Situated Dialogue and Spatial Organization: What, Where… and Why?

By Geert-Jan M. Kruijff, Hendrik Zender, Patric Jensfelt, Henrik Iskov Christensen. 2007

155 citations recorded by the source corpus.

The paper presents an HRI architecture for human-augmented mapping, which has been implemented and tested on an autonomous mobile robotic platform. Through interaction with a human, the robot can augment its autonomously acquired metric map with qualitative information about locations and objects in the environment. The system implements various interaction strategies observed in independently performed Wizard-of-Oz studies. The paper discusses an ontology-based approach to multi-layered conceptual spatial mapping that provides a common ground for human-robot dialogue. This is achieved by combining acquired knowledge with innate conceptual commonsense knowledge in order to infer new knowledge. The architecture bridges the gap between the rich semantic representations of the meaning expressed by verbal utterances on the one hand and the robot's internal sensor-based world representation on the other. It is thus possible to establish references to spatial areas in a situated dialogue between a human and a robot about their environment. The resulting conceptual descriptions represent qualitative knowledge about locations in the environment that can serve as a basis for achieving a notion of situational awareness.

Subfields: Artificial Intelligence, Control and Systems Engineering, Natural Language Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Control and Systems Engineering

In Natural Language Processing

Citations in the corpus

A simple neural network module for relational reasoning (2017, background): “Although grounding language in spatial percepts has a long-standing tradition, the majority of previous research has focused on either rule-based spatial representations or hand-engineered spatial features [8, 10, 20, 21, 24, 29, 38, 39].”
A Multi-World Approach to Question Answering about Real-World Scenes based on Uncertain Input (2014, background): “Integrated systems that execute commands : Others [9, 10, 11, 12, 13] focus on the task of learning the representation of natural language in the restricted setting of executing commands.”
Jointly Learning to Parse and Perceive: Connecting Natural Language to the Physical World (2013, background): “Some work constructs integrated systems which execute plans in response to natural language commands (Winograd, 1970; Hsiao et al., 2003; Roy et al., 2003; Skubic et al., 2004; MacMahon et al., 2006; Levit and Roy, 2007; Kruijff et al., 2007).”
Conceptual spatial representations for indoor mobile robots (2008, background): “In [24] we present the cognitive architecture of our robotic system and give details on its dialogue capabilities .”
A Pooling Approach to Modelling Spatial Relations for Image Retrieval and Annotation (2014, background): “Grounding Spatial Relations Although research on grounding of spatial language has a long standing tradition, previous methods mostly focus on rule-based spatial representation [24, 16] or more recently on a set of hand-crafted spatial features with learnt weights [31, 9, 17, 10].”
Semantic mapping for mobile robotics tasks: A survey (2015, background): “More specifically, a situated dialogue is introduced, which is a functionality that enables the robot to carry out a natural language dialogue with a human [99].”
Robot task planning using semantic maps (2008, unknown): “In this section, we describe how we represent and maintain this knowledge (spatial and semantic) in form of a semantic map.”