Research on Vehicle Active Steering Stability Control Based on Variable Time Domain Input and State Information Prediction
, Jianbo Feng 2, Chao Wang 1, Yu Cao 1, Bonan Qin 1, Tao Zhang 1,*, Senqi Tan 3, Riya Zeng 1, Hongbin Ren 4, Tongxin Ma 1, Youshan Hou 1 and Jie Xiao 1
Round 1
Reviewer 1 Report
According to the content of the manuscript, the author's team has conducted sufficient research and reasonable research on the active steering control of vehicles. The research process and results are consistent with the process of scientific research. I suggest that journals may consider accepting this article. But before the paper is accepted, I think there are still the following problems that need to be further corrected by the author team.
1. The author needs to further standardize the language expression in the manuscript.
2. For the relationship between LSTM network and variable scale input, whether the author needs to further explain in detail.
3. The author needs to explain the necessity of the set coupling conditions.
4. Figure 8(a) and Figure 8(b) are both shown as road information input. Do the authors need to further explain the difference between them?
5. For the in-depth discussion of the research results and their impact on the actual vehicle control, the author should be more in-depth.
Author Response
Reply to review 1
According to the content of the manuscript and the comments of the reviewer, we have revised and replied item by item.
According to the content of the manuscript, the author's team has conducted sufficient research and reasonable research on the active steering control of vehicles. The research process and results are consistent with the process of scientific research. I suggest that journals may consider accepting this article. But before the paper is accepted, I think there are still the following problems that need to be further corrected by the author team.
- The author needs to further standardize the language expression in the manuscript.
Answer: According to your comments, we have carefully reviewed the contents of the manuscript and made corresponding modifications.
Revision:
1) The position of the modified content is from line 33 on page 1 to line 35 on page 1.
……and brain-inspired intelligence technology [3], vehicle chassis system control is gradually changing from expected control of single performance to synchronous optimization of multiple performance.
2) The modified content is located on line 39 on page 1.
In intelligent transportation and information network……
3) The position of the modified content is from line 45 on page 2 to line 51 on page 2.
Although there have been in-depth research on the state information fusion prediction [23] and chassis active integrated control [24], further exploration and verification are still needed for corresponding deficiencies. In the case of external input and excitation disturbance, whether the prediction of target vehicle state can be realized also requires investigation [25]. On the other hand, on the premise of obtaining existing data and conducting adequate training, how to realize vehicle stability robust control based on online prediction of artificial intelligence algorithm data has become a scientific question that……
4) The position of the modified content is from line 8 on page 6 to line 9 on page 6.
In order to avoid steering wheel jitter induced by small external disturbances at high speed, the dead zone range of steering wheel angle is set. Only when the steering wheel angle of the vehicle is outside the dead zone……
5) The position of the modified content is from line 36 on page 6 to line 37 on page 6.
Since real data do not exist in isolation from each other, there is an inevitable correlation among different series of data and between the front and back data in the time series [30] ……
- For the relationship between LSTM network and variable scale input, whether the author needs to further explain in detail.
Answer: Thank you for your pertinent comments. We have explained the corresponding contents in the manuscript.
Revision: The position of the modified content is from line 12 on page 8 to line 15 on page 8.
In the prediction process, the length of the data state is optimized according to the real-time calculation error. The calculation process is shown in Section 3.2 below. According to the real-time dynamic prediction results, the system state variable prediction results under the confidence interval can be obtained.
- The author needs to explain the necessity of the set coupling conditions.
Answer: Thank you for your comments. We have further explained the corresponding contents in the manuscript.
Revision: The position of the modified content is from line 35 on page 13 to line 3 on page14.
In order to validate the robustness and effectiveness of the system steering control strategy under external disturbance and excitation, different working conditions of road elevation information varying with driving mileage are constructed, so as to fully verify the system stability control effect under non ideal road conditions. At the same time, the influence of external stochastic excitation on the steering system also needs to be fully considered.
- Figure 8(a) and Figure 8(b) are both shown as road information input. Do the authors need to further explain the difference between them?
