CN107176098B - An automatic monitoring and early warning device and control method for inner wheel differential blind zone - Google Patents

Abstract

The invention discloses an automatic monitoring and early warning device for an inner wheel difference blind area, which comprises: the driving motor is arranged above the front wheel and the rear wheel on the right side of the vehicle body and is provided with a transmission mechanism; the rear end of the sensor arm is connected with the transmission mechanism, and the front end of the sensor arm is hinged with the surface of the vehicle body; the infrared light sensor is used for detecting the infrared light, which is fixedly connected with the front end of the sensor arm; and the microcontroller is connected with the driving motor through a control circuit. The automatic monitoring and early warning device for the dead zone of the inner wheel difference can monitor the dead zone information of the inner wheel difference in real time when a vehicle turns rightwards and send early warning information when danger is found. Meanwhile, the invention also provides a control method based on the BP neural network, which can adjust the angle of the infrared sensor according to different dead zones of the inner wheel difference of each right turn of the vehicle, so that the acquisition range of the infrared sensor is covered to the whole dead zone of the inner wheel difference, and more accurate early warning information is provided.

Description

An automatic monitoring and early warning device and control method for inner wheel differential blind zone

technical field

The invention belongs to the technical field of automotive electronics, and in particular relates to an automatic monitoring and early warning device and control method for an inner wheel differential blind zone.

Background technique

In recent years, with the rapid development of our country's economy, our country has become a veritable automobile power. However, the increase of automobiles has also brought more traffic accidents, among which the accident of large vehicles is one of the main causes of major traffic accidents. The inner wheel difference refers to the difference between the turning radius of the front inner wheel and the turning radius of the rear inner wheel when the vehicle is turning. Due to the existence of the inner wheel difference, when the vehicle is turning, the trajectories of the front and rear wheels do not coincide. If you only pay attention to the passage of the front wheels and forget the inner wheel difference during driving, it may cause accidents where the rear inner wheels drive off the road or collide with other objects.

Due to the high body of large trucks, the driver sits on the left driver's seat. When the vehicle turns right, it is easy to have a visual blind spot. However, only the rearview mirror is used to observe the right rear, and the visual range is limited. The vehicle body is easy to drive at high speed with the inner lane Due to the large inner wheel difference in the right steering of the truck, and sometimes the driver's misjudgment and negligence, the truck is very likely to collide with the non-motorized vehicle. Therefore, when the vehicle turns right, it is very necessary to give the driver an early warning of the inner wheel difference blind zone. Most of the existing inner wheel difference early warning devices use cameras, sensors, etc. to collect information on whether there are obstacles such as vehicles and pedestrians in the inner wheel difference blind area, and remind the driver. However, due to the different steering angles of the vehicle, the position of the blind area of the inner wheel difference will also change accordingly. Because the radiation range of the sensor is limited, if the sensor is installed in a fixed position, some positions will not be able to radiate, resulting in deviations in the collected information. In addition, the theoretical inner wheel differential deduced by the geometric model of the inner wheel differential will have certain errors. Therefore, it is very necessary to control the angle of the infrared sensor in real time to cover the entire blind area of the inner wheel differential.

Contents of the invention

The purpose of the present invention is to provide an automatic monitoring and early warning device for the inner wheel differential blind area, which can monitor the information of the inner wheel differential blind area in real time when the vehicle turns right, and send out early warning information when danger is found.

Another object of the present invention is to provide a control method based on BP neural network, which can adjust the angle of the infrared sensor according to the difference in the blind area of the inner wheel difference of each right turn of the vehicle, so that the acquisition range of the infrared sensor can cover the entire inner wheel Poor blind area to provide more accurate early warning information.

The technical scheme provided by the invention is:

Steering wheel angle sensor;

Infrared photoelectric sensor rotation mechanism, which includes:

A drive motor, which is arranged above the front wheel and the rear wheel on the right side of the vehicle body, and the drive motor is provided with a transmission mechanism;

A sensor arm, the rear end of which is connected to the transmission mechanism, and the front end is hinged to the surface of the vehicle body;

An infrared photoelectric sensor, which is fixedly connected to the front end of the sensor arm;

A microcontroller is connected to the steering wheel angle sensor and to the drive motor through a control circuit.

