MACHINE LEARNING BASED AUTONOMOUS ROBOT

Robotics is the branch of technology that deals with the design, construction, operation and applications of robots and computer system for their control, sensory feedback, and information processing. These technologies deal with automated machines that can replace humans, perform dangerous manufact

Project Title

MACHINE LEARNING BASED AUTONOMOUS ROBOT

Project Area of Specialization

Artificial Intelligence

Project Summary

Robotics is the branch of technology that deals with the design, construction, operation and applications of robots and computer system for their control, sensory feedback, and information processing. These technologies deal with automated machines that can replace humans, perform dangerous manufacturing processes, or simply just resemble humans. In the years to come artificial intelligence will take over a major chunk of numerous domains that were previously controlled by human, time-consuming labor. An AI-powered smart robot is to be developed that can recognize obstacles, and avoid them and reach a particular destination. This idea is based on machine learning and is essentially a robot that can make decisions on its own. Some collisions and a specific destination will be taught to the robot. It will be efficient enough that if left at a specific location, it will be efficient enough to reach the specified destination while avoiding any collisions. If an obstruction is spotted, the robot will make 900 to 3600 turns and adjust its path to arrive at its destination. The camera mounted on the robot's front side will detect impediments and assist the robot in avoiding them along its path. It run a program that allows it to choose which action to do depending on their environment. All of the components are connected to the Jetson NANO Development kit, which governs these operations.

Comparing it to the previous collision avoidance projects that were based on sensors it is very much smart and is not dependent of any path. It can avoid any collision coming in its way and will reach its particular destination without following a certain path. More than one robot can be operated in a certain environment whereas sensors based project cannot be operated in a certain environment as these projects are dependent of the path. If we operate more than one robot in a certain environment it will not collide with any of the other robot operating there earlier.

Project Objectives

The objective of the project is to design a robot that can

• Detect and avoid collisions using AI
• Reach a particular destination

Project Implementation Method

This project is an AI robot that has two key features of avoiding collisions and reaching a particular destination without following a certain path. The main component of the project is the Jetson Nano developer kit (variant 4gb) that works as a mini computer. Jetson Nano is selected because it has strong and valid infrastructure that allow the execution of diverse deep learning based trials and the collection of useful, meaningful data concerning real-world objects and fields. The robot is a two-wheeled car-shaped system that mounts the Jetson Nano board and other peripheral parts. An 8MP 160O FOV (Field of view) wide angle camera is mounted on the front for capturing the obstacle and destination data. A 0.91inch LED is mounted on the Jetson Nano board for showing some basic information. On the bottoms wheels are attached which are operated by motor drivers. A battery of 12v is installed to run all the components of the robot. Antennas of frequency 2.4GHz/5GHz are installed to catch the Wi-Fi or Bluetooth signals. An SD card of 64 GB memory to install and run all the software part of the project. The OS installed on the Jetson Nano is Ubuntu 18.04 LTS through which we are connecting it to laptop and on which all the software portion is executed.

The project implementation includes various steps. The first step includes assembling all the hardware and checking it whether it is working properly or not. After assembling the robot, the robot will be trained to have basic motions of moving in all four directions. The camera mounted on the front will be brought to work for gather the data of its surrounding (obstacles and destination). With the help of the camera the picture data of all the collisions on which we want to train the robot will be given to the robot. The robot will detect and avoid these collisions when run in real time. The robot will set free in a particular area where we will give it some collisions on which the robot is trained. It will all the collisions coming in its way in real time. The robot will be given some picture data of the destination. After executing all the software data, we will send it for trail. When the robot is trained completely. The obstacle and destination picture data has been given to the robot. Now we'll have some final trails in a specific area, and the robot will navigate to its set destination while avoiding all collisions

Benefits of the Project

This project Machine Learning Based Autonomous Robot has a vast range of benefits. It can be operated in hazardous environment where human presence can be dangerous. It can assist handicap persons for their daily life activities. It can also be used in a restaurant to make it automated and can be used to deliver food and stuff. As the entire world is moving towards automation it can assist the community to automate. This idea can be used to design self-driving cars.

Technical Details of Final Deliverable

This is a prototype on the idea of which we can construct real time working robots. The final deliverable is a mini robot comprising of various components. The main component is the jetson Nano developer kit (variant 4gb) that works as a mini computer. An 8MP 160O FOV (Field of view) wide angle camera is mounted on the front for capturing the obstacle and destination data. A 0.91inch LED is mounted on the circuit board for showing some basic information. On the bottoms wheels are attached which are operated by motor drivers. A battery of 12v is installed to run all the components of the robot. Antennas of frequency 2.4GHz/5GHz are installed to catch the Wi-Fi or Bluetooth signals.

Final Deliverable of the Project

HW/SW integrated system

Core Industry

Others

Other Industries

Telecommunication

Core Technology

Robotics

Other Technologies

Artificial Intelligence(AI)

Sustainable Development Goals

Good Health and Well-Being for People, Decent Work and Economic Growth, Industry, Innovation and Infrastructure

Required Resources

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