面向机器人阵面装配的数字孪生精确碰撞检测方法OA
Digital Twin Accurate Collision Detection Method for Robot Array Assembly
针对工业机器人在执行装配任务时存在的干涉碰撞问题,提出一种面向机器人阵面装配的数字孪生精确碰撞检测方法.根据提出的数字孪生系统框架,构建工业机器人的数字孪生模型,实现物理设备层和虚拟模型层之间的实时映射,基于多层感知机的关节角度预测模型预测工业机器人的未来关节角度,提出的基于混合层次包围盒的碰撞检测方法弥补了 Unity3D 自带碰撞检测的不足,避免机器人在装配过程中可能发生的碰撞问题.结果表明,该数字孪生系统具备可行性和有效性.
To solve the interference and collision problem of industrial robots in assembly tasks,an accurate collision detection method for robot array assembly based on digital twins is proposed.According to the proposed digital twin system framework,the digital twin model of industrial robots is constructed to realize the real-time mapping between the physical equipment layer and the virtual model layer,the joint angle prediction model based on multi-layer perceptron predicts the future joint angle of industrial robots,and the proposed collision detection method based on hybrid hierarchical bounding box makes up for the lack of collision detection in Unity3D.The collision problem that may occur in the assembly process of the robot is avoided.The results show that the digital twin system is feasible and effective.
陈军;刘庭煜
南京理工大学机械工程学院,南京 210094东南大学机械工程学院,南京 211189
信息技术与安全科学
数字孪生工业机器人碰撞检测装配
digital twinindustrial robotcollision detectionassembly
《兵工自动化》 2026 (9)
36-40,47,6
评论