基于航天器可观测性理论的多源融合自主导航技术

Multi-source Fusion Autonomous Navigation Technologies Based on Spacecraft Observability Theory

深空探测任务具有探测目标远、飞行时间长、环境变化大等特点,传统的“地面测控站—航天器”大回路运行导航模式在实时性、安全性和可靠性等方面已无法满足任务需求,而自主导航技术是解决这些问题的有效手段,已成为未来深空探测发展的一个重要方向。为确保任务安全可靠,一般需要对关键的敏感器单机进行冗余容错设计,即采用多源融合的自主导航。文章分析了航天器多源融合自主导航关键技术的5个方面,包括相对动力学建模方法、可观测性理论、多源融合滤波方法、误差补偿方法和地面实验验证技术,结合自主导航工程实施和技术发展需求,提出未来航天器多源融合自主导航技术发展的趋势和重点。

The deep space exploration is characterized by long-range exploration targets, long flight time, and big environmental changes, etc.The traditional large loop navigation operating mode based on the ground telemetry, tracking, and station-spacecraft communication commanding can no longer meet the mission requirements in terms of real-time, safety, and reliability. Autonomous navigation technology is an effective means to solve these problems and has become an important direction for the future development of deep space exploration. To ensure mission safety and reliability, a redundant and fault-tolerant design of the key sensitive individual machines is generally required, i.e., autonomous navigation with multi-source fusion. This paper analyzes five aspects of key technologies towards spacecraft multi-source fusion autonomous navigation, including relative dynamics modeling, observability theory, multi-source fusion filtering, error compensation and ground-based experiment validation technologies, and points out the future development trend and focal point of spacecraft autonomous navigation technology based on the demands of autonomous navigation projects implementation and technology development.