238
Sensemaking in Safety Critical and Complex Situations
Note that this form of characterization of response time is independent of the crew
residing on board or in an RCC, although the times will likely be different in the two
situations due to differences in equipment and access to detailed sensor or situation
information.
Today, most ships have an autopilot or a track pilot. Open sea with no ship or other
objects in the vicinity allows the officer of the watch to be away from the bridge for
relatively long periods. With the above definition of automation and autonomy, this
makes a ship controlled by an autopilot autonomous with respect to the process of
keeping a steady speed and course in open sea. Autonomy in this context may seem
counter-intuitive but is related to the low abstraction level on the involved functions.
In the work presented here, four levels of functional abstraction are used:
1. System objectives: This is the highest abstraction level and is associated
with generating the objectives for the design of the autonomous ship system.
This may in some cases be static or at least have a long horizon, e.g. transport available cargo between ports A and B. This will be the basis for the
design of the control system.
2. Planning: This is also a high abstraction level and will normally be an
external input to the ship control system. It is related to the overall planning
of the voyage or mission within the constraints of the system objectives. In
most cases, this is expected to be supplied by the ship operators.
3. Goal based: These can be seen as a sequence of process goals for the autonomous ship system. Each goal is expected to be associated with one or more
processes or tasks. This is also the most likely abstraction level for commands to the autonomous ship system.
4. Functional: This is specific instructions to a function, such as an autopilot.
This is normally on a low abstraction level and will not normally be used as
commands to the autonomous ship system.
As exemplified above, autonomy on the functional level already has been developed
and is used in well-controlled environments such as autopilots on high sea or car cruise
controls on highways. The goal of the work presented here is to contribute methods to
extend autonomy to higher abstraction levels while giving human operators a better
understanding of the capabilities and limitations of the automation system.
SENSEMAKING AND TRUST IN AUTOMATION
One basic issue in the HAI is its ability to support a proper level of operator’s trust
in the automation system ( Lee & See 2004). This should not be too low, leading to
disuse of the automated functions and neither should it be too high, leading to overreliance and misuse of the automation. In addition, the operator must be able to make
sense of the relationship between automation, his or her responsibilities and the situation at hand. The latter could be called “ sensemaking”, which can be defined as “ a
motivated, continuous effort to understand connections ( which can be among people,
places, and events) in order to anticipate their trajectories and act effectively” ( Klein
et al. 2006).
Sensemaking in Safety Critical and Complex Situations
Note that this form of characterization of response time is independent of the crew
residing on board or in an RCC, although the times will likely be different in the two
situations due to differences in equipment and access to detailed sensor or situation
information.
Today, most ships have an autopilot or a track pilot. Open sea with no ship or other
objects in the vicinity allows the officer of the watch to be away from the bridge for
relatively long periods. With the above definition of automation and autonomy, this
makes a ship controlled by an autopilot autonomous with respect to the process of
keeping a steady speed and course in open sea. Autonomy in this context may seem
counter-intuitive but is related to the low abstraction level on the involved functions.
In the work presented here, four levels of functional abstraction are used:
1. System objectives: This is the highest abstraction level and is associated
with generating the objectives for the design of the autonomous ship system.
This may in some cases be static or at least have a long horizon, e.g. transport available cargo between ports A and B. This will be the basis for the
design of the control system.
2. Planning: This is also a high abstraction level and will normally be an
external input to the ship control system. It is related to the overall planning
of the voyage or mission within the constraints of the system objectives. In
most cases, this is expected to be supplied by the ship operators.
3. Goal based: These can be seen as a sequence of process goals for the autonomous ship system. Each goal is expected to be associated with one or more
processes or tasks. This is also the most likely abstraction level for commands to the autonomous ship system.
4. Functional: This is specific instructions to a function, such as an autopilot.
This is normally on a low abstraction level and will not normally be used as
commands to the autonomous ship system.
As exemplified above, autonomy on the functional level already has been developed
and is used in well-controlled environments such as autopilots on high sea or car cruise
controls on highways. The goal of the work presented here is to contribute methods to
extend autonomy to higher abstraction levels while giving human operators a better
understanding of the capabilities and limitations of the automation system.
SENSEMAKING AND TRUST IN AUTOMATION
One basic issue in the HAI is its ability to support a proper level of operator’s trust
in the automation system ( Lee & See 2004). This should not be too low, leading to
disuse of the automated functions and neither should it be too high, leading to overreliance and misuse of the automation. In addition, the operator must be able to make
sense of the relationship between automation, his or her responsibilities and the situation at hand. The latter could be called “ sensemaking”, which can be defined as “ a
motivated, continuous effort to understand connections ( which can be among people,
places, and events) in order to anticipate their trajectories and act effectively” ( Klein
et al. 2006).