Improving Human Performance in Multi-Human Multi-Robot Interaction

Patel, Jayam U.

Etd

Improving Human Performance in Multi-Human Multi-Robot Interaction

Público

Multi-robot systems are envisioned to assist humans in complex missions such as interplanetary exploration, ocean restoration, underground mining, and forest firefighting. In these missions, along with robot autonomy, human input is required to supervise the operation of the robots, and assign and prioritize the tasks assigned to the robots. By their very nature, multi-robot systems are complex systems composed of many interacting entities. As such, these systems exceed the typical human attention span, which several studies place between 7±2 entities in laboratory conditions. A natural approach to improving human performance is to relieve the burden of individual operators by conceiving supervisory control schemes in which multiple humans cooperate. However, with multiple human users in the system, additional challenges arise. These challenges include unbalanced workload, inhomogeneous awareness, and conflict among operators. This limits the performance of the operators, typically measured in terms of workload, situational awareness, and trust in the system. In my work, I show that the performance of human users can be improved with mixed granularity of control, increased transparency, and human communication. Mixed granularity of control enables an operator to control high-level task goals such as modifying the environment, as well as lower-level goals such as interacting with an individual robot or a group of robots. Increased transparency aids an operator to understand other operators’ and robots’ actions. Operators communicate directly, through verbal and non-verbal communication, and indirectly, through information transparency, to understand other operators’ actions and intentions. The main technological outcomes of my work are a novel mixed-reality interface for proximal interaction and a novel cloud-based interface for remote interaction. These interfaces enable multiple operators to collaborate with multiple robots in local and remote environments. The main scientific outcome of my work is investigating the effects of mixed granularity of control, information transparency, and human communication on the operators’ performance. My experimental evaluation consists of 8 user studies involving a total of 122 participants, in which I analyze operator workload, awareness, trust, and performance.

Creator