Answer: Figure 8(a) shows the change of road elevation with driving mileage, while Figure 8(b) shows the size of random excitation on the left and right sides of the vehicle at the same location, which has certain differences.
In order to further reflect the differences between the two, we also made corresponding modifications to the manuscript content.
Revision: The position of the modified content is from line 14 on page 14 to line 17 on page 14.
……The difference between Figure 8 (a) and Figure 8 (b) is the coupling condition setting of road elevation information and external road excitation. The proposed control strategy is verified by the superposition of multi-source excitation in the time domain.
- For the in-depth discussion of the research results and their impact on the actual vehicle control, the author should be more in-depth.
Answer: Thank you for your comments. We have supplemented the corresponding contents in the manuscript.
Revision: The position of the modified contents are respectively from line 24 on page 15 to line 28 on page 15, and is from line 14 on page 17 to line 21 on page 17.
1) Although according to the data results in Table 2, whether the predictive controller exists or not, its corresponding root mean square value of vehicle state is relatively close. However, it can be seen intuitively from Figure 9(c) that under the input of prediction information, the vehicle state recovers to the stable region faster, which is particularly important for vehicle stability control under complex coupled working conditions.
2) Although the state data of the vehicle system with or without the predictive controller in Table 3 are similar, it can be seen from the data change curve in Figure 11(c) and Figure 11(d) that the vehicle system with the predictive controller can keep the vehicle stable in a safe area. Moreover, because of the input of predictive data, the system state can quickly return to the safe area, which is critical to the safety and stability of vehicles during driving. Table 2 and Table 3 only further quantify the effect of control, but the actual control effect must be reflected according to the response curve in Figure 9 and Figure 11.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Lines 153-155 and 156-158 seems be the same. Autors mast choice the best text and eliminate the other one.
It is not explained the origin of Figure 2. It is taken from some model by other author or experimental data?
Figure 8b must be improved, with another line width that allows compare better both lines.
Figure 9 and 11 must be improved using different colour for the green line, because green is not visible enough.
Table 2 and table 3 exceed margins. Moreover, information of these tables seem not be really relevant because difference between predictive and non predictive control are very reduced for chosen parameters. However, graphic results shown in figures justify the good behaviour of the proposed control. Why these parameters are chosen for tables?
Author Response
Reply to review 2
According to the content of the manuscript and the comments of the reviewer, we have revised and replied item by item.
According to the content of the manuscript, the author's team has conducted sufficient research and reasonable research on the active steering control of vehicles. The research process and results are consistent with the process of scientific research. I suggest that journals may consider accepting this article. But before the paper is accepted, I think there are still the following problems that need to be further corrected by the author team.
- Lines 153-155 and 156-158 seems be the same. Authors must choice the best text and eliminate the other one. It is not explained the origin of Figure 2. It is taken from some model by other author or experimental data?
Answer: Thank you for your careful review. I have corrected and supplemented the relevant contents.
Revision: The position of the modified content is from line 14 on page 5 to line 6 on page 6.
Therefore, by further considering the influence of vertical load on lateral force, the lateral force of vehicle tire corresponding to sideslip angle under different vertical dynamic load can be obtained as shown in the Figure 2 below. The data in Figure 2 are the tire lateral force characteristics of the research object with the corresponding sideslip angle and vertical load in the range of 0-23000N. According to the data curves, the tire lateral force characteristics of the vehicle under different working conditions can be obtained.
Figure 2. Tire lateral force under variable load.
In the static state, the vehicle load is constant. However, during driving, body movements such as pitching and rolling will lead to changes in vertical dynamic loads of different tires, so the corresponding tire lateral forces will also change. The vehicle tire model shown in Figure 2 can adapt to the engineering application and calculation of nonlinear lateral force of vehicle tires, which provides great convenience for online calculation of the model.
- Figure 8b must be improved, with another line width that allows compare better both lines.
Answer: Thank you for your careful review. I have corrected the relevant picture.
