【作者】 凌睿；

【导师】 柴毅；

【作者基本信息】 重庆大学 ， 控制理论与控制工程， 2009， 博士

【摘要】 悬臂式掘进机器人是一种用于巷道掘进的特种机器人。论文针对其工作环境恶劣、危险程度高等问题,根据巷道掘进质量的作业要求,研究了一种悬臂式掘进机器人控制系统,实现了掘进机在巷道或井下等危险、恶劣环境下的无人操作,自动完成其预定的掘进作业。论文以悬臂式掘进机为对象,根据人工操作下的巷道掘进关键技术——悬臂截割控制和行走控制,研究断面成形和轨迹跟踪的控制算法,并研究了控制器及系统实现。为研制生产效率高、掘进速度快、适应性强、调动灵活的智能化、无人化的悬臂式掘进机器人,奠定了理论基础,提供了实验平台。论文提出了悬臂式掘进机器人控制系统结构,分析了机器人的断面成形和轨迹跟踪两个关键技术问题。根据机器人的工作任务和作业工艺要求,结合现有高阶滑模变控制的研究现状,提出了新型的高阶滑模控制方法,实现了悬臂式掘进机器人的断面成形和轨迹跟踪。论文主要进行了如下研究工作:①论文研究悬臂式掘进机器人控制系统的总体结构,分析了悬臂式掘进机器人的机械结构和工作原理。根据工作任务和工艺要求,对机器人的控制系统功能及工作流程进行了分析和设计,并将掘进机器人控制系统分解为截割臂控制系统和行走机构控制系统,建立了截割臂动力学模型和行走机构动力学模型,并提出了控制系统实现的理论问题。②针对掘进机器人截割臂控制系统中存在的多变量、非线性等问题,对高阶滑模控制和一类基于有限状态机策略的二阶滑模控制的算法进行了分析。在已有的基于有限状态机策略的改进的次优二阶滑模变结构控制基础上,将其扩展到几类多输入多输出非线性控制系统,并证明了滑模量渐近稳定的充分条件。通过仿真对提出的控制算法进行了验证。③针对掘进机器人行走控制系统中存在的高阶非线性、不确定等问题,将Anosov Unstable(AU)和改进的次优算法相结合,提出了一种基于有限状态机切换的高阶滑模控制方法,解决了一类相对阶为三的不确定非线性系统的有限时间镇定问题。通过对滑模量和滑模量一阶导数以及滑模量一阶导数和二阶导数两个相平面的分析,证明了算法的稳定性,给出了保证滑模量有限时间收敛的充分条件。提出的高阶滑模控制方法能够在滑模量二阶导数未知且无状态观测器的情况下,通过在多个有限状态机之间进行切换,使得滑模量、滑模量一阶导数和滑模量二阶导数在有限时间内收敛到零。通过仿真对提出的控制算法进行验证。④设计并实现了悬臂式掘进机器人截割臂控制系统。根据工作任务和工艺要求,对机器人断面成形过程中,截割臂的工作行程进行了分析,设计了截割头运动轨迹。根据机器人运动特点,建立了机器人运动的空间几何关系,分析了机器人自动截割过程中多个坐标系的转换关系。根据悬臂式掘进机器人断面成形任务要求,构建了机器人截割臂控制的硬件系统。利用本文提出的基于有限状态机的多输入多输出二阶滑模控制方法设计并实现了机器人截割臂控制器。通过仿真和实验对机器人的截割控制进行了验证。⑤设计并实现了悬臂式掘进机器人履带式行走机构控制系统。分析了履带式行走机构转向和跟踪过程,根据机器人实际的运行环境的限制,提出了机器人轨迹跟踪的控制策略。针对掘进机器人的履带式行走机构及其驱动机构的特点,提出了一种分层控制的系统结构,设计了控制系统实时路径规划策略,搭建了机器人履带行走机构控制的硬件系统。根据路径规划策略,设计并实现了主控制回路。采用本文提出的基于有限状态机切换的高阶滑模控制方法,设计并实现了机器人履带的位置控制器。通过仿真和实验对机器人的行走控制进行了验证。最后,论文对研究工作进行了总结,给出论文主要工作和创新点,对掘进机器人和高阶滑模控制进行了展望。更多还原

