Intent Recognition from Speech and Plan Recognition

Abstract

In multi-agent systems, the ability to infer intentions allows artificial agents to act proactively and with partial information. In this paper we propose an algorithm to infer a speakers intentions with natural language analysis combined with plan recognition. We define a Natural Language Understanding component to classify semantic roles from sentences into partially instantiated actions, that are interpreted as the intention of the speaker. These actions are grounded to arbitrary, hand-defined task domains. Intent recognition with partial actions is statistically evaluated with several planning domains. We then define a Human-Robot Interaction setting where both utterance classification and plan recognition are tested using a Pepper robot. We further address the issue of missing parameters in declared intentions and robot commands by leveraging the Principle of Rational Action, which is embedded in the plan recognition phase.

Two ways to make your robot proactive: Reasoning about human intentions or reasoning about possible futures

“…This limitation is partly addressed by Ramírez and Geffner (2010) , who introduced a more general formulation. Persiani (2020) offers an example of using Bayesian inference in a logic-based approach. The authors use classical planning to generate an action plan for each goal, and then they use a Bayesian prior function to infer human intention.…”

“…In this work, we adopt a logic-based approach as performed by Persiani (2020) : we represent the robot’s knowledge in a symbolic form, which the robot uses to plan its actions. We assume rational humans.…”

“…We select inverse planning since this is a straight-through logic-based approach for fully observable systems. The approach has been widely used in other systems for intention recognition ( Han and Pereira, 2013 ; Farrell and Ware, 2020 ; Persiani, 2020 ). While planning synthesizes a sequence of actions to reach a goal, in inverse planning we observe the execution of a sequence of actions to infer the human’s goal and the corresponding plan.…”

“…Therefore, we take the action list in to be the intention of the human in state s . This strategy has been originally proposed in logic-based approaches by Persiani (2020) .…”