par Khaluf, Yara;Markarian, Christine;Simoens, Pieter;Reina, Andreagiovanni 
Référence Lecture notes in computer science, 10349 LNCS, page (144-156)
Publication Publié, 2017
Référence Lecture notes in computer science, 10349 LNCS, page (144-156)
Publication Publié, 2017
Article révisé par les pairs
| Résumé : | Through this study, we introduce the idea of applying scheduling techniques to allocate spatial resources that are shared among multiple robots moving in a static environment and having temporal constraints on the arrival time to destinations. To illustrate this idea, we present an exemplified algorithm that plans and assigns a motion path to each robot. The considered problem is particularly challenging because: (i) the robots share the same environment and thus the planner must take into account overlapping paths which cannot happen at the same time; (ii) there are time deadlines thus the planner must deal with temporal constraints; (iii) new requests arrive without a priori knowledge thus the planner must be able to add new paths online and adjust old plans; (iv) the robot motion is subject to noise thus the planner must be reactive to adapt to online changes. We showcase the functioning of the proposed algorithm through a set of agent-based simulations. |
