Impact of AI on the HRI Dynamic in Search and Rescue Operations using UAV Swarms

Artificial intelligence (AI) offers significant benefits in search and rescue applications by enhancing the efficiency and effectiveness of the search. However, an over-reliance on AI can hinder the operation due to biases embedded in the underlying algorithms. This partiality, if left un-monitored, can pose a risk to the safety of those in need of disaster relief. typically manifests into inaccuracies in the decision-making processes and if not carefully monitored can cause larger issues. This paper extends the knowledge presented in a previous work, which presents the design and modeling of search and rescue operations using unmanned aerial vehicle (UAV) swarms. The search and rescue operations using UAV swarms (SAROUS) model is modernized by introducing artificial intelligence to the operation framework that ideally will reduce the operator workload. The results suggest that a trade-off exists between operator fatigue, task correctness, and operation timeliness. This is due to the amount of work generated and needing to be processed manually by the operator rather than automatically by an AI algorithm. Both the operator and AI contribute to some level of error when processing tasks, which influence the overall effectiveness of the operation. The performance of the operator and swarm is estimated under this new operation design and the dynamic of the human-robot interaction is analyzed.
^1
1
Research was sponsored by the Army Research Laboratory and was accomplished under Cooperative Agreement Number W911NF-22-2-0199. The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies, either expressed or implied, of the Army Research Laboratory or the U.S. Government. The U.S. Government is authorized to reproduce and distribute reprints for Government purposes notwithstanding any copyright notation herein.