Biped Robot Gait Control With Real-Time ZMP Ankle Compensation

The patent implements a closed-loop control system that continuously monitors the robot's actual posture through sensors and compares it with the planned trajectory. When deviations are detected due to external forces or uneven terrain, the system generates corrective torque commands to adjust the joints and maintain stable walking. This feedback mechanism enables the robot to adapt to environmental disturbances while preserving walking reliability.

The patent transitions from static open-loop gait planning to dynamic closed-loop control. The control system continuously adjusts joint trajectories in real-time based on current posture and external disturbances. The dynamic adjustment allows the robot to maintain stability on uneven ground and respond to external forces, resolving the contradiction between adaptability and reliability.

The patent introduces a posture control processing algorithm that uses feedback from sensors to detect deviations from the planned trajectory. The algorithm calculates corrective actions based on the detected deviations and applies them through the joint actuators. This feedback-based approach ensures posture stability without requiring overly complex control structures, as it builds upon the existing open-loop planning framework.

The patent introduces an intermediate control layer between the open-loop gait planner and the joint actuators. This intermediary posture control module processes the planned trajectory and actual sensor data, generating corrective commands that bridge the gap between simple planning and complex adaptation. This intermediary approach maintains reasonable system complexity while achieving reliable posture stability.