Personalized Collision Avoidance Control for Intelligent Vehicles Based on Driving Characteristics
Round 1
Reviewer 1 Report
The comments from the reviewer are given as follows:
1 1. Figure 1. There is an overlapping of text and line. Please make appropriate changes.
2. Page 7. Please show the schematic of the vehicle model used. In the current version of the manuscript, only equations on the model are provided and the schematic diagram is not provided.
3. Equations (13) and (14). Please define the symbols just after their appearance.
4. In Figure 2, how the range -6 to 6 is chosen should be explained. Also, please show the membership functions used for DeltaK_p, DeltaK_i, and DeltaK_d.
5. How the fuzzy rules in Table 2 are decided? Please explain.
6. What NB, NM, NS, ZO, PS, PM, and PB indicate should be clearly stated.
7. The values of K_p, K_i, and K_d can be provided in a table.
8. There are overlappings of symbols and text in Figure 10. Please make appropriate changes.
9. In Table 4, please includes the values of pitch and roll moments of inertia.
10. Please enlarge the photos provided in Figure 13.
11. Simulation results can be extended to include a case with a lower tire-road friction coefficient (for example, assuming a wet road surface).
12. Figure 20. Please use a lowercase k for kilo in kNm. Also what L1, L2, R1, and R2 represent indicate should be clearly stated.
13. The authors should carefully proofread the manuscript to reduce formatting errors.
The quality of English can be improved by a careful proofreading.
Author Response
Response to Reviewer 1 Comments
Point 1: Figure 1. There is an overlapping of text and line. Please make appropriate changes.
Response 1: Thank you for pointing it out. The manuscript has been checked and revised.
Point 2: Page 7. Please show the schematic of the vehicle model used. In the current version of the manuscript, only equations on the model are provided and the schematic diagram is not provided.
Response 2: Thank you for the suggestion. According to your valuable comments, we have added relevant content and figure, which are marked in red.
Figure R1. Two-DOF vehicle dynamic models.
Assuming the vehicle runs in ideal conditions, the vehicle’s lateral and yaw motion dynamic equations can be expressed, respectively, as
where m is the mass of the vehicle; vx and vy are the longitudinal and lateral velocities; lf and lr are the distances from CG to the front and rear axles, respectively; Fyf and Fyr are the vehicle’s lateral forces; δf and δr are the steering angle of the front and rear wheels; Iz is the yaw moment of inertia, γ is the yaw rate.
Point 3: Equations (13) and (14). Please define the symbols just after their appearance.
Response 3: Thank you for pointing it out. we have added the define of the symbols, which are marked in red of the revision.
Point 4: In Figure 9, how the range -6 to 6 is chosen should be explained. Also, please show the membership functions used for DeltaK_p, DeltaK_i, and DeltaK_d.
Response 4: Thank you for pointing it out. The vehicle is tested in Carsim with a safety range from 0 to 120 m. So the basic theoretical domain of error is set as[-6,6]. In this paper, the same quantization objective is adopted in the establishment of trigonometric membership function, and the theoretical domain of inputs and outputs variables are all set [-6,6]. And the membership functions used for DeltaK_p, DeltaK_i, and DeltaK_d is shown in Figure R3(b)
Figure R3. Membership functions for FCC in fuzzy control: (a) Input variables of e/ec; (b) Output variables of ∆Kp/∆Ki/∆Kd.
Point 5: How the fuzzy rules in Table 2 are decided? Please explain.
Response 5: Thank you for pointing it out. The three parameters of fuzzy PID are determined by the relative distance error and speed difference, with the change rate determining the strength of proportion, integration, and differential effects. Fuzzy rules should consider dynamic error and overshoot in their design. When the relative distance error is very small, quick deviation reduction is necessary to ensure that the vehicle reaches a safe distance as soon as possible.
Point 6: What NB, NM, NS, ZO, PS, PM, and PB indicate should be clearly stated.
Response 6: Thank you for the suggestion! For the design of fuzzy control, seven conditions are considered for fuzzification. The linguistic variables of the input and output variables are classified as negative big (NB), negative middle (NM), negative small (NS), zero (ZO), positive small (PS), positive middle (PM), and positive big (PB).
Point 7: The values of K_p, K_i, and K_d can be provided in a table.
Response 7: Thank you for the suggestion! The initial values for Kp, Ki, Kd are 1, 0.001, and 0.1, respectively.
Point 8: There are overlappings of symbols and text in Figure 10. Please make appropriate changes.
Response 8: Thank you for pointing it out. The figure has been checked and revised.
