We consider the problem of collaboratively delivering a package from a specified source node s to a designated target node t in an undirected graph G = (V, E), using k mobile agents. Each agent i starts at time 0 at a node pi and can move along edges subject to two parameters: Its weight wi, which denotes the rate of energy consumption while travelling, and its velocity vi, which defines the speed with which agent i can travel. We are interested in operating the agents such that we minimize the total energy consumption E and the delivery time T (time when the package arrives at t). Specifically, we are after a schedule of the agents that lexicographically minimizes the tuple (E, T). We show that this problem can be solved in polynomial time O(k|V |2 + APSP), where O(APSP) denotes the running time of an all-pair shortest-paths algorithm. This completes previous research which shows that minimizing only E or only T is polynomial-time solvable [6, 7], while minimizing a convex combination of E and T, or lexicographically minimizing the tuple (T, E) are both NP-hard [7].
CITATION STYLE
Bärtschi, A., & Tschager, T. (2017). Energy-efficient fast delivery by mobile agents. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 10472 LNCS, pp. 82–95). Springer Verlag. https://doi.org/10.1007/978-3-662-55751-8_8
Mendeley helps you to discover research relevant for your work.