2021SpringTeam3

Team Members

• Andres Gutiérrez – Computer Engineering Senior
• Louis Nicaud – Computer Science Engineering Senior
• Yusuf Patel – Mechanical Engineering Senior
• Yikai Huang – Mechanical Engineering Senior

Project Overview

In this project, students are tasked with building and programming a Waveshare Jetbot AI to navigate through a maze. The outcome from this project will allow us to build a robot that will find itself out an unknown area without driving manually. This is accomplished by using Robotic Operating System, SLAM Algorithm, and Jetson Nano.

  Figure 1: Jetbot, Jetson Nano, and Lidar

Must Haves

• Jetson Nano which is the main computer that lets you run multiple neural networks in parallel for applications.
• Waveshare Jetbot AI that is installed with motors and connected to Jetson Nano.
• RP Lidar A1 that uses a system that measures distance data in more than 2000 times' per second and with high resolution distance output, which also provides a 360 degree scan field.
• Detailed physical Maze

Nice to Haves

• Robot scans entire maze and provides a map of which direction to go through the maze.
• Maps out the maze autonomously.
• The implementation of Odometry which is the use of data from motion sensors to estimate change in position over time.

Project Video

(QUICK VIDEO OF PROJECT )

Hardware

  Figure 2: Team 3 Jetbot AI

Mechanical Design

The major components of the mechanical design include the baseplate.

Baseplate

Text about baseplate

(CAD)

  Figure 3: Baseplate CAD Design

Electrical and Hardware Design

The car's electrical assembly consists of three main components:

1. Jetson Nano – The single board computer (SBC) in charge of controlling the Jetbot AI.
2. RP Lidar A1 – A 360 degree 2D laser scanner.
3. DC Motors – Rotates the wheels to provide mobility.

  Figure 4: Car Wiring Diagram with Original Hardware
 

Maze Structure

Figure 5: Maze

Software

Github Repository: (https://github.com/gtierrezandres/maze_buster)

RP Lidar A1 Code

In the Link

Jetson Nano Code

ROS

ROS nodes, topics, etc.

Advice and Suggestions for the Future

Advice

• If using a bigger robot the maze needs to be much bigger.
• With more time, write a code to map the maze autonomously.

Future Suggestions for RP Lidar A1 Use

• The gap around the lidar should be larger then 40 cm
• Add comment

Acknowledgements and References

Acknowledgements

• Dominic and Haoru - Thank you for providing essential advise on how to use implement SLAM and ROS for our project, and also helping us find solutions for our hardware. .
• Professor Silberman and Professor de Oliveira - Thank you for providing a priceless learning opportunity!

References

1. [Link]
2. [link]