In the past several years, grasp analysis of multi-fingered robotic hands has been actively studied through the use of posture synergies. In these grasping planning algorithms, a formulated optimization is usually performed in the hand’s low-dimensional representation together with the hand’s position and orientation. The optimization terminates at a stable grasp, often after repeated trials with different initial guesses. Furthermore, there is no guarantee that the generated grasp leads to a smooth reach-to-grasp trajectory since the grasping planning process mostly concerns hand poses with the fingers proximal to the object. A unified theoretical framework of a gradient-based iterative algorithm is hence proposed in this paper to plan a reach-to-grasp task, predicting the grasp quality and adjusting the hand’s posture synergies, position and orientation during the approaching phase to achieve a stable grasp. The grasp quality measurement is adopted from a highly efficient pseudo-distance formulation. Stable power grasp and precision pinch can be consistently and intentionally planned with different contact conditions specified in the formulation, which means that an intention for planning a power grasp would not generate a pinch result. Several numerical simulation case studies are presented to demonstrate the effectiveness of the proposed algorithm.

中文翻译：

---

基于拟距离公式的协同控制机器人手的掌握规划

在过去的几年里，多指机械手的抓握分析通过使用姿势协同作用得到了积极的研究。在这些抓取规划算法中，通常在手的低维表示中与手的位置和方向一起执行公式化的优化。优化在稳定掌握时终止，通常是在使用不同的初始猜测进行重复试验之后。此外，由于抓取规划过程主要涉及手指靠近物体的手部姿势，因此无法保证生成的抓取会导致平滑的到达抓取轨迹。因此，本文提出了一个基于梯度的迭代算法的统一理论框架来规划一个到达抓取任务，预测抓取质量和调整手的姿势协同作用，在接近阶段的位置和方向，以实现稳定的抓握。抓握质量测量采用高效的伪距离公式。在公式中指定的不同接触条件下，可以一致且有意识地计划稳定的力量抓握和精确捏合，这意味着计划力量抓握的意图不会产生捏合结果。提出了几个数值模拟案例研究来证明所提出算法的有效性。这意味着计划权力掌握的意图不会产生挤压结果。提出了几个数值模拟案例研究来证明所提出算法的有效性。这意味着计划权力掌握的意图不会产生挤压结果。提出了几个数值模拟案例研究来证明所提出算法的有效性。