Journal of Physics: Conference Series
PAPER • OPEN ACCESS You may also like
- MULTI-SPACECRAFT ANALYSIS OF
Modelling and Simulation of Vehicle ESP System ENERGETIC HEAVY ION AND
INTERPLANETARY SHOCK
Based on CarSim and Simulink PROPERTIES IN ENERGETIC STORM
PARTICLE EVENTS NEAR 1 au
R. W. Ebert, M. A. Dayeh, M. I. Desai et al.
To cite this article: Nanyu Jiang and Runqi Qiu 2022 J. Phys.: Conf. Ser. 2170 012032 - Efficient charge generation at low energy
losses in organic solar cells: a key issues
review
Ye Xu, Huifeng Yao, Lijiao Ma et al.
- Dependence of Energetic Storm Particle
View the article online for updates and enhancements. Heavy Ion Peak Intensities and Spectra on
Source CME Longitude and Speed
A. Santa Fe Dueñas, R. W. Ebert, M. A.
Dayeh et al.
This content was downloaded from IP address 89.184.207.180 on 01/02/2023 at 18:42
ISCME-2021 IOP Publishing
Journal of Physics: Conference Series 2170 (2022) 012032 doi:10.1088/1742-6596/2170/1/012032
Modelling and Simulation of Vehicle ESP System Based on
CarSim and Simulink
Nanyu Jiang, Runqi Qiu*
School of International Education, Wuhan University of Technology, Wuhan, Hubei,
430070, China
*
Corresponding author’s e-mail:
[email protected] Abstract. In this paper, the simulation of the vehicle ESP system is carried out, and the linear
two-degree-of-freedom dynamic reference model of the vehicle is established in
MATLAB/Simulink. Combined with the method of direct yaw moment control, the yaw rate and
the yaw angle are selected as the control variables. Based on the fuzzy PID control theory, the
control strategy of the ESP system is designed to calculate the additional yaw moment required
when the vehicle deviates from the desired trajectory. In order to verify the effectiveness of the
ESP system designed in this paper, the vehicle dynamics model in CarSim software is used to
build a joint simulation platform of CarSim and Simulink, and the model is tested at a speed of
120 km/h on a double lane change road with low adhesion rate. The simulation results show that
the designed controller can significantly improve the manoeuvrability and stability of the vehicle.
1. Introduction
ESP is an automobile active safety system developed on the basis of automobile anti-lock braking
system (ABS) and traction control system (TCS). The system measures the steering wheel input angle,
yaw rate, lateral acceleration and wheel speed by sensors, calculates the expected driving trajectory of
the driver and the actual motion state of the vehicle by ECU, and identifies the road conditions.
By controlling the vehicle traction and applying brakes to the corresponding wheels, the system
corrects the vehicle dynamics, especially for the lateral stability of the vehicle driving at high speed on
low adhesion road, so that the vehicle can still maintain stability under extreme working conditions,
which significantly improves the handling stability and driving safety of the vehicle[1].
This paper proposes an ESP system with yaw rate and yaw angle as control variables. The fuzzy PID
algorithm with strong robustness and high accuracy is used to solve the additional yaw moment when
the vehicle deviates from the expected trajectory. The 2DOF model of the vehicle and the simulation
scene of the double lane change road are built by CarSim software. The ESP system is designed in
MATLAB/Simulink, and the performance of the system is verified by the joint simulation of the two.
2. ESP Algorithm Design
2.1. Working Principle of ESP System
ESP monitors the movement of the vehicle in real time according to the feedback information of various
sensors on the vehicle. When the vehicle has oversteering or understeering, it applies braking force to
one or more wheels to ensure the steering stability of the vehicle and prevent sideslip.
