Epistemic Multi-agent Planning Using Monte-Carlo Tree Search

Reifsteck, Daniel; Engesser, Thorsten; Mattmüller, Robert; Nebel, Bernhard

doi:10.1007/978-3-030-30179-8_24

Cited by 4 publications

(4 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In particular, we take an implicit coordination approach, following from the work of Engesser et al, where the agent takes into account the spontaneous cooperation of other agents in achieving the goal, which requires recursive perspective-taking in order to predict their actions (Engesser et al 2017). In (Bolander et al 2018), the authors further dis-cussed the impact of eager and lazy agents in the framework, and in (Reifsteck et al 2019), an MCTS algorithm is developed that shares similar insights as our work. Compared to the work by Engesser et al, we differ in that our framework based on conditional doxastic logic allows the modeling of false beliefs and the revision of false beliefs, and our explanations refer directly to the plan space instead of states as a result of extending the logic to knowledge bases.…”

Section: Related Workmentioning

confidence: 79%

“…Back Up We take a more customized approach to computing the utility score of each node during the back-up phase. The split node takes the minimum score of the children predict nodes since it cares about the worst-case outcome, similar to the work of (Reifsteck et al 2019), then multiplies it by the penalty factor of the action that leads to the split node.…”

Section: Search Treementioning

confidence: 99%

See 1 more Smart Citation

Adaptation and Communication in Human-Robot Teaming to Handle Discrepancies in Agents’ Beliefs about Plans

Zhang

Williams

2023

ICAPS

View full text Add to dashboard Cite

When agents collaborate on a task, it is important that they have some shared mental model of the task routines -- the set of feasible plans towards achieving the goals. However, in reality, situations often arise that such a shared mental model cannot be guaranteed, such as in ad-hoc teams where agents may follow different conventions or when contingent constraints arise that only some agents are aware of. Previous work on human-robot teaming has assumed that the team has a set of shared routines, which breaks down in these situations. In this work, we leverage epistemic logic to enable agents to understand the discrepancy in each other's beliefs about feasible plans and dynamically plan their actions to adapt or communicate to resolve the discrepancy. We propose a formalism that extends conditional doxastic logic to describe knowledge bases in order to explicitly represent agents' nested beliefs on the feasible plans and state of execution. We provide an online execution algorithm based on Monte Carlo Tree Search for the agent to plan its action, including communication actions to explain the feasibility of plans, announce intent, and ask questions. Finally, we evaluate the success rate and scalability of the algorithm and show that our agent is better equipped to work in teams without the guarantee of a shared mental model.

show abstract

Section: Related Workmentioning

confidence: 79%

Section: Search Treementioning

confidence: 99%

Adaptation and Communication in Human-Robot Teaming to Handle Discrepancies in Agents’ Beliefs about Plans

Zhang

Williams

2023

ICAPS

View full text Add to dashboard Cite

show abstract

“…Engesser and Miller (2020) describe a planner which compiles a decidable fragment of epistemic planning into fully observable nondeterministic planning. Reifsteck et al (2019) describe an algorithm and implementation capable of finding epistemic planning policies using a Monte-Carlo Tree Search over epistemic states.…”

Section: Introductionmentioning

confidence: 99%

DEL-based Epistemic Planning for Human-Robot Collaboration: Theory and Implementation

Bolander

Dissing

Herrmann

2021

Proceedings of the Eighteenth International Conference on Principles of Knowledge Representation and Reasoning

View full text Add to dashboard Cite

Epistemic planning based on Dynamic Epistemic Logic (DEL) allows agents to reason and plan from the perspective of other agents. The framework of DEL-based epistemic planning thereby has the potential to represent significant aspects of Theory of Mind in autonomous robots, and to provide a foundation for human-robot collaboration in which coordination is achieved implicitly through perspective shifts. In this paper, we build on previous work in epistemic planning with implicit coordination. We introduce a new notion of indistinguishability between epistemic states based on bisimulation, and provide a novel partition refinement algorithm for computing unique representatives of sets of indistinguishable states. We provide an algorithm for computing implicitly coordinated plans using these new constructs, embed it in a perceive-plan-act agent loop, and implement it on a robot. The planning algorithm is benchmarked against an existing epistemic planning algorithm, and the robotic implementation is demonstrated on human-robot collaboration scenarios requiring implicit coordination.

show abstract

“…They have to use actions with which they can hint the colors and values of cards to the others. While the concept of implicit coordination by epistemic perspective taking has been defined only for achievement goals so far, we think that it can be generalized to optimization problems like maximizing the expected score in Hanabi (Reifsteck et al 2019).…”

Section: Introductionmentioning

confidence: 99%

Implicit Coordination Using FOND Planning

Engesser

Miller

2020

AAAI

Self Cite

View full text Add to dashboard Cite

Epistemic planning can be used to achieve implicit coordination in cooperative multi-agent settings where knowledge and capabilities are distributed between the agents. In these scenarios, agents plan and act on their own without having to agree on a common plan or protocol beforehand. However, epistemic planning is undecidable in general. In this paper, we show how implicit coordination can be achieved in a simpler, propositional setting by using nondeterminism as a means to allow the agents to take the other agents' perspectives. We identify a decidable fragment of epistemic planning that allows for arbitrary initial state uncertainty and non-determinism, but where actions can never increase the uncertainty of the agents. We show that in this fragment, planning for implicit coordination can be reduced to a version of fully observable nondeterministic (FOND) planning and that it thus has the same computational complexity as FOND planning. We provide a small case study, modeling the problem of multi-agent path finding with destination uncertainty in FOND, to show that our approach can be successfully applied in practice.

show abstract

Epistemic Multi-agent Planning Using Monte-Carlo Tree Search

Cited by 4 publications

References 8 publications

Adaptation and Communication in Human-Robot Teaming to Handle Discrepancies in Agents’ Beliefs about Plans

Adaptation and Communication in Human-Robot Teaming to Handle Discrepancies in Agents’ Beliefs about Plans

DEL-based Epistemic Planning for Human-Robot Collaboration: Theory and Implementation

Implicit Coordination Using FOND Planning

Contact Info

Product

Resources

About