Control method and system for improving riding comfort of automatic driving double-deck bus

Abstract

The invention belongs to the technical field of automatic driving, and relates to a control method and system for improving riding comfort of an automatic driving double-deck bus. Establishing a relation function between the body swing amplitude of each type of passengers and the vehicle deceleration and the deceleration duration through the collected data, selecting the minimum optimized per capita body swing amplitude T0-n-min from the optimized per capita body swing amplitude T0-n, and selecting the maximum optimized body swing amplitude sum Tymax from the optimized body swing amplitude sum Ty, the optimization result is verified, and if the requirement is met, optimization is completed; the posture of the passenger is comprehensively taken into consideration of the comfort level, the actual feeling of the passenger standing in the compartment is taken into consideration, and the riding experience is improved; when the system is applied to the ground, an existing automatic driving bus body structure and hardware facilities do not need to be changed completely, and cost is saved.

Description

Technical Field

[0001] The present invention belongs to the technical field of autonomous driving, and relates to a control method and system for improving the riding comfort of an autonomous driving double-decker bus. Background Art

[0002] At present, with the emergence and development of autonomous driving technology, it has a huge impact on the public transportation sector. Autonomous driving technology can optimize the operational efficiency of the public transportation system by reducing driver errors and improving vehicle utilization efficiency. It can also accurately execute route planning and strictly abide by traffic rules, reducing the probability of accidents, thereby improving overall safety.

[0003] However, with the popularization and development of this technology in the field of public transportation, since self-driving buses are electric vehicles, their deceleration during braking is large, which can easily make passengers feel uncomfortable. Summary of the invention

[0004] The purpose of the present invention is to overcome the shortcomings of the above-mentioned prior art and to propose a control method and system for improving the riding comfort of an autonomous double-decker bus.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] A control method for improving the riding comfort of an autonomous double-decker bus is provided. The relationship function between the body swing amplitude of each type of passenger and the vehicle deceleration magnitude and deceleration duration is established through the collected data. The original average body swing amplitude T before optimization for multiple expected decelerations a is calculated through the relationship function. 0-x , optimize the sum of body swing amplitude T y , optimize the average body swing amplitude T 0-n , from optimizing the average body swing amplitude T 0-n Select the minimum optimal average body swing amplitude T 0-n-min , from optimizing the sum of body swing amplitude T y Select the maximum optimized body swing amplitude T ymax , verify the optimization results. If the minimum optimized average body swing amplitude T 0-n-min Smaller than the original average body swing amplitude T 0-xm And the maximum optimized body swing amplitude T ymax If it is smaller than the threshold S in the data, the optimization is completed.

[0007] Furthermore, the method specifically includes the following steps:

[0008] Step 1, collecting data: collecting the deceleration magnitude and duration of the deceleration when the vehicle brakes, as well as the body swing amplitude and threshold S of each of the four types of passengers; wherein: the four types of passengers include: sitting passengers in the lower compartment, sitting passengers in the upper compartment, standing passengers in the upper compartment, and standing passengers in the lower compartment;

[0009] Step 2: Establishing a relationship function: obtaining a relationship function between the body swing amplitude of each type of passenger and the vehicle deceleration magnitude and deceleration duration through the collected data;

[0010] Step 3: Obtain the original average body swing amplitude T before optimization 0-x :Select multiple expected decelerations a that meet the optimization conditions from the data provided by the automatic driving system; calculate the original average body swing amplitude T for each expected deceleration a through the relationship function 0-x ;

[0011] Step 4: Obtain the optimized average body swing amplitude T 0-n : According to a specific optimization method, multiple expected decelerations a are optimized to obtain a deceleration a corresponding to each expected deceleration a y ; Calculate the relationship between each type of passenger and each deceleration a through the relationship function y The optimal body swing amplitude T yn , and then find the value for each deceleration a y Optimized average body swing amplitude T 0-n ;

[0012] Step 5: Select the value: Optimize the average body swing amplitude T from multiple 0-n Select the minimum optimal average body swing amplitude T 0-n-min , the minimum optimal average body swing amplitude T 0-n-min Compared with the original average body sway amplitude T 0-x The first primitive average body sway amplitude T 0-xm Correspondingly, according to the minimum optimized average body swing amplitude T 0-n-min Select the sum of the optimized body swing amplitudes T of four types of passengers yn The maximum optimal body swing amplitude T ymax ;

[0013] Step 6: Determine the optimized deceleration a y Whether it meets the requirements: If the minimum optimized average body swing amplitude T 0-n-min Less than the first primitive average body swing amplitude T 0-xm And the maximum optimized body swing amplitude T ymax If it is less than the threshold S, then the minimum optimized average body swing amplitude T 0-n-minThe corresponding optimized deceleration a y Meets the requirements and optimization is completed.

