In this edition of our posts about the Jetson Community, we’re featuring two projects and their developers who won the NVIDIA Jetson Project of the Month award…
Overview
This article highlights two innovative projects recognized as NVIDIA Jetson Projects of the Month: an AI fever screening thermometer by Tomasz Lewicki and a self-navigating robot for search and rescue by Maggie Xu. Both projects utilize the Jetson Nano platform for advanced video and sensor data analysis in robotics applications.
What You'll Learn
1
How to build an AI thermometer using Jetson Nano
2
Why using thermal and RGB cameras enhances fever detection
3
How to implement autonomous navigation in a robot using ROS
4
When to use LIDAR for object detection in robotics
Prerequisites & Requirements
- Basic understanding of robotics and AI concepts
- Familiarity with ROS and Jetson Nano
Key Questions Answered
How does the AI thermometer measure skin temperature?
The AI thermometer uses an RGB camera for detecting people and a thermal camera for measuring skin temperature. It processes this data in real-time using the SSD object detection algorithm on the Jetson Nano, providing a visual output that includes both RGB and thermal views.
What technologies are used in the self-navigating robot?
The self-navigating robot utilizes ROS Melodic for its operating system, a USB camera for visual input, and LIDAR for scanning the environment. It employs the gmapping and move_base packages for mapping and navigation, respectively.
What is the purpose of the AI thermometer in public spaces?
The AI thermometer serves as an efficient screening tool for detecting fever, a common symptom of Covid-19 and other viral illnesses, particularly when placed at the entrances of shared buildings.
How does the robot retrieve objects autonomously?
The robot autonomously navigates to a specified location, detects an object using OpenCV, and retrieves it using a robotic arm controlled by an Arduino board. This process is facilitated by ROS's navigation capabilities.
Key Statistics & Figures
FPS of object detection
9 FPS
This is the performance of the SSD object detection algorithm running on Jetson Nano.
Technologies & Tools
Hardware
Jetson Nano
Used as the processing unit for both the AI thermometer and the self-navigating robot.
Software
Ros Melodic
Framework used for building the self-navigating robot's navigation and mapping functionalities.
Software
Opencv
Used for object detection in the self-navigating robot.
Hardware
Lidar
Employed for scanning the environment and detecting objects.
Key Actionable Insights
1Developers can leverage the open-source design of the AI thermometer to create their own fever screening solutions.This project not only provides a practical application for health monitoring but also serves as a learning tool for those interested in AI and sensor integration.
2Utilizing LIDAR in robotics enhances object detection and navigation accuracy.In environments where precise mapping is crucial, LIDAR provides detailed spatial information, making it an essential tool for autonomous robots.
3Integrating multiple camera types can improve data richness in AI applications.By combining RGB and thermal imaging, developers can create more robust systems capable of handling diverse real-world scenarios.
Common Pitfalls
1
Many developers may overlook the importance of combining different sensor types for enhanced data accuracy.
Relying solely on one type of sensor can lead to incomplete data and ineffective solutions, especially in complex environments.
Related Concepts
AI And Machine Learning Applications In Robotics
Sensor Fusion Techniques
Autonomous Navigation Systems