Action Graphs for Performing Goal Recognition Design on Human-Inhabited Environments.

Helen Harman,Pieter Simoens

doi:10.3390/s19122741

Abstract

Goal recognition is an important component of many context-aware and smart environment services; however, a person’s goal often cannot be determined until their plan nears completion. Therefore, by modifying the state of the environment, our work aims to reduce the number of observations required to recognise a human’s goal. These modifications result in either: Actions in the available plans being replaced with more distinctive actions; or removing the possibility of performing some actions, so humans are forced to take an alternative (more distinctive) plan. In our solution, a symbolic representation of actions and the world state is transformed into an Action Graph, which is then traversed to discover the non-distinctive plan prefixes. These prefixes are processed to determine which actions should be replaced or removed. For action replacement, we developed an exhaustive approach and an approach that shrinks the plans then reduces the non-distinctive plan prefixes, namely Shrink–Reduce. Exhaustive is guaranteed to find the minimal distinctiveness but is more computationally expensive than Shrink–Reduce. These approaches are compared using a test domain with varying amounts of goals, variables and values, and a realistic kitchen domain. Our action removal method is shown to increase the distinctiveness of various grid-based navigation problems, with a width/height ranging from 4 to 16 and between 2 and 14 randomly selected goals, by an average of 3.27 actions in an average time of 4.69 s, whereas a state-of-the-art approach often breaches a 10 min time limit.

Highlights

Through the deployment of numerous IoT sensors, smart environments can attempt to recognise the goal of a human from the actions they perform, and become more context-aware
We compare our solution to their pruned-reduce algorithm, and show that for navigation domains we have greatly reduced the time required to solve goal recognition design problems
The run-time and number of required state changes, for an increasing number of variables, values and goals are shown in Figures 14–16, respectively; the ACDdep and WCDdep reduction comparisons are shown in Figure 17

Summary

Introduction

Through the deployment of numerous IoT sensors, smart environments can attempt to recognise the goal of a human from the actions they perform, and become more context-aware. Despite recent advances in Goal Recognition (GR) techniques [1,2], a person’s goal often cannot be determined until their plan nears completion. This is because the plans to reach different goals can initially be identical. A planning problem P can be defined as P = ( F, I, A, G ), where F is a set of fluents, I ⊆ F is the initial state, G ⊆ F is a goal state, and A is a set of actions along with their preconditions a pre ⊆ F and effects ( a add ⊆ F, adel ⊆ F ) [7,10,23,24]. Effects are denoted ae f f , fluents are interpreted as variables with values, and actions change the values of variables, e.g., change the value of a variable that represents a human’s location, or change variables that indicate which items have been taken from cupboards from false to true

Objectives

Methods

Results

Conclusion