DOI:
Abstract:
To program a robot to solve a simple shape-sorter puz- zle is trivial. To devise a Cognitive System Architec- ture, which allows the system to find out by itself how to go about a solution, is less than trivial. The development of such an architecture is one of the aims of the COSPAL project, leading to new techniques in vision based Artificial Cognitive Systems, which al- low the development of robust systems for real dynamic environments. The systems developed under the project itself remain however in simplified scenarios, likewise the shape-sorter problem described in the present paper. The key property of the described system is its robust- ness. Since we apply association strategies of local fea- tures, the system behaves robustly under a wide range of distortions, as occlusion, colour and intensity changes. The segmentation step which is applied in many sys- tems known from literature is replaced with local as- sociations and view-based hypothesis validation. The hypotheses used in our system are based on the antic- ipated state of the visual percepts. This state replaces explicit modeling of shapes. The current state is cho- sen by a voting system and verified against the true vi- sual percepts. The anticipated state is obtained from the association to the manipulator actions, where rein- forcement learning replaces the explicit calculation of actions. These three differences to classical schemes allow the design of a much more generic and flexible system with a high level of robustness. On the technical side, the channel representation of in- formation and associative learning in terms of the chan- nel learning architecture are essential ingredients for the system. It is the properties of locality, smoothness, and non-negativity which make these techniques suitable for this kind of application. The paper gives brief descrip- tions of how different system parts have been imple- mented and show some examples from our tests.