Kinematics analysis of a novel 7-DOF humanoid manipulator for table tennis

A novel 7-DOF humanoid manipulator used to play table tennis is presented and the kinematics of the manipulator is analyzed according to its architecture. Both the forward kinematics and the inverse kinematics are established and the Jacobian matrix is calculated. Especially, an analytic method for the inverse kinematics which can obtain anthropomorphic solutions is proposed. Motion characteristics of human arms are taken into account in the inverse kinematics and the relationship between the elbow position and the operation space is built up offline using an artificial neural network (ANN). The elbow position can be calculated according to the ANN model when solving the inverse kinematics and thus the anthropomorphic solutions are derived. The method for the inverse kinematics is simple and beneficial to the real-time motion control of the manipulator. The validation and simulation on the kinematics are carried out and the results prove the correctness and feasibility.