Revision: The modified content is located on line 18 on page 14.
|
(a) |
(b) |
Figure 8. Road information input: (a) driving road elevation dynamic quantity; (b) Roughness of left and right roads.
- Figure 9 and 11 must be improved using different colour for the green line, because green is not visible enough.
Answer: Thank you for your careful review. I have corrected the relevant pictures.
Revision: The position of the modified contents are respectively from line 29 on page 14 to line 1 on page 15, and is from line 6 on page 16 to line 8 on page 16.
|
(a) |
(b) |
|
(c) |
(d) |
Figure 9. System state responses under condition 1: (a) Driving track; (b) Lateral acceleration of the centroid; (c) Sideslip angle of the centroid; (d) Yaw angular velocity of the centroid.
|
(a) |
(b) |
|
(c) |
(d) |
Figure 11. System state responses under condition 2: (a) Driving track; (b) Lateral acceleration of the centroid; (c) Sideslip angle of the centroid; (d) Yaw angular velocity of the centroid.
- Table 2 and table 3 exceed margins.
Answer: Thank you for your careful review. The format of the form is specified in the journal template, and I edit it completely according to the paper template. Therefore, the form should meet the requirements of the magazine.
- Moreover, information of these tables seem not be really relevant because difference between predictive and non predictive control are very reduced for chosen parameters. However, graphic results shown in figures justify the good behaviour of the proposed control. Why these parameters are chosen for tables?
Answer: Thank you for your insightful suggestions. We have also carefully considered the problems you raised. Since this paper aims to discuss the effectiveness of the system controller under the predictive control, our main purpose is to verify whether the predictive controller has an impact on the control. Due to the limited space of the journal, we first list the corresponding results to explain. We have further discussed how the nonlinear vehicle model can be adapted to the predictive controller under complex working conditions, and how the predictive controller can distinguish from the vehicle system without the predictive controller under the premise of maximum time step.
In order to maximize the effectiveness of the predictive controller, we select variables related to vehicle handling stability and safety from vehicle states for visual explanation. The data in the table is only used to quantify the system control differences under different modes. Although these differences seem small, it can be seen from the observation of the change curves in Figure 9(c), Figure 11(c) and Figure 11(d) that the system under predictive control can approach the stable region more quickly because of the input state as a reference, which cannot be reflected by the data in the table.
In response to your question, we have also revised the content of the article to further illustrate the relevance and effectiveness of data.
Revision: The position of the modified contents are respectively from line 24 on page 15 to line 28 on page 15, and is from line 14 on page 17 to line 21 on page 17.
1) Although according to the data results in Table 2, whether the predictive controller exists or not, its corresponding root mean square value of vehicle state is relatively close. However, it can be seen intuitively from Figure 9(c) that under the input of prediction information, the vehicle state recovers to the stable region faster, which is particularly important for vehicle stability control under complex coupled working conditions.
2) Although the state data of the vehicle system with or without the predictive controller in Table 3 are similar, it can be seen from the data change curve in Figure 11(c) and Figure 11(d) that the vehicle system with the predictive controller can keep the vehicle stable in a safe area. Moreover, because of the input of predictive data, the system state can quickly return to the safe area, which is critical to the safety and stability of vehicles during driving. Table 2 and Table 3 only further quantify the effect of control, but the actual control effect must be reflected according to the response curve in Figure 9 and Figure 11.
Author Response File: Author Response.pdf
Reviewer 3 Report
In the paper titled “Research on vehicle active steering stability control based on variable time-domain input and state information prediction”, the authors propose a four-wheel active steering vehicle system built over a nonlinear dynamic model and the Long Short Term Memory (LSTM) network architecture. The former actively corrects the steering angle of the front/rear wheels in real-time by dynamically adjusting them according to external needs; while the latter, well adapted to data processing in complex and highly dynamic systems, allows anticipating parameterization information for the prediction of the vehicle confidence state in the time domain.
