AI Based Environment Learning Robot

Designing a stable and accurately moving robot that provides an efficient and reliable system. ? Less expensive, more accurate can be used for security and surveillance purposes. ? Can navigate in a user-defined environment and develop a 3D map of its path and the surroundings.

Project Title

AI Based Environment Learning Robot

Project Area of Specialization

Robotics

Project Summary

Designing a stable and accurately moving robot that provides an efficient and reliable system.

• Less expensive, more accurate can be used for security and surveillance purposes.

• Can navigate in a user-defined environment and develop a 3D map of its path and the surroundings.

 The robot will be able to:

• Capturing images through cameras

• 3-D Mapping using Stereo Vision

• Path Finding

• Detecting abnormalities in the environment using sensors

• Controlling Speed using Motors

Project Objectives

Designing an  Artificial Intelligence-based environment learning robot to be placed in the user-defined environment to do the tasks:

• Move around in a user-defined environment while avoiding all obstacles.
• Determine a complete obstacles-free path.
• Creating a 3-D map of the environment.
• Sensing the abnormality in the environment via sensors.
• Share information with the user of Threat/No Threat status via alarm/ light. 

Project Implementation Method

1. Using Stereo-vision Technique, to reconstruct the environment in computer memory. • 3D-Image-Reconstruction Used to plan the path for the robot. • Image Recognition: Used to recognize all objects and human beings in the environment and differentiate between friendly and unfriendly objects and human beings.

2. Autonomous Robot After the implementation of the stereo vision system, our next take will base on making it self-intelligent. So that it will be able to detect obstacles around and create the best path for itself without our instructions. This is done by the analysis of the 3-D image, which helps in getting mathematical information of the coordinates, and then by applying various programming techniques to it using programming languages, and strategies, a complete coded system is established. The computer (Jetson Nano) collects data from cameras and saves it in memory for processing. Object Detection Use to detect all objects and human beings in the environment.

3. Obstacle Avoidance is done by using stereo vision. A depth processing is done on Jetson Nano Images captured from the stereo cameras are rectified using the sum of absolute differences in order to produce a disparity image. The disparity image is then divided into 3 regions left, middle, and right (aligned horizontally and vertically) and after analysis, decisions are made for object detection in the specific regions. Motors are used to control the caterpillar wheels of the robots to avoid obstacles that are controlled by PWM.

4. Sensing System Use to inform central server in real-time of Threat/No Threat status Construction and Working It consists of the following sensor: • Flame Sensor: (used to detect the presence of a flame or fire in its surroundings). • Smoke/Gas Sensor:(used to detect the presence of smoke in its surroundings).

5. Self-Charging Use to inform central server in real-time of Threat/No Threat status Using Wireless Charger: Whenever the battery of the robot is low. By using a power bank as a battery for our robot, we will use the wireless receiver/connector (charging element) of the robot for charging. A wireless charger is used, on which whenever the receiver will contact, the power bank will start to charge the robot.

6. Code Implementation For the code implementation, the language we are using is Python, which is an interpreted, object-oriented, high-level programming language designed to be easy to read and simple to implement in automation, artificial intelligence, creating websites, and many other applications. The details are • Python Interpreter: A program used to execute python code. • Python Version to Install: Python 3.7.2, Extension: .py. • Code Editor (Used to write code): PyCharm (Integrated development environment).

Benefits of the Project

Several real-life problems which enforce the demand of the project are

• Inaccessible Sites: During exploration like in archeological research, there are several locations that a human being cannot access due to which many of the ancient places are left uninvestigated.

• Human Hazard: In emergency assistance or surveillance, it becomes a dangerous and life-risking task for a human being to analyze the situation and work in accordance.

• Man Power: In order to collect information of a user-defined environment for monitoring the situation and then reacting to that situation, it becomes a time and energy-consuming task for a human being.

• Security Issue: Security of specified things in the surroundings.

• Invention: As technology is growing, autonomous systems are becoming more frequent due to which deep networks are required to train machines for certain situations. So, it has become necessary to design a system that helps humanity and serves the field of science.

Technical Details of Final Deliverable

• The Raspberry Pi 4 Model B tech specs are;
  • 1.5GHz 64-bit quad-core CPU.
  • 2GB RAM - LPDDR4
  • 2.4 GHz and 5.0 GHz IEEE 802.11ac wireless, Bluetooth 5.0, BLE.
  • Gigabit Ethernet.
  • 2x USB 3.0 ports; 2x USB 2.0 ports.
  • Raspberry Pi standard 40 pin GPIO header.

Sensors: Perfectly compatible with Arduino

Weight of the system: Arduino Uno 25g + Raspberry Pi 66g + Sensors 25g + Cells 300g

Arduino Uno to Raspberry Pi • Connected via Bluetooth• 

Cameras Specifications • Pioneer web cam • 640×480 resolution • 2.0 USB support 30Fps

Final Deliverable of the Project

Hardware System

Core Industry

Security

Other Industries

Manufacturing

Core Technology

Robotics

Other Technologies

Artificial Intelligence(AI)

Sustainable Development Goals

Industry, Innovation and Infrastructure

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com