As shown in Figure 1, when the vehicle understeers, it tends to deviate to the outside of the ideal
trajectory curve, and the system applies braking force to the inside rear wheel, which increases the
Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution
of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd 1
ISCME-2021 IOP Publishing
Journal of Physics: Conference Series 2170 (2022) 012032 doi:10.1088/1742-6596/2170/1/012032
longitudinal force, so the lateral force provided decreases, and subsequently generates a steering-
promoting torque to the body; when oversteering, the car's trajectory deviates to the inside, and the
system applies braking force to the outside front wheel, which increases the longitudinal force, so the
lateral force provided decreases, and thus generates a steering-promoting torque to the body. The system
inhibits the steering moment by applying braking force to the outside front wheel, which increases the
longitudinal force and thus provides less lateral force to the body. This returns the wheels to the desired
trajectory curve, thus ensuring the stability of the car.
Figure 1. Working Principle of ESP System
2.2. Vehicle Dynamics Model
In order to reflect the dynamic characteristics of the vehicle as accurately as possible with as little
calculation as possible, and calculate the desired state quantity of the vehicle based on this, this paper
establishes a two-degree-of-freedom (2DOF) linear dynamic model of the vehicle according to extant
literature[2]. The influence of the suspension on the vehicle is ignored, and it is assumed that the
compartment only performs plane motion parallel to the ground. The vehicle is simplified to a model
with only lateral and yaw motion. The model can well reflect the driver 's steering input and the linear
relationship between the lateral acceleration of the vehicle and the yaw angle and yaw rate. The
differential equation is as follows:
1
𝑘 𝑘 𝛽 𝑎𝑘 𝑏𝑘 𝜔 𝑘 𝛿 𝑚 𝑣 𝑢𝜔
𝑢 1
1
𝑎𝑘 𝑏𝑘 𝛽 𝑎 𝑘 𝑏 𝑘 𝜔 𝑎𝑘 𝛿 𝐼 𝜔
𝑢
Where 𝑚 stands for vehicle quality; 𝛿 is the front wheel angle; 𝛽 is the yaw angle; 𝑎 and 𝑏 are the
distances from the front and rear axles to the vehicle centroid, respectively; 𝑢 and 𝑣 represent the vehicle
speed along 𝑋 axis and 𝑌 axis respectively; 𝑘 and 𝑘 are the lateral stiffness of the front and rear
wheels, respectively; 𝜔 represents vehicle yaw rate; 𝐼 is the rotational inertia of the vehicle around
the 𝑍 axis of the centroid.
The yaw rate and the yaw angle are selected as the state variables to describe the vehicle stability.
The real value of the state variables can be measured by the yaw rate sensor and the yaw angle sensor
carried by the vehicle. By comparing the real value with the ideal state variables calculated by the vehicle
dynamics model, the stability of the vehicle running state can be determined.
By combining the expression of the 2DOF dynamic model and eliminating the sideslip angle 𝛽 of
the centroid, the expected value 𝜔 of the yaw rate under the steady-state response can be obtained:
1 𝑢
𝜔 ⋅ ⋅𝛿 2
𝑚 𝑎 𝑏 𝐿
1 𝑢
𝐿 𝑘 𝑘
By eliminating the yaw rate 𝜔 , the expected sideslip angle 𝛽 under steady state response can be
obtained:
2
ISCME-2021 IOP Publishing
