CASPER++- Cognitive Architecture for Social Perception and Engagement in Robots for ethical and transparent collaboration

Abstract

The field of social robotics has advanced significantly in the last decades, up to the point where its real-world adoption is expected in years to come. The driving motivation for this project is to provide trustworthy ethical social robots with dynamic social awareness, so they can blend naturally into cooperative groups of both humans and other robots. The aim to develop a novel cognitive architecture for social perception and engagement in robots with ethical and transparent collaboration in heterogeneous multi-agent teams. This aim will be realized through the following projects objectives- (i) To design and develop a computational cognitive architecture for transparent share goal and intention reading in heterogenous multi-agent teams; (ii) To validate the architecture for trust and task arbitration mechanism in shared human-robot interaction and trust tasks; (iii) To evaluate the use of transparent and explainable systems for enhanced ethical value of trustworthy humanrobot teams; (iv) To extend the cognitive architecture for ethically-grounded joint decision making. This project builds on previous research, funded by AFSOR, which investigated intention reading and theory of mind capabilities in autonomous social robots under different levels of abstraction for human-robot dyads. The new CASPER++ architecture will allow synergistic behavior in heterogenous team of robots and people, modeling complex social dynamics that make use of several sensor modalities including behavioral (e.g. postural configurations, gaze direction), social (mutual interactions between agents) and linguistic (both explanatory and descriptive utterances) strategies.

Document Details

Document Type
DoD Grant Award
Publication Date
Feb 05, 2025
Source ID
FA86552417047

Entities

People

  • Angelo Cangelosi

Organizations

  • Air Force Office of Scientific Research
  • United States Air Force
  • University of Manchester

Tags

Fields of Study

  • Computer science

Readers

  • Agent-Based Social Robotics and Mobile-Assisted Learning in Virtual Environments.
  • Team-Based Human-Centered Cognitive Task Decision Making and Information Performance.

Technology Areas

  • AI & ML
  • AI & ML - Autonomous Systems
  • Autonomy
  • Autonomy - Autonomous System Control
  • Autonomy - Human-Robot Interaction