An autonomous control technology based on multiple-agent system ( MAS) was proposed to study the un-manned underwater vehicle ( UUV) , which carries out the oceanographic survey mission. An agent-based distribu-ted architecture was used to construct the autonomous architecture. Communication and cooperation between all a-gents were realized in efforts to accomplish the complex mission. The mission of oceanographic survey requires that an underwater vehicle is able to survey oceanography at several local areas in dynamic and uncertain environments using different survey sensors. This method appears convenient for the mission description and management by using waypoints with multiple properties. The mission control was modeled by discrete events driven Petri net formalism. To solve the problem of a lack of central control for synchronous events handling in event management, discrete e-vents were processed by priority to eliminate the events conflict. The feasibility and effectiveness of the proposed al-gorithm were checked by lake and sea trials.%针对水下无人航行器的海洋勘测任务，提出了一种基于多智能体（ MAS）的自主控制技术。自主控制结构采用分布式的智能体结构，智能体之间通过通信和协作来共同完成复杂的整体操作。海洋勘测使命要求航行器在动态不确定的环境中，利用不同的勘测传感器进行多个局部区域的勘测。引入基于多属性路径点的使命描述方法，方便使命描述和控制。使命控制过程采用离散事件驱动的Petri网形式建模。针对事件管理过程中并发事件缺乏中央控制的问题，离散事件被赋予优先级以消除冲突。通过湖海试验验证了所提自主控制技术的可行性和有效性。
展开▼