Point 9: In Table 4, please includes the values of pitch and roll moments of inertia.
Response 9: Thank you for pointing it out. The values for pitch and roll moments of inertia have been added to Table 4 and highlighted in red.
Point 10: Please enlarge the photos provided in Figure 13.
Response 10: Thank you for pointing it out. Figure 13 has been revised.
Point 11: Simulation results can be extended to include a case with a lower tire-road friction coefficient (for example, assuming a wet road surface).
Response 11: Thank you for the suggestion! The tire-road friction coefficient is set as 0.85. Driving in lower tire-road friction coefficient, it is easy to slideslip for IV. This main work is to design a personalized collision avoidance control strategy. Futher work will consider this issue.
Point 12: Figure 20. Please use a lowercase k for kilo in kNm. Also what L1, L2, R1, and R2 represent indicate should be clearly stated.
Response 12: Thank you for the suggestion. We fixed this issue and marked it in red. ‘L1’, ‘R1’, ‘L2’ and ‘R2’ denote the front left wheel, front right wheel, rear left wheel and rear right wheel, respectively.
Point 13: The authors should carefully proofread the manuscript to reduce formatting errors.
Response 13: Thank you for pointing it out. The manuscript has been checked and revised.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The authors presented the article " Personalized collision avoidance control for intelligent vehicles based on driving characteristics." It is a topic of interest in the scientific and industrial community. However, the authors have failed to make their article represent a scientific study that can be considered for the following reasons.
Majors
- Add the lumped system of the system two-DOF linear vehicle model because the references gave the authors do not correspond to the exact variable representation of the equation in state space. Therefore it is not possible to verify the equations (13,14 and 15) proposed by the authors. Consequently, what they are proposing to the authors cannot be confirmed.
- It needs to rewrite the abstract because it gives the key content, research purpose, relevance, and main outcomes.
- The introduction needs to reformulate the motivation.
- Add a table of all the variables involved in the document.
- To attract a young audience (Masters or Ph.D. students), please provide a methodology (flow diagram) section. It means how you solve the case study of your article.
- It needs to check all the figures because there are some incomplete words, and putting in a correct distribution is essential.
- The formulas are developed by the authors?
- In section 4, the vehicle parameter. What is the vehicle class? And Where are authors took that values?
- In section 4.1, why 16 subjects? And scenario 11 is developed by the authors?
- It is necessary to reformulate the conclusions.
To summarize, the article must be reviewed in detail regarding its writing, approach, motivation, and the details mentioned before. For the moment, the article presented does not represent scientific novelty. Therefore, I do not recommend the publication of this article.
English should be improved: There are some typos and grammatical writing issues. So please have a meticulous check of the writing.
Author Response
Response to Reviewer 2 Comments
Point 1: Add the lumped system of the system two-DOF linear vehicle model because the references gave the authors do not correspond to the exact variable representation of the equation in state space. Therefore it is not possible to verify the equations (13, 14and 15) proposed by the authors. Consequently, what they are proposing to the authors cannot be confirmed.
Response 1: Thank you for pointing it out. We have added the two-DOF linear vehicle model, as.
Figure R1. Two-DOF vehicle dynamic models.
Assuming the vehicle runs in ideal conditions, the vehicle’s lateral and yaw motion dynamic equations can be expressed as
Point 2: It needs to rewrite the abstract because it gives the key content, research purpose, relevance, and main outcomes.
Response 2: Thank you for the suggestion. According to your valuable comments, we have rewritten the abstract and marked it in red as.
“Collision avoidance has been widely researched in the field of intelligent vehicles (IV). However, majority researches neglect the individual driver differences. This paper introduced a novel personalized collision avoidance control (PCAC) strategy for IV based on driving characteristics (DC), which can better satisfy various scenarios and can improve drivers’ acceptance. Firstly, the driver’s DC is classified into four types by K-means clustering, and the analytic hierarchy process (AHP) method is used to construct the DC identification model for the design of the PCAC. Then, a novel PCAC is integrated with a preview-follower control (PFC) module, an active rear steering (ARS) module, and a forward collision control (FCC) module to ensure individual requirements and driving stability. Moreover, simulations verified the validity of the developed PCAC in terms of path tracking, lateral acceleration, and yaw rate. The research results indicate that DC can be identified effectively through APH and PCAC based on DC can facilitate the development of intelligent driving vehicles with superior human acceptance performance.”
Point 3: The introduction needs to reformulate the motivation.
Response 3: Thank you for the suggestion! We have rewritten the motivation of the introduction as follows.
