Dimensions of Adjustable Autonomy and Mixed-Initiative Interaction

Abstract

Abstract. Several research groups have grappled with the problem of characterizing and developing practical approaches for implementing adjustable autonomy and mixed-initiative interaction in deployed systems. However, each group takes a little different approach and uses variations of the same terminology in a somewhat different fashion. In this chapter, we will describe some common dimensions in an effort to better understand these important but ill-characterized topics. We are developing a formalism and implementation of these concepts as part of the KAoS framework in the context of our research on policy-governed autonomous systems.

Beyond Cooperative Robotics: The Central Role of Interdependence in Coactive Design

“…7 The fi rst sense, self-suffi ciency, is about the degree to which an entity operates without outside help. For example, a Roomba robot can vacuum a room without assistance.…”

“…For these and other reasons, it is more productive to think about autonomy in terms of multiple task-specific dimensions rather than in terms of a single, unidimensional scale. 7 The perspective in which we view a system can also affect our assessment of autonomy. For example, ambiguity about the term autonomy comes into play in Figure 1.…”

Web Rule Languages to Carry Policies

“…They also legitimatize the behavior of an agent by identifying its liberties and suppressions during the process of authorization. The main advantage in using policies is the possibility to dynamically change the behavior of the system by adjusting the policies without interfering with the internal code of the system [3]. A policy-aware system can be simply conducted to act based on the role of the requesting entities and or the context of its performance.…”

Implementing Policy Management through BDI

“…Bradshaw et al [11] discuss the notion of policy in the context of autonomous entities that cannot always be trusted to regulate their own behaviour appropriately, because poorly designed, buggy or malicious. In this context, they introduce KAoS, a policy language that allows them to externally adjust the bound of autonomous behaviour in order to ensure safety and effectiveness of the system.…”