作为接收机自主完好性监测（RAIM）的核心功能，严格的完好性风险估计是保障导航定位可靠性的关键。全球导航卫星系统的互操作使得多星故障概率上升，导致传统基于单星最差故障偏差的完好性风险估计严格性降低。为此，提出顾及多星故障的完好性风险估计方法。多星故障完好性风险的严格估计受限于多星最差故障偏差的精确确定，针对此建立确定多星最差故障偏差的“两步法”，即通过最差故障单位向量的确定以及最差故障幅值的有界搜索，实现导航定位完好性风险的保守估计。采用星座仿真软件模拟卫星故障场景测试完好性风险估计方法的严格性。实验结果表明：与传统单星故障完好性风险估计方法相比，所提方法在多星座组合场景下可显著降低漏检比率。此外，与传统顾及多星座多星故障的高级RAIM相比，所提方法在全球范围内满足99.5%可用性指标的比例至多可提升9.42%。

As the core function of Receiver Autonomous Integrity Monitoring （RAIM）， strict integrity risk estimation is the key to accurately measure navigation and positioning integrity. The interoperation of global navigation satellite systems increases the probability of multi-satellite faults dramatically， which reduces the rigor of the traditional integrity risk estimation based on the worst fault bias of a single satellite. In this paper， an integrity risk estimation method considering multi-satellite faults is proposed. As the rigorous estimation of multi-satellite fault integrity risk is limited by the accurate determination of the multi-satellite worst fault bias， a two-step method for determining the worst fault bias is established by determining the worst-case fault unit vector and search interval of the worst-case fault magnitude. The conservative estimation of the integrity risk of navigation and positioning is realized. Constellation simulation software is used to simulate satellite fault scenarios to test the rigor of the integrity risk estimation method. Experimental results show that compared with the traditional integrity risk estimation method， the proposed method can significantly reduce the rate of missed detection in multi-constellation scenarios. In addition， when compared to advanced RAIM method that consider multi-constellation and multi-satellite faults， the proposed method can improve the proportion meeting a global availability of 99.5% by 9.42% at most.