Journal of Physics: Conference Series 2170 (2022) 012032 doi:10.1088/1742-6596/2170/1/012032
𝑚𝑎
1 ⋅𝑢 𝑏
𝐿𝑏𝑘
𝛽 ⋅ ⋅𝛿 3
𝑚 𝑎 𝑏 𝐿
1
𝐿 𝑘 𝑘
The maximum yaw rate corresponding to the road surface with different adhesion coefficients can
be expressed as:
𝜇𝑔
High Adhesion Road: 𝜔 𝜔
𝑢 4
𝜇𝑔
Low Adhesion Road: 𝜔 𝜔
𝑢
Thus, the ideal yaw rate can be written as:
𝜔 min |𝜔 |, |𝜔 | 5
Similarly, the maximum yaw angle corresponding to adhesion conditions are also different:
𝑏 𝑚𝑎
⎧High Adhesion Road: 𝛽 𝜇𝑔
𝑢 𝑘 𝐿
6
⎨Low Adhesion Road: 𝛽 𝑏 𝑚𝑎
𝜔 𝑢
⎩ 𝑢 𝑘 𝐿
And the ideal yaw angle is:
𝛽 min |𝛽|, |𝛽 | 7
2.3. Braking Torque Distribution
The steering characteristics of the vehicle can be expressed by the difference between the front and rear
wheel sideslip angle, which can be expressed as:
0, understeering
𝛼 𝛼 0, neutralsteering 8
0, oversteering
The front and rear wheel sideslip angles can be expressed as:
𝛼𝜔
𝛼 𝛽 𝛿
𝑢
𝑏𝜔 9
𝛼 𝛽
𝑢
Substituting (9) into (8), there is the following expression:
𝑙𝜔 𝛿𝑢 𝑙𝜔
𝛼 𝛼 𝛿 10
𝑢 𝑢
When the lateral stiffness of rear wheel is large, the ideal yaw rate is:
𝛿𝑢
𝜔 11
𝑙
Substituting (11) into (10), we have:
𝜔 𝜔 𝑙
𝛼 𝛼 12
𝑦
Therefore, the steering characteristics of the vehicle can be expressed by the difference between the
ideal yaw rate and the true value:
0, understeering
Δ𝜔 𝜔 𝜔 0, neutralsteering 13
0, oversteering
According to the extant literature[3], when understeering and oversteering, the relationship between
the yaw moment Δ𝑀 that required by the vehicle and the longitudinal force and lateral force of the four
wheels is as follows:
When the vehicle is understeering:
3
ISCME-2021 IOP Publishing
Journal of Physics: Conference Series 2170 (2022) 012032 doi:10.1088/1742-6596/2170/1/012032
𝐵 𝐵
Δ𝑀 𝐹 𝑏 sin 𝛿 cos 𝛿 Δ𝐹 sin 𝛿 𝑏 cos 𝛿 14
2 2
When the vehicle is oversteering:
𝐵 𝐵
Δ𝑀 𝐹 𝑎 sin 𝛿 cos 𝛿 Δ𝐹 sin 𝛿 𝑎 cos 𝛿 15
2 2
In the equation, 𝐹 and 𝐹 represent the longitudinal and transverse forces on the j-side tire of the
i-axle when braking is applied. 𝑖 ∈ 𝑓, 𝑟 , respectively representing the front axle and the rear axle, 𝑗 ∈
𝑖, 𝑜 , respectively representing the inside and outside; 𝐵 and 𝐵 are front and rear axle wheel spacing
respectively. To simplify the model, only the front wheel steering 𝛿 is considered, which means 𝛿
0.
Assuming that the tire braking force is approximately proportional to its vertical load when the wheel
is not locked, the braking force of the four wheels of the vehicle can be dynamically distributed as
follows:
If understeering occurs, the braking torque of the inner wheel is:
𝐹 4∆𝑀
⎧𝑇 ∙ ∙𝑅
⎪ 𝐹 𝐹 𝐵 𝐵
16
⎨𝑇 𝐹 4∆𝑀
⎪ ∙ ∙𝑅
⎩ 𝐹 𝐹 𝐵 𝐵
If oversteering occurs, the braking torque of the inner wheel is:
𝐹 4∆𝑀
⎧𝑇 ∙ ∙𝑅
⎪ 𝐹 𝐹 𝐵 𝐵
17
⎨𝑇 𝐹 4∆𝑀
⎪ ∙ ∙𝑅
⎩ 𝐹 𝐹 𝐵 𝐵
Where, 𝐹 represents the vertical load of the 𝑖-axle 𝑗-side wheel, and 𝑅 is the tire radius.
3. ESP Control Strategy Design
3.1. Working Principle of Fuzzy PID
Fuzzy control is a control strategy imitating the fuzzy concept of human brain in artificial control. The
fuzzy PID established according to this theory is a parameter self-tuning intelligent PID control, which
can adaptively adjust the PID parameters according to the input error 𝑒 and the error change rate 𝑒𝑐.
