Design of AGV speed control system based on Fuzzy PID

  • A+
Categories:AGV Data

Automated Guide Vehicle(AGV)requires high control precision in the process of automatic tracking. Based on this,firstly,the structure of AGV system is introduced in this paper. Secondly, aiming at the situation that the speed control system of DC motor in AGV is non⁃linear,time⁃varying and susceptible to external interference,a speed fuzzy PID controller is designed,and the ladder diagram is used to realize the fuzzy control algorithm in AB PLC. The operation results show that the designed fuzzy controller has the advantages of fast response and small overshoot,which can drive the AGV car to move along the set track at the optimum speed.

AGV (automated guided vehicle), as an important part of the current industrial automation system, plays an important role in the logistics scheduling system of factories and workshops [1], and has a very broad application prospect.

The traditional AGV car control is mainly based on the classical PID control, but the accuracy of PID control depends on the accuracy of the system mathematical model and parameter setting [2]. The working environment of AGV car in the factory and workshop is complex, and it will be interfered by various working conditions in the driving process. The traditional PID control is sometimes difficult to achieve the effect [3].

Based on this, the fuzzy control mechanism is analyzed. Aiming at the speed control system of DC motor in AGV, the fuzzy PID control is adopted, which greatly improves the accuracy, stability and response speed of the speed control of intelligent car.

1 Composition of AGV system

The AGV system is equipped with magnetic navigation, CCD navigation, laser navigation, etc. [4], which enables the AGV to run according to the path planned in advance during the driving process. The operator only needs to schedule the scene in real time according to the upper computer, and for different field operation conditions, the AGV The real-time on-site scheduling, therefore, greatly reduces the dependence on human resources, and improves the work efficiency and safety factor.

The AGV system uses the magnetic navigation sensor attached to the ground to navigate, real-time monitor the location of the AGV car on the spot, and dispatch it on the spot to avoid the collision between AGVs. The power supply system of AGV is provided by batteries, two batteries are installed at both ends of AGV (generally symmetrical installation), and magnetic induction sensors are installed at both ends of AGV, generating a closed-loop system to adjust the actual position of the car on site [5]. There are many types of AGV sizes. For the scheduling of large AGV, it is necessary to consider whether its safety range is enough and whether it meets the site requirements when turning. The parts drawing of AGV system is shown in Figure 1, including ① start, stop, reset and emergency stop buttons; ② and ③ are 12 V batteries; ④ wheels; ⑤ tricolor lights; ⑥ conveyor belt; ⑦ touch screen; ⑧ obstacle avoidance sensor; ⑨ opposite sensor. The AGV servo installation is shown in Figure 2, where ① is the front magnetic navigation sensor; ② is the rear magnetic navigation sensor; ③ is the servo motor; ④ is the servo driver.

Design of AGV speed control system based on Fuzzy PID

Figure 1 AGV system parts drawing

Design of AGV speed control system based on Fuzzy PID

Figure 2 AGV servo installation diagram

2 Design of fuzzy PID control for AGV system

2.1 AGV speed control strategy

The speed control of AGV based on Fuzzy PID is shown in Figure 3 [6].
The driving controller controls the infrared sensor to work. According to the collected tracking information of road conditions, the given speed value of the car is obtained. At the same time, the wheel speed detection module detects the car speed for isolation, amplification and a / d After conversion, compared with the given speed, if the logistics scheduling path of the plant has upward, downward or turning conditions, the speed detection module will show up or down phenomenon, and send this deviation signal to the fuzzy PID The controller, the controller carries on the fuzzy rule choice, outputs the precise speed pulse signal to the power electronic actuator, drives the DC motor to run accurately according to the actual road condition trace speed.

Design of AGV speed control system based on Fuzzy PID

Figure 3 speed control block diagram of AGV

2.2 Design of AGV speed fuzzy PID controller

The structure of the fuzzy PID controller is shown in Figure 4, which is mainly divided into two modules. One is the traditional PID controller, the other is the fuzzy controller. The fuzzy controller realizes the on-site determination of the PID controller parameters The PID control effect is more ideal. The fuzzy PID controller mainly realizes the relationship between the three parameters of PID and the change rate △ e of error E. in the operation process, the final proportion, integral and differential parameters are constantly judged and calculated according to the change rate of error and error [7].

Design of AGV speed control system based on Fuzzy PID

Figure 4 structure of fuzzy PID control

Based on the above structure, AGV speed control fuzzy PID design can be divided into the following four steps [8-9].

