VOICE AND IMAGE FEEDBACK CONTROL ROBOTIC ARM
Robotics always remains a major interesting field in the scientific community. Robots are able to perform different tasks. There are many different types of robotic arms made for different purposes having different applications. Robotic arm reported in this paper will be able to grab the object acco
2025-06-28 16:36:40 - Adil Khan
VOICE AND IMAGE FEEDBACK CONTROL ROBOTIC ARM
Project Area of Specialization RoboticsProject SummaryRobotics always remains a major interesting field in the scientific community. Robots are able to perform different tasks. There are many different types of robotic arms made for different purposes having different applications. Robotic arm reported in this paper will be able to grab the object according to the given instructions/commands. The robotic arm will be given command through voice and robotic arm will perform the given task as per given instructions. It will be using image processing techniques to identify the object, grab it according to its strength without damaging and place it on a predefined location. Grabbing mechanism of this robotic arm will be based upon the nature of object.
For example, it would grab an egg with less force and grab an iron ball with more force since surface sensitivity is different for both objects. As this robotic arm will grab the object according to its strength, hence it has wide applications in many field like heavy industry to precise medical procedures.
Project ObjectivesCompare to other robotic arms our main focus and motive is to grab object according to the specifications and sensitivity of the objects.
The project is “VOICE AND IMAGE FEEDBACK CONTROL ROBOTIC ARM”.As the name represents, arm will interact with the surrounding through voice command, image processing and will grab the object according to its strength and sensitivity
- Project Features
- 6 degree of freedom (6 motors)
- Voice service.
- Object detection (Image processing)
- Grab object by specification and sensitivity.
The main controller guides all the movements of the robotic arm which acts as the brain of the system, and various sensors are to be embedded to acquire data from the environment. The most important task of the arm is to gripe and pick the object.
It is provided with the voice service , image processing. Information for different sensors, controller and drivers will be provided through camera, pressure sensor, mic etc. The robotic arm performs movements controlled by feedback mechanism.
The required hardware is:
- Raspberry Pi 3B & Arduino UNO (Microcontrollers)
- Force Sensitive Resistor (Pressure sensor)
- PCA 9685 driver.
- 6 Analog DC servo motors ( PRO tower 950 )
- Microphones
- Camera
- LCD for monitoring purpose
- Couple of power supplies
- Couple of jumper wires.
The required software is:
- Python
- Tensor flow
- Open CV
- Google assistant
- Arduino IDE
- Raspbian
There are many robotic arms made till now for different purposes. But they are manufactured for either heavy loading purposes or very specific purposes. Also many of them are manually controllable. Mostly robotic arms are not efficient in picking sensitive objects.
Our motive is to grab the object depending on the sensitivity and other physical specification.
Robotic arm with similar purpose of sensitive object picking is publicized by LEGO as Pro-Challenge. The arm is based on the feedback mechanism.
The Lego Pro-Challenge is too costly.Secondly, our project will be cost efficient.
Technical Details of Final DeliverableGoogle assistant take Voice signal from microphone and depending on the voice commands it will enable/disable the raspberry pi GPIOs.These GPIOs are the input for the Arduino and raspberry pi itself (for activating the tensor flow).Google assistant outputs 5 pins , 3 pins for Arduino and 2 pins for raspberry pi.
As the input is received by Arduino from pi, it will run the motors in a predefine manner depending on the voice command. Simultaneously tensor Flow is activated to detect the image from the live camera frames. Once the object is detected and it will provide a signal to check whether the detected object matches the given voice command or not. If this voice signal and detected image are synchronized, the tensor flow will give signal to Arduino to pickup the object according to the code. The gripper motor will take feedback from the pressure sensor to pick the object according to the sensitivity. Furthermore, PCA driver is used to drive the motors , using Arduino as it controller.
Signals and feedbacks for controllers for specific task are taken by the following transducers:
- Google assistant takes feedback by microphone.
- Tensor flow takes feedback by camera
- gripper motor is controlled by taking feedback through resistive pressure sensors.
| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 30500 | |||
| Raspberry Pi 3B | Equipment | 6000 | 1 | 6000 |
| Arduino UNO | Equipment | 650 | 1 | 650 |
| Force sensitive resistor | Equipment | 1200 | 1 | 1200 |
| PCA 9685 driver. | Equipment | 300 | 1 | 300 |
| robotic arm + 6 Analog DC servo motors ( PRO tower 950 ) | Equipment | 8000 | 1 | 8000 |
| Microphones | Equipment | 100 | 1 | 100 |
| pi Camera | Equipment | 1200 | 1 | 1200 |
| LCD for monitoring purpose | Equipment | 5000 | 1 | 5000 |
| Couple of power supplies | Equipment | 600 | 1 | 600 |
| Couple jumper wires. | Equipment | 250 | 1 | 250 |
| arduino nano | Miscellaneous | 450 | 2 | 900 |
| testing components | Miscellaneous | 300 | 1 | 300 |
| acrylic sheet | Miscellaneous | 2000 | 1 | 2000 |
| spacer | Miscellaneous | 2000 | 1 | 2000 |
| other | Miscellaneous | 2000 | 1 | 2000 |