Application Study on Compact Autonomous Underwater Vehicle

紧凑型AUV(自主水下航行器)能够执行几项具有挑战性的功能，并使其能够被有限数量的人使用。水下航行器通常配备许多推进器。推进器控制机构复杂，重量比大，使得AUV体积更大、重量更重。我们通过使用一个VARIVEC螺旋桨开发了一款紧凑、轻型的AUV。VARIVEC螺旋桨系统是利用无线电控制直升机元件进行配置的，这些元件是斜盘和直流伺服。VARIVEC螺旋桨可以通过周期性地改变螺旋桨的桨叶角度，在一圈内产生推力、侧向力和力矩六个分量。在该控制系统中，我们实现了三个模块，包括：接收器控制模块：该模块控制支持船任务指令的接收和定位系统应答器的接收。传感器控制模块：该模块控制各种传感器的数据采集，实现自主导航。螺旋桨控制模块：该模块控制任务指令的最终决策，并利用可编程逻辑器件电路生成VARIVEC螺旋桨系统的控制数据。每个模块采用单片微型计算机(日立SH-2)作为AUV的分布式控制系统。我们已经使用电子罗盘、避碰声纳、深度传感器和GPS接收器进行自主导航。AUV还可以通过NTT DoCoMo的DOPA网络与海面上的支持团队进行通信。我们已经实现了模糊和神经网络程序作为软计算控制器。此外，为了考察AUV的导航能力，我们设计了PID控制器。为了对紧凑型AUV进行评估，我们进行了浅海自主导航实验。基于软计算的实验结果表明，该方法是有效的，AUV可以沿目标路径移动。

A compact AUV(Autonomous Underwater Vehicle) is capable of performing several challenging functions and enables it to be used by a limited number of persons. Usually underwater vehicles are equipped with many thrusters. The thruster control mechanism is complicated and this takes a larger weight ratio, making the AUV larger and heavier. We developed a compact and light AUV by employing one VARIVEC propeller. The VARIVEC propeller system is configured utilizing the radio control helicopter elements, which are swash plate and DC servos. VARIVEC propeller can generate six components including thrust, lateral force and moment by changing periodically the blade angle of the propeller during one revolution. In this control system, we have implemented three modules, which include : Receiver control module: This module controls the reception of the mission command from support vessel and the transponder for positioning system. Sensor control module: This module controls the various sensors data acquisition for autonomous navigation. Propeller control module : This module controls the final decision of mission command and generates the control data for VARIVEC propeller system using the programmable logic device circuits. Each module utilizes one-chip microcomputer (Hitachi SH-2) as the distributed control system for the AUV. We have used the electronic compass, collision avoidance sonar, depth sensor and GPS receiver for autonomous navigation. The AUV also can communicate with the support team at the surface of the sea via NTT DoCoMo's DOPA network. We have implemented fuzzy and neural network program as a soft computing controller. Furthermore, we designed the PID controller in order to investigate the navigation ability of the AUV. To evaluate the compact AUV, we have carried out the autonomous navigation experiments in the shallow sea. Experimental results using soft computing are indicated the effectiveness of our approach and the AUV could move along the target path.