Preferably, the transmission mechanism adopts a rack and pinion structure.

Preferably, the rear end of the sensor arm is fixedly connected to the rack of the transmission mechanism.

Preferably, the automatic monitoring and early warning device for the inner wheel difference blind zone also includes an alarm module, which is arranged close to the instrument panel and connected to the microprocessor, and the alarm module includes:

Chinese liquid crystal display module, which uses a 12864 liquid crystal display with a Chinese character library;

Voice warning module, which adopts ISD0 series recording and playback chip;

The photoelectric warning module uses a bright light-emitting diode as a warning light.

Preferably, the infrared photoelectric sensor is a reflective photoelectric sensor.

Preferably, said microcontroller adopts STM32F103RCT6 single-chip microcomputer.

Preferably, the automatic monitoring and early warning device for inner wheel differential blind zone also includes a power supply module, which adopts AMS1117-5.0 and L6932-3.3 chips to stabilize two voltages required by the system, 5V and 3.3V.

A control method for an automatic monitoring and early warning device for an inner wheel differential blind zone, characterized in that when the vehicle turns right, according to the angle information transmitted by the steering wheel angle sensor, the minimum turning radius of the vehicle, the front and rear wheelbase of the vehicle, and the vehicle wheelbase are based on BP neural network Adjusting the angle of the infrared photoelectric sensors on the front and rear sides of the vehicle includes the following steps:

Step 1. Normalize the front and rear wheelbase l, vehicle wheelbase s, vehicle minimum turning radius r, and steering wheel sensor steering angle θ in sequence, and determine the input layer vector x of the three-layer BP neural network = {x ₁ ,x ₂ , x ₃ , x ₄ }; Among them, x ₁ is the front and rear wheelbase coefficient of the vehicle, x ₂ is the vehicle wheelbase coefficient, x ₃ is the minimum turning radius coefficient of the vehicle, and x ₄ is the steering angle coefficient of the steering wheel sensor;

Step 2, the input layer vector is mapped to the intermediate layer, the intermediate layer vector y={y ₁ , y ₂ ,...,y _m }; m is the number of intermediate layer nodes;

Step 3. Obtain the output layer vector z={z ₁ , z ₂ }; wherein, z ₁ is the angle adjustment coefficient of the infrared photoelectric sensor on the front side of the vehicle, and z ₂ is the angle adjustment coefficient of the infrared photoelectric sensor on the rear side of the vehicle, so that

α _i+1 = z ₁ ⁱ α _max ,

β _i+1 = z ₂ ⁱ β _max ,

Among them, z ₁ ⁱ and z ₂ ⁱ are the output layer vector parameters of the i-th sampling period respectively, α _max and β _max are the maximum angle of the infrared photoelectric sensor on the front side of the vehicle and the maximum angle of the infrared photoelectric sensor on the rear side of the vehicle respectively , α _i+1 , β _i+1 are respectively the angle of the infrared photoelectric sensor on the front side of the vehicle and the angle of the infrared photoelectric sensor on the rear side of the vehicle at the i+1th sampling period; and

In the first step, the front and rear wheelbase l of the vehicle, the wheelbase s of the vehicle, the minimum turning radius r of the vehicle, and the steering wheel sensor steering angle θ are normalized and the normalization formula is:

Among them, x _j is the parameter in the input layer vector, X _j is the parameter l, s, r, θ respectively, j=1, 2, 3, 4; X _jmax and X _jmin are the maximum and minimum values of the corresponding parameters respectively .

Preferably, in the step three, in the initial running state, the angle of the infrared photoelectric sensor on the front side of the vehicle and the angle of the infrared photoelectric sensor on the rear side of the vehicle meet the empirical values:

α ₀ =0.85α _max ,

β ₀ =0.67β _max ,

Among them, α ₀ and β ₀ are the initial angle of the infrared photoelectric sensor on the front side of _the vehicle and the initial angle of _the infrared photoelectric sensor on the rear side of the vehicle respectively; The maximum angle of the rear infrared photoelectric sensor.

