PSO Implementation On Hardware

The project is based on artificial intelligence. In this project we are going to use AI based algorithm (Particle Swarm Optimization) PSO for finding the shortest path or best path for this project  four equiped roborts will be used the pso algorithm will be implemented on these robor

Project Title

PSO Implementation On Hardware

Project Area of Specialization

Artificial Intelligence

Project Summary

The project is based on artificial intelligence. In this project we are going to use AI based algorithm (Particle Swarm Optimization) PSO for finding the shortest path or best path for this project  four equiped roborts will be used the pso algorithm will be implemented on these roborts. The roborts will move without humman interataction and wiil find the shortest path according to the algorithm this project will also have real time analysis through web API that will show the speed postion direction distancce etc of the robort on the screen  

Project Objectives

• To improve the ability of robots to evaluate the shortest path
• To create an echo system for robots to communicate effectively and efficiently to solve complex problems
• Each robot in the group must reach the same target by following the best immediate neighbor to the target
• Each robot must avoid all obstacles including each robot and the wall of an arena

Project Implementation Method

We will build car robots that will act as particles, each of them will have the following components:

1. Raspberry Pi
2. Arduino
3. Li-ion Battery
4. Acrylic Boards
5. DC Motor
6. Proximity Sensor
7. Wires

Assembly:

         The robots will have the similar structure of a RC car (tiers and motor) on top of that we will mount an Acrylic board. There will be space for a battery as per need and distance. There will be proximity sensors mounted at front of the robot it will sense obstacles and reduce or stop accordingly. The battery will power Raspberry Pi and Arduino, the motors power will come through the Arduino so the AI can signal the Arduino and control the power to the dc motor.

Functionality:

             The robots will run on a path chosen by itself, the path will lead to the destination. All the functionalities will be controlled by the AI like speed, turns and choosing paths to the destination. In case of obstacle the sensor will check the range and input it in the AI the program will trigger the Arduino to minimize the current flowing to motor (powering the tires). Robot will be stopped or slowed and then it will evaluate the next moves. Robots will broadcast information to the monitoring station and relevant data to the other Robots on that pool to solve the quest. Robots will announce it position or distance covered since the starting point, each robot will compare it self to the other’s robot data if there is no one better it will be considered as global best and broadcast to the all others including the monitoring station.

If any one of the particle/robots arrives the destination it will broadcast the result. In the beginning, one prototype will be assembled. Test run will be conducted on it after satisfying results the other 3 will be assembled

Anaconda Python distribution will be used for the implementation of algorithm and Artificial Intelligence. The reason of using Python is that it is a high-level programming language. A high-level language offers higher level abstraction of machine language with a focus on usability over optimal program efficiency.

Once code is successfully compiled and trained on the simulations it be integrated in the Raspberry Pi.

The program will provide APIs through which we will receive the data and shown on the interface in Real-Time. The interface will show the following things:

1. Current Position of each robot.
2. Speed of each robot
3. Battery remaining
4. Global best
5. Local best
6. Distance covered since the beginning point.
7. Time since it left the starting point
8. Obstacle dodged

The interface will be Web built with HTML CSS and JavaScript and D3 a data visualizing Library.

Benefits of the Project

Mazes are a simplified form of navigation can be used as a form of navigation to achieve the shortest path for the situation like.  In factories to optimize the delivery system of robots.  Optimizing delivery systems in autonomous stores and shipyard.

Technical Details of Final Deliverable

In the end we would have four well equiped and tested roborts that could find the best solution among different solution and which will work according to AI based algorithm PSO and we will also have a Web API that will display the real time analysis 

Final Deliverable of the Project

HW/SW integrated system

Type of Industry

Food , Transportation

Technologies

Artificial Intelligence(AI), Robotics

Sustainable Development Goals

Good Health and Well-Being for People, Quality Education, Peace and Justice Strong Institutions

Required Resources

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