Hybrid Momentum Compensation Control by Using Arms for Bipedal Dynamic Walking

Area of Science:

• Robotics
• Control Systems
• Biomechanics

Background:

• Biped robots require dynamic walking for task performance.
• Leg swinging during walking can disrupt balance and ground interaction.
• Adapting to unknown disturbances is crucial for stable bipedal locomotion.

Purpose of the Study:

• To develop a control method for torque-controlled biped robots to adapt to unknown disturbances during dynamic walking.
• To enable robots to maintain stability using intrinsic motion characteristics, similar to humans.
• To improve the robot's ability to resist disturbances affecting its lower limbs.

Main Methods:

• A hybrid angular and linear momentum regulator was designed to compensate for disturbances caused by leg swinging.
• A mixed-momentum quadratic programming controller was developed for stable dynamic walking based on real-time leg state changes.
• The control method focuses on utilizing arm movements to resist external disturbances.

Main Results:

• The proposed hybrid momentum compensation control method enables biped robots to adapt to unknown disturbances.
• The robot maintained balance while walking down an unknown platform.
• The method ensured good straightness in the forward direction of dynamic motion.
• Experimental verification on the BHR-B2 robot platform confirmed the method's effectiveness.

Conclusions:

• The hybrid momentum compensation control method is effective for enhancing the stability of biped robots during dynamic walking.
• The approach allows robots to maintain balance and directional control when facing unexpected environmental changes.
• This research contributes to more robust and adaptable bipedal locomotion systems.