Adaptive Automation for Human Supervision of Multiple Uninhabited Vehicles: Effects on Change Detection, Situation Awareness, and Mental Workload

Abstract

Human operators supervising multiple uninhabited air and ground vehicles (UAVs and UGVs) under high task load must be supported appropriately in context by automation. Two experiments examined the efficacy of such adaptive automation in a simulated high workload reconnaissance mission involving four subtasks: (a) UAV target identification; (b) UGV route planning; (c) communications, with embedded verbal situation awareness probes; and (d) change detection. The results of the first "baseline" experiment established the sensitivity of a change detection procedure to transient and nontransient events in a complex, multi-window, dynamic display. Experiment 1 also set appropriate levels of low and high task load for use in Experiment 2, in which three automation conditions were compared: manual; static automation, in which an automated target recognition (ATR) system was provided for the UAV task; and adaptive automation, in which individual operator change detection performance was assessed in real time and used to invoke the ATR if and only if change detection accuracy was below a threshold. Change detection accuracy and situation awareness were higher and workload was lower for both automation conditions compared to manual performance. In addition, these beneficial effects on change detection and workload were significantly greater for adaptive compared to static automation. The results point to the efficacy of adaptive automation for supporting the human operator tasked with supervision of multiple uninhabited vehicles under high workload conditions.

Co-Adaptive Aiding and Automation Enhance Operator Performance

“…In addition, limited automation was invoked that paired vehicles and targets, subject to operator override. This builds on a similar study performed by Parasuraman, Cosenzo, and De Visser (2009) by testing attentional aiding in addition to automation.…”

Development of an Initial Model of Human-Automatio

“…Their results showed that the participant's performance with adaptive automation was better compared to fully manual operation, but that adaptive automation also was more effective when supporting less cognitively demanding stages of information processing (e.g., information acquisition and action implementation) than more cognitively demanding stages (e.g., information analysis and decision making). Parasuraman et al (2009) investigated the effect of an adaptive automation system for a human operator supervising multiple unmanned vehicles. The adaptive automation was adjusted based on the human operator's performance in a change detection task.…”

“…Though some studies have successfully employed adaptive automation using a specific triggering condition (e.g., Parasuraman et al, 2009), there has not been a great deal of research on the relative merits of the various triggering conditions for adaptive automation (Kaber, 2012;), yet their importance to human-automation collaboration is widely acknowledged (Kaber, 2012;Sheridan & Parasuraman, 2005). Since there are relative benefits and disadvantages to each type of techniques mentioned in Section 2.2.7, several authors have recommended the use of hybrid methods to ensure that automation is initiated (or changed) when it should be (Parasuraman et al, 1996;Sheridan & Parasuraman, 2005).…”

“…This disorientation can cause transient performance decrements. The cognitive cost of initiating automation cannot outweigh its benefits or operator will not use it (Parasuraman et al, 2009). Thus when operator initiated triggers are used, designer should seek to minimize the workload associated with it.…”

“…However, a number of studies have shown that the dynamic allocation of functions and tasks between automated systems and operators can positively and negatively affect HAC and system performance (e.g, Parasuraman et al, 2009;Clamann et al, 2002). Another example is if the adaptation mechanism fails and functions can no longer be dynamically allocated or the operator is unable to regain control of key safety functions, this will not only adversely affect the operator's trust and future use of automation, but also overall system performance.…”