The first step is to determine the input and output signals of the system. The AGV car data sampling is conducted by the 16 bit digital sampling point of the magnetic navigation sensor in the mode of I / O input. The actual position of the AGV car is determined by analyzing the different input signals [10]. The magnetic navigation is 16 channels of 16 point digital output, and the signal normally checked is 5 channels of 5 point digital output. The theory is divided into the following 20 cases:
Take the angle velocity error E and error change rate EC of DC motor as the system input variables, and Δ KP, Δ Ki, Δ KD as the output variables. The fuzzy basic domain of error E and error rate of change EC is taken as [- 6, 6].
Mapping to the domain by scale and quantization factors: set the 16 channel output of the magnetic navigation sensor from left to right as (1) → - 6, (2) → - 5, (3) (4) → - 4, (5) (6) → - 3, (7) → - 2, (8) (9) → - 1, (10) (11) → 0, (12) (13) → 1, (14) → 2, (15) (16) → 3, (17) (18) → 4, (19) (20) → 5
In the second step, the input and output variables of the system are fuzzed.

In the design, the quantization levels of E, EC, Δ KP, Δ Ki and Δ KD are all set to 13 levels, that is, the domain of two input variables and one output variable on the fuzzy set is: {- 6, - 5, - 4, - 3, - 2, - 1,0, + 1, + 2, + 3, + 4, + 5, + 6}. The corresponding fuzzy language is {Nb, nm, NS, Zo, PS, PM, Pb}. The elements in the set represent negative large, negative medium, negative small, zero, positive small, positive medium, and positive large respectively [11-12]. The above set of elements are described as the acceleration and deceleration conditions of AGV trolley, which are: NB for large deceleration, nm for medium deceleration, ns for small deceleration, Zo for maintaining the current speed, PS for small acceleration, PM for medium acceleration, Pb for large acceleration. Considering that the error of AGV car in actual operation is random, triangle membership function is adopted, as shown in Figure 5.

Design of AGV speed control system based on Fuzzy PID

Figure 5 triangle membership function

The third step is the establishment of fuzzy rule base.

The establishment of fuzzy rule base is to find out the fuzzy relationship between P, I, D, error E and error change rate EC. According to the change law of E and EC, the fuzzy control rules are applied to adjust the three parameter values of Δ KP, Δ Ki and Δ KD, so that the AGV car has good dynamic performance and stability in the operation process [13-14]. The control law of speed error E is particularly important when it is applied in the intelligent car. The inappropriate control law will make the speed of AGV oscillate, and has no good adaptability to different paths of different workshops. When the deviation between the actual running speed of the car body and the expected speed e is large, in order to accelerate the tracking speed of the system, a larger KP should be taken; however, in order to avoid the differential supersaturation that may occur due to the instantaneous increase of the deviation e at the beginning and make the control effect beyond the allowable range, a smaller KD should be taken, and in order to prevent the system speed response from large overshoot, resulting in integral saturation and response product It is limited by the action, usually ki = 0, etc. According to the operation experience of AGV, the fuzzy rule base of speed control output parameters (Δ KP, Δ Ki, Δ KD) can be obtained as shown in Table 1 [15-17].

Design of AGV speed control system based on Fuzzy PID

Table 1 Fuzzy Control Law of Δ KP, Δ Ki and Δ KD

The output value of fuzzy controller is fuzzy value, which can not be used to control the speed of DC motor directly. These results should be converted into accurate values in the actual control of AGV. Considering that the barycenter method can better reflect the real distribution of the control quantity, the barycenter method is used to transform the fuzzy variables in this design.

3 Comparative analysis of experiments

In this paper, AB PLC is used as the control equipment, PID controller and fuzzy controller are designed with ladder diagram. In order to respond to the actual position of AGV car quickly, the sampling time is set as 100mms. After the circuit and gas circuit are connected and checked to be correct, power on to download the program and monitor the program operation online. Connect the router and set the IP of AB PLC. The specific commissioning process is as follows:

1) In the main menu screen, use the up and down keys on the keyboard to select advanced set.

2) Click OK on LCD keyboard to open advanced setting operation interface, as shown in Figure 6. There are up and down flip keys on the interface to select eNet function. Click OK after selection to enter.

3) Use "up" and "down" up and down flip keys to set the IP address. Click OK for the set IP address.