The beneficial effects of the present invention are: the automatic monitoring and early warning device for the inner wheel differential blind area provided by the present invention can monitor the information of the inner wheel differential blind area in real time when the vehicle turns right, and issue early warning information when danger is found. At the same time, the present invention also provides a control method based on BP neural network, which can adjust the angle of the infrared sensor according to the difference in the blind area of the inner wheel difference each time the vehicle turns right, so that the collection range of the infrared sensor covers the entire inner wheel difference Blind spots to provide more accurate early warning information.

Description of drawings

Fig. 1 is a schematic diagram of the infrared photoelectric sensor rotating mechanism of the automatic monitoring and early warning device for inner wheel differential blind area according to the present invention.

Fig. 2 is a schematic diagram of module connection of the automatic monitoring and early warning device for inner wheel differential blind zone according to the present invention.

Fig. 3 is a schematic diagram of the working flow of the automatic monitoring and early warning device for inner wheel differential blind area according to the present invention.

FIG. 4 is a schematic diagram of the pin distribution of the microcontroller according to the present invention.

Fig. 5 is the overall structure of the automatic monitoring and early warning device for inner wheel difference blind zone according to the present invention.

Detailed ways

The present invention will be further described in detail below in conjunction with the accompanying drawings, so that those skilled in the art can implement it with reference to the description.

As shown in Figures 1-4, the present invention provides an automatic monitoring and early warning device for the inner wheel differential blind zone, which can adjust the angle of the infrared sensor according to the different steering angles of the vehicle. At the same time, the early warning information collected by the infrared sensor is displayed on the Chinese LCD screen, and the voice and light flashing warning are given, and when an accident occurs when the vehicle turns, the user's geographical location and time information will be sent in the form of a short message via the GSM network. Go to the traffic police monitoring center. The automatic monitoring and early warning device for the inner wheel differential blind zone includes a steering wheel angle sensor, which is connected to a microcontroller 110, and the steering wheel angle sensor transmits the steering wheel steering angle signal to the microcontroller 110 when the vehicle turns; the microcontroller 110 adopts the STM32F103RCT6 type microcontroller. Drive motor 120, it is arranged on vehicle body right side front wheel, rear wheel top and is connected with microcontroller 110 by control circuit, drive motor 120 is provided with transmission mechanism, and described transmission mechanism comprises column gear 121 and rack 122; The rear end of the arm 130 is fixedly connected to the rack 122 , and the front end is hinged to the surface of the vehicle body; the infrared photoelectric sensor 140 is fixedly connected to the front end of the sensor arm 130 . When the vehicle is turning right, the steering wheel angle sensor transmits the steering wheel angle signal to the microcontroller 110, and the microcontroller 110 sends a signal to the drive motor control circuit to start and control the drive motor 120 to drive the teeth meshed with the cylindrical gear 121. The bar 122 moves laterally, thereby driving the front end of the sensor arm 130 fastened to the rack 122 to rotate around the base point of its hinge with the surface of the vehicle body, thereby driving the infrared photoelectric sensor 140 to rotate around the base point. The sensor arm 130 is a retractable structure. When the infrared photoelectric sensor 140 rotates at a larger angle, the sensor arm 130 is extended accordingly. When the infrared photoelectric sensor 140 rotates at a smaller angle, the sensor arm 130 is shortened.

The infrared photoelectric sensor 140 is mainly used to generate sensing signals to obstacles (pedestrians, vehicles, etc.) in the "blind field of vision" area when the truck is turning, and then send the signals to the microcontroller 110 for judgment processing. The infrared photoelectric sensor 140 adopts a reflective photoelectric sensor, which has a pair of infrared emitting and receiving tubes. The transmitting tube emits infrared rays of a certain frequency. When the detection direction encounters an obstacle (reflecting surface), the infrared rays are reflected back and received by the receiving tube. After being processed by the comparator circuit, the green indicator light will light up, and at the same time, the signal output interface outputs a digital signal (a low-level signal). The data pin of the infrared photocell on the front side of the truck is connected to the PC0 pin of the microcontroller, and is set as a jump edge external interrupt; the data pin of the infrared photocell on the rear side of the truck is connected to the PC1 pin of the microcontroller, and is set to jump Along with the external interrupt, when there are pedestrians, vehicles and other obstacles within the detection range of the sensor, a jump signal can be output to trigger the microcontroller 110 to receive sensor data.

