Uncertain action planning method for linear temporal logic containing soft and hard constraints

Abstract

The invention discloses an uncertain action planning method for linear temporal logic containing soft and hard constraints, which is characterized in that a control strategy meeting the optimal cost is solved, the whole process of a hard constraint part task is free of risks, and the existence of risks is allowed for a soft constraint part task. The method comprises the steps of firstly, establishing a double-layer automaton task model, wherein the task state Q<-> of each layer is a state set formed by arbitrary combination of task states in a hard constraint task model and a soft constraint task model in the model, the initial state is on the first layer, the normally-returned state of the hard constraint task is placed on the first layer, the normally-returned state of the soft constraint task is placed on the second layer, the non-normally-returned state is placed on the first layer and the second layer at the same time, by setting the state transition relationship, the conversion condition between the two layers is that one layer can be converted into the other layer only when the task corresponding to the layer is completed; modifying the task model for the soft constraint task; and finally, combining the double-layer automaton model and an intelligent agent model, designing an optimization problem, and obtaining an optimal control strategy.

Description

technical field [0001] The invention belongs to the technical field of intelligent body control, and in particular relates to an uncertain action planning method for linear sequential logic with soft and hard constraints. Background technique [0002] Linear sequential logic is a description method used to describe system specifications, which is composed of environmental information expressed in the form of atomic propositions and agent information through a series of Boolean symbols defined by grammatical specifications and sequential symbols. The linear temporal logic formula is composed of atomic proposition set AP, Boolean operators, and temporal operators. The formula form is: here p∈AP is a proposition in the set of atomic propositions, Temporal operators included in Linear Temporal Logic (LTL) formulas include ○ (next), ∪ (until), □ (always), ◇ (finally), (indicator). [0003] The soft and hard task constraints are based on the LTL task formula to strengthen ...

AI Technical Summary

Problems solved by technology

[0009] In view of this, the present invention aims at the long-term uncertain action planning problem of the agent oriented to LTL tasks, and proposes a two-layer automaton task model to represent the soft and hard constraints in the LTL task formula, and solves the control strategy that satisfies the optimal cost, and some tasks with hard constraints There is no risk in the whole process, and some tasks with soft constraints allow risks

Images