Welcome Back to My Research Journey: Introducing ICON
Anshul B -
Hello everyone! Welcome back to my blog! My name is Anshul Baddi, and I am excited to share an update on my research journey. If you read my last post, you might remember that my previous project was about ethical hacking an AI-based cybersecurity model. However, I have since transitioned into my new research initiative, the Intelligent Classification and Optimized Navigator (ICON).
What is ICON?
Imagine a car that sees the world as we do, understands its surroundings through the power of AI, and navigates flawlessly without the need for complex mapping of GPS. That’s ICON—Intelligent Classification and Optimized Navigator—a revolutionary project pushing the boundaries of autonomous driving by teaching a car to drive solely based on image classification. Our research will answer these key questions:
- How can AI-driven image classification enable real-time autonomous navigation without GPS?
- What are the biggest limitations to training an AI car solely based on images?
I will work with Professor Ryan Woodward at GCU to develop and test ICON. Our main goal is to create autonomous driving that solely relies on image processing. We will achieve this by using a convolutional neural network, a type of deep learning model. Our main experiment location will be the GCU campus, where we will test ICON’s real-time navigation.
Challenges
- Lighting: Changes in lighting conditions, weather, and obstacles could affect ICON’s accuracy.
- Computational Limitations: Since we are working with a small-scale model powered by a Raspberry Pi, we won’t have much computational power.
- Data Collection: Gathering a diverse dataset while ensuring ethical usage and privacy compliance will be a difficulty.
What’s Next?
In the upcoming weeks, I will be involved in creating the actual machine learning model that ICON runs on, conducting performance tests, and comparing it to GPS-based autonomous systems. I will make sure to document my findings and challenges, so stay tuned for updates!
Comments:
All viewpoints are welcome but profane, threatening, disrespectful, or harassing comments will not be tolerated and are subject to moderation up to, and including, full deletion.