Effects of Safety Protocols on Unmanned Vehicle Ground Operations

form such as “taxi to this waypoint” or “monitor this area;” these commands are broken down into more specific inputs by the autonomy onboard the vehicle or contained in the operator’s workstation. These systems can be classified into two general forms. The first form is referred to here as Vehicle-based Human Supervisory Control (VHSC), in which supervisors supply commands to a single vehicle at a time, working iteratively through the queue of vehicles awaiting tasks (e.g. [24]). Here, the operator serves as a processing server in the system, managing a queue of vehicles awaiting new instructions that the operator processes at some distribution of service times under a certain queuing policy (e.g. first-in, first out). The larger the number of vehicles in the system, the longer the time between successive interactions with any individual vehicle. For systems with greater than 8-12 vehicles, additional autonomy is often introduced in the form of planning and scheduling algorithms that simultaneously replan all tasks for all vehicles (ex. [25]). These are referred to as System-based Human Supervisory Control (SHSC) systems. The operator typically provides goals and constraints for the scheduling algorithm, which will calculate a new schedule and present it to the operator for review. The operator then has the option of accepting, rejecting, or modifying the schedule before it is transmitted to the vehicles for execution. Once accepted, new task schedules are uploaded for all vehicles, which begin executing their individual task lists. In the context of flight deck operations and models of agent behavior, the important activities of a pilot that must be replicated by these three unmanned vehicle control architectures (GC, VHSC, and SHSC) can be summarized in terms of three key aspects: visual detection of an Aircraft Director that is attempting to provide instructions, correctly interpreting task instructions, and correctly executing the commanded tasks. Each of these corresponds to a related variable: the probability of observing the Director if within the vehicle’s field of view, the probability of correctly