【摘要】 为解决汽车电动座椅出厂前位姿调整过程中的耦合现象,提出了一种基于机器学习的座椅位姿调整方法。根据座椅内部机械结构证明了座椅表面特定点之间的位置是线性相关的;利用激光测距传感器采集这些特定点的位置信息,在此基础上由线性分类器对岭回归模型进一步训练,得出分段岭回归模型并给出调整方案,通过实验验证了所提方法的可行性。实验结果表明:经过分段岭回归模型的预测,每个开关仅需要拨动一次就能使座椅位姿达到出厂标准,并降低了座椅位姿调整的误差,提高座椅位姿调整效率。更多还原

【Abstract】 To solve the coupling phenomenon in the adjustment process of auto electric seat before delivery,a method of seat pose adjustment based on machine learning is proposed. According to the internal mechanical structure of the seat,the position between specific points on the seat surface is linearly dependent. The location information of these points is collected by laser range sensor,based on the information,the piecewise ridge regression model,which is derived from the further training of ridge regression model by linear classifier is established and the adjustment scheme is given. The feasibility of the proposed method is verified by experiments,and the experimental results show that,according to the prediction of the piecewise ridge regression model,each switch only needs to be moved once to make the seat posture meet the factory standard and the error of seat posture adjustment can be reduced,so the proposed method can significantly improve the efficiency of seat posture adjustment.更多还原