Simulation theory of mind for heterogeneous human-robot teams

Area of Science:

• Robotics
• Artificial Intelligence
• Human-Robot Interaction

Background:

• Coordinating multiple robots and humans in complex tasks is challenging.
• Existing systems struggle with dynamic environments and heterogeneous teams.

Purpose of the Study:

• Develop a real-time distributed architecture for collaborative task execution.
• Enable dynamic coordination between humans and heterogeneous robots.

Main Methods:

• Simulation Theory of Mind (ToM)-inspired approach.
• Novel concurrent coordination for human-robot teams.
• Pipeline for automatic grasping of unknown objects.
• Continuous-valued metric for robot task capability assessment.

Main Results:

• Validated architecture with heterogeneous teams (robots only, human-robot).
• Demonstrated adaptive, real-time task execution in varying environmental conditions.
• Showcased dynamic task allocation based on changing robot capabilities.

Conclusions:

• The proposed architecture effectively supports adaptive, real-time collaborative task execution.
• Dynamic task allocation by heterogeneous human-robot teams is feasible for complex tasks.
• The system handles hierarchical task representations and multiple constraints.