Inspiration
We wanted to create a system that shows how autonomous drones can make intelligent decisions in unpredictable environments. Inspired by real-world challenges like disaster response and aerial delivery, we aimed to simulate realistic drone missions with events such as weather hazards and flock interference.
What it does
KDKR simulates multiple drones navigating a 3D environment while reacting to random dynamic events. It uses optimization algorithms to find efficient flight paths based on distance, battery life, and mission goals, all visualized through an interactive dashboard.
How we built it
We used Python and OR-Tools for mission planning and route optimization, and Flask to power a live dashboard for monitoring simulations. The system integrates real distance matrices and mission data to visualize drone performance and decision-making in real time.
Challenges we ran into
Getting the path optimization to scale efficiently with multiple drones and missions was difficult. Integrating live simulation data with the Flask dashboard also required restructuring how we processed updates between the backend and frontend.
Accomplishments that we're proud of
We built a working autonomous drone simulation from scratch with dynamic routing and mission visualization. Our system handles multiple assets and adapts intelligently to randomized environmental conditions.
What we learned
We learned how to combine optimization, simulation, and visualization in one system. We also gained experience with real-world constraints like computational limits, live data updates, and visual feedback design.
What's next for KDKR
We plan to expand KDKR with reinforcement learning to improve decision-making and add 3D visualization for drone flight paths. Eventually, we want to make it a modular framework for testing AI-driven drone behaviors in realistic mission scenarios.
Log in or sign up for Devpost to join the conversation.