Tire Contact Force Equations for Vision-Based Vehicle Weight Identification

Round 1

Reviewer 1 Report

The authors in the paper present a method to determine the weight of a vehicle based on tire mechanics and computer vision. Two cars and two trucks were used for the experiments. The authors developed a set of equations, improving the generally accepted equations for calculating mechanical loads produced on tires. A big plus is the perfect theoretical presentation of the subject and the excellent presentation of the whole reasoning.

However, to be honest, I am still not convinced about the practicality of the proposed approach. In my opinion, the paper, in general, is fascinating, but there are many possibilities that the approach would result in the wrong prediction (e.g., low pressure in tires, different pressures in double tires in trucks, and many others). 

I think that removing some phrases that the proposed method is "non-contact" (because you have to measure pressure in all tires to make a good prediction) is quite important.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

The article has been significantly improved and I think it can be published in the Journal.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

The paper is very well written and all aspects of the study are well described and discussed. Although, a number of issues need further exploration for practical applications, the proposed approach is a good first step towards a low cost method for WIM systems. A couple of minor grammar/presentations errors have been found on lines: 84, 280-281.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

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.

Round 1

Reviewer 1 Report

This study tries to solve a practical problem, vehicle weight estimation, by utilizing computer vision to detect tire deflection. The author first derives some theoretical equations of tire-contact force with some reasonable assumptions. Then FE models of different tires are established and analyzed under different conditions in simulation to support the author’s claim. A non-contact vehicle weight identification method was proposed which is a combination of the proposed theoretical model and computer vision techniques. Finally, the proposed method is validated with some field experiments. Overall, this research is interesting. Structure of the manuscript is clear. Equations and assumptions are reasonable. Data and other supporting materials are also quite convincing. 

 

However, a very fundamental problem of this proposed method is that it is not applicable in real world scenarios. The weight identification method relies on a measurement of inflation pressure (P), which can be obtained (according to the author) by TPMS, or pressure detector. Both approaches would require some kind of manual work to get the pressure reading from the vehicle, and put it in the weight calculation equation. It is not realistic to do this in real life as it will be very time consuming. It will be no better than just installing a weight scale to measure the weight. Unless the author can justify the use case for his/her proposed method, I don’t think this method is appropriate for any real world applications.

 

Some additional comments:

1. Since this method heavily relies on the computer vision technique to extract information of the tires, it will be beneficial to introduce the method/model the author is using. For example, is it a deep learning based model, or a rule-based method? 
2. Also, any computer vision method, in its nature, will be impacted by lighting conditions, weather conditions, and many other factors. The author should spend some efforts demonstrating or explaining how the proposed method will perform under different conditions.
3. A total of four vehicles in the validation section seems not enough. There are many different types of vehicles running on the road. And even the same model of vehicles will have different loadings. How will this method perform for some common vehicle models and loadings? Author should explain more clearly.
4. Typo: page 17, section 6, (1) “The theorectical equation of” should be “The theoretical equation of”.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 2 Report

The authors in the paper present a method to determine the weight of a vehicle based on tire mechanics and computer vision. Two cars and two trucks were used for the experiments. The authors developed a set of equations, improving the generally accepted equations for calculating mechanical loads produced on tires. A big plus is the perfect theoretical presentation of the subject and the excellent presentation of the whole reasoning.

The biggest drawback of the current solution, in my opinion, is its low practicality. To use the developed method, it is necessary to measure the pressure in each tire, which disqualifies the solution from practical application. 

Moreover, it is a huge mistake to say that the proposed method is non-contact, low cost, convenient, and efficient compared with the static weighing and WIM methods since measuring the pressure in each vehicle's tires is necessary.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

Vehicle weight identification method based on tire mechanics and computer vision is presented in the article. The topic, vehicles overloading and problems of identification of them weight, is relevant. The article is original.

The literature review is comprehensive enough, the authors use many sources. The authors describe the problem in detail and provide a theoretical analysis of wheel-road contact. Numerical analysis and verification of tire model are presented in Chapter 3 as well as modification of tire contact force equations. Authors should make some corrections (lines 274 and 276) because there is information about Error (reference source not found). An experimental verification of the proposed method is presented in Chapter 5. Tests performed on different vehicle types. Authors’ research shows that non-contact vehicle weight identification method can effectively enough identify the axle weight and gross weight of the vehicle. Identification error is quite high. In the future, the authors could refine their proposed method.

The authors could comment in more detail on their planned researches to refine this approach in the future.

Author Response

Please see the attachment.

Author Response File: Author Response.docx