To reduce the grinding trajectory deviation caused by the absolute positioning accuracy of robot, a trajectory compensation method based on Co-Kriging space interpolation method is proposed. Meanwhile, an adaptive iterative constant force control method based on one-dimensional force sensor is proposed to improve the processing quality and efficiency of robot belt grinding. Firstly, an error model based on 6 DOF robot is constructed. Then, considering the workspace of robot belt grinding and the similarity of robot position error, the Co-Kriging compensation algorithm is used to compensate the grinding trajectory, which makes the compensation process convenient and accurate. Then, a grinding dynamics model based on deformation is established, and an adaptive iterative constant force control is proposed for complex robot belt grinding process, which overcomes the instability of grinding force and shortens its convergence time. Finally, the grinding trajectory compensation experiment and the force control experiment of spherical workpiece are carried out. The results show that the space interpolation compensation algorithm based on Co-Kriging method can significantly improve both the space position error of grinding trajectory and the actual error of workpiece, which proves the feasibility of compensation algorithm. Through force control algorithm, the grinding force fluctuation is maintained within 2 N, the mean value, standard deviation and variance of absolute value of force error are significantly reduced, the convergence rate of grinding force and the roughness of workpiece are much better than before, which shows the effectiveness of the proposed force control algorithm.

为了减少机器人绝对定位精度引起的磨削轨迹偏差，提出了一种基于协同克里格空间插值方法的轨迹补偿方法。同时，提出一种基于一维力传感器的自适应迭代恒力控制方法，以提高机器人砂带磨削的加工质量和效率。首先，建立了基于六自由度机器人的误差模型。然后，考虑到机器人皮带磨削的工作空间以及机器人位置误差的相似性，采用协克里金补偿算法对磨削轨迹进行补偿，补偿过程方便，准确。然后，建立了基于变形的磨削动力学模型，并针对复杂的机器人皮带磨削过程提出了自适应迭代恒力控制，克服了磨削力的不稳定性，缩短了收敛时间。最后，进行了球形工件的磨削轨迹补偿实验和力控制实验。结果表明，基于Co-Kriging方法的空间插补补偿算法可以显着改善磨削轨迹的空间位置误差和工件实际误差，证明了该补偿算法的可行性。通过力控制算法，将磨削力波动保持在2 N以内，显着减小了磨削力误差的均值，标准偏差和绝对值方差，磨削力的收敛速度和工件的粗糙度均大大提高。这说明了所提出的力控制算法的有效性。