[0014] Furthermore, the magnitude of the vehicle deceleration and the duration of the deceleration in the step one are collected by the vehicle's built-in IMU, and the body sway amplitude of each type of passenger is collected by human body posture estimation method, optical flow method, or deep learning model.

[0015] Furthermore, the relationship function between the body swing amplitude of each type of passenger and the vehicle deceleration magnitude and deceleration duration in step 2 includes:

[0016] Function F1: relationship between the amplitude of body swing of sitting passengers in the lower-level carriage and the magnitude of vehicle deceleration and the duration of deceleration; function F2: relationship between the amplitude of body swing of sitting passengers in the upper-level carriage and the magnitude of vehicle deceleration and the duration of deceleration; function F3: relationship between the amplitude of body swing of standing passengers in the upper-level carriage and the magnitude of vehicle deceleration and the duration of deceleration; function F4: relationship between the amplitude of body swing of standing passengers in the lower-level carriage and the magnitude of vehicle deceleration and the duration of deceleration.

[0017] Furthermore, the specific formulas of the relationship function F1, the relationship function F2, the relationship function F3, and the relationship function F4 are as follows:

[0018] F1=K 11 *α+K 12 *β (1)

[0019] F2=K 21 *α+K 22 *β (2)

[0020] F3=K 31 *α+K 32 *β (3)

[0021] F4=K 41 *α+K 42 *β (4)

[0022] Among them: α is the normalized value of deceleration, β is the normalized value of deceleration duration, K 11 is the deceleration influence coefficient of the sitting passengers in the lower compartment, K 12 K is the deceleration duration influence coefficient of the sitting passengers in the lower compartment; 21 is the deceleration influence coefficient of the sitting passengers in the upper compartment, K 22 K is the deceleration duration influence coefficient of the sitting passengers in the upper compartment; 31 is the deceleration influence coefficient of the standing passengers in the upper compartment, K 32 K is the deceleration duration influence coefficient of the standing passengers in the upper compartment;41 is the deceleration influence coefficient of standing passengers in the lower deck carriage, K 42 is the deceleration duration influence coefficient of standing passengers in the lower deck carriage.

[0023] Furthermore, the optimization conditions in step three are:

[0024]

[0025] Where: a is the expected deceleration; M is the minimum expected deceleration threshold; N is the maximum expected deceleration threshold.

[0026] Furthermore, the sum of the optimized body swing amplitudes T in step 4 is yn Specifically:

[0027]

[0028] Among them, F1 is the body sway amplitude of the sitting passengers in the lower-level carriage obtained according to the expected deceleration and the expected deceleration duration, and n1 is the total number of sitting passengers in the lower-level carriage; F2 is the body sway amplitude of the sitting passengers in the upper-level carriage obtained according to the expected deceleration and the expected deceleration duration, and n2 is the total number of sitting passengers in the upper-level carriage; F3 is the body sway amplitude of the standing passengers in the lower-level carriage obtained according to the expected deceleration and the expected deceleration duration, and n3 is the total number of standing passengers in the lower-level carriage; F4 is the body sway amplitude of the standing passengers in the upper-level carriage obtained according to the expected deceleration and the expected deceleration duration, and n2 is the total number of standing passengers in the upper-level carriage.

[0029] Furthermore, the original average body swing amplitude T 0-x The specific formula is:

[0030]

[0031] Among them: T1 is the sum of the body swing amplitudes of all sitting passengers in the lower compartment; T2 is the sum of the body swing amplitudes of all sitting passengers in the upper compartment; T3 is the sum of the body swing amplitudes of all standing passengers in the upper compartment; T4 is the sum of the body swing amplitudes of all standing passengers in the lower compartment; n is the total number of passengers in the compartment.

[0032] Furthermore, the specific optimization method in step 4 is specifically:

[0033] The duration of the expected deceleration a is divided into two stages. Based on the expected deceleration a, the deceleration of the first stage is ax%, and the deceleration of the second stage is a+x%. Alternatively, the deceleration of the first stage is a+y%, and the deceleration of the second stage is ay%, where the value range of x and y is [10, 30].

[0034] The present invention also provides a system based on a control method for improving the riding comfort of an autonomous driving double-decker bus, comprising:

[0035] Collection module: used to collect the vehicle deceleration magnitude and duration of deceleration, the body swing amplitude of each type of passenger, and the threshold value S when the vehicle brakes for four types of passengers, namely, sitting passengers in the lower compartment, sitting passengers in the upper compartment, standing passengers in the upper compartment, and standing passengers in the lower compartment;

[0036] Calculation module: Calculate the original average body swing amplitude T before optimization of multiple expected decelerations a through the relationship function 0-x , optimize the sum of body swing amplitude T yn , optimize the average body swing amplitude T 0-n , from optimizing the average body swing amplitude T 0-n Select the minimum optimal average body swing amplitude T 0-n-min , from optimizing the sum of body swing amplitude T yn Select the maximum optimized body swing amplitude T ymax ; From the original average body swing amplitude T 0-x Select and minimize the average body swing amplitude T 0-n-min The corresponding first primitive average body sway amplitude T 0-xm ;

[0037] Verification module: Verify the optimization results. If the minimum optimized average body swing amplitude T 0-n-min Smaller than the first original average body swing amplitude T 0-xm And the maximum optimized body swing amplitude T ymax If it is less than the threshold S, the optimization is completed.