Design of AGV speed control system based on Fuzzy PID

Figure 6 PID parameter program debugging interface

4) Input the password operation interface. In the AGV car control system, set the master password through the left, right, up and down keys. The maximum length of the password is 10 digits. In the operation system, set 1234 as the master password.

5) Password verification interface: if the input password is correct, the Ethernet network type selection interface will be displayed. Click up and down to select the appropriate network type. If the input password is not correct, the operation interface will prompt the error message of the wrong password.

6) Set the IP address and subnet mask of the network. After the debugging, Figure 7 and figure 8 show the results of traditional PID and fuzzy PID. It can be seen that the fuzzy PID control has a shorter response time, a smaller overshoot, and can quickly enter the steady state, so it can better track the speed control of AGV car.

Design of AGV speed control system based on Fuzzy PID

Figure 7 PID response diagram

Design of AGV speed control system based on Fuzzy PID

Figure 8 response diagram of fuzzy PID

4 Conclusion

This paper introduces the structure of AGV car, on this basis, for the speed control system of the car, design the speed fuzzy PID controller, and carry out the actual verification on the PLC equipment. Compared with the traditional PID control, AGV fuzzy PID speed controller has shorter response time, no oscillation and overshoot in the response process, and has better dynamic and steady-state performance.


[1] Zeng Xuan. Application Research on path correction of magnetic guided AGV system based on fuzzy control [D]. Hangzhou: Hangzhou University of Electronic Science and technology,2015.
[2] Mendes J,Araujo R,Matias T,et al. Automatic extraction of the fuzzy control system by a hierar⁃ chical genetic algorithm[J]. Engineering Applica⁃ tions of Artificial Intelligence,2014,29(3):70-78.
[3] Baturone I,Moreno-Velo FJ,Śnchez-Solano S,et al. Automatic design of fuzzy controllers for carlike autonomous robots[J]. IEEE Transactions on Fuzzy Systems,2004,12(4):447-465.
[4] Song Xuexian, Tan Honghua, pan Zhengchun, et al. Research on AGV high precision tracking system based on Adaptive Fuzzy PID [J]. Mechatronics technology at home and abroad,2012(1):29-31.
[5] Wu Xing, Lou peihuang, Tang dunbing. Hybrid motion control of path tracking and servo control for AGV [J]. Journal of mechanical engineering, 2011, 47 (3): 43-48

[6] Ye Jinxin. Research on intelligent car control algorithm based on Fuzzy PID [J]. Experimental science and technology, 2016, 14 (1): 46-47

[7] Wang Shuyan, Shi Yu, Feng Zhongxu. Research on control method based on Fuzzy PID controller [J]. Mechanical science and technology, 2011, 30 (1): 166-172

[8] Huang yourui, Qu Liguo. Parameter setting and Realization of PID controller

[9] Tang Hongcheng, Li zxin, Wang Zhengtao, et al. A fuzzy PID control system [J]. Journal of electrical machinery and control, 2005, 9 (2): 136-138

[10] Ji Yafang, Zhang Zhigang. Liquid level control system design of double tank based on Fuzzy PID [J]. Journal of Shanxi Normal University (NATURAL SCIENCE EDITION), 2019, 33 (2): 37-40

[11] Li Yanmin, Miao Xin, Jiang Fujie, et al. Design of multi motor cross coupling control synchronous control system based on Fuzzy PID controller [J]. Electrical technology, 2005 (3): 118-120

[12] Wang Jianhong, Chen Yaozhong, Chen Gui, et al. Research on dual motor synchronous control system based on cross coupling control [J]. Journal of Nanjing University of technology, 2017, 41 (6): 693-697

[13] Zhang Xiaojuan. Design of deaerator temperature control system based on AT89S51 single chip computer [J]. Journal of textile university basic science, 2010 (3): 372-376

[14] Ju Shenghui, Li Yangmin. Research on domain self-adjusting fuzzy control of nonholonomic constrained mobile robot [J]. Electronic design engineering, 2016, 24 (5): 103-106

[15] Xu Zhenglin, Zhu Jianbo, Zheng Weijie, et al. Research on electric vehicle charger based on fuzzy control [J]. Electronic design engineering, 2017, 25 (12): 85-88

[16] Wang Lu. Research on a multi-sensor management algorithm based on fuzzy control [J]. Electronic design engineering, 2018, 26 (21): 47-51

[17] Su Yang, Liu Wenyan, Liu Xiao, et al. Design of obstacle avoidance system for blind people based on fuzzy control [J]. Information network security, 2014 (6): 67-71


You must beto post a comment.