基于模糊神经网络的海底采矿车路径跟踪行走控制研究

【作者】 张敏；

【导师】 李力；

【作者基本信息】 中南大学 ， 机械电子工程， 2010， 硕士

【摘要】 富钴结壳是一种赋存于800-2500m海山侧表面上的海底矿产资源。钴结壳海底采矿车采用具有良好越障性能和爬坡能力及机动性能的铰接式履带车作为其行走机构。海底采矿车按预定路径行走控制为采矿作业关键技术之一。本文以海底采矿车多刚体动力学模型为基础,提出了一种基于ANFIS的模糊神经网络行走控制方法,并建立了基于Simulink、ADAMS和LabVIEW的海底采矿车协同仿真模型,从而为海底采矿车自动行走控制研究提供了新的方法和思路。首先,设计了以Compact RIO控制器为核心的海底采矿车控制系统方案,提出了在LabVIEW测控平台上采用内环速度控制和外环路径控制的智能自动行走控制方案。在此基础上,开展了海底采矿车运动学分析和建模,构建了基于ANFIS的模糊神经网络外环路径控制和PID内环速度控制的按预定路径行走控制模型；同时,基于ANFIS神经模糊推理系统,根据训练数据,完成了模糊神经网络路径控制器设计；设计了PID控制器,并对参数进行了整定；建立了速度分配模块、滑转率控制模块、滑转率实时计算模块和延时模块等模型。其次,建立了海底采矿车动力学机械模型与按预定路径行走控制模型的协同仿真模型,开展了海底采矿车越单边障碍、偏离路径跟踪、爬坡滑转率控制的直线路径跟踪控制的仿真研究,仿真结果表明,海底采矿车行走路径跟踪控制和滑转率控制效果良好,表明所设计的基于ANFIS的模糊神经网络的行走控制算法是有效的。然后,利用SIT Sever,建立了基于Simulink、ADAMS和LabVIEW的海底采矿车协同仿真模型,实现了Simulink、ADAMS和LabVIEW三个软件的协同仿真,为海底采矿车自动行走控制提供了技术支持。最后,开展了海底采矿车模型机行走实验,实现了海底采矿车行走控制算法与LabVIEW测控系统的联接,验证了所设计的基于ANFIS的模糊神经网络外环控制和PID内环速度控制的算法是有效的。更多还原

【Abstract】 Cobalt-rich crust is the seabed mineral resource which exists on the surface of the marine hills.Beacause articulated tracked vehicle is of good obstacle performance,climbing ability and advanced mobility, articulated tracked vehicle is selected as the running machine for cobalt crust mining.It is a key technology that cobalt crust mining walks on the Scheduled path.On the foundation of multi-body dynamic model of seabed mining vehicle,a fuzzy neural network running control method is presented based on ANFIS,and a co-simulation model is established which is based on Simulink, ADAMS and Lab VIEW. So new methods and ideas are provided for walking control research of seabed mining vehicle.First of all,the control scheme of seabed mining vehicle is designed which takes Compact RIO controller as a core.And the intelligent automatic running control program is proposed which is of inner loop using speed control and outer loop using path control with Lab VIEW. On this basis,the kinematic analysis and modeling of seabed mining vehicle is carried out. And the walking control model of outer loop using path control and inner loop using PID speed control is established based on ANFIS. According to the training data, the path controller of fuzzy neural network is designed based on ANFIS.Then, the PID controller is designed and parameters is tuned.At the same time, the model of speed distribution module,slip rate control module,slip rate calculation module and real-time delay module is established.Secondly,the co-simulation model of a dynamic mechanical model and a scheduled path walking control model is established.And the simulation study of straight path tracking control about steping over unilateral barriers,deviating from the path and slip rate control when the seabed mining vehicle climbing is carried out. A set of simulation results is presented,showing the validity of the proposed control method based on ANFIS.Afterwards, a co-simulation model is established which is based on Simulink,ADAMS and Lab VIEW by SIT Sever,, achieving the purpose of three softwares co-simulation, so the technical support is provided for the automatic control of.Finally,the model experiments about seabed mining vehicle are carried out, which achieves the connection of the control algorithm and LabVIEW.So it is verified that the validity of the proposed control method of seabed mining vehicle.更多还原