We contribute a novel approach to understand, dialogue, plan, and execute complex sentences to command a mobile service robot. We define a complex command as a natural language sentence consisting of sensing-based conditionals, conjunctions, and disjunctions. We introduce a flexible template-based algorithm to extract such structure from the parse tree of the sentence. As the complexity of the command increases, extracting the right structure using the template-based algorithm decreases becomes more problematic. We introduce two different dialogue approaches that enable the user to confirm or correct the extracted command structure. We present how the structure used to represent complex commands can be directly used for planning and execution by the service robot.

As a bridge between humans and computers, the use of natural language processing in compliance demonstrates how this branch of AI is adding value across the financial services industry. Natural language processing (NLP) falls under the wider umbrella of artificial intelligence (AI) and essentially uses algorithms to help computers understand the everyday language of humans -- both spoken and written. As such, NLP is a fundamental bridge enabling interaction between computers and humans, and allowing machines to understand commands and input from humans in a seamless and streamlined manner. Once a machine can understand a human's everyday means of communication, the potential to add value becomes almost limitless. The Thomson Reuters Center for Cognitive Computing is constantly researching ways to perfect and advance different areas of AI, including machine perception, reasoning, knowledge management, and human-computer interfaces.

We contribute a novel approach to understand, dialogue, plan, and execute complex sentences to command a mobile service robot. We define a complex command as a natural language sentence consisting of sensing-based conditionals, conjunctions, and disjunctions. We introduce a flexible template-based algorithm to extract such structure from the parse tree of the sentence. As the complexity of the command increases, extracting the right structure using the template-based algorithm decreases becomes more problematic. We introduce two different dialogue approaches that enable the user to confirm or correct the extracted command structure. We present how the structure used to represent complex commands can be directly used for planning and execution by the service robot. We show results on a corpus of 100 complex commands