Miller Robotic Interface II Manual de usuario descargar pdf (Pagina 10)

In Drury et al. [2007], human-robot awareness is further broken up into five types to aid in

assessing the operator’s SA. The categories are location awareness, activity awareness,

surroundings awareness, status awareness and overall mission awareness. The two categories that

will be mentioned often in this document are location awareness and surroundings awareness.

Location awareness is the operator’s knowledge of where the robot is situated on a larger scale.

For instance, knowing where the robot is from where it started or that it is in room 314.

Surroundings awareness is the knowledge the user has of the robot’s circumstances in a local

sense. This could be the knowledge that there is an obstacle two feet away from the right side of

the robot, or that the area directly behind the robot is completely clear. Location awareness is

good for mission planning for a general idea of where the robot is and where it needs to go.

Surroundings awareness, on the other hand, is more important for maneuvering the robot safely

in a real-time manner.

In a study of safety related problems occurring in a USAR competition, Drury et al. [2003] noted

that “all critical incidents [such as collisions] were due to some type of awareness violation.” A

critical incident, in this case, is a significant event such as the robot bumping an obstacle, a

hardware failure, or the injury of a victim. An awareness violation occurs when “HRI awareness

information that should be provided is not provided.”

Situation awareness is arguably the main factor in completing a remote robot task effectively.

Unfortunately, it is a challenge to design interfaces to provide good SA. Situation awareness

with respect to robots is a recent area of study, so in previous interfaces the emphasis of the

designs have typically not been concerned with providing sufficient SA. For example, two

studies examined twelve separate USAR interfaces [Yanco, Drury, 2002, Scholtz et al., 2004].

Throughout the studies, various things transpired which lead to critical incidents resulting from

poor SA. For instance, as described in Yanco, Drury and Scholtz [2004], “During the first run

[of the user study], the Team B operator moved the robot’s video camera off-center to look at a

victim for identification and also switched to his thermal camera to verify that it was a live

victim. After the victim identification, the operator switched to shared mode to allow the robot

to get out of a tight space with less operator intervention. At this point, the operator forgot that

he had turned his camera to the left. When he switched back to safe mode, he found that the

results of his actions did not correspond to the video image he saw. This confusion resulted in

1 2 ... 5 6 7 8 9 10 11 12 13 14 15 ... 79 80

Sin comentarios

Miller Robotic Interface II Manual de usuario Pagina 10