【作者】 褚刚秀；

【导师】 边信黔；

【作者基本信息】 哈尔滨工程大学 ， 控制理论与控制工程， 2003， 硕士

【摘要】 自治式水下机器人作为重要的潜水设备，它的研究不仅有重要的经济意义，也是军事和政治的迫切需要。路径规划是水下机器人导航中最重要的问题之一，因为未知障碍物有可能出现在事先规划好的路径上，这样在机器人的导航中需要基于传感器信息的以在线的方式进行避碰规划。水下机器人的避碰能力是是否具有环境适应能力的一种具体体现。由于水下机器人在海底是空间运动，本文研究的主要是3维的实时路径规划。基于前视声纳的视区和水下机器人的姿态动力学特性，根据基于规则的路径规划算法来规划机器人接近最短的路径。 本文分析了水下机器人的水动力模型，并基于水下机器人的运动数据确定出整个运动闭环控制系统的数学模型，设计出基于规则的避碰控制系统。设计和实现了整个仿真系统的系统软件，进行了大量的仿真试验，并对实验数据进行了系统的分析。系统仿真试验结果验证了基于规则的局部算法是可靠的，并且适用于任何机器人的3维实时路径规划。 QNX是实时、多任务、多用户、可嵌入的操作系统。为了满足实时性的要求，路径规划在用C语言在QNX操作系统下仿真，本文简单介绍了QNX操作系统的体系结构。 本文简单的介绍了六自由度潜器模拟试验平台，并对潜器的基于传感器实时路径规划算法进行了物理模拟实验验证。更多还原

【Abstract】 Autonomous Underwater Vehicle(AUV) is one kind of important underwater equipment . It has profound meaning to develop the technology of the AUV at econmic territory and in matrial and politics struggle. Path planning is a important subject in the navigation system of the robotic. The robotic may meet with the unknown obstacle at the path planned, so it is necessary to plan path on-line based on sensor in the system of the navigation. The ability avoiding obstacles is the reflection of the adaption to the environment.The paper studies path planning strategies on real-time because of robitic moving in the space under the ocean. Based on the sensor vision zone and AUV attitude dynamics, a 3D near-shortest path planning algorithm is developed for the AUV in the unknown environment.The paper studies the AUV hydrodynamics model , the mathmaticals model of the closed control system based on the data from AUV moving and the obstacles avoidance control system based on th rule. The result of simulation demonstrate that the algorithm are powerful and reliable for the 3-D path planning of the AUV in the unknown environment and it can be generally used to on-line plan 3-D path for other kind robots.The operating system of QNX is the real-time, muti-task, muti-user embedding system .The path planning is simulated on the operating system of QNX for the meeting of the real-time. The paper describes the structure of operating system of QNX.The paper describes the experiment of the vehicle six dimension. That validates the real-time path planning for sensor-based AUV in unknown environment更多还原