Design principles for biologically inspired cognitive robotics

作者：Krichmar Jeffrey L^*

来源：Biologically Inspired Cognitive Architectures, 2012, 1: 73-81.

DOI：10.1016/j.bica.2012.04.003

• Cognition
• Computational neuroscience
• Embodiment
• Neurorobots

The goals of cognitive robotics are to better understand cognition through the construction of physical artifacts, and to create practical systems that demonstrate cognitive capabilities. I believe for cognitive robotics to move forward, a balanced approach that emphasizes the interaction of brain, body, and environment is necessary. In general, cognitive robots and cognitive architectures focus too much on brain control, and overlook the contributions of morphology to intelligent behavior. On the other hand, the behavior based robotics approach is unbalanced in the opposite direction. For cognitive robotics to move forward, these disparate research communities need to come into balance. The materials, morphology, sensors, actuators, and the nervous system should be balanced and coordinated in their action. In their book, "How the body shapes the way we think: A new view of intelligence" (MIT Press, 2007), Pfeifer and Bongard have suggested that intelligent agents should follow a set of design principles that highlight the importance of embodiment and physical interaction with the environment. In the present paper, I apply each of these principles to biologically inspired cognitive robotics and suggest how the field can shift toward better cognitive architectures by adherence to these principles.

• 出版日期2012-7