Reliability is a prerequisite for spacecraft applications, and in some highly reliable application scenarios, requiring their control computers to possess the capability to tolerate Byzantine faults. Rapid detection of soft-faults and autonomous recovery are common methods to enhance the reliability of on-board control computer systems, especially during long-duration space missions like space stations and manned lunar missions. In this paper, a fast detection and autonomous recovery method for soft fault of on-board Byzantine fault-tolerant control computer is presented. The method realizes rapid fault detection, location and self-recovery by dividing the memory of the control computer and marking processing, combining with data comparison. The experimental results of typical control computers show that the proposed method has short time of fault detection, location and recovery, good flexibility and higher reliability, and high engineering practical value.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

An On-Board Byzantine Fault-Tolerant Computer Soft-Fault Rapid Detection and Autonomous Recovery Method

  • Yang Xiao,
  • Bo Liu,
  • Chaowei Liu,
  • Jian Xu,
  • Mengdan Cao,
  • Duo Gong,
  • Mingyu Zhang

摘要

Reliability is a prerequisite for spacecraft applications, and in some highly reliable application scenarios, requiring their control computers to possess the capability to tolerate Byzantine faults. Rapid detection of soft-faults and autonomous recovery are common methods to enhance the reliability of on-board control computer systems, especially during long-duration space missions like space stations and manned lunar missions. In this paper, a fast detection and autonomous recovery method for soft fault of on-board Byzantine fault-tolerant control computer is presented. The method realizes rapid fault detection, location and self-recovery by dividing the memory of the control computer and marking processing, combining with data comparison. The experimental results of typical control computers show that the proposed method has short time of fault detection, location and recovery, good flexibility and higher reliability, and high engineering practical value.