摘要
为了克服时延对空间机器人遥操作的影响,提出了利用视觉技术辅助遥操作进行卫星自维护的方法.针对遥操作数据传输滞后的问题,设计并实现了一种基于视觉的共享控制策略,该策略将维护任务分为自由空间控制与自主视觉伺服控制.在自由空间控制阶段,通过3D图形预测仿真以及实时反馈的视频流,提高了遥操作者的视觉临场感.在自主控制阶段,采用冗余特征视觉伺服控制,克服了系统对标定精度的依赖性,从而提高了整个系统的智能化程度.最后,通过对典型的卫星维护任务实例的分析,验证了该方法的有效性.
To overcome effect of time delay on space robot teleoperation, a novel method of vision aid for teleoperation applied to satellite on-orbit self-servicing was presented. A vision-based shared control scheme was designed and implemented to deal with data transmission delay. The scheme consists of free space control and auto visual servo control. In the process of free space control, 3 D graphic prediction simulation and real feedback video stream make teleoperator achieve the best telepresence experience. In the process of auto visual servo control, the visual servoing of redundant features eliminates system dependency on calibration accuracy and improves system intelligence. Finally, a typical satellite servicing experiment demonstrates the effectiveness of this method.
出处
《北京航空航天大学学报》
EI
CAS
CSCD
北大核心
2009年第7期882-886,共5页
Journal of Beijing University of Aeronautics and Astronautics
基金
国家"863"高科技资助项目(2006AA04Z225)
长江学者和创新团队发展计划资助项目
关键词
卫星自维护
空间机器人
3D图形预测仿真
视觉伺服
satellite self-servicing
space robot
3D graphic predictive simulation
visual servoing
