Unified Algorithmic Framework for High Degree of Freedom Complex Systems and Humanoid Robots

The degrees of freedom (DOF) of robots and complex systems have been increasing increasing exponentially since the early 20th century. Today it is common place for complex control systems to have 40 DOF. This number is projected to be 70 DOF by the year 2020. Robots with high DOF allows for complex tasks such as tool manipulation, greater human-robot interaction and agile full-body locomotion. More DOF require greater attention to local communication delays, bandwidth, system configuration and stability. In addition different tasks being performed by separate parts of the robot in tandem bring on greater issues including controller timing and priorities. The increase in DOF on single system requires that the traditional methods of controller design be re-examined.;This dissertation describes a Unified Algorithmic Framework for High Degree of Freedom Complex Systems and Humanoid Robots that allows a user to develop controllers using a three tier infrastructure. The Unified Algorithmic Framework called Hubo-Ach is a multi-process based system that allows for robust multi-rate simultaneous control and seamless implementation between virtual, miniature, and full-size robots with no modification. The three tier infrastructure provides different levels of cost to entry and testing. Examples of this field tested framework functioning on simulated, miniature, and full-size high DOF robots is given as well as validation by external researchers.