[0038] Compared with the prior art, the present invention has the following beneficial effects:

[0039] First, compared with the existing technology, the present invention, on the one hand, takes the passenger posture into consideration in all aspects of comfort, taking into account the riding experience of all passengers in the car; on the other hand, when it is put into practical application, the present invention does not need to change the existing body structure and hardware facilities of the self-driving bus, and can achieve accurate optimization and intelligent control of deceleration only by relying on the original cameras in the two cars and basic equipment such as the speed controller. In other words, it only needs to be debugged at the algorithm level, without the need to introduce additional expensive hardware equipment, which greatly saves costs.

[0040] Second, this method can be applied to a variety of vehicle types. Not only double-decker buses can use this method, but other vehicles such as ordinary buses, airport shuttle buses, etc. can also use this method, and it has extremely high universality. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The accompanying drawings are incorporated in and constitute a part of this specification and, together with the description, serve to explain the principles of the present invention.

[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0043] Figure 1 It is a flow chart of the control method in the present invention;

[0044] Figure 2 This is an optimization flow chart of the expected deceleration in the present invention. DETAILED DESCRIPTION

[0045] Here, exemplary embodiments will be described in detail, and the embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Instead, they are only examples of devices consistent with some aspects of the present invention as detailed in the appended claims.

[0046] Example

[0047] like Figure 1-2 As shown, a control method for improving the riding comfort of an autonomous double-decker bus is used to establish a relationship function between the body swing amplitude of each type of passenger and the vehicle deceleration magnitude and deceleration duration through the collected data;

[0048] The original average body swing amplitude T before optimization of multiple expected decelerations a is calculated through the relationship function. 0-x , optimize the sum of body swing amplitude T y , optimize the average body swing amplitude T 0-n ;

[0049] From optimizing the average body swing amplitude T 0-n Select the minimum optimal average body swing amplitude T 0-n-min , from optimizing the sum of body swing amplitude T y Select the maximum optimized body swing amplitude T ymax , from the original average body swing amplitude T 0-x Select and minimize the average body swing amplitude T 0-n-min The corresponding first primitive average body sway amplitude T 0-xm ;

[0050] Verify the optimization results. If the minimum optimized average body swing amplitude T 0-n-min Less than the first primitive average body swing amplitude T0-xm And the maximum optimized body swing amplitude T ymax If it is less than the threshold S, the optimization is completed.

[0051] Furthermore, a control method for improving the riding comfort of an autonomous driving double-decker bus specifically comprises the following steps:

[0052] Step 1, collecting data: collecting the deceleration magnitude and duration of the deceleration when the vehicle brakes, as well as the body swing amplitude and threshold S of each of the four types of passengers; wherein: the four types of passengers include: sitting passengers in the lower compartment, sitting passengers in the upper compartment, standing passengers in the upper compartment, and standing passengers in the lower compartment;

[0053] Step 2: Establishing a relationship function: obtaining a relationship function between the body swing amplitude of each type of passenger and the vehicle deceleration magnitude and deceleration duration through the collected data;

[0054] Step 3: Obtain the original average body swing amplitude T before optimization 0-x :Select multiple expected decelerations a that meet the optimization conditions from the data provided by the automatic driving system; calculate the original average body swing amplitude T for each expected deceleration a through the relationship function 0-x ;

[0055] Step 4: Obtain the optimized average body swing amplitude T 0-n : According to a specific optimization method, multiple expected decelerations a are optimized to obtain a deceleration a corresponding to each expected deceleration a y ; Calculate the relationship between each type of passenger and each deceleration a through the relationship function y The optimal body swing amplitude T yn , and then find the value for each deceleration a y Optimized average body swing amplitude T 0-n ;

[0056] Step 5: Select the value: Optimize the average body swing amplitude T from multiple 0-n Select the minimum optimal average body swing amplitude T 0-n-min , the minimum optimal average body swing amplitude T 0-n-min Compared with the original average body sway amplitude T 0-x The first primitive average body sway amplitude T 0-xm Correspondingly, according to the minimum optimized average body swing amplitude T 0-n-min Select the sum of the optimized body swing amplitudes T of four types of passengers yn The maximum optimal body swing amplitude T ymax ;

[0057] Step 6: Determine the optimized deceleration ay Whether it meets the requirements: If the minimum optimized average body swing amplitude T 0-n-min Less than the first primitive average body swing amplitude T 0-xm And the maximum optimized body swing amplitude T ymax If it is less than the threshold S, then the minimum optimized average body swing amplitude T 0-n-min The corresponding optimized deceleration a y Meets the requirements and optimization is completed.

