27th International Conference on Automated Planning and Scheduling - IntEx 2017 - Working Notes of the Workshop on Integrated Execution of Planning and Acting

Oct 4, 2026

Upload by Aran

27th International Conference on Automated Planning and Scheduling (IntEx 2017) working notes compile papers on integrating planning and acting for online execution. The notes address how planners generate plans that must be adapted when dynamic-world assumptions break, including whether acting or planning should respond, when to abandon or repair goals, and how executives should adjust, switch, or learn policies. They emphasize computational-budgeted online deliberation. The included material also surveys k-robust multi-agent path finding.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages57

Reading Time2 h 24 min

AccessFree to download

Foreword

Contents

k-Robust Multi-Agent Path Finding

Integrating Execution and Rescheduling

Autonomous Search-Detect-Track for Small UAVs

An Architecture for Integrated Timeline Planning and Model-based Execution

Goal Reasoning as Multilevel Planning

Automated Planning with Goal Reasoning in Minecraft

Towards Planning With Hierarchies of Learned Markov Decision Processes

Abstract

Introduction and Overview

27th International Conference on Automated Planning and Scheduling June 19-23, 2017, Pittsburgh, USA IntEx 2017 Working Notes of the Workshop on Integrated Execution of Planning and Acting Edited by: Mark Roberts, Sara Bernardini, Tim Niemueller, and Tiago Vaquero Organization Mark Roberts, Naval Research Laboratory, USA Sara Bernardini, Royal Holloway University, London, UK Tim Niemueller, RWTH, Aachen, Germany Tiago Vaquero, MIT, USA Program Committee Ron Alford, Mitre Corporation, USA J. Benton, NASA Ames, USA Mark Boddy, Adventium Labs, USA Michael Cashmore, King's College London, UK Jeremy Frank, NASA Ames, USA Andy Hertle, University of Freiburg, Germany Nir Lipovetzky, University of Melbourne, Australia Julie Porteous, Teesside University, UK Wheeler Ruml, University of New Hampshire, USA Scott Sanner, University of Toronto, Canada Vikas Shivashankar, Knexus Research, USA Foreword Automated planners are increasingly being integrated into online execution systems. The integration may, for example, embed a domain-independent temporal planner in a manufacturing system (e.g., the Xerox printer application) or autonomous vehicles. The integration may resemble something more like a ”planning stack” where an automated planner produces an activity or task plan that is further reﬁned before being executed by a reactive controller (e.g., robotics) . Or, the integration may be a domain-speciﬁc policy that maps states to actions (e.g., reinforcement learning) . Online learning may or may not be involved, and may include adjusting or augmenting the model, determining when to repair versus replan, learning to switch policies, etc. A speciﬁc focus of these integrations involves online deliberation, bringing to the foreground concerns over how much computational effort planning should invest over time. In any of these systems, a planner generates action sequences that are eventually dispatched to an executive, yet taking action in a dynamic world rarely proceeds according to plan. When planning assumptions are challenged during execution, it raises a number of interesting questions about how the system should respond. Is the ”acting” side of the system responsible fora response or the ”planning” side? Or do the two need to cooperate and how much? When should the activity planner abandon or preempt the current goals? Should the task planner repair a plan or replan from scratch? Should the executive adjust its current policy, switch to a new one, or learn a new policy from more relevant experience? This set of working notes consolidates the papers to be presented at IntEx 2017 . These notes will be updated after the workshop is held with details of the workshop discussions. Mark Roberts, Sara Bernardini, Tim Niemueller, and Tiago Vaquero June 2017 Contents k-Robust Multi-Agent Path Finding DorAtzmon, Roni Stern, Ariel Felner, Roman Bartak, Neng-Fa Zhou and Glenn Wagner 1 Integrating Execution and Rescheduling Jeremy Frank 10 Autonomous Search-Detect-Track for Small UAVs Bob Morris, Anjan Chakrabarty, Joshua Baculi, Xavier Bouyssounouse and Rusty Hunt 19 An Architecture for Integrated Timeline Planning and Model-based Execution Tiago Nogueira and Simone Fratini 27 Goal Reasoning as Multilevel Planning Alison Paredes and Wheeler Ruml 36 Automated Planning with Goal Reasoning in Minecraft Mark Roberts, Wiktor Piotrowski, Pyrce Bevan, David Aha, Maria Fox, Derek Long and Daniele Magazzeni 43 Towards Planning With Hierarchies of Learned Markov Decision Processes John Winder, Shawn Squire, Matthew Landen, Stephanie Milani and Marie desJardins 50 k-Robust Multi-Agent Path Finding Dor Atzmon, Roni Stern, Ariel Felner and Roman Bartak and Neng-Fa Zhou Ben Gurion University of the Negev Charles University in Prague City University of New York Be'er Sheva, Isreal Prague, Czech Republic New York City, NY, USA Abstract In the multi-agent path-􀀌nding (MAPF) problem

FAQs

What is the k-robust MAPF problem introduced in the included paper?

It requires a multi-agent path-finding plan that remains robust to up to k unexpected delays per agent. The approach converts a conflict-based search solver to support k-robustness and analyzes policies for cases with more than k delays.

Oct 4, 2026

Upload by Aran