Predictive Model for Human-Unmanned Vehicle Systems

Abstract

Advances in automation are making it possible for a single operator to control multiple unmanned vehicles (UVs). However, the complex nature of these teams presents a difficult and exciting challenge for designers of human-UV systems. To build such systems effectively, models must be developed that describe the behavior of the human-UV team and that predict how alterations in team composition and system design will affect the system's overall performance. In this paper, we present a method for modeling human-UV systems consisting of a single operator and multiple independent UVs. Via a case study, we demonstrate that the resulting models provide an accurate description of observed human-UV systems. Additionally, we demonstrate that the models can be used to predict how changes in the human-UV interface and the UVs' autonomy alter the system's performance.

Computing the Effects of Operator Attention Allocation in Human Control of Multiple Robots

“…RESCU was initially used to analyze how changes in team size, robot-autonomy characteristics, and the human-robot interface affect the HMRS's performance [19], [20]. In this paper, we are interested in how OAASs, or the way operators allocate their attention among multiple robots, affect HMRS's effectiveness.…”

“…Previous studies have demonstrated that a stochastic discreteevent simulation can provide reasonably good predictions of how changes in team size, robot autonomy, and human-robotinterface characteristics alter system effectiveness [19], [20], [24]. In this paper, we evaluate the ability of this same modeling method to estimate the effects of OAASs on system effectiveness in RESCU.…”

“…Crandall et al [19] model an HMRS with the fourtuple M = (II, N I, ST , OAAS) of stochastic structures. We briefly describe each stochastic structure 1) Interaction impact.…”

“…In this paper, we address these questions using the HMRS and user study described in Section II. In Section III, we review a previously introduced computational modeling method for HMRSs [19] and introduce how this model can be used to compute the effects of OAASs. We also show how the model can be used to estimate an "optimal" OAAS under certain assumptions.…”

Evaluation of USV system autonomy based on group decision making

“…The human-centric method is to evaluate the system autonomy on the basis of human performance or human model. Crandall proposed analogous concepts to allocate operator ' s attention, in order to maximize team performance with the given appropriate threshold of human workload [4], [5]. The robot-centric method builds the system autonomy evaluation on the capa bility of autonomous vehicles under the situations.…”