Stability and Control of a Pacing Quadruped Using Stabilizing Switching Design

Area of Science:

• Robotics
• Control Systems
• Dynamical Systems

Background:

• Quadruped robots utilize a pacing gait, alternating between left and right leg pairs for locomotion.
• This gait involves two discrete, inherently unstable stance configurations.
• Dynamic stability is achieved by skillfully switching between these stances.

Purpose of the Study:

• To develop and present a state feedback switching control method for stabilizing a pacing quadruped robot.
• To actively control the switching between two sets of non-linear dynamical equations modeling the robot.
• To achieve dynamic stability in the roll direction for the quadruped robot.

Main Methods:

• Modeling the quadruped robot system using two sets of non-linear dynamical equations.
• Developing a state feedback switching law based on linearized models of the system's configurations.
• Evaluating controller performance via simulation on a complete dynamical model and verifying stability with a Poincaré map.

Main Results:

• Successfully stabilized the quadruped robot in the roll direction using the proposed state feedback switching law.
• The control method, combined with passive leg design, resulted in a stable pacing gait.
• Simulation and Poincaré map analysis confirmed the stability of the controlled system.

Conclusions:

• The proposed state feedback switching method is effective for controlling and stabilizing pacing quadruped robots.
• The integration of active switching control with passive design elements enhances overall gait stability.
• This approach offers a viable solution for achieving dynamic stability in legged robotic locomotion.