【摘要】 针对自动装配规划和交互式规划都存在不足,将虚拟现实和仿生算法结合起来,提出一种生成优化装配序列的新方法。建立基于几何约束的虚拟装配环境,在该环境中根据经验和知识进行交互式拆卸,定义优先约束表来表达零件间的优先约束关系。应用蚁群算法规划出初始优化的装配顺序,再在虚拟环境下进行仿真、评价和优化,考虑装配位置可达性和工具操作等因素,识别新的优先约束和评价准则,重新规划出更优的装配顺序,不断反复和完善,直到得到满意的最佳装配顺序为止。通过实例验证该方法有效性。更多还原

【Abstract】 Assembly sequence planning can be accomplished by using automated or interactive approaches, both kinds of approach have limitations. A new method to generate optimizedassembly sequence is presented by combining virtual reality and bionic algorithm. Geometry constraint based virtual assem- bly environment is set up, and the precedence constraint rela- tionships between parts are generated by virtual disassembly and represented by defining a precedence constraint table. Then, Ant colony optimization algorithm is applied to generate initial optimized assembly sequence. This sequence is simulated, evaluated and optimized in virtual environment. During the assembly process new precedence constraints and evaluation rules will be recognized considering part accessibility and tool verification, a more practical and optimal assembly sequence is then replaned, and once again evaluated by virtual assembly simulation. This is an iterative process until the best assembly sequence is obtained. An application case is studied to demon- strate the effectiveness of the method.更多还原