Perception Methods for Autonomous Aerial Docking with Moving Ground Vehicles
Docking with moving ground vehicles has long been seen as a seminal challenge for aerial autonomy. Combining high-speed perception on embedded systems with precision control in challenging flow fields, it remains an open problem at typical driving speeds. This talk will highlight prior and ongoing work in the Toronto Robotics and Artificial Intelligence Lab (TRAILab) on three aspects of the problem. First, a complete solution for low speed docking will be described and demonstrated, followed by a discussion of the limitations inherent in the method, such as viewpoint restrictions and reliance on a known target. Second, advances in dynamic camera cluster perception will be presented, which can resolve viewpoint limitations without losing state estimation accuracy. In particular, active calibration approaches for gimballed cameras and their utility for visual odometry will be presented. Finally, a recurrent neural network approach to multi-step motion predictions for quadrotors will be described, enabling precise motion planning over varied operating conditions. This work sets the stage for high-speed landings on moving vehicles, an ongoing project in the TRAILab.
Professor Steven Waslander is a leading authority on autonomous aerial and ground vehicles, including multirotor drones and autonomous driving vehicles. Simultaneous Localization and Mapping (SLAM) and multi-vehicle systems. He received his B.Sc.E. in 1998 from Queen’s University, his M.S. in 2002 and his Ph.D. in 2007, both from Stanford University in Aeronautics and Astronautics, where as a graduate student he created the Stanford Testbed of Autonomous Rotorcraft for Multi-Agent Control (STARMAC), the world’s most capable outdoor multi-vehicle quadrotor platform at the time. He was a Control Systems Analyst for Pratt & Whitney Canada from 1998 to 2001. He was recruited to Waterloo from Stanford in 2008, where he founded and directs the Waterloo Autonomous Vehicle Laboratory (WAVELab), extending the state of the art in autonomous drones and autonomous driving through advances in localization and mapping, object detection and tracking, integrated planning and control methods and multi-robot coordination. In 2018, he joined the University of Toronto Institute for Aerospace Studies (UTIAS), and founded the Toronto Robotics and Artificial Intelligence Laboratory (TRAILab).