3D printed structural designing | Miscellaneous | 1 | 10000 | 10000 |