【作者】 刘晓恒；

【导师】 牛志刚；

【作者基本信息】 太原理工大学 ， 机械制造及其自动化， 2010， 硕士

【摘要】 煤矿井下探测机器人的主要作用是替代人员进入灾后的矿井,探测井下的环境信息以及人员的伤亡情况,并将其发送到指挥中心。因此,煤矿井下探测机器人的研究对减少人员伤亡和煤矿救灾工作的顺利开展有重要意义。课题来源于山西省科技攻关项目“煤矿井下搜救探测机器人技术与系统研究”,本文研究的路径规划和运动控制技术是该课题研究的关键技术,是煤矿井下探测机器人顺利完成其探测任务的保障。煤矿井下探测机器人路径规划的主要任务是在灾后井下障碍物不确定的环境中,寻找一条从起始点到目标点的安全避碰路径,且所走路径较短、能源消耗最少。结合煤矿井下探测机器人的特点,首先,本文采用了基于栅格的机器人局部环境建模方法;其次,根据标准微粒群算法容易陷入局部最优的缺点,通过对微粒的重新随机初始化,改进了微粒群的算法,并通过仿真验证了改进算法的可行性;再次,在研究微粒群算法的基础上,提出了基于行为学和微粒群算法的煤矿井下探测机器人的路径规划方法,根据行为学把机器人的路径规划行为分解为驶向目标的行为和避障行为,同时根据滚动扫描区域里障碍物信息的不同来确定机器人的行为,并通过应用兼顾全局的子目标的确定方法,也使该路径规划在保证局部最优的同时兼顾了全局最优;最后,通过对基于行为学和微粒群算法的路径规划方法的MATLAB仿真,验证了本文算法的有效性。根据煤矿井下探测机器人的机构特点,把煤矿井下探测机器人的运动控制主体分为了主驱动模块和摇臂驱动模块。首先,根据轮式机器人的运动学模型,推导出了主驱动模块的运动学方程和在直行、转弯时两个履带的速度公式;其次,通过把机器人的摇臂简化为空间连杆开链机构,推导出了机器人摇臂模块的运动学方程;再次,结合煤矿井下探测机器人的工作条件和自身特点,对其运动控制系统的硬件和软件系统进行了选型和设计,其中,经过对软件系统中的“路径动态显示”窗口和“图像采集窗口”的调试,不仅实现了机器人合理路径、障碍物信息和井下环境的可视化,并也为机器人的远程监控与控制奠定了基础;最后,应用Pro-Engineer软件建立了煤矿井下探测机器人的运动学仿真模型,并通过ADAMS仿真软件对主驱动模块的运动学方程进行了仿真,验证了运动学方程的有效性和规划路径的可行性。更多还原

【Abstract】 Main effect of Coal Mine Detection Robot is taking place of human to go into the post-disaster coal mine for detecting information of the environment and sending the information to the command center. Therefore, it is of great significance to research on Coal Mine Detection Robot for reducing casualties and unfolding of the saving work smoothly. The issue of“Research on technologies and systems of coal mine rescue and detection robot”is scientific and technological project in Shanxi Province. Researches on path planning and motion control are not only the key technologies of the issue but also the guarantee on completing the detection task of Coal Mine Detection Robot.Main work of path planning for Coal Mine Detection Robot are not only to find a safe and collision-free path from the starting point to end point in the environment of post-disaster coal mine with the uncertain obstacles in it but also to make the path shorter and energy consumed less. Firstly, in this paper, grid-based method is used in robot local environmental modeling. Secondly, because the exist of disadvantage of standard Particle Swarm Optimization is easy to fall into local optimality. Particle Swarm Optimization is improved by random initialization of the particles again. And simulation results show the feasibility of the algorithm. Thirdly, basing on the research on the Particle Swarm Optimization, a path planning method of the Coal Mine Detection Robot is propounded which base on Behavior and Particle Swarm Optimization. Behavior of the Robot’s path planning is decomposed into Trending To Target behavior and Obstacle Avoidance behavior according to the Behavior. Robot’s current behavior is established by information of obstacle in the rolling scan region. And the method of path planning take into account the global optimality at the same time of ensuring the local optimality by applying to the method of establishing the sub-goals with taking into account the overall situation. At last, validate the effectiveness of the algorithm by simulation of the Path Planning algorithm base on Behavior and Particle Swarm Optimization in MATLAB.According to mechanism’s characteristics of Coal Mine Detection Robot, the body of motion control is divided into Master Module and Rocker Arm Module. Firstly, the kinematics equations of Master Module and the formula of speed on going straight and turning are derived according to the kinematics equations of wheeled mobile robot. Secondly, the kinematics equations of Rocker Arm Module are derived according to simplifying the robot’s rocker arms as open chain mechanisms of space connecting rod. Thirdly, hardware system and software system of Motion Control are selected and designed. According to debugging the windows of“Path Dynamic Display”and“Images Capture”successfully, robot’s reasonable path, obstacles’information and environmental information in coal mine are viewed immediately, and it is the foundation of remote monitoring and remote control. At last, kinematics simulation model of Coal Mine Detection Robot is established by applying to software of Pro-Engineer, and the kinematics equations of Master Module are simulated by applying to software of ADAMS successfully. They not only testify to the valid of the kinematics equations but also the feasibility of the path planned.更多还原