[0058] In this embodiment, each expected deceleration a (before optimization) in step 4 corresponds to a deceleration a y (After optimization), each deceleration a y Corresponding to an optimized body swing amplitude T yn (including the sum of the body swing amplitudes of the four types of passengers), find the value of each deceleration a y Optimized average body swing amplitude T 0-n Then, the average body swing amplitude T of each person is optimized from multiple 0-n Select the minimum optimal average body swing amplitude T 0-n-min (corresponding to a deceleration a y value), from this deceleration a y The inverse is inferred from the deceleration a y Optimize the sum of body swing amplitude T yn The value T of the category with the largest sum of body swing amplitude among the four categories of passengers ymax .

[0059] By optimizing the expected deceleration sent by the automatic driving system (the specific optimization procedure is as described above), a series of expected average body swing amplitudes and maximum body swing amplitudes of passengers are obtained. For example, for the expected deceleration a, the interval is 0.2m / s in the range of 0.7a to 1.3a. 2 Get multiple possible expected decelerations, where the value is a in the range of 0.7a to a i (i=1,2,3,…), and the value in the range of a~1.3a is b j (j = 1, 2, 3, ...), in order to ensure safety (i.e. to ensure that the speed drops to the required range within the corresponding deceleration duration and the braking distance does not change much) and to simply and clearly explain the content of this step, we will i With b j One-to-one correspondence, a deceleration change is obtained in which the deceleration gradually increases during the deceleration duration, which is recorded as A ij Similarly, we can get the deceleration change with gradually decreasing deceleration, denoted as A ji .

[0060]

[0061] Substitute the corresponding deceleration distribution into the functional relationship of the corresponding personnel, and calculate the corresponding average body swing amplitude by the method in step 3, denoted as A ′ ij and A j ′ i , corresponding to A ij and A ji The average body sway amplitude of the person with deceleration changes is shown below.

[0062]

[0063] Specifically, the obtained results are compared and sorted, and the average body swing amplitude T 0-n Sort from small to large, and put the smallest average body swing amplitude T 0-n Compared with the original average body sway amplitude T 0-x Compare and compare the corresponding maximum optimized body swing amplitude T ymax Compare with the threshold S, if it is less than the maximum optimized body swing amplitude T ymax and S, the corresponding deceleration distribution is the optimization result; if the maximum optimized body swing amplitude is T ymax If the value of the average body swing amplitude per person is greater than the threshold S, the suboptimal result of the average body swing amplitude per person is taken for the above comparison. If both groups are less than, the corresponding deceleration distribution is the optimization result; otherwise, repeat the above process; if the average body swing amplitude per person after optimization is greater than that of the unoptimized one, the deceleration range and interval are adjusted, and the calculation and comparison are recalculated until the optimization result is obtained. The result is sent to the automatic driving control system, and the system determines whether to adopt the deceleration value to achieve the minimum average body swing amplitude per person under the premise of ensuring safety.

[0064] Furthermore, the magnitude of the vehicle deceleration and the duration of the deceleration in the step one are collected by the vehicle's built-in IMU, and the body sway amplitude of each type of passenger is collected by human body posture estimation method, optical flow method, or deep learning model.

[0065] In this embodiment: First, the passengers are divided into four categories, namely, passengers sitting and standing in the lower compartment and passengers sitting and standing in the upper compartment. Of course, the distinction can be made more detailed to obtain more targeted results. For example, passengers can be divided by more specific rules such as age group, gender, and the front, middle, and rear of the lower compartment of the vehicle. Through actual experiments, the amplitude of body swing of the four types of passengers when the vehicle brakes, the magnitude of vehicle deceleration at this time, and the duration of deceleration are collected. The experiment was carried out on an actual double-decker bus and collected naturally. In order to ensure the diversity and representativeness of the data, it is recommended to ensure that the experimental time is sufficient, such as 3 months, to ensure that all types of data reach a certain amount to meet the requirements of subsequent analysis. During the data collection process, the passenger's body swing amplitude data can be collected by the vehicle's own camera through a certain algorithm, such as: human posture estimation method, optical flow method, deep learning model, etc., and the magnitude and duration of deceleration are collected by the vehicle's built-in IMU.

[0066] Furthermore, the relationship function between the body swing amplitude of each type of passenger and the vehicle deceleration magnitude and deceleration duration in step 2 includes:

[0067] Function F1: relationship between the amplitude of body swing of sitting passengers in the lower-level carriage and the magnitude of vehicle deceleration and the duration of deceleration; function F2: relationship between the amplitude of body swing of sitting passengers in the upper-level carriage and the magnitude of vehicle deceleration and the duration of deceleration; function F3: relationship between the amplitude of body swing of standing passengers in the upper-level carriage and the magnitude of vehicle deceleration and the duration of deceleration; function F4: relationship between the amplitude of body swing of standing passengers in the lower-level carriage and the magnitude of vehicle deceleration and the duration of deceleration.

