This paper describes an architecture that combines the complementary strengths of declarative programming and probabilistic graphical models to enable robots to represent, reason with, and learn from, qualitative and quantitative descriptions of uncertainty and knowledge. An action language is used for the low-level (LL) and high-level (HL) system descriptions in the architecture, and the definition of recorded histories in the HL is expanded to allow prioritized defaults. For any given goal, tentative plans created in the HL using default knowledge and com-monsense reasoning are implemented in the LL using probabilistic algorithms, with the corresponding observations used to update the HL history. Tight cou-pling between the two levels enables automatic selection of relevant variables and generation of suitable action policies in the LL for each HL action, and supports reasoning with violation of defaults, noisy observations and unreliable actions in large and complex domains. The architecture is evaluated in simulation and on physical robots transporting objects in indoor domains; the benefit on robots is a reduction in task execution time of 39 % compared with a purely probabilistic, but still hierarchical, approach. 1

This paper describes an architecture that combines the com-plementary strengths of declarative programming and proba-bilistic graphical models to enable robots to represent, reason with, and learn from, qualitative and quantitative descriptions of uncertainty and knowledge. An action language is used for the low-level (LL) and high-level (HL) system descrip-tions in the architecture, and the definition of recorded histo-ries in the HL is expanded to allow prioritized defaults. For any given goal, tentative plans created in the HL using default knowledge and commonsense reasoning are implemented in the LL using probabilistic algorithms, with the corresponding observations used to update the HL history. Tight coupling between the two levels enables automatic selection of rele-vant variables and generation of suitable action policies in the LL for each HL action, and supports reasoning with vio-lation of defaults, noisy observations and unreliable actions in large and complex domains. The architecture is evaluated in simulation and on physical robots moving objects to specific places in indoor domains; the benefit on robots is a reduc-tion in task execution time of 39 % compared with a purely probabilistic, but still hierarchical, approach. 1

This paper describes an architecture that combines the com-plementary strengths of declarative programming and proba-bilistic graphical models to enable robots to represent, reason with, and learn from, qualitative and quantitative descriptions of uncertainty and knowledge. An action language is used for the low-level (LL) and high-level (HL) system descriptions in the architecture, and the definition of recorded histories in the HL is expanded to allow prioritized defaults. For any given goal, tentative plans created in the HL using default knowl-edge and commonsense reasoning are implemented in the LL using probabilistic algorithms, with the corresponding obser-vations used to update the HL history. Tight coupling be-tween the two levels enables automatic selection of relevant variables and generation of suitable action policies in the LL for each HL action, and supports reasoning with violation of defaults, noisy observations and unreliable actions in large and complex domains. The architecture is evaluated in sim-ulation and on physical robots transporting objects in indoor domains; the benefit on robots is a reduction in task execu-tion time of 39 % compared with a purely probabilistic, but still hierarchical, approach. 1