ePartners for Dynamic Task Allocation and Coordination

This dissertation promotes a focus shift from automation extending human capabilities to automation partnering with the human. Highly automated systems tend to be incompatible with humans, leading to degradations in performance that may lead to serious problems in high-risk professional domains, such as the Navy and Urban Search and Rescue. This dissertation shows how an electronic partner (ePartner) can act as task allocator and coordination facilitator to substantially improve the human-automation teamwork in these high-risk professional domains. Adaptive automation concerns the dynamic allocation of tasks between the human and the automation, based on ePartner’s proposal or decision. The effects of adaptive automation using the object-oriented task model and the hybrid triggering mechanism have been investigated with Navy officers. The results revealed a large performance improvement, especially in the more complex condition. This latter finding makes adaptive automation a likely candidate to incorporate in future combat management systems as military scenarios are expected to become complex (asymmetrical) while manning reduction initiatives stress the military system. Observability displays are proposed to supports the coordination process while working in a distributed setting. A number of experimental studies have demonstrated that observability displays improve backing-up behavior, performance, and resilience while coordinating activities compared with conventional communication technologies. These results clearly show the power of observability displays as a mean to support the team process between human actors and have the potential to support Urban Search & Rescue missions.MediamaticsElectrical Engineering, Mathematics and Computer Scienc