Locomotion Stability Control of Quadruped Robot using the Combination of Ankle, Hip, and Step Strategies and Fuzzy Logic Controller

This paper proposes a whole-body controller for a trotting quadruped robot. By combining hip, ankle, and step strategies, the developed controller allows both tracking desired trajectory and restoring balance from external disturbances. The control strategy is based on simple template models and Capture Point (CP) concept. A fuzzy logic controller (FLC) is used for adjusting gain values of trajectory tracking control and balance recovery control according to the magnitude of external disturbances. For the ankle and hip strategies, ankle and hip torques are calculated using the CP criteria to avoid balance loss. For the step strategy, the controller computes footholds to maintain the desired Center of Mass (CoM) velocity against external disturbances. A simulation was conducted on a 3D quadruped robot to evaluate the effectiveness of the proposed controller. Forces in both longitudinal and lateral directions were applied to the robot body, with the controller active and inactive. The simulation results showed that the controller reduced CP, CoM position, and CoM velocity errors in both directions. The results demonstrate that the controller, which integrates hip, ankle, and step strategies, improves a quadruped robot's trajectory following and balance recovering capabilities.