首页> 外国专利> REAL-TIME ROBOTIC GRASP PLANNING

REAL-TIME ROBOTIC GRASP PLANNING

机译:实时机器人抓斗计划

摘要

A method of determining an optimal grasp pose of an object by an end-effector of a robot in the execution of a task includes receiving a set of inputs via a controller, including a descriptive parameter of the object, a task wrench having a frame of reference, and a commanded end-effector grasp force. The method includes calculating a grasp wrench in the frame of reference of the task wrench, rotating and shifting the grasp pose, and then computing a resultant new grasp wrench via the controller until the task wrench and the grasp wrench are optimally balanced. Additionally, the method includes recording the optimal grasp pose as the grasp pose at which the grasp and task wrenches are optimally balanced, and then executing a control action via the controller using the recorded optimal grasp pose. A robotic system includes the robot and a controller programmed to execute the method.
机译:一种在执行任务时由机器人的末端执行器确定对象的最佳抓握姿势的方法,包括经由控制器接收一组输入,该输入包括对象的描述性参数,具有框架的任务扳手。参考,以及指令的末端执行器抓取力。该方法包括在任务扳手的参考系中计算抓握扳手,旋转和移动抓握姿势,然后通过控制器计算所得的新抓握扳手,直到任务扳手和抓握扳手达到最佳平衡为止。另外,该方法包括:将最佳抓握姿势记录为抓握姿势和任务扳手最佳平衡的抓握姿势,然后使用所记录的最佳抓握姿势经由控制器执行控制动作。机器人系统包括机器人和被编程为执行该方法的控制器。

著录项

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号