Fredrik Heintz and Patrick Doherty
Any autonomous system embedded in a dynamic and changing environment must be able to create qualitative knowledge and object structures representing aspects of its environment on the fly from raw or preprocessed sensor data in order to reason qualitatively about the environment. These structures must be managed and made accessible to deliberative and reactive functionalities which are dependent on being situationally aware of the changes in both the robotic agent’s embedding and internal environment. DyKnow is a software framework which provides a set of functionalities for contextually accessing, storing, creating and processing such structures. In this paper, we focus on the use of DyKnow in supporting the representation and reasoning about dynamic objects such as road vehicles in the external environment of an autonomous unmanned aerial vehicle. The representation of complex objects generally consists of simpler objects with associated features that are related to each other via linkages. These linkage structures are constructed incrementally as additional sensor data is acquired and integrated with existing structures. The resulting linkage structures represent complex objects at many levels of abstraction. Many issues related to anchoring and symbol grounding can be approached by taking advantage of the versatility of these linkage structures. Examples are provided in the paper using an experimental UAV research platform.