| 摘要 |
A method of controlling a redundant manipulator for assigning one or more redundant joints from a plurality of joints and obtaining the solution to an inverse kinematics problem at high-speed. The joints are arbitrarily classified into redundant joints and non-redundant joints, and an initial value is set for the joint angle of the classified redundant joint as a parameter. Then based on an evaluating function or a constraint condition defined by the joint angle of the redundant joint provided as a parameter and the joint angle of the non-redundant joint, which is determined by the inverse kinematics calculation according to the change of the parameter, an optimum solution of a set of joint angles is determined, and until the optimum solution covers the target range of the hand position, the procedure to determine the optimum solution is repeated with relaxing the constraint conditions.
|