“Intelligent vehicles (IV), equipped with advanced driving assistant systems (ADAS), are the best way to improve traffic efficiency, reduce collisions, and achieve self-driving. However, complex traffic situations, unexpected objects and different driving styles could easily induce crash accidents. Therefore, collision avoidance is one essential issue in the development of IV. Early collision avoidance strategies, such as active emergency brake (AEB), forward collision warning (FCW), and adaptive cruise control (ACC) are mainly focused on longitudinal control. It cannot achieve all driving conditions at this stage. Meanwhile, driver and IV will exist at the same time in the long future. The driving characteristics (DC) of different individuals are an essential factor that affects the performance of collision avoidance controller (CAC). For this reason, the DC is becoming a mainly considered aspect of CAC design.”
Point 4: Add a table of all the variables involved in the document.
Response 4: Thank you for the suggestion!
Point 5: To attract a young audience (Masters or Ph.D. students), please provide a methodology (flow diagram) section. It means how you solve the case study of your article.
Response 5: Thank you for the suggestion! The architecture of the personalized collision avoidance control (PCAC) system is shown in Fig. R1, and it is comprised of perception sensors together with decision layer, collision avoidance control layer, and a chassis actuators.
Figure r1. General framework of PCAC.
Point 6: It needs to check all the figures because there are some incomplete words, and putting in a correct distribution is essential.
Response 6: Thank you for pointing it out. All the figures with formatting issues have been checked and revised.
Point 7: The formulas are developed by the authors?
Response 7: Thank you for pointing it out. Not all formulas are developed by the authors, however, unauthored formulae are cited.
Point 8: In section 4, the vehicle parameter. What is the vehicle class? And Where are authors took that values?
Response 8: Thank you for the comments! The test vehicle is a CarSim SUV model with the parameters listed in Table 4. CarSim simulation sofware is used to validate the control strategy proposed in this article.
Figure r1. CarSim modeling interface.
Point 9: In section 4.1, why 16 subjects? And scenario 11 is developed by the authors?
Response 9: Thank you for pointing it out. In the first phase, only 16 driveris are selected and more samples will be test in future work. Also, scenario 11 is developed by the author (Lina Gao).
Point 10: It is necessary to reformulate the conclusions.
Response 10: Thank you for the suggestion! We have rewritten the conclusions as follows.
“The designed novel personalized collision avoidance control (PCAC) strategy for intelligent vehicles (IV) based on driving characteristics (DC) can not only improve the driving stability but also improve driver adaptability. The PCAC consists of a decision part and a control part. Specifically, the CAC is integrated with preview-follower control (PFC), active rear steering (ARS), and forward collision control (FCC). As a result, we came to the conclusions as follows:
The DC of 16 drivers can be classified into four types: steady, general, general radical, and radical. The analytic hierarchy process (AHP) can be used to recognize the DC accurately.
The PFC model with different DC based on lateral acceleration feedback (LAC) show that the evaluation indexes are smaller, more stable, and steering earlier by steady and general drivers compared with radical drivers, which can meet individual requirements.
For steady and general drivers, the acceptance of the PCAC is still lacking by PFC as the lateral acceleration still exceeds its threshold. The PCAC with PFC+ARS show that the lateral acceleration was reduced 11% compared with PFC only. Also, the yaw rate peak value reduction exceeds 30%.
The PCAC integrated with PFC, ARS, and FCC can greatly reduce lateral acceleration and yaw rate greatly, and the acceptance index is improved by more than 30% compared with the PCAC with PFC. The proposed PCAC can effectively improve driving stability and acceptance for steady and general drivers, especially in high-speed obstacle avoidance conditions.”
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
The manuscript is revised based on the comments by reviewer.
There are no further comments by the reviewer.
Author Response
Thank you for the comments and kind recommendation.
Reviewer 2 Report
In the model presented in Equation 13 and corresponding to Figure 8, the major observations are :
- The authors refer to the two degrees of freedom model with reference 13. In the reference given, the relation of tire forces only concerns the steering angle front. However, the authors additionally place the steering angle rear. Additionally, the authors place it as part of the input vector (u). Therefore, the authors must justify why the steering angle rear is established as an independent entry in the vector (u).
- the yaw acceleration equation (second equation 13) does not consider the moments generated concerning O' (vehicle's center of gravity) of the force of the tires.
- The vector of Vy is in the wrong direction in Figure 8.
- Equation 16, where matrix B is represented with its elements, is the wrong put transpose. The authors must check before sending for a second time an article because it gives credibility to the work done.
The authors must add a table with all the variables involved in the document, only there are abbreviations.
