The assumption of complete domain knowledge is not warranted for robot planning and decision-making in the real world . The problem is more challenging than partial observability in the sense that agent is unaware of certain knowledge, in contrast to it being partiallyobservable: the difference between known unknowns and unknown unknowns . In thiswork, we formulate it as the problem of Domain Concretization, an inverseproblem to domain abstraction . We tested our approach on IPC domains and a simulatedrobotics domain where incompleteness was introduced by removing domain features from the complete model . Results show that our planning algorithm increases theplan success rate without impacting the cost much. Results were found to increase the planning success rate by increasing the plan success rate with an online version of it to improve its search time. In addition to a standard search formulation in the model-space, we also propose a sample-based search method and also an online versions of it in addition to the

Author(s) : Akshay Sharma, Piyush Rajesh Medikeri, Yu Zhang

Links : PDF - Abstract

Code :

Keywords : domain - planning - rate - success - search -

Leave a Reply

Your email address will not be published. Required fields are marked *