基于改进SSA-DNN的工业机器人定位误差补偿研究

Research on Industrial Robot Positioning Error Compensation Based on Improved SSA-DNN

针对工业机器人定位精度不高和传统误差补偿方法复杂的问题,提出了一种改进SSA-DNN神经网络的工业机器人定位误差补偿模型。首先,在笛卡尔空间优化空间网格分割采样规划,获得目标点位与其定位误差的规律;其次,提出一种基于Tent混沌映射的ISSA算法,结合Levy飞行机制,提升SSA算法的搜索能力与加快收敛速度;最后,建立ISSA-DNN定位误差补偿模型,为验证补偿模型的有效性,以自由度工业机器人为对象,与其他模型进行对比实验,实现了对机器人实际点位的补偿,提高了机器人的定位精度。结果表明,在机器人定位误差补偿方面,与DNN与SSA-DNN等神经网络模型相比,ISSA-DNN神经网络模型具有更高的补偿精度和稳定性。

Aiming at the low positioning accuracy of industrial robots and the complexity of traditional error compensation methods,an improved SSA-DNN neural network positioning error compensation model for industrial robots was proposed.Firstly,optimize the spatial grid segmentation sampling planning in Cartesian space to obtain the pattern of target point position and its positioning error.Secondly,a ISSA algorithm based on Tent chaotic mapping is proposed,which combines the Levy flight mechanism to enhance the search ability and accelerate the convergence speed of the SSA algorithm.Finally,an ISSA-DNN positioning error compensation model was established.To verify the effectiveness of the compensation model,a comparative experiment was conducted on a degree of freedom industrial robot with other models to achieve compensation for the actual point position of the robot and improve its positioning accuracy.The results show that in terms of robot positioning error compensation,compared with neural network models such as DNN and SSA-DNN,the ISSA-DNN neural network model has higher compensation accuracy and stability.