Dynamic Hybrid Control of a Hexapod Walking Robot: Experimental Verification

In this paper, a walking robot that consists of its body and six legs is considered for trajectory control. The leg mechanism is a 1-UP&2-UPS (U-universal joint, P-prismatic joint, and S-spherical joint) parallel mechanism with the arm-leg conversion function. Due to the nonlinearity and uncertainties of the robot, it is difficult to obtain its exact dynamic model, therefore this paper utilizes the decentralized control strategy to design a dynamic hybrid controller, in which the outer loop is an impedance controller, the middle and inner loops are designed by using the sliding-mode control and the nonlinear proportional-integral-derivative (PID) control, respectively. The inverse kinematics of one leg is derived for mathematic model, based on which three control methods including the classical PID control, position-velocity loops cascade control, and dynamic hybrid control are examined. Contrast tracking experiments were performed with one robotic leg. The experimental results are shown to identify their differences and prove the effectiveness and applicability of the proposed dynamic hybrid control.