Context-Based Intent Understanding for Autonomous Systems in Naval and Collaborative Robot Applications.

摘要

The problem: Understanding intent is a critical aspect of communication among people and for many biological systems. While people are very good at recognizing intentions, endowing an autonomous system (robot or simulated agent) with similar skills is a more complex problem, which has not been sufficiently addressed in the field. The issue of intent recognition is particularly important in situations that involve collaboration among multiple agents or assessment of potential threats. In the former case collaboration can be greatly enhanced, while in the latter case dangerous situations can be detected before any harmful actions can be finalized. In this project, we propose to develop methodologies for intent understanding, with specific focus on autonomous systems for naval and collaborative robotics applications. The main research problems we will address in this project are to: (1) develop tools for understanding the high-level intentions of groups of agents, (2) develop algorithms for intent understanding based on contextual information, (3) develop vision-based techniques for learning of contextual information, and detection and identification of objects of interest.