MP | Devpost

Inspiration

Autonomous flight systems require precision and clear logic to navigate physical spaces safely. We wanted to move beyond manual piloting to give a drone the intelligence to define its own path. The goal was to build a system that verifies its position and makes decisions to reach specific target points.

What it does

The project automates the flight path of a Parrot Minidrone within a simulated 3D space. It uses a control system to guide the quadcopter through a specific sequence of spatial coordinates. The drone continuously monitors its current location and automatically turns to reach the next waypoint.

How we built it

We constructed the core flight logic using MATLAB and the Simulink block environment. We implemented a Stateflow chart to act as the central brain that switches between target coordinates. We routed the drone position data back into the controller to close the loop and ensure accurate travel.

Challenges we ran into

Integrating our custom logic required exact matching of data types and array sizes within the command bus. We experienced several errors where the simulation rejected our control signals due to dimension mismatches. We also had to troubleshoot software rendering crashes caused by local graphics driver conflicts.

Accomplishments that we're proud of

What we learned

We realized how strict bus signals and data structures must be when designing aerospace control systems. We learned how to use Stateflow to build finite state machines for autonomous decision making. We discovered how to diagnose and bypass hardware acceleration issues to keep simulations stable.