Distributed RL for junction-level traffic light phase control for decentralized traffic flow monitoring and optimization
Distributed reinforcement learning for scalable, decentralized multi-agent path finding (e.g., teams of UAVs)
Multi-robot search of an area by a team of (potentially heterogeneous) platforms, by leveraging the different capabilities of the robots.
Communication Learning (reinforced/differentiable communication) for simultaneous communication and action policy learning
Locomotion of a hexapod robot based on onboard torque/visual/3D sensors in challenging terrains (on hardware)