Quantum Self-Driving Neural Networks
Challenge Background
Self-Driving cars like the ones we see in Tesla work with complex deep neural networks for predicting various parameters dependent on the situation. One example a self-driving autopilot mode must watch out for is controlling the car's speed relative to other cars, the steering angle, and thousands of other variables that must be trained. Tesla and many other car companies are perfecting the autopilot technology by training complex deep neural networks.
The Problem
A new avenue of research currently being explored is quantum-classical hybrid neural networks which can be more optimized and efficient. This project entails constructing a neural network that predicts steering angles for cars and testing it on a simulator, then implementing an additional layer, the “quantum” layer, and comparing if it is better in predicting better steering angles.
Goal of the Project
- CNN with transfer learning capabilities for the quantum-classical Neural Network.
- Test both models on a self-driving car simulator.
- Deploy the simulator along with the models on a web application.
Project Timeline
- Organize the team and describe the orientation of the project.
- Familiarize yourself with quantum circuits and how neural networks can play a role.
- Train students in quantum neural networks and how ML can be accelerated using quantum computing.
- Start setting up the dataset of images and steering angles
- Start setting up the dataset of images and car velocities
- Decide on machine learning frameworks
- Start implementing neural networks based on papers
- Set up a self driving car simulator test like AirSim, train all neural networks
Build the quantum neural network and evaluate performance with classical neural networks
Hyper parameter and fine tuning for neural networks for steering angle, velocity, and combined.
Quantum neural newtork implementation should be perfected by this time
Web application development into Streamlit with models and testing formats
Finalize models and final deployment of models
What you'll learn
First Omdena Local Chapter Project?
Beginner-friendly, but also welcomes experts
Education-focused
Duration: 4 to 8 weeks
Open-source
Your Benefits
Address a significant real-world problem with your skills
Build your project portfolio
Access paid projects (as an Omdena Top Talent)
Get hired at top organizations
Requirements
Good English
Suitable for AI/ Data Science beginners but also more senior collaborators
Learning mindset
Application Form
This Challenge is hosted by:
Become an Omdena Collaborator
