Trust in co-robotics, humans and robots working side by side
First, if the robot is meant to understand natural language and take commands
from a human operator, its capacity to understand language correctly is an
important component. The person needs to be able to trust that the robot will
understand the commands given to it. The robot's ability to assess its current
state by sensing its environment are crucial considerations as well.
Second, if the robot is giving information to the person, the person has to trust this
information to be able to use it in decision making. Hence, the robot's ability to
use information gathered from different sensors in problem solving are relevant
to deciding whether to trust the robot or not.
Third, the robot should be able to accomplish tasks, such as grabbing objects
properly, lifting and positioning as needed.
The first aspect describes the sensing ability of the robot, the second aspect
involves its problem solving ability and the third aspect refers to the physical
ability of the robot. These aspects of capability are not sufficient to “trust” the
robot for a specific action. The person needs to believe that the robot is designed
in such a way that it will not cause bodily harm (by accidentally bumping into
people, for example). This involves not only the robot's capability, but also a
person's beliefs about the intentions behind its design.
Let us now look at the situation from the robot's perspective. The robot also has
to trust the human team member in various ways.
When should the robot follow the person's commands? Does the person know
what he is talking about? What if the command will hurt the person or is based
on an unsound move?
If the robot is cooperating with the person, it has to take into account that people
are not always as careful as robots. What if the person is not paying attention?
What if he is getting tired? What if he will not be able to hold a heavy part in