Artificial Intelligence

Scalable Task and Motion Planning for Multi-Robot Systems in Obstacle-Rich Environments

Friday, June 15, 2018, 11:00am - 12:00pm PDTiCal
Conf. Rm #689
This event is open to the public.
AI Seminar
Wolfgang Hönig

Motion planning problems have been studied in both the artificial intelligence AI and robotics communities. AI solvers can compute plans for hundreds of simple agents in minutes with suboptimality guarantees, while robotics solutions typically include richer kinodynamic models during planning, but are very slow when many robots and obstacles are taken into account. We combine the advantages of the two methods by using a two-step approach. First, we use and extend AI solvers for a simplified coordination problem. The output is a discrete plan that cannot be executed on real robots. Second, we apply a computationally efficient post-processing step that creates a continuous plan, taking  kinodynamic constraints into account. We show examples for ground robots in a warehouse domain and quadrotors that are tasked with formation change.

Wolfgang Hönig is a Ph.D. student in the ACT Lab at the University of Southern California. He holds a Diploma in Computer Science from the Technical University Dresden, Germany and an M.S. in Computer Science Intelligent Robotics from USC. His research focuses on enabling large teams of physical robots to collaboratively solve real-world tasks by combining methods from artificial intelligence and robotics

« Return to Events