“…A few of them are related to the exploration of partially known environments [39] and the find-and-follow of people [40] with robots. Others [41], [42] concern robot navigation using only POMCP or hierarchical methods approaches with POMCP for high-level control and neural networks for low-level control. Popular MCTS-based approaches have been recently used also for developing agents with superhuman performance in the game of Go [43], [44].…”