The microcontroller 110 disconnects the Chinese LCD display 150 , and the microcontroller 110 calculates and processes the signal collected and reflected by the infrared electrical sensor 140 , and transmits the alarm information to the Chinese LCD 150 . The Chinese liquid crystal display module 150 adopts a 12864 liquid crystal display screen with a Chinese character library, which can display the safety situation of the blind area of the front side of the truck and the safety situation of the blind area of the rear side of the truck. If the truck enters the blind zone of the inner wheel difference from obstacles such as pedestrians during the turning process, the system will display an alarm message on the LCD screen.

Voice warning module 160 adopts ISD0 series recording and playback chip. It can record in multiple segments, the sampling rate can be adjusted between 4K and 12K, and the power supply range can be between 2.4V and 5.5V. The microcontroller 110 disconnects the data pin of the voice warning module 160, and when pedestrians are detected in the blind area of the truck’s turning field of vision, the system can prompt the driver with an on-site voice warning: “Attention, the truck is turning dangerously!”

The photoelectric warning module 170 is connected to the microcontroller 110, which uses a bright LED as a warning light, and further reminds the driver that turning at this time is dangerous by photoelectric flashing.

The power supply module 180 adopts AMS1117-5.0 and L6932-3.3 chips to stabilize the required voltages of 5V and 3.3V.

The automatic monitoring and early warning device for the inner wheel difference blind area also includes a GMS automatic alarm module 190 based on GPS positioning, which uses a SIM300GSM module as a wireless remote communication interface device. The module integrates a standard SIM card holder, which can facilitate users to use the SIM card to access the network, and use the AT command set to control the GSM module to communicate with the owner's mobile phone through a computer or a single-chip microcomputer. The microcontroller 110 uses the serial port to drive the GSM automatic alarm module 190 . The GMS automatic alarm module 190 is connected to the vibration sensor installed in the truck, which outputs a digital signal through a NOT gate. When there is no vibration signal, the sensor is turned on, the input terminal of the non-gate is high level, and reversed after the TTL non-gate, so the output terminal is low level, when there is a vibration signal, the sensor is cut off, and the input terminal of the non-gate is low level Level, the output end is high level after being reversed, and the longer the vibration time, the sensor cut-off time will also increase. The vibration sensor is connected with the PA2 interface of the single-chip microcomputer, and the data is input into the microcontroller 110 . The P1 port of the microcontroller 110 is used as the output terminal of the alarm signal, because the P1 port is through the I/O bidirectional static interface and has an output latch function, which is convenient for realizing alarm control through software.

When the vibration monitor does not output a digital pulse signal, the microcontroller 110 has been looping in the main program. When the vibration monitor outputs a digital pulse signal, the microcontroller 110 program jumps to the interrupt subroutine EXT0 to execute, triggering the GPS positioning system to detect the alarm position, and the current position information output by the GPS positioning system is sent to The traffic department, call the police.

The automatic monitoring and early warning device for the inner wheel difference blind zone also includes a serial port debugging module circuit 210. Through the serial port, variables that need to be observed in the software program, GSM module return results, execution results, etc. can be printed on the upper computer software, so that it can be more fully Understand the software execution process to speed up software writing and stabilization.

As shown in Figure 5, the present invention provides an automatic monitoring and early warning device for the inner wheel difference blind zone, with a microcontroller (STM32 single-chip microcomputer) as the core, and using an infrared photoelectric sensor to collect front-end signals in real time. An infrared photoelectric sensor is installed at the front and rear positions on the right side. When a car or pedestrian enters the driver's "blind field of vision" area caused by the inner wheel difference, the two infrared photoelectric sensors collect the reflected signal and send it to the microcontroller for calculation. ,deal with. The alarm information will be displayed on the Chinese LCD screen, and the voice and light flashing warning will be given to remind the driver to make an early judgment on the dangerous situation. If there is an accident when the vehicle turns, the vibration sensor installed on the vehicle will transmit the collision signal to the microcontroller, and the GPS module receives the satellite signal from the GPS antenna unit, and obtains the user's geographic location and time information through processing, and sends it to the GSM The module is transmitted by the GSM antenna and sends information such as the user's geographic location and time to the traffic police monitoring center in the form of short messages through the GSM wireless communication network.

