Talk-to-Resolve: Combining scene understanding and spatial dialogue to resolve granular task ambiguity for a collocated robot

TL;DR

Talk-to-Resolve is a system that enables collocated robots to use scene understanding and spatial dialogue to resolve task ambiguities, improving interaction naturalness and task success rate.

Contribution

The paper introduces a novel dialogue-based approach combining scene understanding and language to resolve task stalemates in robot instruction execution.

Findings

82% accuracy in identifying and resolving stalemates

Questions from the system are rated more natural (4.02/5) than state-of-the-art

System effectively uses dense scene captions and instructions for decision-making

Abstract

The utility of collocating robots largely depends on the easy and intuitive interaction mechanism with the human. If a robot accepts task instruction in natural language, first, it has to understand the user's intention by decoding the instruction. However, while executing the task, the robot may face unforeseeable circumstances due to the variations in the observed scene and therefore requires further user intervention. In this article, we present a system called Talk-to-Resolve (TTR) that enables a robot to initiate a coherent dialogue exchange with the instructor by observing the scene visually to resolve the impasse. Through dialogue, it either finds a cue to move forward in the original plan, an acceptable alternative to the original plan, or affirmation to abort the task altogether. To realize the possible stalemate, we utilize the dense captions of the observed scene and the given instruction jointly to compute the robot's next action. We evaluate our system based on a data set of initial instruction and situational scene pairs. Our system can identify the stalemate and resolve them with appropriate dialogue exchange with 82% accuracy. Additionally, a user study reveals that the questions from our systems are more natural (4.02 on average on a scale of 1 to 5) as compared to a state-of-the-art (3.08 on average).