Marcus V. Santos and Paulo E. Santos
This paper extends a logic-based framework for robot scene interpretation by using the notion of path semantics. The main contributions of this approach are two fold. First, a logic language designed to account for the phenomenon of state changes in databases is extended with the concept of abduction and further applied on the task of robot sensor-data assimilation. Secondly, the present framework provides the theoretical foundations for symbolically interpreting long sequences of transitions of sensor data. In practice, each snapshot taken by the robot camera represents a database state. Transitions (i.e., changes) between consecutive snapshots are modeled by changes in the database and interpreted by a state transition oracle, which encodes axioms about commonsense spatial reasoning.