This preview shows pages 1–2. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: Role Allocation and Reallocation in Multiagent Teams: Towards A Practical Analysis Ranjit Nair Computer Science Dept Univ. of Southern California Los Angeles, CA 90089 firstname.lastname@example.org Milind Tambe Computer Science Dept Univ. of Southern California Los Angeles, CA 90089 email@example.com Stacy Marsella Information Sciences Institute Univ. of Southern California Marina del Rey, CA 90292 firstname.lastname@example.org ABSTRACT Despite the success of the BDI approach to agent teamwork, initial role allocation (i.e. deciding which agents to allocate to key roles in the team) and role reallocation upon failure remain open chal- lenges. What remain missing are analysis techniques to aid human developers in quantitatively comparing different initial role alloca- tions and competing role reallocation algorithms. To remedy this problem, this paper makes three key contributions. First, the paper introduces RMTDP (Role-based Markov Team Decision Problem), an extension to MTDP , for quantitative evaluations of role al- location and reallocation approaches. Second, the paper illustrates an RMTDP-based methodology for not only comparing two com- peting algorithms for role reallocation, but also for identifying the types of domains where each algorithm is suboptimal, how much each algorithm can be improved and at what computational cost (complexity). Such algorithmic improvements are identified via a new automated procedure that generates a family of locally optimal policies for comparative evaluations. Third, since there are combi- natorially many initial role allocations, evaluating each in RMTDP to identify the best is extremely difficult. Therefore, we introduce methods to exploit task decompositions among subteams to signifi- cantly prune the search space of initial role allocations. We present experimental results from two distinct domains. 1. INTRODUCTION The belief-desire-intention (BDI) approach to agent teamwork has led to many practical multiagent applications [12, 14, 13]. Ini- tial role allocation, i.e. which agents to allocate to the various roles in the team, and role reallocation upon failures or new tasks, are two continuing challenges for building teams [15, 7]. For instance, in mission rehearsal simulations , we need to select the numbers and types of helicopter agents to allocate to different roles in the team, and to decide how role substitution should occur upon fail- ures. Similarly, in disaster rescue , initial role allocations (e.g. which brigades for each fire) can greatly impact team performance. Critically needed now are analysis techniques to aid human de- velopers in quantitatively comparing and evaluating different ini- Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific...
View Full Document
- Fall '08
- Computer Science