“…A number of works have tried to deal with the problem of adapting a robot's behavior to the order of execution of a task by a human and many techniques have been proposed to generate behaviors more or less dynamically. These are the techniques (in nonexhaustive list): Planning and Scheduling [4,14,15,17], AND-OR Graphs [8], Behavior-Tree [18,22], Hierarchical Task Networks (HTN) [2,6,13], Game Theory [5], Markov Decision Process (MDP), or a combination of them [21]. Among these techniques, task planning and more precisely hierarchical planning is widely used and very effective to generate symbolic plans in a similar way of the human decomposition of complex tasks into simpler actions [3].…”