Autonomous VIO-based Quadcopter

This project integrates Visual Inertial Odometry (VIO) with trajectory planning and control to achieve fully autonomous quadcopter flight. The key challenge: making the drone navigate using only its own state estimates instead of ground-truth data, as would be required in any real-world deployment.

We developed and tested everything in the flightsim environment, creating a closed-loop system that flies based purely on onboard sensing. The pipeline consists of:

• A VIO module providing real-time state estimates (position, velocity, orientation) from fused camera and IMU data.
• An A*-based motion planner generating collision-free paths through the environment.
• A trajectory controller that tracks smooth reference trajectories from the planner using the VIO state feedback.

Minor hardware tests were also conducted as proof of concept, validating that the simulation pipeline transfers to a physical platform. This brings us one step closer to real-world autonomous flight.