针对夹矸与片帮共存的大断面巷道掘进难题，本团队研发了护盾式煤矿巷道掘进机器人系统。为了提高护盾式掘进机器人系统的定位和纠偏精度，本文借助于惯性测量技术、多传感器信息融合技术、先进PID控制技术等，研究煤矿护盾式掘进机器人系统的精确定位与自主纠偏方法。

研究了护盾式掘进机器人的系统构成和工作原理，提出了煤矿护盾式掘进机器人系统的精确定位与自主纠偏总体方案；分析了掘进机器人系统的工作环境，提出了“捷联惯导+数字全站仪+位移传感器”的组合定位方案；分析了掘进机器人系统的行走特点，提出了基于定位误差的推移油缸位移精确控制的纠偏方案。

分析并确定了“捷联惯导+数字全站仪+位移传感器”型号及其安装位置，研究了捷联惯导、数字全站仪及位移传感器的位姿测量原理，分别建立了捷联惯导、数字全站仪及位移传感器的位姿解算数学模型，并推导了各自的位姿测量误差数学模型。

研究了“捷联惯导+数字全站仪+位移传感器”的组合定位方法，通过数字全站仪和位移传感器共同测量的位置信息辅助修正捷联惯导系统解算得到的掘进机器人的位置及姿态角信息。研究了“捷联惯导+数字全站仪+位移传感器”的组合定位滤波算法与结构，确定了联邦滤波器设计步骤，建立了基于联邦滤波算法的组合定位方法的状态方程和量测方程，设计了联邦滤波器，实现了三者位姿测量信息的融合。

针对护盾式掘进机器人系统在其掘进过程中偏离巷道设计中线的问题，研究了护盾式掘进机器人系统位姿纠偏机理，建立了护盾式掘进机器人系统的位姿纠偏控制数学模型，提出了基于“捷联惯导+数字全站仪+位移传感器”多信息融合的位姿纠偏控制策略和基于模糊PID算法的推移油缸伸缩量精确控制的方法，通过多个推移油缸行程控制实现护盾式掘进机器人系统的纠偏。

最后，在煤矿井下针对护盾式掘进机器人系统进行了组合定位方法与纠偏功能实验，实验表明：该组合定位方法抑制了纯惯性导航位置解算误差，提高了护盾式掘进机器人系统的定位与纠偏精度，满足煤矿井下巷道掘进工艺要求。

The team has developed a shield-type coal mine tunneling robot system to address the problem of large section roadway excavation where parting and piece coexist. In order to improve the positioning and deflection accuracy of the shield roadheader robot system, this paper investigates the precise positioning and autonomous deflection correction method of the shield roadheader robot system in coal mines with the help of inertial measurement multi-sensor information fusion technology and advanced PID control technology.

It studies the system composition and working principle of the shield roadheader robot and proposes a general scheme for precise positioning and autonomous deflection correction of the shield roadheader robot system in coal mines. The working environment of the roadheader robot system is analyzed, and the combined positioning scheme of "Strapdown inertial navigation + digital total station + displacement sensor" is proposed. It explores the walking characteristics of the roadheader robot system and proposes a correction scheme for accurate control of nudge cylinder displacement based on positioning error.

The article analyzes and determines the type of "Strapdown inertial navigation + digital total station + displacement sensor" and its installation position, studies the positional measurement principle of Strapdown inertial navigation, digital total station, and displacement sensor, establishes the positional solution mathematical model of Strapdown inertial navigation, digital total station, and displacement sensor respectively, and derives the positional measurement error mathematical model of each.

It investigates the combined positioning method of "Strapdown inertial navigation + digital total station + displacement sensor" The position and attitude angle information of the roadheader robot is corrected by the position information measured by the digital total station and the displacement sensor together. The article investigates the combined positioning filtering algorithm and structure of " Strapdown inertial navigation + digital total station + displacement sensor" determines the design steps of the federal filter, establishes the state equation and measurement equation of the combined positioning method based on the federal filtering algorithm, designs the federal filter, and realizes the fusion of the three positional measurement information.

The article researches the mechanism of the shield roadheader robot system's attitude correction mechanism and establishes the mathematical model of the shield roadheader robot system's attitude correction control for the problem that the shield roadheader robot system deviates from the centerline of the tunnel design during its digging process. It proposes a multi-information fusion control strategy based on "Strapdown inertial navigation + digital total station + displacement sensor" and an accurate control method of pushing cylinder expansion and contraction based on a fuzzy PID algorithm to realize the deflection correction of shield roadheader robot system through multiple pushing cylinder stroke control.

Finally, the paper conducts a combined positioning and deflection control experiment for underground coal mines' shield roadheader robot system. The experiments show that the combined positioning method suppresses the pure inertial navigation position solution error, improves the positioning and deflection correction accuracy of the shield type tunneling robot system, and meets the underground coal mine tunneling process requirements.