[0068] Furthermore, the specific formulas of the relationship function F1, the relationship function F2, the relationship function F3, and the relationship function F4 are as follows:

[0069] F1=K 11 *α+K 12 *β (1)

[0070] F2=K 21 *α+K 22 *β (2)

[0071] F3=K 31 *α+K 32 *β (3)

[0072] F4=K 41 *α+K 42 *β (4)

[0073] Among them: α is the normalized value of deceleration, β is the normalized value of deceleration duration, K 11is the deceleration influence coefficient of the sitting passengers in the lower compartment, K 12 K is the deceleration duration influence coefficient of the sitting passengers in the lower compartment; 21 is the deceleration influence coefficient of the sitting passengers in the upper compartment, K 22 K is the deceleration duration influence coefficient of the sitting passengers in the upper compartment; 31 is the deceleration influence coefficient of the standing passengers in the upper compartment, K 32 K is the deceleration duration influence coefficient of the standing passengers in the upper compartment; 41 is the deceleration influence coefficient of standing passengers in the lower deck carriage, K 42 is the deceleration duration influence coefficient of standing passengers in the lower deck carriage.

[0074] In this embodiment: by analyzing the data collected from the experiment, a relationship model between the body swing amplitude of each type of personnel and the deceleration magnitude and deceleration duration of the vehicle is determined. The input of the model is the deceleration magnitude and deceleration duration of the vehicle, and the output is the body swing amplitude of each type of personnel. By establishing a model, each type of personnel will obtain a relationship function F1, F2, F3, F4 corresponding to the deceleration, deceleration duration and body swing amplitude. After obtaining the relationship function of each type of personnel, the expected deceleration and expected deceleration duration of the next stage can be obtained through the automatic driving system, and it can be brought into the functional relationship to obtain the expected body swing amplitude corresponding to each person. Among them, F1 is the relationship function of the sitting passengers in the lower compartment, and F2 is the relationship function of the sitting passengers in the upper compartment. Similarly, F3 and F4 are the relationship functions of the standing passengers in the upper and lower compartments respectively. The body swing amplitude of each type of personnel under different decelerations and deceleration durations can be obtained through the corresponding functions.

[0075] Furthermore, the optimization conditions in step three are:

[0076]

[0077] Where: a is the expected deceleration; M is the minimum expected deceleration threshold; N is the maximum expected deceleration threshold.

[0078] Furthermore, the sum of the optimized body swing amplitudes T in step 4 is yn Specifically:

[0079]

[0080] Among them, F1 is the body sway amplitude of the sitting passengers in the lower-deck carriage obtained according to the expected deceleration and the expected deceleration duration, n1 is the total number of sitting passengers in the lower-deck carriage, and T1 is the sum of the body sway amplitudes of all sitting passengers in the lower-deck carriage; F2 is the body sway amplitude of the sitting passengers in the upper-deck carriage obtained according to the expected deceleration and the expected deceleration duration, n2 is the total number of sitting passengers in the upper-deck carriage, and T2 is the sum of the body sway amplitudes of all sitting passengers in the upper-deck carriage; F3 is the body sway amplitude of the standing passengers in the lower-deck carriage obtained according to the expected deceleration and the expected deceleration duration, n3 is the total number of standing passengers in the lower-deck carriage, and T3 is the sum of the body sway amplitudes of all standing passengers in the lower-deck carriage; F4 is the body sway amplitude of the standing passengers in the upper-deck carriage obtained according to the expected deceleration and the expected deceleration duration, n2 is the total number of standing passengers in the upper-deck carriage, and T4 is the sum of the body sway amplitudes of all standing passengers in the upper-deck carriage.

[0081] In this embodiment: based on the collected data and other reference materials, determine the optimal range of deceleration. During the operation of the autonomous driving bus, there will be different expected deceleration requirements depending on the scenario. When the expected deceleration a is too small, it has little effect on passenger comfort and can be ignored. When the expected deceleration a is too large, passenger comfort cannot be considered too much due to safety issues, and control of the expected deceleration is not considered at this time. When the expected deceleration a is in the medium area, the expected deceleration a and passenger comfort need to be considered at the same time. The expected deceleration a refers to the deceleration that the autonomous driving vehicle needs to take next in order to avoid an accident. Where a is the expected deceleration, M is the expected deceleration threshold value when the deceleration is too small and passenger comfort does not need to be considered, which can be obtained through experiments, and N is the expected deceleration threshold value when the deceleration is too large and passenger comfort cannot be considered, which can be obtained through experiments.

[0082] Furthermore, the original average body swing amplitude T 0-x The specific formula is:

[0083]

[0084] Among them: T1 is the sum of the body swing amplitudes of all sitting passengers in the lower compartment; T2 is the sum of the body swing amplitudes of all sitting passengers in the upper compartment; T3 is the sum of the body swing amplitudes of all standing passengers in the upper compartment; T4 is the sum of the body swing amplitudes of all standing passengers in the lower compartment; n is the total number of passengers in the compartment.

