Dual control network method for efficient real-time humanoid robot control system

Abstract

In this paper, a new control network method for real-time humanoid robot control system is proposed. The network method consists of dual structure and it is implemented with two communication protocol: high speed over lOOMbps IEEE 1394, and Controller Area Network (CAN) strong in noisy environment. The main advantages of the control network are represented by a high communication speed and low noise sensitivity. In addition the communication delay in control network is considerably reduced. The control network method is validated in robot control system humanoid robot MAHRU. The robot equipped by several motors represents the best experimental platform for testing high power motor control and correlated noise. ？2009 IEEE.