Research On Task Decomposition,Allocation And Action Realization Technology Of Vessel Equipment Multi-Person Cooperative Maintenance

Maintenance simulation is an important means to verify the maintainability of equipment.Because of the special task type and complex component structure of the multiperson collaborative maintenance operation for vessel equipment in the virtual environment,there are some problems in the process of conducting maintenance verification and simulation analysis of vessel equipment,such as difficulty in fully describing the maintenance process,inability to allocate maintenance tasks reasonably,and deficiency in establishing a connection between maintenance planning and maintenance simulation.In combination with the domestic and foreign theoretical achievements,this paper conducts in-depth research on fundamental technologies that include vessel equipment maintenance task decomposition,multi-person collaborative task assignment,and virtual human simulation action generation.This paper mainly includes the following aspects:To begin with,according to the actual maintenance operation types of vessel equipment,a maintenance task decomposition approach based on hierarchical task network is proposed so that the maintenance tasks are decomposed to the meta-task level.Then the mapping relationship between the maintenance task decomposition and the hierarchical task network is established.Through HTN planning and matching,the vessel equipment maintenance tasks are decomposed into an orderly meta-task sequence.Secondly,the mathematical model of multi-person collaborative maintenance task allocation is developed after the constraint relationship of different meta-task types is expounded.This chapter subsequently explains the encoding method,objective function and initial solution generation method,and puts forward a meta-task allocation solution method that is built on particle swarm algorithm.The defect of particle swarm algorithm that is easy to fall into local optimum is improved combing the analysis of solving process using particle swarm algorithm.Taking the repair of a certain power equipment as an example,the effectiveness and superiority of the algorithm are verified through paired comparison.Subsequently,the execution process of meta-task in virtual environment is studied.By dividing the maintenance workspace,an adaptive behavior judgment mechanism for different meta-tasks is established,and the virtual human motion is proposed based on inverse kinematics and bounding box collision detection.After the modular division of the kinematic chain,the adaptive generation of key actions when the virtual human performs meta-tasks is completed.The superiority of the algorithm is compared and analyzed at the end of this chapter.Finally,the corresponding functional modules are completed using the secondary development of DELMIA software.The usage of different functional modules is explained through a case study of maintenance task execution in the virtual environment.At the same time,the generated multi-person collaborative virtual maintenance simulation process is evaluated to verify the rationality and accuracy of the virtual human actions.The above research results show that the method and theory proposed in this paper can easily help to generate the simulation process of warship equipment multi-person cooperative maintenance,which provides a new idea and method for warship equipment maintainability verification and simulation analysis.