Machine Learning in Hybrid Hierarchical and Partial-Order Planners for Manufacturing Domains

Abstract

The application of AI planning techniques to manufacturing systems is being widely deployed for all the tasks involved in the process, from product design to production planning and control. One of these problems is the automatic generation of control sequences for the entire manufacturing system in such a way that final plans can be directly used as the sequential control programs which drive the operation of manufacturing systems. Hybis is a hierarchical and nonlinear planner whose goal is to obtain partially ordered plans at such a level of detail that they can be used as sequential control programs for manufacturing systems. Currently, those sequential control programs are being generated by hand using modelling tools. This document describes a work whose aim is to improve the efficiency of solving problems with Hybis by using machine learning techniques. It implements a deductive learning method that is able to automatically acquire control knowledge (heuristics) by generating bounded explanations of the problem solving episodes. The learning approach builds on Hamlet, a system that learns control knowledge in the form of control rules.

PLTOOL: A knowledge engineering tool for planning and learning

“…We have used the same deductive approach for automatically acquiring control rules for other types of planners, such as hybrid HTN-POP planners as HYBIS (Castillo et al, 2001) 5 , though we have not yet implemented its inductive counterpart (Fern andez et al, 2005). Within the general framework of learning heuristics for different types of planners, we are currently generating a common language for describing heuristics for different planning paradigms, independently of the planner, such as GRAPHPLAN or FF based planners.…”

Unsupervised Learning of HTNs in Complex Adversarial Domains

“…HTNs are a very popular planning framework, and have been used in many real-world domains, from manufacturing to medical care (Fernández, Aler, and Borrajo 2005) (Musen 1989). However, there is one great drawback traditionally cited to them whenever they are considered, and this is the cost of expert labor involved in creating a useful HTN.…”