CASUS Institute Seminar, Dr. Wolfgang Hönig, Independent Junior Research Group Leader, Computer Science, TU Berlin
Wolfgang is also heading the Intelligent Multi-Robot Coordination Lab. Previously, he was a postdoctoral scholar at the California Institute of Technology, USA. He received the diploma in Computer Science from TU Dresden, Germany in 2012, and the M.S. and Ph.D. degrees from the University of Southern California (USC), USA in 2016 and 2019, respectively. His research focuses on enabling large teams of physical robots to collaboratively solve real-world tasks, using tools from informed search, optimization, and machine learning. Wolfgang has been the recipient of several awards, including Best Paper in Robotics Track for a paper at ICAPS 2016 and the 2019 Best Dissertation Award in Computer Science at USC.
Motion coordination (namely planning and control) is a fundamental building block for deploying robots to execute useful tasks safely. Particular coordination challenges arise if many obstacles or robots are operating in a small space. Wolfgang will present techniques that compute near-optimal solutions efficiently in such challenging scenarios. First, he considers the centralized case and introduces a hybrid planner that combines informed multi-agent search and distributed optimization to compute smooth and safe trajectories for hundreds of robots within minutes.
Second, Wolfgang presents a general method that uses function approximation (specifically, supervised learning) to automatically synthesize a distributed, provably safe controller. Third, he introduces the Neural-Swarm, a method to efficiently reason about the aerodynamic interaction forces between multirotors that fly in close proximity to each other. All results are experimentally validated on multirotor teams using the Crazyswarm framework.