...
首页> 外文期刊>Advanced Robotics: The International Journal of the Robotics Society of Japan >A Method for Sustaining Consistent Sensory-Motor Coordination under Body Property Changes Including Tool Grasp/Release
【24h】

A Method for Sustaining Consistent Sensory-Motor Coordination under Body Property Changes Including Tool Grasp/Release

机译:在包括工具抓握/释放在内的身体属性变化下维持一致的感觉-运动协调的方法

获取原文
获取原文并翻译 | 示例
           

摘要

The body's physical and spatial properties change because of unexpected accidents or objects it holds (e.g., tools). To maintain stable and consistent behaviors in acting time, it is necessary to adapt body representation to the unexpected changes. This adaptation is achieved when a robot sustains its coordination between sensation and body motion, i.e., observation and identification of the discrepancy between its knowledge and actually obtained sensory feedback. For sustainability, the robot should autonomously judge the reliability and the tolerance of the identification. Based on basic mathematics, we propose a method with an indirect but clear criterion of the reliability for the sustainable coordinations. We also implement the method on visual-motor coordination and on kinesthetic-motor coordination. Remarkably, our method achieves a marker-free, easily convergent and sufficiently accurate (i.e., easily applicable) hand-eye calibration method with irrelevant objects in view. The evaluation of the method is provided with experiments in a real robot. Our experimental results show the novelty of the concept of sustainable coordination and the availability of our method for the concept. We hope this paper be a powerful approach for building autonomous robots.
机译:身体的物理和空间特性会因意外事故或所持有的物体(例如工具)而发生变化。为了在动作期间保持稳定和一致的行为,有必要使身体表示适应意外的变化。当机器人维持其在感觉和身体运动之间的协调,即观察和识别其知识与实际获得的感觉反馈之间的差异时,就可以实现这种适应。为了实现可持续性,机器人应自主判断标识的可靠性和公差。在基础数学的基础上,我们提出了一种具有间接但明确的可持续协调可靠性标准的方法。我们还实现了视觉-运动协调和动觉-运动协调的方法。值得注意的是,我们的方法实现了无标记,易于收敛并且足够准确(即易于应用)的手眼校准方法,并且没有相关物体。该方法的评估在真实的机器人中进行了实验。我们的实验结果表明了可持续协调概念的新颖性和适用于该概念的方法。我们希望本文是构建自主机器人的有力方法。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

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