Multi-Mode Active Suspension Control Based on a Genetic K-Means Clustering Linear Quadratic Algorithm
Round 1
Reviewer 1 Report
The authors proposed a new GKL method for suspension control in this work. In the reviewer’s opinion, the manuscript is of great interest to the automotive researchers and manufacturers, and the structure is well organized and has enough novelty. I recommend accepting this paper with the following comments.
(1). The clustering idea for weight values is very interesting. But please add more discussions on its contribution, especially in the introduction section.
(2). Most vibration simulations only take a certain road input for the 1/4 model. However, in this work, four different roads are put together in turn. Please justify why using these combined road surfaces.
(3). Besides the K-means clustering method, there are many other machine learning algorithms available. Why do the authors not select relatively popular algorithms, such as Support Vector Machine or Deep Convolution Network? If applicable, please also discuss different machine learning algorithms.
(4). If possible, please compare the simulation results with the results in the existing literature. This is optional for authors.
(5). In the conclusion section, please discuss the potential applications of the proposed method.
(6). The parameter n0 in Table 1 is missing units.
(7). In the parameter definition for Equation 13, the variable names are incorrect.
(8). The quality of several figures in the manuscript should be improved. For example, the axis labels, legends, and data points in Figure 5 are not clear.
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The Author's have done extensive work and documented well. However there are few minor comments given here.
- How parameters in GA are taken? any reference? (Table2)
- The weighting co-efficients values change with change in road input or change in simulation speed? The weights are independent of expert experience, these can not be generalized based on parametric study?
- In Figure 6, the response of the 1/4th vehicale model to each road class (A,B,C,D) needs to be presented seperate graphs for better understanding other wise most of the curves are overlapping.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
In this paper, a new Genetic K-means clustering Linear quadratic method is proposed for suspension control. Based on the
traditional linear–quadratic regulator control and genetic algorithm, the machine learning idea is introduced to
obtain more objective weighting coefficients for different driver types.
This paper is well written focusing on the important theoretical aspects and their application .
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.
