Keynote speech IV

Summary form only give. This talk introduces the recent progress and development of computational intelligence for Robots That Think (RTT). Considering the next generation of robotics [1-5], robot intelligence can be classified into six categories such as cognitive intelligence, social intelligence, behavioural intelligence, ambient intelligence, collective intelligence and genetic intelligence. The robot intelligence can be realized through intelligence technology (IT) using cognitive architecture in which computational intelligence (CI) is embedded. This talk focuses on cognitive architecture to implement IT for RTT. The IT is based on the Mechanism of Thought (MoT) using CI embedded in the cognitive architecture. To make RTT deliberately interact with the surrounding environment like living creatures, the behavior selection method mimicking living creatures\´ thinking mechanism is needed. For this purpose, the probability-based mechanism of thought was proposed based on probabilistic knowledge links between input (assumed fact) and target (behavior) symbols for reasoning. However, real intelligent creatures including human beings select a behavior based on the multi-criteria decision making process considering the degree of consideration (DoC) for input symbols, i.e. will and context symbols, in their memory. In this talk, the DoC-based Mechanism of Thought (DoC-MoT) is introduced, which is applied to the behavior selection of RTT [6], [7]. The knowledge links between input and behavior symbols are represented by the partial evaluation values of behaviors over each input symbol, and the degrees of consideration for input symbols are represented by the fuzzy measures [8]. The DoC-MoT selects a behavior through global evaluation by the fuzzy integral, as a multi-criteria decision making process, of knowledge link strengths with respect to the fuzzy measure values. As a couple of applications, this talk introduces a synthetic character "Rity" in the 3D virtual envi- onment and the gaze control of the humanoid robot HanSaRam (HSR) based on DoC-MoT [9], [10], [11].