The present invention also provides a control method for the automatic monitoring and early warning device of the inner wheel differential blind zone. When the vehicle turns right, according to the angle information transmitted by the steering wheel angle sensor, the minimum turning radius of the vehicle, the front and rear wheelbase of the vehicle, and the vehicle wheelbase are based on the BP neural network. The network regulates the angle of the infrared photoelectric sensors on the front and rear sides of the vehicle, which specifically includes the following steps:

Step 1, establishing a BP neural network model;

The BP network architecture adopted by the present invention is composed of three layers. The first layer is the input layer, with n nodes in total, corresponding to n detection signals representing the working status of the equipment, and these signal parameters are given by the data preprocessing module. The second layer is the hidden layer, with a total of m nodes, which is determined in an adaptive manner by the training process of the network. The third layer is the output layer, with a total of p nodes, determined by the actual output response of the system.

The mathematical model of the network is:

Input layer vector: x=(x ₁ ,x ₂ ,…,x _n ) ^T

Middle layer vector: y=(y ₁ ,y ₂ ,…,y _m ) ^T

Output layer vector: z=(z ₁ ,z ₂ ,…,z _p ) ^T

In the present invention, the number of input layer nodes is n=4, and the number of output layer nodes is p=2. The number of hidden layer nodes m is estimated by the following formula:

According to the sampling period, the input 4 parameters are, x ₁ is the front and rear wheelbase coefficient of the vehicle, x ₂ is the vehicle wheelbase coefficient, x ₃ is the minimum turning radius coefficient of the vehicle, and x ₄ is the steering angle coefficient of the steering wheel sensor;

The four factors that determine the blind zone of vehicle inner wheel differential belong to different physical quantities, and their dimensions are different. Therefore, before the data is fed into the neural network, the data needs to be normalized to a number between 0-1.

Specifically, after normalizing the front and rear track l of the vehicle, the front and rear track coefficient x ₁ of the vehicle is obtained:

Among them, _lmin and _lmax are the minimum and maximum values of the front and rear wheelbases of the vehicle, respectively.

Similarly, after normalizing the vehicle wheelbase s, the vehicle wheelbase coefficient x ₂ is obtained:

Among them, s _min and s _max are the minimum and maximum values of the vehicle wheelbase, respectively.

After normalizing the minimum turning radius r of the vehicle, the minimum turning radius coefficient x ₃ of the vehicle is obtained:

Among them, r _min and r _max are the minimum and maximum values of the minimum turning radius of the vehicle, respectively.

After normalizing the steering wheel sensor steering angle θ, the steering wheel sensor steering angle coefficient x ₄ is obtained:

Among them, θ _min and θ _max are the minimum and maximum steering angles of the steering wheel sensor, respectively.

The two parameters of the output signal are respectively expressed as: z ₁ is the angle adjustment coefficient of the infrared photoelectric sensor on the front side of the vehicle, and z ₂ is the angle adjustment coefficient of the infrared photoelectric sensor on the rear side of the vehicle;

The angle adjustment coefficient z ₁ of the infrared photoelectric sensor on the front side of the vehicle is expressed as the ratio of the angle of the infrared photoelectric sensor on the front side of the vehicle in the next sampling period to the maximum angle set in the current sampling period, that is, in the ith sampling period, the collected The angle is α _i , after outputting the angle adjustment coefficient z ₁ ⁱ of the ith sampling period through the BP neural network, the angle in the i+1 sampling period is controlled to be α _i+1 so that it satisfies α _i+1 = z ₁ ⁱ α _max ;

The angle adjustment coefficient z ₂ of the infrared photoelectric sensor on the rear side of the vehicle is expressed as the ratio of the angle of the infrared photoelectric sensor on the rear side of the vehicle in the next sampling period to the maximum angle set in the current sampling period, that is, in the ith sampling period, the collected The nozzle angle of is β _i , after the nozzle angle adjustment coefficient z ₂ ⁱ of the i sampling period is output through the BP neural network, the nozzle angle in the i+1 sampling period is controlled to be β _i+1 so that it satisfies β _{i+ 1} = z ₂ ⁱ β _max ;

Step 2: Carry out the training of BP neural network.

