Design of lightweight hydraulic power unit for legged robots based on the Sobol sensitivity analysis
Bin Yu, Huashun Li, Chengze Gu, Ao Shen, Shuai Zhang, Xu Liu, Jingbin Li, Kaixian Ba, Guoliang Ma, Xiangdong Kong
- Year
- 2025
- Citations
- 20
Abstract
• A lightweight hydraulic power unit is developed for legged robots. • Key weight-influencing parameters are identified using Sobol sensitivity analysis. • Motor-pump matching is optimized, and tank, filter, and block are improved. • Over 40% weight reduction is achieved through component optimization. • The work leads to the realization of high-performance hydraulic systems. Reducing the weight of the legged robot improves its endurance, maneuverability, and carrying capacity, enabling it to perform more complex tasks, such as transitioning from walking on flat terrain to executing challenging jumps. As the core energy source of a hydraulic legged robot, the weight of hydraulic power unit (HPU) accounts for more than 50% of the entire robot, offering significant potential for weight reduction. However, the design and sizeing parameters of the HPU are complex, and the key factors influencing its weight are not fully unclear, making targeted optimization challenging. To solve the above problems, this paper designs a lightweight hydraulic power unit (LHPU) for the legged robot based on the Sobol sensitivity analysis method. First, a weight model for the LHPU’s key components is established, and the key parameters influencing its weight are determined using the Sobol sensitivity analysis method. Then, key parameters matching is optimized for the motor-pump, and design improvements are made to components like the low-pressure tank, filter, and integrated block. Finally, the LHPU is tested for static and dynamic characteristics. The findings demonstrate that the LHPU can stabilize the required pressure and flow outputs for the legged robot and achieve a weight reduction exceeding 40%. This research offers significant insights for the lightweight design of advanced mobile systems, enhancing energy efficiency, mobility, and load capacity.
Keywords
Related papers
Statistical Learning Theory
Yuhai Wu, Vladimir Vapnik
1999
Artificial intelligence: a modern approach
1995
Applied Nonlinear Control
Jean-Jacques Slotine, Weiping Li
1991
A new optimizer using particle swarm theory
R.C. Eberhart, James Kennedy
2002