Planning sub-target merging method based on deep learning

Abstract

The invention discloses a planning sub-target merging method based on deep learning, which mainly solves the problem of granularity upper limit selection of sub-target merging in the preprocessing process of a planning system, and comprises the following steps: generating a data set, and performing data expansion and data cleaning on samples; training the obtained sample as a deep learning model,selecting a Resnet deep residual network as the model, and separately training planning problems in different fields to generate a feature model of each field; modifying a preprocessing code of the ASOP planning system, adding an updating function of a sub-target granularity threshold value. Through a trained model, the planning system can determine the upper limit of granularity according to a specific planning problem. According to the method, the upper limit threshold value of the granularity is determined from the macroscopic level through a deep learning method, so that the efficiency andthe quality of the planning system during problem solving are balanced, the planning capability of the planning system is effectively improved, and the labor cost is greatly reduced through dynamic autonomous adjustment of a computer.

Description

technical field [0001] The invention relates to the fields of intelligent planning and deep learning, in particular to a method for merging planning sub-goals based on deep learning. Background technique [0002] Research on intelligent planning began in the 1960s, stemming from studies in state-space search, theorem proving, and control theory, as well as practical needs in robot scheduling and other fields. In the past few decades of development, planning systems in the field of intelligent planning have significantly improved in solution efficiency and solution quality. [0003] However, when the planning system faces large-scale complex problems, there is still a big bottleneck. People try to decompose the original task and complete the target state through the idea of ​​​​divide and conquer, so as to simplify the original planning problem. Relevant scholars have designed Hierarchical task network planning system (HTN), the core of this type of planning system is to gra...

AI Technical Summary

Problems solved by technology

However, in actual use, when the ASOP planning system based on the admissible sorting algorithm handles planning problems in some fields, the quality of the planning solution may not be ideal, and the solution efficiency may also appear when solving planning problems of the same scale in the same field. Obvious inconsistencies make the planning solver efficiency unstable

The reason is that the admissible sorting relationship does not always conform to the optimal sub-goal realization order, and the wrong sorting may cause abnormalities in the subsequent search methods

[0006] The existing problems can be alleviated by merging the sub-target parts, but the merging will increase the number of propositions of a single target set, so that the search space of the state will increase, resulting in a decrease in the solution ability

Images