With the increasing complexity of tasks assigned to humanoid robots, planning dynamically stable and energy-efficient motions remains a significant challenge. This study addresses the problem of generating collaborative, multi-joint movements for humanoid robots, exemplified by a complex dance performance representing the coordinated limb movements required in advanced garment manufacturing, while simultaneously ensuring stability and optimizing energy consumption. We propose a unified analytical framework that combines precise kinematic modeling with integrated stability control. The methodology employs Homogeneous Transformation Matrices (HTM) for accurate end-effector positioning and motor torque validation. C²-continuous 5th-order polynomial interpolation is utilized for trajectory generation to ensure smoothness and minimize jerk. Critically, we integrate the Zero-Moment Point (ZMP) stability criterion directly into the motion planning phase, enabling proactive leg adjustments to compensate for upper-body dynamics. This integrated model facilitates the analytical calculation of total energy consumption, including motor torque and thermal effects, allowing for subsequent optimization. The significance of this research lies in providing a computationally lightweight, deterministic, and holistic solution for real-time motion planning. It offers a practical alternative to computationally expensive, iterative optimization algorithms, which is essential for the real-time deployment of humanoid platforms in dynamic environments such as smart textile assembly lines.

humanoid robots, motion planning, zero-moment point (ZMP), energy optimization, textile automation