基于RBF-BP算法的机械臂多自由度分拣控制系统设计

Design of multi-degree-of-freedom sorting control system for manipulator based on RBF-BP algorithm

摘要:

为提升机械臂设备的准确分拣能力，设计基于RBF-BP算法的机械臂多自由度分拣控制系统。利用RBF-BP前馈神经网络，规划机械臂设备的运动轨迹路线，实现基于RBF-BP算法的机械臂运动轨迹建模。确定工业相机在总功能框架中所处连接位置，根据机械臂选型情况，确定可编程逻辑控制器、变频控制器对于所选机械臂元件的调节能力，完成对系统功能模块的开发。借助传输信道，将Sorting分拣指令、ToolControl控制指令反馈回核心控制主机，建立完整的指令执行回路，再联合相关硬件设备结构，实现基于RBF-BP算法的机械臂多自由度分拣控制系统设计。实验结果表明，在抓取能力相同的情况下，应用RBF-BP算法控制系统机械臂成功分拣的工件数量为19件，机械臂抓取成功率为95%，说明所设计系统满足提升机械臂准确分拣能力的设计初衷。

Abstract:

In order to improve the accurate sorting capability of the robotic arm equipment, a robotic arm multi-degree-of-freedom sorting control system based on the RBF-BP algorithm is designed. The RBF-BP feedforward neural network is used to plan the motion trajectory of the manipulator equipment, and the motion trajectory modeling of the manipulator based on the RBF-BP algorithm is realized. Determine the connection position of the industrial camera in the overall functional framework. According to the selection of the manipulator, determine the adjustment capability of the programmable logic controller and the variable frequency controller for the selected manipulator components, and complete the development of the system function module. With the help of the transmission channel, the Sorting order and ToolControl control order are fed back to the core control host, and a complete order execution loop is established, and then combined with the relevant hardware equipment structure to realize the design of the robotic arm multi-degree-of-freedom sorting control system based on the RBF-BP algorithm . The experimental results show that under the condition of the same grasping ability, the number of workpieces successfully sorted by the robotic arm controlled by the RBF-BP algorithm is 19, and the grasping success rate of the robotic arm is 95%, indicating that the designed system meets the requirements of lifting the robotic arm. The original intention of the design of accurate sorting capabilities.