PID algorithm is a commonly used control strategy in engineering. Although it has the advantages
of simple calculation, good robustness and high reliability, it often requires personnel with a lot of
engineering experience to manually adjust its parameters 𝐾 , 𝐾 and 𝐾 . Fuzzy control can be used to
adjust these three parameters in a certain range.
The functional relationship between PID parameters and 𝑒 and 𝑒𝑐 can be established by fuzzy
control theory. When the error is large, 𝐾 is appropriately increased and 𝐾 is reduced. When the error
is small, 𝐾 and 𝐾 increase; decrease 𝐾 when the error is in the proper range to avoid overshoot.
3.2. Design of Fuzzy PID Controller
In order to ensure the accuracy and stability of the control system, seven fuzzy levels are selected for
the corresponding output variables, namely, the fuzzy subset is {NL, NM, NS, Z, PS, PM, PL},
representing {Negative Large, Negative Medium, Negative Small, Zero, Positive Small, Positive
Medium, Positive Large} respectively[4].
The fuzzy rule is expressed as the form of "if 𝑒 and 𝑒 then M", and the decision results are inferred
based on the premises of the opposite yaw angle error and the yaw rate error, and the fuzzy rule is shown
in Table 1.
4
ISCME-2021 IOP Publishing
Journal of Physics: Conference Series 2170 (2022) 012032 doi:10.1088/1742-6596/2170/1/012032
Table 1. Fuzzy Rules
𝑒 𝑒
PB PS Z NS NB
PB PS PS PM PB PB
PS PS PS PM PM PB
Z NM NS Z PS PM
NS NB NM NS NS NS
NB NB NB NM NS NS
4. CarSim Parameter Settings.
4.1. The Acquisition of Tire Cornering Stiffness
CaSim does not directly provide tire cornering stiffness, but provides the tire cornering characteristic
curve, so it is necessary to use MATLAB to identify the tire model cornering stiffness.
According to extant literature[5], the relationship between tire cornering force and cornering angle
can be fitted by Pacejka's magic formula:
Where,𝐹 𝛼 is the lateral force corresponding to the tire cornering angle 𝛼, and the coefficients 𝐵,
𝐶 and 𝐷 are determined by the vertical load and camber angle of the tire in turn. The product of 𝐵, 𝐶
and 𝐷 corresponds to the slope at the origin of the cornering characteristic curve. Because the tire is
assumed to be in the linear region in the 2DOF vehicle dynamics model, the tire cornering stiffness 𝑘
𝐵𝐶𝐷.
Based on the above principle, the curve fitting toolbox in MATLAB can be used to fit the sideslip
characteristic data of B-Class vehicle provided by CarSim, and he fitting parameters of the front wheel
lateral force are: 𝐵 20.9, 𝐶 1.4, 𝐷 8718, 𝐸 0.5489; the fitting parameters of rear wheel
lateral force are: 𝐵 15.17, 𝐶 1.833, 𝐷 6413, 𝐸 0.9657; it can be calculated that the front
wheel lateral stiffness 𝑘 102176 N/rad, and the rear wheel lateral stiffness 𝑘 14650 N/rad.
4.2. Test Road Settings
In order to test the performance of ESP system, a double lane-change test road scene on low adhesion
ground is built in CarSim to simulate the extreme working conditions when the vehicle overtakes at a
high speed (120 km/h) on the low adhesion road and returns to the correct lane quickly, so as to test the
handling stability of the vehicle. In the experiment, the road adhesion coefficient is set to 0.5. The test
road is shown in Figure 2.
