We consider how algorithms allow for increasing system robustness, both in the sense of low-level feedback control, and motion planning under sensor uncertainty. We will present two projects exploring this. The first considers the low level feedback control of a system with significant uncertainty in physical parameters. By using machine learning techniques, we are able to design (and validate in flight on a subscale system) a feedback controller capable of stably flying a multicopter in hover, even under extreme uncertainty in parameters (such as factor 2 uncertainty in vehicle linear size). This is achieved with a low-level controller trained in simulation.
The second project considers the problem of robust navigation without GPS. We specifically study the problem of navigation using cameras and inertial data (so-called Visual Inertial Odometry), and how this perception modality is affected by the motion of the platform. This allows us then to do motion planning with explicitly optimizing for uncertainty along the trajectory, in addition to more typical metrics like flight speed.
Mark Mueller leads the High Performance Robotics Laboratory (HiPeRLab), where research focuses on design and autonomy of aircraft. The group has expertise in building and experimentally evaluating aerial robotics. Methods of interest include optimal and robust control, as well as learning-based methods. Mueller received a B.Eng. (Mechanical and Aeronautical Engineering) from the University of Pretoria, and then M.Sc. and Dr.Sc. both from ETH Zurich (Switzerland). Mueller is lead inventor on multiple commercially used patents, and currently serves as Associate Faculty Director of Aerospace Programs at Berkeley.
Relevant Expertise:
- Learning Highly Adaptive Controllers for Autonomous Aircraft: Creating learning-based systems to predict and execute emergency maneuvers for autonomous operations, including emergency landings and fault tolerance
- Estimation and planning: Algorithms for state estimation, including GPS-denied autonomy, with a focus on both robustness/safety, as well as performance
Attendees are invited to join the session in person at NASA Ames Research Center, Building N210, Room 115 (a NASA badge is required) or online:
Microsoft Teams Need help?
Join the meeting now
Meeting ID: 219 053 318 025 74
Passcode: sM6Qo6to
About Air Time by NAMS-2
Air Time is a series of seminars on advanced aviation hosted by Crown Innovations, Inc., in collaboration with the University of California’s CITRIS and the Banatao Institute. The seminars feature leading experts on cutting-edge research who share interesting ideas on pertinent topics and innovative methodologies. Air Time speakers include subject matter experts from UC Berkeley, Merced, Davis, and Santa Cruz. The seminars take place weekly.
Crown Innovations, Inc. is the prime contractor for the NASA Academic Mission Services 2 (NAMS-2) contract. Contact the program management at nams2pmo@crownci.com for more information or to arrange a collaboration in your field.