课题基金 / 基金详情

A worksite surveyor for Canadarm3: ensuring safe operation of Canada's next-generation space manipulator

A worksite surveyor for Canadarm3: ensuring safe operation of Canada's next-generation space manipulator
Canadaarm3的现场测量员:确保加拿大下一代太空机械手的安全运行
批准号:
576630-2022
负责人:
Kelly, JonathanJS
金额:
$2.73万
依托单位:
依托单位国家:
加拿大
项目类别:
Alliance Grants
财政年份:
2022
资助国家:
加拿大
项目状态:
已结题
起止时间:
2022-01-01 至 2023-12-31

项目摘要

项目成果

相似基金

相关文献

中文摘要
翻译
加拿大被国际公认为空间机器人技术的领导者。这一领导地位始于最初的加拿大臂的开发,该臂于1981年在哥伦比亚号航天飞机上首次发射。加拿大对太空探索的下一个重大贡献将是Canadarm 3,这是美国主导的月球门户的先进机器人机械手。网关是一个月球前哨站,将使人类能够持续探索月球。Canadarm3将在Gateway上运行,Gateway将从2020年代中期开始绕月球运行,支持月球表面的任务,作为科学实验室,并作为探索太空更深处任务的试验场。由于网关的位置距离地球约40万公里,通信将受到限制和延迟。因此,Canadarm3需要高度自主,在没有地面持续监督的情况下运作。联盟研究计划将在基于视觉的异常检测领域开发新的功能,以使Canadarm3能够在网关结构周围安全移动。与加拿大第三臂的主要承包商MDA公司合作,我们将利用机器学习,从月球轨道上可变和困难的照明条件下获取的单目相机图像中创建精确的3D占用地图。将占用图与结构的策划图进行比较,以自动识别可能存在安全隐患的偏差。同样的危险识别技术也可应用于地球轨道运行卫星的维修。
英文摘要
Canada is internationally recognized as a leader in space robotics. This leadership began with the development of the original Canadarm, first launched in 1981 on board Space Shuttle Columbia. The next major Canadian contribution to space exploration will be Canadarm3, an advanced robotic manipulator for the US-led Lunar Gateway. The Gateway is a lunar outpost that will enable sustained human exploration of the Moon. Canadarm3 will operate aboard Gateway, which will orbit the Moon beginning in the mid-2020s, supporting missions to the lunar surface, serving as a science laboratory, and acting as a proving ground for exploration missions deeper into space. Because of Gateway's location some 400,000 kilometres from Earth, communication will be limited and delayed. Canadarm3 will therefore need to be highly autonomous and to operate without constant supervision from the ground. The Alliance research program will develop new capabilities in the area of vision-based anomaly detection to enable safe movement of the Canadarm3 around the Gateway structure. In collaboration with the primary Canadarm3 contractor, MDA Inc., we will leverage machine learning to create accurate 3D occupancy maps from monocular camera images acquired under variable and difficult lighting conditions in lunar orbit. The occupancy maps will be compared with curated maps of the structure to automatically identify deviations that may be a safety hazard. The same hazard-identification technology has applications for the servicing of satellites operating in Earth orbit.
期刊论文(0)
专著(0)
科研奖励(0)
会议论文
海外基金