[0085] Furthermore, the specific optimization method in step 4 is specifically:

[0086] The duration of the expected deceleration a is divided into two stages. Based on the expected deceleration value, the deceleration of the first stage is ax%, and the deceleration of the second stage is a+x%. Alternatively, the deceleration of the first stage is a+y%, and the deceleration of the second stage is ay%, where the value range of x is [10, 30].

[0087] In this embodiment: get T 0-x After that, the expected deceleration and duration are optimized, and the average body swing amplitude T is recalculated. 0-n , compared with T 0-x With T 0-n The size of the body swing can be used to determine whether the average body swing amplitude has decreased.

[0088] Here, we list two optimization methods.

[0089] First, according to the natural reaction characteristics of the human body, in a naturally moving bus, passengers sitting on seats are in a more relaxed state than standing passengers, while most standing passengers will hold the handrails tightly and be more alert to the surrounding environment. Therefore, when facing braking with different decelerations, the reactions of passengers are also different, and the feelings of passengers when the deceleration continues and when the braking ends are also different. When braking begins, if the deceleration at the beginning of braking is small, the body swing amplitude of passengers sitting on seats is small due to the low center of gravity and the support of the seats, while the body swing amplitude of standing passengers is greater than that of passengers sitting on seats due to factors such as the high center of gravity and less support. If the deceleration at the beginning of braking is large, passengers sitting on seats may not be able to react in time due to being more relaxed, grabbing the seat in front or the handrail next to them, so the body swing amplitude is large. On the contrary, standing passengers may be more alert to the surrounding environment and always hold the handrails, so the body swing amplitude may be smaller than that of passengers sitting on seats. Different deceleration distributions will have different effects during the braking period. If the braking deceleration is large at the beginning and then gradually decreases, the large deceleration at the beginning will cause a large body swing. At the same time, the sudden change of state will also cause discomfort and psychological tension to the passengers. As the deceleration decreases, the passengers sitting on the seats will gradually relax, which is specifically manifested in gradually loosening the armrests or seats that they held tightly at the beginning. Standing passengers need to hold the armrests with greater force even if the deceleration gradually decreases, and in order to resist the forward inertia, the body will gradually lean forward, causing a large body swing and increasing the physical burden on the passengers. Similarly, if the deceleration is small at the beginning of braking and then gradually increases, for the passengers sitting on the seats, it is a process of gradually feeling uncomfortable and gradually holding the seats and armrests. The same is true for standing passengers, but at the end of braking, the large deceleration suddenly disappears, causing the body to suddenly swing backwards, causing a large body swing. Our optimization solution is based on the natural human reaction characteristics of passengers during braking and the natural reaction of people when they are continuously subjected to deceleration. Specifically, in order to more conveniently describe the optimization process, we divide the entire braking process into two stages and perform different optimization operations on them. Note that the two stages are divided here only for the convenience of description, and the specific ones can be modified according to actual conditions.

[0090] The first method, when optimizing, first divides the expected deceleration duration into two stages. The deceleration magnitudes of the first and second stages are reduced by x% in the first stage and increased by x% in the second stage while meeting the expected braking effect. The value range of x is [10,30], with an interval of 10%, and the deceleration of the second stage is required not to exceed N. For the above different methods, the average body swing is calculated respectively, and each method obtains a new average body swing amplitude T 0-1 , T 0-2 , T 0-3 The sum of the maximum optimized body swing amplitude T of each type of passenger corresponding to the sum of the maximum optimized body swing amplitude T ymax etc. (When calculating the body swing amplitude of each type of passenger, by comparing the calculation results, a maximum optimized body swing amplitude sum T corresponding to the deceleration distribution can be obtained. ymax .

[0091] The second method divides the expected deceleration duration into two stages. The deceleration of the first and second stages is increased by y% in the first stage and decreased by y% in the second stage while meeting the expected braking effect. The value range of y is [10,30], with an interval of 10%. At the same time, the deceleration of the first stage must not exceed N. For the above different methods, the per capita swing amount is calculated respectively, and each method obtains a new per capita swing amount T 0-4 , T 0-5 , T 0-6 Equally and record the maximum optimized body swing amplitude T ymax .

