Perception planning as a distributed decision-making process

It is desirable that the perception system for an autonomous vehicle is implemented as a reactive process which has the necessary expertise to autonomously adjust its perception strategies according to the state of achievement of local perception and navigation goals, and to suggest alterations to vehicle speed or attitude to improve safety, or to maximize the collection of relevant environment data. In this paper, a perception planner is presented, which is based in a distributed decision-making architecture, used to deal with the complex set of decisions involved in the design of such a system. >