Page 27 - Innovations in Intelligent Machines
P. 27

Predicting Operator Capacity for Supervisory Control of Multiple UAVs  15
                           3.1 Wait Times

                           Modeling interaction and neglect times are critical for understanding human
                           workload in terms of overall management capacity. However, there remains
                           an additional critical variable that must be considered when modeling human
                           control of multiple robots, regardless of whether they are on the ground or in
                           the air, and that is the concept of Wait Time (WT). In HSC tasks, humans
                           are serial processors in that they can only solve a single complex task at a time
                           [3, 4], and while they can rapidly switch between cognitive tasks, any sequence
                           of tasks requiring complex cognition will form a queue and consequently wait
                           times will build. Wait time occurs when a vehicle is operating in a degraded
                           state and requires human intervention in order to achieve an acceptable level
                           of performance. In the context of a system of multiple vehicles or robots, wait
                           times are significant in that as they increase, the actual number of vehicles that
                           can be effectively controlled decreases, with potential negative consequences
                           on overall mission success.
                              Equation 2 provides a formal definition of wait time. It categorizes total
                           system wait time as the sum of the interaction wait times, which are the
                           portions of IT that occur while a vehicle is operating in a degraded state
                           (WTI), wait times that result from queues due to near-simultaneous arrival of
                           problems (WTQ), plus wait times due to operator loss of situation awareness
                           (WTSA). An example of WTI is the time that an unmanned ground vehicle
                           (UGV) idly waits while a human replans a new route. WTQ occurs when a
                           second UGV sits idle, and WTSA accumulates when the operator doesn’t even
                           realize a UGV is waiting. In (2), X equals the number of times an operator
                           interacts with a vehicle while the vehicle is in a degraded state, Y indicates the
                           number of interaction queues that build, and Z indicates the number of time
                           periods in which a loss of situation awareness causes a wait time. Figure 2
                           further illustrates the relationship of wait times to interaction and neglect
                           times.
                              Increased wait times, as defined above, will reduce operator capacity, and
                           Equation 3 demonstrates one possible way to capture this relationship. Since



                                                             Robot 1
                             Robot 1                                     IT
                                    WTQ 1  IT`
                             Robot 2                         Robot 2
                                        WTQ     IT``                       WTSA  IT```
                                           2                 Robot 3
                             Robot 3

                                         IT+NT                             IT+NT
                                           (a)                              (b)

                              Fig. 2. Queuing wait times (a) versus situational awareness wait times (b)
   22   23   24   25   26   27   28   29   30   31   32