Mess is a substrate for building simulation environments suitable for testing plans and on-line or real-time planners. The article describes the design of Mess,howsimu- lations are built and how on-line planners integrate with the substrate. Mess supports activities, defined as processes over

Symbolic representations have been used successfully in off-line planning algorithms for Markov decision processes. We show that they can also improve the performance of online planners. In addition to reducing computation time, symbolic generalization can reduce the amount of costly real

In this paper we present a novel planning approach, based on well-known techniques such as goal decomposition and heuristic planning, aimed at working in highly dynamic environments with time constraints. Our contribution is a domain-independent planner to incrementally generate plans under a

the first successful industrial application of embedded domain-independent temporal planning. Our system handles execution failures and multi-objective preferences. At its heart is an on-line algorithm that combines techniques from state-space planning and partial-order scheduling. We suggest

We describe a real manufacturing problem that lies between job shop scheduling and temporal planning. The setting is on-line in the sense that new jobs arrive asynchronously, perhaps several per second, while plans for previous jobs are being executed. We formalize the problem as a variant

In contrast to most academic work on AI planning, many potential applications have an on-line character. For example, the true objective may not be to find the fastest plan, but to have solving plus plan execution end as soon as possible. Or additional goals may arrive while previous plans

, but challenge conventional plan recognition algorithms for inferring students ’ activities with the software. This paper presents a new algorithm for recognizing stu-dents ’ activities in ELEs that works on-line during the student’s interaction with the software. Our approach, called CRADLE, re-duces the amount

This paper presents an agent-based approach to online assembly task allocation and scheduling for flexible assembly systems. We propose an anytimescheduling algorithm to handle dynamic changes during planning. This algorithm is the center of an agent-based architecture in which agents

Abstract. SimPlanner is an integrated tool for planning and execution-monitoring which allows to interleave planning and execution. In this paper we present the on-line planner incorporated in SimPlanner. This is a domain-independent planner for STRIPS domains. SimPlanner par-ticipated in the IPC

In on-line planning an agent executes a sequence of actions to maximise some time dependent objective criterion. Deciding which actions to execute takes time, which the agent must manage in order to maximise overall performance. This is the metacontrol problem. Incremental search algorithms select