Regulariation and Control of Multibody Systems with Irregular Modes of Operation

某些机械系统，特别是在有障碍物的情况下运行的机器人系统，在避开障碍物的过程中会受到运动突然变化的影响。这样的变化可能会引起振动，从而导致不准确，从而严重影响系统的性能。机器人避碰的研究大多集中在障碍物的描述和避碰几何(运动学)条件的推导上。与之相关的动力学问题还没有引起人们的高度重视。这项研究的目的是开发一种控制算法，能够控制系统动态，以适应运动的快速变化。本项目使用的方法是基于动力学的变分公式。所开发的控制器将提供满足约束条件的最优控制力，可以在指定任务间隔的任何指定时间被激活，并且将在计算上高效地实现。本研究将针对冗余度刚体机器人系统展开。如果成功，后续研究将探索该方法在高速柔性系统中的更广泛应用。与该控制器相关的更平滑的运动可以提高系统的性能、可靠性和位置精度。

Certain mechanical systems, especially robotic systems operating in the presence of obstacles, are subjected to sudden changes in their motion in avoiding the obstacles. Such changes can severely affect the system's performance by inducing vibrations and thereby inaccuracies. Most research on robot collision avoidance has focused on description of the obstacles and derivation of the geometric (kinematic) conditions for avoiding collision between bodies. The related dynamics problem has not received a great deal of attention. The objective of this research is to develop a control algorithm that will be able to control the system dynamics to accommodate the rapid changes in motion. The methodology used in this project is based upon the variational formulation of dynamics. The controller developed will provide the optimum control forces to satisfy constraint conditions, can be activated at any specified time of a prescribed task interval, and will be computationally efficient to implement. This research will be carried out for redundant rigid robotic systems. If successful, subsequent research will explore broader application of the method to high speed flexible systems. Smoother motion associated with this controller could improve system performance, reliability, and position accuracy.