Abstract. Foraging tasks, such as search and rescue or reconnaissance, in which UVs are either relatively sparse and unlikely to interfere with one another or employ automated path planning, form a broad class of applications in which multiple robots can be controlled sequentially in a round-robin fashion. Such human-robot systems can be described as a queuing system in which the human acts as a server while robots presenting requests for service are the jobs. The possibility of improving system performance through well-known scheduling techniques is an immediate consequence. Unfortunately, real human-multirobot systems are more complex often requiring operator monitoring and other ancillary tasks. Improving performance through scheduling (jobs) under these conditions requires minimizing the effort expended monitoring and directing the operator's attention to the robot offering the most gain. Two experiments investigating scheduling interventions are described. The first compared a system in which all anomalous robots were alarmed (Open-queue), one in which alarms were presented singly in the order in which they arrived (FIFO) and a Control condition without alarms. The second experiment employed failures of varying difficulty supporting an optimal shortest job first (SJF) policy. SJF, FIFO, and Open-queue conditions were compared. In both experiments performance in directed attention conditions was poorer than predicted. A possible explanation based on effects of volition in task switching is proposed.
Abstract-Decision aiding sometimes fails not because following guidance would not improve performance but because humans have difficulty in following guidance as it is presented to them. This paper presents a new analysis of data from multi-robot control experiments in which guidance in a demonstrably superior robot selection strategy failed to produce improvement in performance.We had earlier suggested that the failure to benefit might be related to loss of volition in switching between robots being controlled. In this paper we present new data indicating that spatial, and hence cognitive proximity, of robots may play a role in making volitional switches more effective.Foraging tasks, such as search and rescue or reconnaissance, in which UVs are either relatively sparse and unlikely to interfere with one another or employ automated path planning, form a broad class of applications in which multiple robots can be controlled sequentially in a round-robin fashion. Such human-robot systems can be described as a queuing system in which the human acts as a server while robots presenting requests for service are the jobs. The possibility of improving system performance through well-known scheduling techniques is an immediate consequence. Two experiments investigating scheduling interventions are described. The first compared a system in which all anomalous robots were alarmed (Alarm), one in which alarms were presented singly in the order in which they arrived (FIFO) and a Control condition without alarms. The second experiment employed failures of varying difficulty supporting an optimal shortest job first (SJF) policy. SJF, FIFO, and Alarm conditions were compared. In both experiments performance in directed attention conditions was poorer than predicted. This paper presents new data comparing the spatial proximity in switches between robots selected by the operator (Alarm conditions) and those dictated by the system (FIFO and SJF conditions).
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.