[0092] Furthermore, the present invention also provides a system based on a control method for improving the riding comfort of an autonomous driving double-decker bus, comprising:

[0093] Collection module: used to collect the vehicle deceleration magnitude and duration of deceleration, the body swing amplitude of each type of passenger, and the threshold value S when the vehicle brakes for four types of passengers, namely, sitting passengers in the lower compartment, sitting passengers in the upper compartment, standing passengers in the upper compartment, and standing passengers in the lower compartment;

[0094] Calculation module: Calculate the original average body swing amplitude T before optimization of multiple expected decelerations a through the relationship function 0-x , optimize the sum of body swing amplitude T yn , optimize the average body swing amplitude T 0-n , from optimizing the average body swing amplitude T 0-n Select the minimum optimal average body swing amplitude T 0-n-min , from optimizing the sum of body swing amplitude T yn Select the maximum optimized body swing amplitude Tymax ;

[0095] Verification module: Verify the optimization results. If the minimum optimized average body swing amplitude T 0-n-min Smaller than the original average body swing amplitude T 0-xm And the maximum optimized body swing amplitude T ymax If it is less than the threshold S, the optimization is completed.

[0096] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention.

[0097] It should be understood that the present invention is not limited to what has been described above and that various modifications and changes may be made without departing from its scope. The scope of the present invention is limited only by the appended claims.

Claims

1. A control method for improving the riding comfort of an autonomous driving double-decker bus, characterized in that: The relationship function between the body swing amplitude and the vehicle deceleration magnitude and deceleration duration of each type of passenger is established through the collected data; The original average body swing amplitude T before optimization of multiple expected decelerations a is calculated through the relationship function. 0-x , optimize the sum of body swing amplitude T yn , optimize the average body swing amplitude T 0-n ; From optimizing the average body swing amplitude T 0-n Select the minimum optimal average body swing amplitude T 0-n-min , from optimizing the sum of body swing amplitude T yn Select the maximum optimized body swing amplitude T ymax , from the original average body swing amplitude T 0-x Select and minimize the average body swing amplitude T 0-n-min The corresponding first primitive average body sway amplitude T 0-xm ; Verify the optimization results. If the minimum optimized average body swing amplitude T 0-n-min Less than the first primitive average body sway amplitude T 0-xm And the maximum optimized body swing amplitude T ymax If it is smaller than the threshold S in the data, the optimization is completed.

2. A control method for improving the riding comfort of an autonomous driving double-decker bus, characterized in that: The specific steps include: Step 1, collecting data: collecting the deceleration magnitude and duration of the deceleration when the vehicle brakes, as well as the body swing amplitude and threshold S of each of the four types of passengers; wherein: the four types of passengers include: sitting passengers in the lower compartment, sitting passengers in the upper compartment, standing passengers in the upper compartment, and standing passengers in the lower compartment; Step 2: Establishing a relationship function: obtaining a relationship function between the body swing amplitude of each type of passenger and the vehicle deceleration magnitude and deceleration duration through the collected data; Step 3: Obtain the original average body swing amplitude T before optimization 0-x :Select multiple expected decelerations a that meet the optimization conditions from the data provided by the automatic driving system; calculate the original average body swing amplitude T for each expected deceleration a through the relationship function 0-x ; Step 4: Obtain the optimized average body swing amplitude T 0-n : According to a specific optimization method, multiple expected decelerations a are optimized to obtain a deceleration a corresponding to each expected deceleration a y ; Calculate the relationship between each type of passenger and each deceleration a through the relationship function y The optimal body swing amplitude T yn , and then find the value for each deceleration a y Optimized average body swing amplitude T 0-n ; Step 5: Select the value: Optimize the average body swing amplitude T from multiple 0-n Select the minimum optimal average body swing amplitude T 0-n-min , the minimum optimal average body swing amplitude T 0-n-min Compared with the original average body sway amplitude T 0-x The first primitive average body sway amplitude T 0-xm Correspondingly, according to the minimum optimization of average body swing amplitude T 0-n-min Select the sum of the optimized body swing amplitudes T of four types of passengers yn The maximum optimal body swing amplitude T ymax ; Step 6: Determine the optimized deceleration a y Whether it meets the requirements: If the minimum optimized average body swing amplitude T 0-n-min Less than the first primitive average body sway amplitude T 0-xm And the maximum optimized body swing amplitude T ymax If it is less than the threshold S, then the minimum optimized average body swing amplitude T 0-n-min The corresponding optimized deceleration a y Meets the requirements and optimization is completed.

3. A control method for improving riding comfort of an autonomous driving double-decker bus according to claim 2, characterized in that: The vehicle deceleration magnitude and deceleration duration in the step one are collected by the vehicle's built-in IMU, and the body sway amplitude of each type of passenger is collected by human body posture estimation method, optical flow method, or deep learning model.

4. A control method for improving riding comfort of an autonomous driving double-decker bus according to claim 2, characterized in that: The relationship function between the body swing amplitude of each type of passenger and the vehicle deceleration magnitude and deceleration duration in step 2 includes: Function F1: relationship between the amplitude of body swing of sitting passengers in the lower-level carriage and the magnitude of vehicle deceleration and the duration of deceleration; function F2: relationship between the amplitude of body swing of sitting passengers in the upper-level carriage and the magnitude of vehicle deceleration and the duration of deceleration; function F3: relationship between the amplitude of body swing of standing passengers in the upper-level carriage and the magnitude of vehicle deceleration and the duration of deceleration; function F4: relationship between the amplitude of body swing of standing passengers in the lower-level carriage and the magnitude of vehicle deceleration and the duration of deceleration.