【Abstract】 Boom-type roadheader robot was a kind of advanced robot used for laneway excavation. The dissertation aim at the poor working environment and high level of risk problems, according to the quality of excavation operation requirements, to study a kind of boom-type roadheader robot control system, implement the unmanned operation under harsh environment such us laneway, underground and so on, and complete its scheduled operations automatically.The dissertation took boom-type roadheader as the target, according to excavation under the manual operation key technologies-cutting control and walking control, studied the cross-section forming and trajectory tracking control algorithm, and achieved the controller and the implemention of system. It had laid a theoretical foundation and provided an experimental platform in order to develop and produce a high efficiency, rapid excavation, robustness, mobility and flexible intelligent, unmanned boom-type roadheader robot.In this dissertation, boom-type roadheader robot control system architecture was proposed. Two key technical problems, including the cross-section forming and trajectory tracking, were analyzed. According to the robot tasks and operational process requirements, combined with the existing Higher Order Sliding Mode Control, some novel high-order sliding mode control methods were presented to achieve a cross-section forming and the robot trajectory tracking. The main study work can be generalized as follows:①A general structure for robot control system was presented. The mechanical structure and working principle of boom-type roadheader robot were analyzed. According to tasks and process requirements, functions and work processes of the robot’s control system were analyzed and designed. The robot control system was decomposed into cutting arm control system and walking control system. The dynamics models of cutting arm and walking mechanism were presented, and some theoretical problems needed to be solved for control system were proposed.②Aim at the multiple variable, nonlinear problems and so on of the cutting arm control of boom-type roadheader robot system, higher-order sliding mode control and a class of second-order sliding mode control based on finite state machine strategy were analyzed. On the basic of an improved single-input single-output sub-optimal second-order sliding mode variable structure control based on finite state machine, several classes of second-order multi-variable sliding mode control for uncertain nonlinear systems were studied to extend it to several types of multi-input multi-output nonlinear control system effectively, and the sufficient condition for asymptotic stability of sliding mode variables were proved. Algorithms proposed were validated by simulation.③Aim at the high order nonlinear, uncertainty problems and so on of the walking control system of boom-type roadheader robot, Anosov Unstable (AU) and improved sub-optimal algorithm were combinated to present high-level sliding-mode control method based on finite state machine which solved the finite-time stabilization problem of a class of uncertain nonlinear systems with relative dregree three. The phase plane of sliding mode variable and its derivative and the phase plane of its derivative and its second-order derivative was analyzed to prove the stability of the algorithm and present the sufficient condition for sliding mode variable to converge in finite time. Proposed high-order sliding mode control method can force the sliding mode variable, its derivative and its second-order derivative to zero in finite time with the knowledge of the sliding mode variable and its derivative. Algorithms proposed were validated by simulation.④The cutting arm control system of boom-type roadheader robot was designed and implemented. According to tasks and process requirements, the cutting arm stroke was analyized during the cross-section forming process and the cutting header trajectory was designed. According to robot characteristics, geometric relationship of robot was established. The conversion between various coordinate systems was analyzed. According to robot’s task requirements, the hardware system of cutting arm control system was built. Cutting arm controllers were designed and achieved by the proposed multiple-input multiple-output second-order sliding mode control method based on finite state machine. The simulation and experiments were carried out to validate the control system.⑤The track type walking control system of boom-type roadheader robot was designed and implemented. The steering and walking process of track type walking mechanism was analyzed. According to the actual robot operating environment constraints, the trajectory tracking control strategy was proposed. According to the characteristic of the track type walking mechanism, a layered control system architecture was proposed. The path planning strategy was designed. The hardware system of track type walking mechanism control system was built. The track postion controller was designed and achieved by the proposed high-level sliding-mode control method based on finite state machine for a class of uncertain nonlinear systems with relative dregree three.The simulation and experiments were carried out to validate the control system.Finally, the paper was summarized and the main research work and innovation point was presented. Boom-type roadheader robot and high-order sliding mode control were prospected.更多还原