月面表层无人采样控制技术  

Control Technology for Unmanned Sampling of Lunar Surface

作　　者：张宽[1] 于天一[1] 胡晓东 刘传凯[1] 李立春[1] 赵焕洲 ZHANG Kuan;YU Tianyi;HU Xiaodong;LIU Chuankai;LI Lichun;ZHAO Huanzhou(Beijing Aerospace Control Center,Beijing 100094,China)

机构地区：[1]北京航天飞行控制中心,北京100094

出　　处：《深空探测学报（中英文）》2022年第2期173-182,共10页Journal Of Deep Space Exploration

基　　金：国家自然科学基金青年科学基金资助项目(62003025);国家自然科学基金资助项目(61972020)。

摘　　要：针对“嫦娥五号”月面表层无人采样任务过程中机械臂精确控制难度大、器地协同工作复杂导致常规遥操作工作模式难以在有限时间内完成任务的问题,提出了月面表层无人采样高效控制方法和新的工作模式。通过建立采样区量化评价指标体系实现采样点规划、采用预先规划方法进行任务规划和机械臂运动控制规划、在线学习机械臂末端精调路径与多视角融合图像测量分析精调控制量、基于状态自动发令与实时管控工作进度等技术,实现了环境感知、任务规划与验证、机械臂末端精调规划、控制实施等环节提高工作效率。“嫦娥五号”月面表层采样任务在轨实施情况表明控制方法和新的工作模式能够极大提升采样工作效率,为未来火星和小行星采样工程项目奠定了基础。Aiming at the difficulty of accurate controlling the robotic arm,and the complexity of probe-ground cooperation for the Chang’E-5 unmanned lunar surface sampling mission in a finite time period,a new control method and teleoperation mode which improves the efficiency of environmental perception,task planning and verification,fine-turning planning of manipulator end,and control implementation are proposed in this paper.Many technologies are applied into this method,such as establishing quantitative evaluation index system of sampling area to realize sampling points planning,using preplanning methods to complete mission planning and mechanical arm motion control planning,online learning to bring about the fine-tuning path in the manipulator end and analyzing fine-tuning using multi-view fusion images,automatic command based on state and controlling work progress real-time.The implementation results of Chang’E-5 lunar surface sampling mission in orbit indicate that new control method and teleoperation mode can greatly improve the sampling efficiency.It lays a foundation for the Mars and asteroid sample return mission in the future.

关 键 词：月面无人表层采样 嫦娥五号 遥操作 任务规划 机械臂精确控制 

分 类 号：V448.2[航空宇航科学与技术—飞行器设计]