Human Interaction with Autonomous Systems in Complex Environments
Papers from the AAAI Spring Symposium
David Kortenkamp and Michael Freed, Cochairs
Autonomous systems can greatly enhance human effectiveness in complex environments by handling routine or cognitively challenging operations. However, autonomy changes the nature of human tasks and can introduce new risks. Mitigating those risks raises issues in autonomous systems research such as: 1) How to accept task inputs from humans; 2) How to adjust the level of autonomy and/or change the distribution of roles and responsibilities between autonomous systems and humans; 3) How to model humans and their tasks and to what level of detail and; 4) How to facilitate human understanding of the goals, tasks and contexts of autonomous systems to reduce the potential that anomalies would lead to unexpected responses from the system or inappropriate responses by the human.
How do we make people more effective and safe in performing tasks in cooperation with an autonomous system? It is our view that effective human interaction with autonomous systems requires more than just good user interface design. It involves substantial challenges in the design of the autonomous systems themselves and in the representation and use of the cognitive models underpinning human interaction with autonomous systems. Specifically, it affects the design requirements for autonomous system functions ranging from planning and modeling to natural language and intent inference.
The objective of this symposium was to promote technology development for improved human-autonomy interaction by facilitating collaboration between researchers in autonomy and researchers in human-computer interaction. It was especially interested in technology and case studies relevant to complex, applied environments in which people interact with autonomous systems regularly and in-depth. Such systems include autonomous control of buildings or spacecraft, robots that interact with people, and software for assisting complex human tasks, such as logistics planning.