Improving Human-Robot Teamwork in Urban Search and Rescue Through Episodic Memory of Prior Collaboration
arXiv:2606.18836v1 Announce Type: cross Abstract: Effective human-robot teamwork requires robots to adapt to partners, situations, and task dynamics from the start of an interaction. In the MATRX Urban Search and Rescue (USAR) environment, people can externalize collaboration patterns (CPs) they discover during teamwork through a chat and reflection interface. We study whether a robot can use such prior team experience to become a better teammate in future interactions. To this end, we represent historical CPs as knowledge-graph episodic memories and use graph representation learning with a no
The increasing sophistication of AI and robotics necessitates more seamless human-robot collaboration, and current research is focused on developing methods for robots to learn and adapt from prior interactions.
Improving human-robot teamwork is crucial for deploying autonomous systems in complex, high-stakes environments like urban search and rescue, enhancing operational efficiency and safety.
Robots will become more adaptive and effective teammates, learning from past collaborative experiences rather than starting fresh, leading to faster integration and better performance in dynamic situations.
- · Robotics companies
- · Emergency services
- · AI software developers
- · Defense and security sectors
Enhanced human-robot teaming capabilities will improve the efficiency and safety of dangerous operations.
The development of adaptive robots will accelerate their deployment into more diverse and unconstrained environments.
This could lead to a broader societal acceptance and integration of autonomous systems in everyday tasks, blurring the lines between human and machine roles.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.AI