| Description |
This thesis provides details on the development of automatic collision avoidance for manually tele-operated unmanned aerial vehicles. We note that large portions of this work are also reprinted with permission, from 2014 IEEE International Conference on Robotics and Automation, \Automatic Collision Avoidance for Manually Tele-operated Unmanned Aerial Vehicles", by J. Israelsen, M. Beall, D. Bareiss, D. Stuart, E. Keeney, and J. van den Berg c 2014 IEEE. We provide a method to aid the operator of unmanned aerial vehicles. We do this by automatically performing collision avoidance with obstacles in the environment. Our method allows the operator to focus on the overall motion of the vehicle rather than requiring the operator to perform collision avoidance. Where other currently existing systems override the controls of the operator only as a last resort, our approach was developed such that the operator can rely on the automatic collision avoidance for maneuverability. Given the current operator control input, our approach continually determines the future path of the vehicle. If along the future path a collision is predicted, then our algorithm will minimally override the operator's control such that the vehicle will not collide with the obstacles in the environment. Such an approach ensures the safety of the operator's controls while simultaneously maintaining the original intent of the operator. We successfully implemented this approach in a simulated environment, as well as on a physical quadrotor system in a laboratory environment. Our experiments show that, even when intentionally trying to do so, the operator failed to crash the vehicle into environment obstacles. |
