A supervision system and method for starting standard operation of a slag truck driver

Abstract

The present invention discloses a system and method for supervising the standardized operation of a muck truck driver when starting, comprising: an infrared camera is used to capture an image of whether the driver observes a small wide-angle round mirror during the starting operation, and transmit the image to a first memory; a CAN acquisition module is used to capture vehicle operation information and transmit the image to the first memory; the first memory is used to store the image of whether the driver observes a small wide-angle round mirror during the starting operation and the vehicle operation information, and transmit the image to a processor; the processor is used to process the image of whether the driver observes a small wide-angle round mirror during the starting operation and the vehicle operation information, obtain the number of vehicle starts and the safety score of the driver's starting operation, and transmit the image to a second memory; the second memory is used to store the number of vehicle starts and the safety score of the driver's starting operation. The present invention helps to cultivate the driver's safe and standardized starting operation habits by evaluating the driver's operating behavior when starting and proposing management suggestions.

Description

Technical Field

[0001] The invention belongs to the technical field of driver safety driving supervision, and in particular relates to a system and method for supervising the starting standard operation of a muck truck driver. Background Art

[0002] Due to the serious nature of traffic accidents caused by bad driving behavior of muck trucks, it is necessary to regulate the driving operation behavior of muck truck drivers. Muck truck drivers have a large blind spot. When starting, the driver must observe the driving environment in the blind spot of the front of the vehicle by looking at the wide-angle small round mirror on the upper right side of the vehicle. Therefore, the driver's behavior of observing the wide-angle small round mirror when starting is monitored to ensure that the driver can take measures when there are pedestrians, vehicles and other obstacles in the blind spot of the front of the vehicle. Educating and retraining drivers who are not accustomed to looking at the wide-angle small round mirror when starting can reduce accidents when muck trucks start. However, there is currently a lack of systems and methods for supervising the safe and standardized operations of muck truck drivers when starting. Summary of the invention

[0003] In view of the problems existing in the prior art, the purpose of the present invention is to provide a system and method for supervising the standardized operation of dump truck drivers when starting. By monitoring and recording the driver's operating behavior of whether he looks at the wide-angle small circular mirror when starting, the driver's operating behavior when starting is evaluated, scored and management suggestions are made, which helps to cultivate the driver's safe and standardized operating habits when starting, and is suitable for wide use.

[0004] In order to achieve the above object, the present invention adopts the following technical solutions:

[0005] (1) A system for monitoring the starting standard operation of a muck truck driver, comprising: an infrared camera, a CAN acquisition module, a first memory, a processor, and a second memory;

[0006] The infrared camera is installed facing the driver at a position on the upper right side of the vehicle near the wide-angle small round mirror, and is used to capture an image of whether the driver is observing the wide-angle small round mirror during the starting operation, and transmit the image to the first memory;

[0007] The CAN acquisition module is used to collect vehicle operation information and transmit it to the first memory;

[0008] Among them, the vehicle operation information includes the ignition signal, the accelerator pedal opening and the turn signal opening status;

[0009] The first memory is used to store whether the driver observes the image of the wide-angle small circular mirror and the vehicle operation information during the starting operation, and transmit it to the processor;

[0010] The processor is used to process whether the driver observes the image of the wide-angle small circular mirror and the vehicle operation information during the starting operation, obtain the number of vehicle starts and the safety score of the driver's starting operation, and transmit it to the second memory;

[0011] The second memory is used to store the number of vehicle starts and the safety score of the driver's starting operation. The second memory is provided with a USB interface, and the USB interface can be used to export the number of vehicle starts and the safety score of the driver's starting operation stored in the second memory.

[0012] (II) A method for supervising the driver of a muck truck to start a standardized operation, comprising the following steps:

[0013] Step 1: Formulate the starting operation specifications for muck trucks;

[0014] Step 2, obtaining a driver's head motion image, and calibrating the driver's head motion image using a convolutional neural network model to obtain a calibration result;

[0015] Step 3, collecting vehicle operation information, comparing the vehicle operation information and the calibration result with the starting operation specification of the muck truck, determining whether each driver's starting operation complies with the starting operation specification of the muck truck, if so, no points are deducted; otherwise, determining the point deduction value during each driver's starting operation;

[0016] Among them, the vehicle operation information includes the ignition signal, the accelerator pedal opening and the turn signal opening status;

[0017] Step 4, obtaining the total deduction value of the driver's starting process within a period of time, and determining whether to stop the driver's driving work by comparing the total deduction value G of the driver's starting process within a period of time with a preset threshold value T. If so, conduct special training for the driver according to the starting operation specification of the muck truck; otherwise, the driver's starting operation process complies with the starting operation specification of the muck truck;

[0018] Here, a period of time is one day, one month or one year.

[0019] Optionally, step 1 includes the following sub-steps:

[0020] Sub-step 1.1, start the engine of the muck truck, step on the clutch, and shift the gear lever to a low gear;

[0021] Sub-step 1.2, according to the actual situation, turn on the left turn signal or the right turn signal, honk the horn, release the parking brake operating handle, and release the clutch;

[0022] Sub-step 1.3: Hold the steering wheel and look at the wide-angle small round mirror once before pressing the accelerator pedal;

[0023] Sub-step 1.4: During the starting and turning process, you must look at the wide-angle small round mirror once every 4 seconds, and the time for each look at the wide-angle small round mirror must not be less than 0.5 seconds.

[0024] Optionally, the convolutional neural network model is used to calibrate the driver's head motion image as follows:

[0025] First, a training sample set and a test sample set are obtained, and a convolutional neural network model is trained by the training sample set to obtain a trained convolutional neural network model. Then, the trained convolutional neural network model is tested by the test sample set. When the average value of the difference between the output of the convolutional neural network model and the test sample label is less than a preset error threshold, it is determined that the training of the convolutional neural network model is completed.

