Mixed-Initiative Planning and Execution for Multiple Drones in Search and Rescue Missions

We present a mixed-initiative planning and execution system for human multi-drones interaction during search and rescue missions. The proposed system should allow a single operator to supervise and orchestrate the operations of a set of UAVs by means of a natural multimodal communication. In particular, we consider the task of searching for missing persons in a realworld alpine scenario. In this context, we assume that the human operator is an alpine rescuer, involved in the scene and co-located with the drones, hence not fully dedicated to the robotic platforms, but only able to provide sparse and sketchy interventions. This scenario requires a framework that supports adjustable autonomy, from explicit teleoperation to a complete autonomy, and an effective and natural mixed-initiative interaction between the human and the robotic team. In this paper, we illustrate the domain and the overall framework discussing the system at work in a simulated case study.