The problem EthicaDrive solves

EthicaDrive effectively addresses the following challenges:

• Simulation Complexity: Provides a simplified 2D simulation environment for self-driving car testing, reducing computational requirements.
• Map Customization: Utilizes OpenStreetMap API for customizable map scenarios, enhancing testing flexibility.
• Autonomous Control: Employs feed-forward neural networks for efficient vehicle control, demonstrating potential for real-world applications.
• Visualization: Offers MiniMap and bird's eye view for improved situational awareness and user experience.

Challenges I ran into

During the development of EthicaDrive, the following challenges were encountered:

• Integration Complexity: Combining TensorFlow, PyTorch, and OpenStreetMap API presented technical integration hurdles.
• Neural Network Optimization: Training and optimizing feed-forward neural networks for vehicle control required significant computational resources and tuning efforts.
• Simulation Realism: Balancing simulation simplicity with realism to accurately model real-world driving scenarios.
• Scalability: Managing the trade-off between simulation complexity and performance to ensure smooth operation.