After the BP neural network node model is established, the BP neural network can be trained. Obtain training samples according to the empirical data of the product, and given the connection weight w _ij between input node i and hidden layer node j, the connection weight w _jk between hidden layer node j and output layer node k, implicit The threshold θ _j of layer node j, and the thresholds w _ij , w _jk , θ _j , θ _k of output layer node k are all random numbers between -1 and 1.

During the training process, the values of w _ij and w _jk are constantly revised until the system error is less than or equal to the expected error, and the training process of the neural network is completed.

As shown in Table 1, a set of training samples and the values of each node in the training process are given.

Table 1 The values of each node in the training process

Step 3, collecting data and operating parameters into the neural network to obtain the regulation coefficient;

The trained artificial neural network is solidified in the chip, enabling the hardware circuit to have prediction and intelligent decision-making functions, thus forming intelligent hardware. After the intelligent hardware is powered on, the angle of the infrared photoelectric sensor on the front side of the vehicle is α ₀ =0.85α _max , and the angle of the infrared photoelectric sensor on the rear side of the vehicle is β ₀ =0.67β _max ;

通过BP神经网络的运算得到初始输出向量/>

At the same time, normalize the initial data of vehicle front and rear wheelbase l, vehicle wheelbase s, vehicle minimum turning radius r, and steering wheel sensor steering angle θ to obtain the initial input vector of BP neural network

The initial output vector is obtained through the operation of the BP neural network />

后，即可调车辆前侧红外光电传感器的角度及车辆后侧红外光电传感器的角度，车辆前侧红外光电传感器的角度及车辆后侧红外光电传感器的角度分别为：Step 4: Get the initial output vector

After that, the angle of the infrared photoelectric sensor on the front side of the vehicle and the angle of the infrared photoelectric sensor on the rear side of the vehicle can be adjusted. The angles of the infrared photoelectric sensor on the front side of the vehicle and the infrared photoelectric sensor on the rear side of the vehicle are respectively:

α ₁ =z ₁ ⁰ α _max

β ₁ = z ₂ ⁰ β _max

Obtain the front and rear wheelbase l, vehicle wheelbase s, vehicle minimum turning radius r, and steering wheel sensor steering angle θ in the i-th sampling period through the sensor, and obtain the input vector x ⁱ of the i-th sampling period through normalization =( x ₁ ⁱ , x ₂ ⁱ , x ₃ ⁱ , x ₄ ⁱ ), the output vector z ⁱ = (z ₁ ⁱ , z ₂ ⁱ ) of the i-th sampling period is obtained through the operation of the BP neural network, and then control and adjust the vehicle front The angle of the side infrared photoelectric sensor and the angle of the vehicle rear infrared photoelectric sensor, so that the angle of the vehicle front infrared photoelectric sensor and the vehicle rear infrared photoelectric sensor angle are respectively:

α _i+1 = z ₁ ⁱ α _max ,

β _i+1 = z ₂ ⁱ β _max ,

Through the above settings, the BP neural network algorithm is used to regulate the angle of the infrared photoelectric sensor on the front side of the vehicle and the angle of the infrared photoelectric sensor on the rear side of the vehicle to achieve the best operating state, thereby improving the accuracy of early warning information.

Although the embodiment of the present invention has been disclosed as above, it is not limited to the use listed in the specification and implementation, it can be applied to various fields suitable for the present invention, and it can be easily understood by those skilled in the art Therefore, the invention is not limited to the specific details and examples shown and described herein without departing from the general concept defined by the claims and their equivalents.

Claims (7)

