Intelligent control architecture for mobile robot based on fusion methods

摘要

An autonomous mobile robot must provide various functions such as perception, location and control in order to understand the environment and react accordingly to the task to be performed. If the environment is barely known, then the robot must have decisional abilities that allow an understanding of the mission and a determination of the appropriate actions to be taken. The authors propose a multisensor perception system allowing the control of an autonomous mobile robot when uncertainty and incomplete knowledge are the major characteristics of the environment. Their robot control architecture is designed to support capabilities needed by many autonomous mobile robot systems, including concurrent planning, execution, perception and reacting to changes in the environment. This architecture is based on the combination of local and absolute location (fusion methods) for an autonomous mobile robot. The robot task is described by an AND/OR graph in a propositional logic and decomposed in primitive actions. The control is in charge of achieving each task in the best possible way, taking into account the context and sensory feedback.