摘要
机器人与机器学习每日新闻-机器人与自动化的最新研究结果已经发表。根据NewsRx编辑在意大利皮萨的新闻报道，研究表明，“选择好的抓取是完成机器人抓取和操作任务的基础。通常情况下，G RASP合成与规划阶段分开处理，这可能导致任务执行过程中的失败。”我们的新闻记者从Scuola Superiore Sa Nt'anna的研究中获得了一句话，"此外，目前大多数抓取方法都是特权稳定指标。为后续任务的执行提供了不合适的抓握，提出了一个高级推理框架，根据任务选择最佳抓握，通过求解一个优化问题，考虑环境约束，保证任务的终态舒适性和置信度，从一组抓握中选择最佳抓握。与人类行为相似。该框架利用广义Bender Dec Omposition将主要非线性优化问题解耦为子问题S，从而呈现出模块化结构。该方法通过使用三种不同的最先进抓取算法和三种不同类型任务的低级运动规划器的实验活动得到了验证：在受限环境中拾取和放置、移交/工具使用和对象重新定向。
Abstract
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News – Current study results on Robotics - Ro botics and Automation have been published. According to news reporting out of Pi sa, Italy, by NewsRx editors, research stated, “Choosing a good grasp is fundame ntal for accomplishing robotic grasping and manipulation tasks. Typically, the g rasp synthesis is addressed separately from the planning phase, which can lead t o failures during the execution of the task.” Our news journalists obtained a quote from the research from Scuola Superiore Sa nt’Anna, “In addition, most of the current grasping approaches privilege stabili ty metrics, providing unsuitable grasps for executing subsequent tasks. The prop osed work presents a framework for high-level reasoning to select the best-suite d grasp depending on the task. The best grasp is chosen among a set of grasp can didates by solving an optimization problem, considering the environmental constr aints, and guaranteeing the end-state comfort and the confidence effects for the task, similar to human behavior. The framework leverages Generalized Bender Dec omposition to decouple the main non-linear optimization problem into sub-problem s, thus presenting a modular structure. The method is validated with an experime ntal campaign using three different state-of-the-art grasping algorithms and thr ee low-level motion planners in three different types of tasks: pick-and-place i n a constrained environment, handover/tool-use, and object re-orientation.”
NETL
NSTL