SUMMARYDynamic modeling is a fundamental step in analyzing the movement of any mechanical system. Methods for dynamical modeling of constrained systems have been widely developed to improve the accuracy and minimize computational cost during simulations. The necessity to satisfy constraint equations as well as the equations of motion makes it more critical to use numerical techniques that are successful in decreasing the number of computational operations and numerical errors for complex dynamical systems. In this study, performance of a variant of Kane’s method compared to six different techniques based on the Lagrange’s equations is shown. To evaluate the performance of the mentioned methods, snake-like robot dynamics is considered and different aspects such as the number of the most time-consuming computational operations, constraint error, energy error, and CPU time assigned to each method are compared. The simulation results demonstrate the superiority of the variant of Kane’s method concerning the other ones.

中文翻译：

---

Kane 和 Lagrange 方法在约束动力系统分析中的比较

摘要动态建模是分析任何机械系统运动的基本步骤。约束系统的动态建模方法已被广泛开发，以提高模拟过程中的准确性和最小化计算成本。满足约束方程和运动方程的必要性使得使用能够成功减少复杂动力系统的计算操作次数和数值误差的数值技术变得更加关键。在这项研究中，显示了与基于拉格朗日方程的六种不同技术相比的 Kane 方法变体的性能。为了评估上述方法的性能，考虑了蛇形机器人动力学和不同方面，例如最耗时的计算操作的数量、约束误差、比较了分配给每种方法的能量错误和 CPU 时间。仿真结果证明了凯恩方法的变体相对于其他方法的优越性。