基于 PSO-BP 神经网络的分拣机器人视觉反馈跟踪

针对分拣机器人视觉反馈跟踪精度差、耗时较长的问题，研究基于粒子群算法-反向传播(particle swarm optimization-back propagation, PSO-BP)神经网络的分拣机器人视觉反馈跟踪方法，以提升视觉反馈跟踪效果。依据分拣机器人的视觉反馈信息，建立分拣机器人运动学模型，并求解分拣机器人机械臂输出位置和输入位置的误差函数；利用PSO算法优化BP神经网络的权值与偏置；在权值与偏置优化后的BP神经网络内，输入误差函数，预测分拣机器人视觉反馈跟踪控制量；利用预测视觉反馈跟踪控制量，在线调整增量式比例-积分-微分(proportional-integral-derivative, PID)的参数，输出高精度的分拣机器人视觉反馈跟踪控制量，实现分拣机器人视觉反馈跟踪。实验结果表明，该方法可有效视觉反馈跟踪分拣机器人机械臂的关节角；存在干扰情况下，在运行时间为10 s左右时，阶跃响应趋于稳定；有干扰情况下，视觉反馈跟踪的平均误差为0.09 cm,耗时平均值为0.10 ms;无干扰情况下，平均误差为0.03 cm,耗时平均值为0.04 ms。

Visual feedback tracking of sorting robot based on PSO-BP neural network

To address the issues of poor visual feedback tracking accuracy and long time consumption in sorting robots,a visual feedback tracking method based on PSO-BP neural network for sorting robots is studied to improve the visual feedback tracking effect.Based on the visual feedback information of the sorting robot,the kinematics model of the sorting robot was established,and the error function of the output position and input position of the sorting robot arm was solved.PSO is used to optimize the weight and bias of BP neural network.In the BP neural network optimized by weight and bias,the error function is input to predict the visual feedback tracking control quantity of sorting robot.The predictive visual feedback tracking control quantity is used to adjust the parameters of incremental PID online,and the high-precision visual feedback tracking control quantity of sorting robot is output to realize the visual feedback tracking of sorting robot.Experimental results show that this method can effectively visually track and feedback the joint angles of the sorting robot''s mechanical arm.In the presence of interference,the step response stabilizes around 10 seconds of operation time.When there is interference,the average error of visual feedback tracking using this method is 0.09 cm, with an average time consumption of 0.10 ms.Without interference,the average error is 0.03 cm,and the average time consumption is 0.04 ms.