5. A control method for improving riding comfort of an autonomous driving double-decker bus according to claim 4, characterized in that: The specific formulas of the relationship function F1, relationship function F2, relationship function F3, and relationship function F4 are as follows: F1=K 11 *a+K 12 *b (1) F2=K 21 *α+K 22 *β (2) F3=K 31 *a+K 32 *b (3) F4=K 41 *α+K 42 *β (4) Among them: α is the normalized value of deceleration, β is the normalized value of deceleration duration, K 11 is the deceleration influence coefficient of the sitting passengers in the lower compartment, K 12 K is the deceleration duration influence coefficient of the sitting passengers in the lower compartment; 21 is the deceleration influence coefficient of the sitting passengers in the upper compartment, K 22 K is the deceleration duration influence coefficient of the sitting passengers in the upper compartment; 31 is the deceleration influence coefficient of the standing passengers in the upper compartment, K 32 K is the deceleration duration influence coefficient of the standing passengers in the upper compartment; 41 is the deceleration influence coefficient of standing passengers in the lower deck carriage, K 42 is the deceleration duration influence coefficient of standing passengers in the lower deck carriage.

6. A control method for improving riding comfort of an autonomous driving double-decker bus according to claim 2, characterized in that: The optimization conditions in step 3 are: Where: a is the expected deceleration; M is the minimum expected deceleration threshold; N is the maximum expected deceleration threshold.

7. A control method for improving riding comfort of an autonomous driving double-decker bus according to claim 2, characterized in that: The original average body swing amplitude T in step 3 0-x The specific formula is: Among them: T1 is the sum of the body swing amplitudes of all sitting passengers in the lower compartment; T2 is the sum of the body swing amplitudes of all sitting passengers in the upper compartment; T3 is the sum of the body swing amplitudes of all standing passengers in the upper compartment; T4 is the sum of the body swing amplitudes of all standing passengers in the lower compartment; n is the total number of passengers in the compartment.

8. A control method for improving riding comfort of an autonomous driving double-decker bus according to claim 2, characterized in that: The sum of the optimized body swing amplitudes T in step 4 yn Specifically: Among them, F1 is the body sway amplitude of the sitting passengers in the lower-deck carriage obtained according to the expected deceleration and the expected deceleration duration, and n1 is the total number of sitting passengers in the lower-deck carriage; F2 is the body sway amplitude of the sitting passengers in the upper-deck carriage obtained according to the expected deceleration and the expected deceleration duration, and n2 is the total number of sitting passengers in the upper-deck carriage; F3 is the body sway amplitude of the standing passengers in the lower-deck carriage obtained according to the expected deceleration and the expected deceleration duration, and n3 is the total number of standing passengers in the lower-deck carriage; F4 is the body sway amplitude of the standing passengers in the upper-deck carriage obtained according to the expected deceleration and the expected deceleration duration, and n2 is the total number of standing passengers in the upper-deck carriage.

9. A control method for improving riding comfort of an autonomous driving double-decker bus according to claim 2, characterized in that: The specific optimization method in step 4 is specifically: The duration of the expected deceleration a is divided into two stages. Based on the expected deceleration a, the deceleration of the first stage is ax%, and the deceleration of the second stage is a+x%. Alternatively, the deceleration of the first stage is a+y%, and the deceleration of the second stage is ay%, where the value range of x and y is [10, 30].

10. A system based on the control method for improving the riding comfort of an autonomous driving double-decker bus according to any one of claims 1 to 9, characterized in that: include: Collection module: used to collect the vehicle deceleration magnitude and duration of deceleration, the body swing amplitude of each type of passenger, and the threshold value S when the vehicle brakes for four types of passengers, namely, sitting passengers in the lower compartment, sitting passengers in the upper compartment, standing passengers in the upper compartment, and standing passengers in the lower compartment; Calculation module: Calculate the original average body swing amplitude T before optimization of multiple expected decelerations a through the relationship function 0-x , optimize the sum of body swing amplitude T yn , optimize the average body swing amplitude T 0-n , from optimizing the average body swing amplitude T 0-n Select the minimum optimal average body swing amplitude T 0-n-min , from optimizing the sum of body swing amplitude T yn Select the maximum optimized body swing amplitude T ymax ; From the original average body swing amplitude T 0-x Select and minimize the average body swing amplitude T 0-n-min The corresponding first primitive average body sway amplitude T 0-xm ; Verification module: Verify the optimization results. If the minimum optimized average body swing amplitude T 0-n-min Less than the first original average body swing amplitude T 0-xm And the maximum optimized body swing amplitude T ymax If it is less than the threshold S, the optimization is completed.

Images