Neural Network Control of two Electrically Driven Cooperating 6 DOF Robot Manipulators

In this paper, the problem of dynamic modeling and neural network control of two 6 DOF elbow manipulators handling a rigid object is studied when actuator dynamics is also considered in the system dynamics. First, kinematics and dynamics of the mechanical subsystem and the relation between different forces acting on the object using different Jacobians are derived. It is shown that the dynamic model of two robots has the same properties that the dynamic model of a single robot has; this permits to use the control schemes, which are used for control of a robot, to control two cooperating robots. Then the actuator dynamics of each robot is included into dynamic equation of two robots and a third order matrix equation is obtained. At the end, a neural network controller scheme is developed for this third order dynamics and the related Lyapunov function is established. This controller guarantees asymptotic tracking of the object desired trajectory.