Adaptive walking control for quadruped robot by using oscillation patterns.

Area of Science:

• Robotics
• Control Systems
• Bio-inspired Engineering

Background:

• Quadruped robots require advanced locomotive systems for adaptability in diverse environments.
• Existing systems often struggle with seamless gait transitions and stable movement on irregular surfaces.

Purpose of the Study:

• To propose an adaptive locomotive system for quadruped robots using a double-layered central pattern generator (CPG).
• To enhance robot adaptability, enable seamless gait switching, and achieve stable movement in complex terrains.

Main Methods:

• Developed a novel CPG network with double master and slave units, utilizing gyroscope (yaw, pitch) and ultrasonic sensor data.
• Master units control the first joints; slave units generate symmetry signals for the second and third joints.
• Implemented parameter adjustments for steering behavior and defined stable angle ranges for adaptive irregular surface movement.

Main Results:

• The double-layered CPG network facilitates seamless switching between locomotion gaits, including stopping and starting.
• The system allows for steering behavior through joint angle adjustments.
• Experimental results demonstrate stable trot patterns in complex environments, validating the system's effectiveness.

Conclusions:

• The proposed double-layered CPG system significantly improves quadruped robot adaptability in multiple scenarios.
• The integration of gyroscope and ultrasonic sensors enables robust control and stable locomotion.
• This approach offers a promising solution for advanced robotic locomotion in challenging terrains.