Since the mathematical model with which the basis of the article presented corresponds has points to be discussed, I do not suggest publication.
English should be improved: There are some typos and grammatical writing issues. So please have a meticulous check of the writing. For example, The authors should check every sentence. They should insert a comma to separate elements. In addition, there are words overused. The authors should use specific synonyms to improve the writing.
Author Response
Point 1: The authors refer to the two degrees of freedom model with reference 13. In the reference given, the relation of tire forces only concerns the steering angle front. However, the authors additionally place the steering angle rear. Additionally, the authors place it as part of the input vector (u). Therefore, the authors must justify why the steering angle rear is established as an independent entry in the vector (u).
Response 1: Thank you for the comments. We have revised the paper based on the reviewer comments. Please find in blue the responses to the reviewer’s comments.
Reference 13 presents the dynamic equations of a two-DOF linear vehicle model with two-wheel steering (2WS). However, four-wheel steering (4WS) refers to a system in which all four wheels turn simultaneously during steering maneuvers. This technology enables the vehicle to achieve a smaller turning radius at low speeds, while also enhancing its agility and maneuverability. Additionally, it provides faster response times when driving at high speeds. The schematic diagram of a four-wheel steering system model with two-DOF is shown in Figure R1.
Figure R1. Schematic diagram of a four-wheel steering system model with two-DOF
For the purpose of analysis, let us assume a simplified two-DOF model as:
- The vehicle maintains constant speed and moves in a straight line.
- Direct front and rear wheel angles are used as inputs.
- This model does not include suspension; instead, spring-loaded and unsprung masses are considered as part of the system.
- Changes in tire sidewise characteristics caused by tangential forces on the ground are disregarded.
- The effect of aerodynamics is neglected, and tire changes caused by load variations are disregarded.
- The lateral characteristics of the tires are assumed to be linear, and the vehicle's weight distribution is symmetric.
- The righting torque influence is not taken into account in this two-DOF steering system model that only considers lateral motion on the y-axis and yaw motion on the z-axis.
Point 2: The yaw acceleration equation (second equation 13) does not consider the moments generated concerning O' (vehicle's center of gravity) of the force of the tires.
Response 2: Thank you for the comment. We make a certain simplifying assumption as: This model does not include suspension; instead, spring-loaded and unsprung masses are considered as part of the system. The lateral characteristics of the tires are assumed to be linear, and the vehicle's weight distribution is symmetric. The righting torque influence is not taken into account and only considers lateral motion on the y-axis and yaw motion on the z-axis.
Point 3: The vector of Vy is in the wrong direction in Figure 8.
Response 3: Thank you for the comment. The reviewer is correct. The vector of vy in Fig. 8 of the revised l version should be in the opposite direction. We have corrected it in the revised paper.
Figure R2. Schematic diagram of a four-wheel steering system model with two-DOF
Point 4: Equation 16, where matrix B is represented with its elements, is the wrong put transpose. The authors must check before sending for a second time an article because it gives credibility to the work done.
Response 4: Thank you for pointing it out. We have corrected it in the revised paper
where, the state vector x= [vy γ Y ψ]T, and the input vector u= [δf, δr]T,
Point 5: The authors must add a table with all the variables involved in the document, only there are abbreviations.
Response 5: Thank you for the suggestion! We add a table with all the variables involved in the revised paper.
Point 6: Since the mathematical model with which the basis of the article presented corresponds has points to be discussed, I do not suggest publication.
Response 6: Thank you for the comments. We have studied your comments carefully and have made several changes/additions to the paper.
Author Response File: Author Response.pdf
Round 3
Reviewer 2 Report
The global position mentioned before of equation in 14 does not have the same symbol in the explanation, neither in state vector x. The authors must put the same symbol. Also, it must be added in the list of symbols.
- Some sentences could be more precise or easier to follow. Consider rephrasing.
- Some phrases could be improved the reception.
- The authors should add transition phrases to improve the flow of the paragraphs.
- The authors need to check some words which appear repeatedly. Consider using synonyms.
Author Response
Point 1: The global position mentioned before of equation in 14 does not have the same symbol in the explanation, neither in state vector x. The authors must put the same symbol. Also, it must be added in the list of symbols.
Response 1: Thank you for the comments. We have revised the paper based on the reviewer comments.
Point 2: Some sentences could be more precise or easier to follow. Some phrases could be improved the reception. The authors should add transition phrases to improve the flow of the paragraphs. The authors need to check some words which appear repeatedly. Consider using synonyms.
Response 2: Thank you for the comment. We have studied your comments carefully and have made several changes/additions to the paper.
