Automation in Handling Uncertainty in Semantic-knowledge based Robotic Task-planning by Using Markov Logic Networks

Generating plans in real world environments by mobile robot planner is a challenging task due to the uncertainty and environment dynamics. Therefore, task-planning should take in its consideration these issues when generating plans. Semantic knowledge field has been planned as a source of information for deriving implicit information and generating semantic procedure. This paper extend the SemanticKnowledge Based (SKB) plan generation to take into account the uncertainty in accessible of objects, with their type and property, and proposes a new approach to construct plans based on probabilistic values which are resultant from Markov Logic Networks (MLN). An MLN module is established for probabilistic knowledge and inferencing jointly with semantic information to provide a basis for plausible learning and reasoning armed forces in at the bottom of of robot task-planning. In addition, an algorithm has been devised to construct MLN from semantic information. By provided that a means of model uncertainty in system archit ecture, task-planning serves as a behind tool for programmed applications that can profit from probabilistic conclusion within a semantic domain. This come up to is illustrate by means of test scenario run in a domestic environment using a mobile robot.