Description

This project aims to leverage AI and machine learning technologies to enhance the design of realistic driving scenarios in a simulation environment utilizing the new UA driving simulator. By contributing their expertise in machine learning and AI, students will have the opportunity to develop advanced algorithms that generate dynamic and interactive scenarios, elevating the immersive experience of the driving simulator. They will gain valuable hands-on experience in AI model development and real-time system integration, while contributing to the advancement of AI-driven simulation technologies in the field of driving research.

Duration

Total project length: 175 hours

Task ideas

1. Data Collection and Preprocessing:
   • Gather a diverse dataset of real-world driving behaviors and scenarios.
   • Preprocess the collected data to remove noise, outliers, and ensure data quality.
2. AI Model Development:
   • Research and select appropriate machine learning algorithms for scenario generation.
   • Train AI models on the collected dataset to learn and mimic real-world driving behaviors.
   • Fine-tune the models to optimize performance and accuracy.
3. Scenario Generation:
   • Develop algorithms to generate diverse driving scenarios based on user-defined parameters.
   • Implement AI models to generate realistic traffic patterns, road conditions, and behaviors.
   • Ensure the scenarios adapt to user inputs and dynamically respond to changes in real-time.
4. Integration with Driving Simulator:
   • Integrate the AI-driven scenario generation algorithms with the existing driving simulator infrastructure.
   • Establish seamless communication between the AI models and the simulator for scenario enactment.
   • Test and refine the integration to ensure accurate representation of scenarios within the driving simulator.
5. User Interface and Customization:
   • Design a user-friendly interface for scenario creation, editing, and customization.
   • Implement features that allow users to modify specific aspects of the generated scenarios.
   • Enable real-time customization and interactivity using JavaScript.
6. Evaluation and Validation:
   • Develop metrics and evaluation criteria to assess the realism and effectiveness of the generated scenarios.
   • Conduct user testing and gather feedback to refine and improve the generated scenarios.
   • Validate the effectiveness of the AI-driven scenario design approach through comparative analysis with traditional methods.
7. Documentation and User Support:
   • Prepare comprehensive documentation on the usage and customization of the scenario design software.
   • Create user guides and tutorials to assist users in designing and implementing custom driving scenarios.
   • Provide technical support and assistance to users during the project duration and beyond.
8. Performance Optimization:
   • Identify and implement techniques to optimize the performance and efficiency of the AI models and scenario generation algorithms.
   • Explore methods to reduce computational requirements and improve real-time responsiveness.

Expected results

• Development of intelligent algorithms that can generate diverse and realistic driving scenarios in real-time.

Requirements

Proficiency in programming languages such as Python, Java, or C++, knowledge of machine learning frameworks, solid understanding of statistical analysis, data visualization, and data preprocessing.

Project difficulty level

Intermediate

Test

Please use this link to access the test for this project.

Mentors

• Dr. Despina Stavrinos (University of Alabama)
• Dr. Benjamin McManus (University of Alabama)
• Mr. Piyush Pawar (University of Alabama)

Please DO NOT contact mentors directly by email. Instead, please email human-ai@cern.ch with Project Title and include your CV and test results. The mentors will then get in touch with you.

Corresponding Project

• ISSR

Participating Organizations

• Alabama

---