Abstract: Aiming at the problems of low efficiency and low applicability of virtual human pose generation technology used in the current maintenance verification analysis of civil aviation maintenance,a method for generating virtual human poses based on multi-objective optimization algorithm was proposed.The skeleton model of the virtual human was established according to the Euler quaternion operation rules.Then the target position of the maintenance was determined through the given engine design layout.The constraint objective function of each machine factor was constructed by comprehensively considering the accessibility,balance,joint rotation range and field of view factors in maintenance.On this basis,the virtual human attitude was optimized by using particle swarm optimization algorithm,so as to realize the generation of maintenance work posture of the virtual human.The experimental results showed that the proposed method could realize the generation of the virtual human working posture when the target working point was known in the maintenance work,and had good effectiveness,applicability and feasibility.

Key words: aircraft maintenance, virtual human, posture generation, multi-objective optimization

摘要： 针对现阶段民航机务维修性验证分析中使用的虚拟人姿态生成技术存在效率低下、适用性较低等问题,提出一种基于多目标优化算法的虚拟人姿态生成方法。根据欧拉四元数运算规则建立虚拟人的骨架模型,通过给定的发动机设计布局确定维修的目标位置,综合考虑机务维修中的可达性、平衡性、关节转动范围以及视域范围因素构建各人机因素的约束目标函数,在此基础上运用粒子群算法对虚拟人姿态进行优化求解,实现虚拟人维修工作姿态的生成。验证实验结果表明,所提方法可以实现机务维修工作中目标作业点已知时虚拟人作业姿态的生成,拥有较好的有效性、适用性和可行性。

关键词: 机务维修, 虚拟人, 姿态生成, 多目标优化