1. An automatic monitoring and early warning device for inner wheel differential blind zone, comprising: Steering wheel angle sensor; Infrared photoelectric sensor rotation mechanism, which includes: A drive motor, which is arranged above the front wheel and the rear wheel on the right side of the vehicle body, and the drive motor is provided with a transmission mechanism; A sensor arm, the rear end of which is connected to the transmission mechanism, and the front end is hinged to the surface of the vehicle body; An infrared photoelectric sensor, which is fixedly connected to the front end of the sensor arm; A microcontroller, which is connected to the steering wheel angle sensor and connected to the drive motor through a control circuit; The control method includes a control method for an automatic monitoring and early warning device for an inner wheel differential blind zone, which is characterized in that when the vehicle turns right, according to the angle information transmitted by the steering wheel angle sensor, the minimum turning radius of the vehicle, the front and rear wheelbase of the vehicle, and the vehicle wheelbase are based on the BP neural network. The network regulates the angle of the infrared photoelectric sensors on the front and rear sides of the vehicle, including the following steps: Step 1. Normalize the front and rear wheelbase l, vehicle wheelbase s, vehicle minimum turning radius r, and steering wheel sensor steering angle θ in sequence, and determine the input layer vector x of the three-layer BP neural network = {x ₁ ,x ₂ , x ₃ , x ₄ }; Among them, x ₁ is the front and rear wheelbase coefficient of the vehicle, x ₂ is the vehicle wheelbase coefficient, x ₃ is the minimum turning radius coefficient of the vehicle, and x ₄ is the steering angle coefficient of the steering wheel sensor; Step 2, the input layer vector is mapped to the middle layer, the middle layer vector y={y ₁ , y ₂ ,...,y _m }; m is the number of middle layer nodes; Step 3. Obtain the output layer vector z={z ₁ ,z ₂ }; wherein, z ₁ is the angle adjustment coefficient of the infrared photoelectric sensor on the front side of the vehicle, and z ₂ is the angle adjustment coefficient of the infrared photoelectric sensor on the rear side of the vehicle, so that α _i+1 = z ₁ ⁱ α _max , β _i+1 = z ₂ ⁱ β _max , Among them, z ₁ ⁱ and z ₂ ⁱ are the output layer vector parameters of the i-th sampling period respectively, α _max and β _max are the maximum angle of the infrared photoelectric sensor on the front side of the vehicle and the maximum angle of the infrared photoelectric sensor on the rear side of the vehicle respectively , α _i+1 , β _i+1 are respectively the angle of the infrared photoelectric sensor on the front side of the vehicle and the angle of the infrared photoelectric sensor on the rear side of the vehicle at the i+1th sampling period; and In the first step, the front and rear wheelbase l of the vehicle, the wheelbase s of the vehicle, the minimum turning radius r of the vehicle, and the steering wheel sensor steering angle θ are normalized and the normalization formula is:

Among them, x _j is the parameter in the input layer vector, X _j is the parameter l, s, r, θ respectively, j=1, 2, 3, 4; X _jmax and X _jmin are the maximum and minimum values of the corresponding parameters respectively ; In the third step, in the initial running state, the angle of the infrared photoelectric sensor on the front side of the vehicle and the angle of the infrared photoelectric sensor on the rear side of the vehicle meet the empirical values: α ₀ =0.85α _max , β ₀ =0.67β _max , Among them, α ₀ and β ₀ are the initial angle of the infrared photoelectric sensor on the front side of _the vehicle and the initial angle of the photoelectric infrared sensor _on the rear side of the vehicle respectively; The maximum angle of the rear infrared photoelectric sensor. 2. The automatic monitoring and early warning device for inner wheel differential blind area according to claim 1, characterized in that, the transmission mechanism adopts a rack and pinion structure. 3. The automatic monitoring and early warning device for inner wheel differential blind zone according to claim 1, characterized in that the rear end of the sensor arm is fixedly connected with the rack of the transmission mechanism. 4. The automatic monitoring and early warning device for inner wheel differential blind area according to claim 1, characterized in that, it also includes an alarm module, which is arranged near the instrument panel and connected to the microcontroller, and the alarm module includes: Chinese liquid crystal display module, which uses a 12864 liquid crystal display with a Chinese character library; Voice warning module, which uses ISD0 series recording and playback chip; The photoelectric warning module uses a bright light-emitting diode as a warning light. 5. The automatic monitoring and early warning device for inner wheel differential blind area according to claim 1, characterized in that the infrared photoelectric sensor is a reflective photoelectric sensor. 6. The automatic monitoring and early warning device for inner wheel differential blind area according to claim 1, characterized in that, the microcontroller adopts STM32F103RCT6 single-chip microcomputer. 7. The automatic monitoring and early warning device for inner wheel differential blind zone according to claim 1, characterized in that it also includes a power supply module, which adopts AMS1117-5.0 and L6932-3.3 chips to stabilize the voltage of 5V and 3.3V. required voltage.

Images