The authors propose two case studies establishing initial conditions to carry out the simulations to show the stability of the vehicle presented by them, based on the coupling relationship between the vehicle dynamics model, the state prediction network architecture, and the mixed sensitivity controller. In both cases, it is clearly observed that with the passage of time there is a loss of performance related to the four elements of analysis (a) Driving Track; (b) Centroid Lateral Acceleration; (c) Centroid Slip Angle; (d) Centroid Angular Rate of Turn, when the vehicle system is uncontrolled
Thus, the authors contribute to the current knowledge by proposing a robust control system for vehicle stability based on real-time prediction provided by an artificial intelligence algorithm. I also find that the article is well structured, with a detailed description of the techniques and methodologies they employ, and an explanation of the strategies used. The level of presentation of the mathematical models, schematics, designs, structures, data, and simulation results is also well enough.
Author Response
Reply to review 3
According to the content of the manuscript and the comments of the reviewer, we have revised and replied item by item.
In the paper titled “Research on vehicle active steering stability control based on variable time-domain input and state information prediction”, the authors propose a four-wheel active steering vehicle system built over a nonlinear dynamic model and the Long Short Term Memory (LSTM) network architecture. The former actively corrects the steering angle of the front/rear wheels in real-time by dynamically adjusting them according to external needs; while the latter, well adapted to data processing in complex and highly dynamic systems, allows anticipating parameterization information for the prediction of the vehicle confidence state in the time domain.
The authors propose two case studies establishing initial conditions to carry out the simulations to show the stability of the vehicle presented by them, based on the coupling relationship between the vehicle dynamics model, the state prediction network architecture, and the mixed sensitivity controller. In both cases, it is clearly observed that with the passage of time there is a loss of performance related to the four elements of analysis (a) Driving Track; (b) Centroid Lateral Acceleration; (c) Centroid Slip Angle; (d) Centroid Angular Rate of Turn, when the vehicle system is uncontrolled
Thus, the authors contribute to the current knowledge by proposing a robust control system for vehicle stability based on real-time prediction provided by an artificial intelligence algorithm. I also find that the article is well structured, with a detailed description of the techniques and methodologies they employ, and an explanation of the strategies used. The level of presentation of the mathematical models, schematics, designs, structures, data, and simulation results is also well enough.
We have also supplemented the references and revised English spelling problems raised by the reviewers.
- Supplementary contents of relevant literature
1) The position of the modified content is from line 14 on page 2 to line 17 on page 2.
……The interaction information between vehicles is helpful to identify vehicle behavior. Combined with the prediction results of vehicle state information, the accurate vehicle driving track can be obtained [13].
2) The position of the modified content is from line 41 on page 2 to line 44 on page 2.
……Through multi-source excitation decomposition, and combining the natural frequency analysis with the vehicle steering system, it can provide an effective reference for the structural design of the steering system [21, 22].
- Improvement of English expression
1) The position of the modified content is from line 33 on page 1 to line 35 on page 1.
……and brain-inspired intelligence technology [3], vehicle chassis system control is gradually changing from expected control of single performance to synchronous optimization of multiple performance.
2) The modified content is located on line 39 on page 1.
In intelligent transportation and information network……
3) The position of the modified content is from line 45 on page 2 to line 51 on page 2.
Although there have been in-depth research on the state information fusion prediction [23] and chassis active integrated control [24], further exploration and verification are still needed for corresponding deficiencies. In the case of external input and excitation disturbance, whether the prediction of target vehicle state can be realized also requires investigation [25]. On the other hand, on the premise of obtaining existing data and conducting adequate training, how to realize vehicle stability robust control based on online prediction of artificial intelligence algorithm data has become a scientific question that……
4) The position of the modified content is from line 8 on page 6 to line 9 on page 6.
In order to avoid steering wheel jitter induced by small external disturbances at high speed, the dead zone range of steering wheel angle is set. Only when the steering wheel angle of the vehicle is outside the dead zone……
5) The position of the modified content is from line 36 on page 6 to line 37 on page 6.
Since real data do not exist in isolation from each other, there is an inevitable correlation among different series of data and between the front and back data in the time series [30] ……
Author Response File: Author Response.pdf
