Tagliabue Michele, Pedrocchi Alessandra, Pozzo Thierry, Ferrigno Giancarlo
Department of Bioengineering, Politecnico di Milano, NITlab, Milano, Italy.
Med Biol Eng Comput. 2008 Jan;46(1):11-22. doi: 10.1007/s11517-007-0252-4. Epub 2007 Sep 11.
In spite of the complexity of human motor behavior, difficulties in mathematical modeling have restricted to rather simple movements attempts to identify the motor planning criterion used by the central nervous system. This paper presents a novel-simulation technique able to predict the "desired trajectory" corresponding to a wide range of kinematic and kinetic optimality criteria for tasks involving many degrees of freedom and the coordination between goal achievement and balance maintenance. Employment of proper time discretization, inverse dynamic methods and constrained optimization technique are combined. The application of this simulator to a planar whole body pointing movement shows its effectiveness in managing system nonlinearities and instability as well as in ensuring the anatomo-physiological feasibility of predicted motor plans. In addition, the simulator's capability to simultaneously optimize competing movement aspects represents an interesting opportunity for the motor control community, in which the coexistence of several controlled variables has been hypothesized.
尽管人类运动行为复杂，但数学建模的困难限制了对中枢神经系统所使用的运动规划标准的识别，目前仅局限于相当简单的运动。本文提出了一种新颖的模拟技术，该技术能够针对涉及多个自由度以及目标达成与平衡维持之间协调的任务，预测对应于广泛运动学和动力学最优标准的“期望轨迹”。结合了适当的时间离散化、逆动力学方法和约束优化技术。将该模拟器应用于平面全身指向运动，展示了其在处理系统非线性和不稳定性以及确保预测运动计划的解剖生理可行性方面的有效性。此外，该模拟器同时优化相互竞争的运动方面的能力，为运动控制领域提供了一个有趣的契机，在该领域中，人们已经假设了多个受控变量的共存。
