面向精细乐器维修的机械臂运动自动规划系统研究

Research on automatic planning system of robotic arm motion for fine musical instrument maintenance

乐器由于其精密、复杂的结构和微妙、多变的音响特性,对维修工作提出了极高的要求。传统的手工维修方法不仅效率低下,还可能因为维修者的技术水平和经验差异导致维修质量不稳定。基于此背景,此次研究首先结合深度学习技术搭建了乐器故障物识别模型,然后对传统的路径优化算法进行改良,搭建了面向乐器维修机械臂的避障路径优化模型。研究结果表明,所设计的故障识别模型与路径优化模型均具有较好的性能。其中,识别模型的最高识别精度可达0.962,路径优化模型的最高避障精度可达0.97。综上,将此次研究所提出的故障识别模型与路径优化模型用于乐器维修机械臂系统中,能够有效改善乐器的自动维修效果。

Musical instruments pose extremely high requirements for repair work due to their precise and complex structures and subtle and variable acoustic characteristics.Traditional manual repair methods are not only inefficient,but also may lead to unstable repair quality due to the difference in the skill level and experience of the repairers.Based on this background,this study first builds a musical instrument fault object recognition model by combining deep learning technology,and then improves the traditional path optimization algorithm and builds an obstacle avoidance path optimization model for the musical instrument repair robotic arm.The research results show that the designed fault recognition model and path optimization model both have better performance.The highest recognition accuracy of the recognition model is up to 0.962,and the highest obstacle avoidance accuracy of the path optimization model is up to 0.97.In conclusion,the fault recognition model and path optimization model proposed in this study can be used in the musical instrument repair robotic arm system to effectively improve the effect of automatic repair of musical instruments.