Abstract: 

This thesis explores motion planning and control strategies for enabling rapid quadrotor navigation in unknown environments using limited field-of-view depth sensors. We propose real-time onboard algorithms that enable agile flight through diverse and cluttered spaces.

First, we present a reactive planner based on forward-arc motion primitives that uses a short history of RGB-D observations to safely navigate near obstacles at speeds exceeding 6m/s. A safe stopping strategy ensures that the quadrotor always maintains a trajectory that allows it to hover safely within known free space.

Second, we build on recent advances in reinforcement learning with differentiable physics to develop a navigation policy that predicts thrusts directly from depth and state observations. We show that by using privileged information during training, our approach is able to navigate around large obstacles.

Through extensive simulation and real-world experiments, we show that our methods outperform baselines in both efficiency and reliability in cluttered environments. These results have the potential to aid in a wide range of real-world applications from disaster response and search-and-rescue missions to infrastructure inspection and exploration of unknown environments.

Committee:

Wennie Tabib (advisor)

Wenshan Wang

Seungchan Kim

Meeting ID: 976 1514 7926 Passcode: 651073

Google Calendar: https://calendar.app.google/g1w649rT6nz8NpLE9