【摘要】 【目的】针对复杂空间多连杆机构刚度分析中计算效率低、与试验结果偏差大的问题，基于宿主-寄生机构原理，设计多连杆平行解耦式轮腿机器人，实现高效、准确的空间刚度建模与分析。【方法】首先，依据宿主-寄生机制构建轮腿机器人构型，为后续刚度建模提供结构基础；其次，利用Matlab软件建立由极少计算单元组成的空间拟合刚度分布（Spatial Fitting Stiffness Distribution,S??FSD）模型，简化刚度计算流程；然后，通过Ansys软件静力学仿真获取参考数据，采用蒙特卡洛法对模型参数进行拟合，提升模型计算精度；最后，开展实物试验，验证所建模型的正确性。【结果】研究表明，S-FSD模型计算效率显著提升，计算时间减少99.35%以上；单位姿最大偏差为4.06%，多位姿综合拟合后最大偏差为19.66%，与实测刚度的最大偏差仅为14.5%。研究揭示了机器人在x、y、z方向刚度分布的差异性规律，可为轮腿机器人结构优化与轨迹规划提供参考。更多还原

【Abstract】 [Objective] To address the problems of low computational efficiency and significant deviation from test results in the stiffness analysis of complex spatial multi-link mechanisms, a parallel decoupled multi-link wheel-legged robot was designed based on the host-parasitic mechanism, aiming to achieve efficient and accurate spatial stiffness modeling and analysis. [Methods] Firstly, the configuration of the wheel-legged robot was constructed according to the host-parasitic mechanism to provide a structural basis for subsequent stiffness modeling; secondly, a spatial fitting stiffness distribution(S-FSD) model composed of very few computational units was established using Matlab software to simplify the stiffness calculation process; then, reference data were obtained through static simulation with Ansys software, and the model parameters were fitted by the Monte Carlo method to improve the computational accuracy of the model; finally, physical tests were conducted to verify the correctness of the established model. [Results] The results show that the computational efficiency of the S-FSD model is significantly improved, with the computation time reduced by more than 99.35%. The maximum deviation for a single pose is 4.06%, the maximum deviation after comprehensive multi-pose fitting is 19.66%, and the maximum deviation from the measured stiffness is only 14.5%. Meanwhile, the differential regularity of stiffness distribution in the x, y, and z directions of the robot is revealed, providing a reference for the structural optimization and trajectory planning of wheel-legged robots.更多还原