Dynamic modeling of robotic systems: a dual quaternion formulation

This thesis proposes a technique for the dynamic modeling of serial and branched robots using dual quaternion algebra. The modeling accounts for all lower-pair kinematic joints and six-degree-of-freedom joints, and the framework enables the systematic modular composition of dynamic models comprising several subsystems, each, in turn, composed of multiple rigid bodies. The proposed strategy is applicable even if some subsystems are regarded as black boxes, requiring only the twists and wrenches at the connection points between different subsystems. To help in the model composition, a unified graph representation that encodes the propagation of twists and wrenches between the subsystems is also proposed. The joint wrenches result from the calculation of the interconnection matrix of the graph, making the modeling procedure straightforward. The framework was validated using serial manipulators of 6-DoF and 50-DoF, a 9-DoF holonomic mobile manipulator, and a 38-DoF branched robot composed of 9 subsystems. The results were compared with Peter Corke's Robotics Toolbox, Roy Featherstone's Spatial V2, and the robot simulator V-REP/CoppeliaSim, demonstrating that the proposed formalism is as accurate as state-of-the-art libraries.