Concurrent Planning and Execution Using Dispatch-Dependent Values

Andrew Coles; Erez Karpas; Solomon E. Shimony; Shahaf Shperberg; Wheeler Ruml

Concurrent Planning and Execution Using Dispatch-Dependent Values

Andrew Coles, Erez Karpas, Solomon E. Shimony, Shahaf Shperberg, Wheeler Ruml

Research output: Chapter in Book/Report/Conference proceeding › Conference paper › peer-review

Abstract

Agents operating in the real world must cope with the fact that time passes while they plan.
In some cases, such as under tight deadlines, the only way for such an agent to achieve its goal is to execute an action before a complete plan has been found. This problem is called Concurrent Planning and Execution (CoPE). Previous work on CoPE relied on a value function that assumes search will finish before actions are executed, causing the agent to be overly pessimistic in many situations.
In this paper, we define a new value function that takes into account the agent's ability to dispatch actions incrementally. This allows us to devise a much simpler algorithm for concurrent planning and execution. An experimental evaluation on problems with time pressure shows that the new method significantly outperforms the previous state-of-the-art.

Original language	English
Title of host publication	Proceedings of the 34th International Joint Conference on Artificial Intelligence
Publisher	AAAI Press
Publication status	Accepted/In press - 30 Apr 2029

Embargoed Document

_IJCAI_25__CoPE_using_dispatch_dependent_values
Accepted author manuscript, 272 KB
Embargo ends: 16/08/2025

COHERENT - COllaborative HiErarchical Robotic ExplaNaTions
Coles, A. (Primary Investigator)
EPSRC Engineering and Physical Sciences Research Council
1/04/2021 → 31/03/2025
Project: Research

Cite this

@inbook{39f82b5fce5647699e27de5a8b336d89,

title = "Concurrent Planning and Execution Using Dispatch-Dependent Values",

abstract = "Agents operating in the real world must cope with the fact that time passes while they plan. In some cases, such as under tight deadlines, the only way for such an agent to achieve its goal is to execute an action before a complete plan has been found. This problem is called Concurrent Planning and Execution (CoPE). Previous work on CoPE relied on a value function that assumes search will finish before actions are executed, causing the agent to be overly pessimistic in many situations.In this paper, we define a new value function that takes into account the agent's ability to dispatch actions incrementally. This allows us to devise a much simpler algorithm for concurrent planning and execution. An experimental evaluation on problems with time pressure shows that the new method significantly outperforms the previous state-of-the-art.",

author = "Andrew Coles and Erez Karpas and Shimony, {Solomon E.} and Shahaf Shperberg and Wheeler Ruml",

year = "2029",

month = apr,

day = "30",

language = "English",

booktitle = "Proceedings of the 34th International Joint Conference on Artificial Intelligence",

publisher = "AAAI Press",

}

TY - CHAP

T1 - Concurrent Planning and Execution Using Dispatch-Dependent Values

AU - Coles, Andrew

AU - Karpas, Erez

AU - Shimony, Solomon E.

AU - Shperberg, Shahaf

AU - Ruml, Wheeler

PY - 2029/4/30

Y1 - 2029/4/30

N2 - Agents operating in the real world must cope with the fact that time passes while they plan. In some cases, such as under tight deadlines, the only way for such an agent to achieve its goal is to execute an action before a complete plan has been found. This problem is called Concurrent Planning and Execution (CoPE). Previous work on CoPE relied on a value function that assumes search will finish before actions are executed, causing the agent to be overly pessimistic in many situations.In this paper, we define a new value function that takes into account the agent's ability to dispatch actions incrementally. This allows us to devise a much simpler algorithm for concurrent planning and execution. An experimental evaluation on problems with time pressure shows that the new method significantly outperforms the previous state-of-the-art.

AB - Agents operating in the real world must cope with the fact that time passes while they plan. In some cases, such as under tight deadlines, the only way for such an agent to achieve its goal is to execute an action before a complete plan has been found. This problem is called Concurrent Planning and Execution (CoPE). Previous work on CoPE relied on a value function that assumes search will finish before actions are executed, causing the agent to be overly pessimistic in many situations.In this paper, we define a new value function that takes into account the agent's ability to dispatch actions incrementally. This allows us to devise a much simpler algorithm for concurrent planning and execution. An experimental evaluation on problems with time pressure shows that the new method significantly outperforms the previous state-of-the-art.

M3 - Conference paper

BT - Proceedings of the 34th International Joint Conference on Artificial Intelligence

PB - AAAI Press

ER -

Concurrent Planning and Execution Using Dispatch-Dependent Values

Abstract

Embargoed Document

Fingerprint

Projects

COHERENT - COllaborative HiErarchical Robotic ExplaNaTions

Cite this