Hierarchical Stochastic Optimization With Application to Parameter Tuning for Electronically Controlled Transmissions
Hierarchical Stochastic Optimization With Application to Parameter Tuning for Electronically Controlled Transmissions
复制标题
DOI：
10.1109/lra.2020.2965085
复制
发表时间：
2020-01
影响因子：
5.2
通讯作者：
H. Karasawa;T. Kanemaki;Kei Oomae;R. Fukui;M. Nakao;Takayuki Osa
中科院分区：
文献类型：
--
作者：
H. Karasawa;T. Kanemaki;Kei Oomae;R. Fukui;M. Nakao;Takayuki Osa
在机械系统中，控制参数通常由专家通过反复试验手动调整，这是劳动密集型和耗时的。此外，这个问题的困难在于，往往存在提供高回报的多种解决方案。由于设计的目标函数在实践中往往不是最优的，因此提供最高回报的解决方案可能不是最优解决方案。因此，通常需要验证解决方案的多个候选方案，以确定最适合实际系统的解决方案。为了解决这个问题，我们提出了一个参数优化系统，使用分层随机优化（HSO），可以处理多峰目标函数。在电控变速器的案例研究中，优化器学习满足所有约束的多组参数，并优于人类工程师手动设计的参数。实验表明，该算法能够识别目标函数的多种模态，比现有的交叉熵方法和协方差矩阵自适应进化策略等方法具有更高的样本效率，并且是一个人类工程师。
In mechanical systems, control parameters are often manually tuned by an expert through trial and error, which is labor-intensive and time-consuming. In addition, the difficulty of this problem is that there often exist multiple solutions that provide high returns. As a designed objective function is often not optimal in practice, the solution that provides the highest return may not be the optimal solution. Therefore, it is often necessary to verify the multiple candidates of the solution to identify the one most suitable for the actual system. To address this issue, we propose a parameter optimization system using hierarchical stochastic optimization (HSO) that can handle multimodal objective functions. In a case study of electronically controlled transmissions, the optimizer learns multiple sets of parameters that satisfy all constraints and outperforms the parameters manually designed by human engineers. We demonstrate experimentally that our HSO can identify several modes of the objective function and is more sample-efficient than the existing methods, such as cross-entropy method and covariance matrix adaptation evolution strategy, as well as a human engineer.