Balance and Jump Control Strategies for Bipedal Wheeled-Legged Robots in Unstructured Environments

The high degrees of freedom (DOF) and complex dynamics of bipedal wheeled-legged robots pose significant challenges for modeling and control design. To address these challenges, this study proposes an integrated control approach that combines Linear Quadratic Regulator (LQR) and Virtual Model Control (VMC). The LQR is employed for full-body posture and balance regulation, while the VMC generates precise hip joint torques to coordinate leg motion. Simulation experiments conducted on complex unstructured terrains, such as bumps, steps, and slopes, demonstrate the feasibility and effectiveness of the proposed method in tasks such as balance control, leg length adjustment, jumping, and slope traversal.