Abstract:Rat cyborg is a rat which can receive cerebral commands and act as human wants. The navigation task is a widely used scenario in which the rat cyborg is required to walk along a specified path according to commands. Mostly the commands are given by human operators, but there's a need for automatic control which is quite challenging since several modules need to be elaborately combined together to work as a whole, among which the monitoring system and the commands model are two major components. Previously, few attempts were made for behavior monitoring of rat cyborg. Existing works often implement an over-simple rat information extractor which is capable of giving only a few parameters of the rat cyborg, which greatly limited the performance of their automatic control accuracy. In response to this requirement, we develop a monitoring system that is capable of giving detailed motion parameters and accurate body postures of the rat cyborg. We explore the possibility of recognizing rat postures using shape information, which is described by the powerful Zernike moments. We propose several simple shape descriptors which are fast to compute and achieve acceptable performance.

A Posture Recognition System for Rat Cyborg Automated Navigation