Figure 2. Test Road
5. Simulation Results and Analysis
After the vehicle dynamics model is established in CarSim, the output of the model is set as: left front
wheel angle (Steer _ L1), right front wheel angle (Steer _ R1), sideslip angular velocity (AVz), sideslip
angle (Beta), longitudinal speed (Vx). The input is the braking torque of four wheels (IMP _ MYBK _
L1, IMP _ MYBK _ R1, IMP _ MYBK _ L2, IMP _ MYBK _ R2), and the vehicle model is exported
to Simulink and connected with the ESP system to obtain a simulation model as shown in Figure 3:
5
ISCME-2021 IOP Publishing
Journal of Physics: Conference Series 2170 (2022) 012032 doi:10.1088/1742-6596/2170/1/012032
Figure 3. Simulation Model
Figures 4 to 6 are the curves of vehicle trajectory, yaw angle and yaw rate of the test vehicle. Figures
show that the vehicle with ESP system restrains sideslip and tail flick well and improves the driving
stability of the vehicle, while the vehicle without ESP is completely unstable and in a very dangerous
state:
Figure 4. Trajectory of the Test Vehicle Figure 5. Yaw Angle of the Test Vehicle
Figure 6. Yaw Rate of the Test Vehicle
6. Conclusion
In this paper, a fuzzy PID controller with strong robustness is designed for ESP, a co-simulation platform
of CarSim and Simulink is built, and the simulation is carried out under the condition of low adhesion
and double lane change. Through experiments, it can be seen that the designed ESP system can
effectively improve the driving trajectory, sideslip angle and yaw rate of the vehicle, verify the
rationality and accuracy of the controller, and greatly improve the driving safety and handling stability
of the vehicle. However, it can also be seen from the test results that there is still a certain deviation
between the vehicle trajectory and the ideal trajectory, so the control algorithm still has some room for
improvement. The feasible solutions include:
6
ISCME-2021 IOP Publishing
Journal of Physics: Conference Series 2170 (2022) 012032 doi:10.1088/1742-6596/2170/1/012032
Because the experiment is carried out at high speed of 120km/h, the 2DOF model may not
accurately describe the dynamic characteristics of the vehicle, so the reference model with
higher DOF can be used to obtain more accurate expected state variables.
Although PID algorithm has many advantages, it does not depend on the accurate model of the
controlled object, so it will inevitably make the adjustment time longer. Therefore, MPC, LQR
and other control algorithms based on the accurate mathematical model of the controlled object
can be used to achieve more detailed yaw angle control.
References
[1] Sun, Y.D., Guo, S., Zou, P. (2018) Joint simulation study of CarSim and Simulink for ESP system,
Machine Design and Manufacture, (3): 16-18
[2] Yu, Z.S. (2009) Vehicle Handling Stability. In: Zhao, A.N. (Eds.), Automobile Theory. China
Machine Press, Beijing. 144-147.
[3] Qin, C.X., Zhang, L., Yan, Y. (2020) CarMaker/Simulink modelling and simulation of automotive
ESP system. Mechanical Science and Technology, 38(12): 1263-1270.
[4] Wang, Z.C., Cheng, J., Zhang, C., Liu, J.W., Zheng, D.Z. (2013) Simulation of automobile ESP
control system based on fuzzy control. Fuzzy Systems and Mathematics, 27(5): 88-94
[5] Li, L., Ma, L., Mou, Y., Xu, Chao. (2016) Parameter identification method for lateral deflection
stiffness of model car tires. Automotive Engineering, 38(12): 1508-1514
[6] Chu, L., Gao, X.Z., Guo, J.H., Liu, H.W., Chao, L.B., Shang, M.L. (2012) Coordinated Control
of Electronic Stability Program and Active Front Steering. Procedia Environmental Sciences,
12:1379-1386.
[7] Gallep, J., Govender, V., Müller, S. (2018) Model Predictive Lateral Vehicle Guidance Using a
Position Controlled EPS System. IFAC-PapersOnLine, 50(2): 265-270.
[8] Nagai, M., Shino, M., Gao, F. (2002) Study on integrated control of active front steer angle and
direct yaw moment. JSAE Review, 23(3): 0389-4304.