[0026] The training sample set and the test sample set are selected from historical data respectively. The training sample set and the test sample set are respectively composed of the driver's head direction during the starting process, the driver looking at the wide-angle small round mirror during the starting operation, and the driver not looking at the wide-angle small round mirror during the starting operation. The labels of the test samples are the driver looking at the wide-angle small round mirror during the starting operation and the driver not looking at the wide-angle small round mirror during the starting operation.

[0027] Secondly, the driver's head direction during the starting process is input into the trained convolutional neural network model, and the trained convolutional neural network model outputs that the driver looks at the wide-angle small circular mirror during the starting process or the driver does not look at the wide-angle small circular mirror during the starting operation.

[0028] Optionally, the calibration result is specifically: the driver looks at the wide-angle small round mirror during the starting operation or the driver does not look at the wide-angle small round mirror during the starting operation.

[0029] Optionally, the deduction value during each driver's starting operation is determined as follows:

[0030] First, before the accelerator pedal opening data becomes non-zero, the driver of the muck truck turns on the turn signal when starting. If yes, no points are deducted; otherwise, the deduction value a=1 point;

[0031] Secondly, before the accelerator pedal opening becomes non-zero when the muck truck driver starts, whether the driver looks at the wide-angle small round mirror once, if yes, no points will be deducted; otherwise, the deduction value b = 2 points;

[0032] Again, after the driver of the muck truck starts turning, if the driver looks at the wide-angle small round mirror once every 4 seconds, no points will be deducted, otherwise, the deduction value c = 2 points; and,

[0033] After the dump truck driver starts turning, if the driver looks at the wide-angle small circular mirror for more than or equal to 0.5s each time, no points will be deducted. Otherwise, the deduction value is d = 2 points.

[0034] Optionally, the total deduction points of the driver during the starting process within a period of time are obtained, and by comparing the total deduction points of the driver during the starting process within a period of time with a preset threshold, determining whether to stop the driver's driving work is specifically as follows:

[0035] If the total deduction value G of the driver's starting process within a period of time is greater than the preset threshold value T, the driver's driving work is stopped and the driver is specially trained according to the starting operation specification of the muck truck; otherwise, the driver's starting operation process complies with the starting operation specification of the muck truck;

[0036] Among them, G=a+b+c+d.

[0037] Optionally, when the period is one day, the preset threshold T d is 9, if Gd>T d , stop the driver's work and provide special training to the driver according to the starting operation specifications of the muck truck. Otherwise, the starting operation process of the muck truck driver complies with the starting operation specifications of the muck truck.

[0038] Optionally, when the period is one month, the preset threshold T m is 50, if Gd>T m , stop the driver's work and provide special training to the driver according to the starting operation specifications of the muck truck. Otherwise, the starting operation process of the muck truck driver complies with the starting operation specifications of the muck truck.

[0039] Optionally, when the period is one year, the preset threshold T y is 480, if Gd>T y , stop the driver's work and provide special training to the driver according to the starting operation specifications of the muck truck. Otherwise, the starting operation process of the muck truck driver complies with the starting operation specifications of the muck truck.

[0040] The beneficial effects that the present invention can produce include:

[0041] 1) The present invention utilizes existing mature hardware systems such as infrared cameras, CAN acquisition modules, as well as processors and memories to form a driver operation supervision system when starting. The hardware composition of the starting standard operation supervision system for the driver of a muck truck is simple, and the behavior recognition is accurate, which helps to cultivate the driver's safe and standardized starting operation habits and is suitable for wide use.

[0042] 2) The standardized operation supervision system for starting a dump truck driver disclosed in the present invention mainly identifies the driver's operation of the turn signal when starting and whether he observes the wide-angle small circular mirror. It relies on image recognition to determine the driver's gaze area, and can accurately identify whether the driver observes the wide-angle small circular mirror, and can accurately identify the driver's operating behavior.

[0043] 3) The present invention manages drivers through a scoring system, scores driving operation behaviors that do not meet safe starting operation specifications, and retrains drivers with higher deduction values, which can correct drivers' unsafe driving operations at the start, and enable drivers to develop the habit of observing the wide-angle small circular mirror when starting, thereby improving drivers' starting safety awareness and improving the starting safety of vehicles.

[0044] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] 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, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying creative work.

[0046] Figure 1 It is a structural diagram of the starting standard operation supervision system for the driver of a muck truck disclosed in the present invention;

[0047] Figure 2 The invention discloses a flow chart of a method for supervising a muck truck driver's standardized starting operation.

[0048] Among them, 100: infrared camera; 200: CAN acquisition module; 300: first memory; 400: processor; 500: second memory. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical scheme and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. The test methods in the following embodiments that do not specify specific conditions are usually carried out under conventional conditions or under conditions recommended by various manufacturers.

[0050] refer to Figure 1 The present invention discloses a starting standard operation supervision system for a muck truck driver, comprising: an infrared camera 100, a CAN acquisition module 200, a first memory 300, a processor 400 and a second memory 500;

[0051] The infrared camera 100 is installed facing the driver at a position on the upper right side of the vehicle near the wide-angle small round mirror, and is used to capture an image of whether the driver is observing the wide-angle small round mirror during the starting operation, and transmit it to the first memory 300;

[0052] The CAN acquisition module 200 is used to collect vehicle operation information and transmit it to the first memory 300;

[0053] Among them, the vehicle operation information includes the ignition signal, the accelerator pedal opening and the turn signal opening status;

[0054] The first memory 300 is used to store whether the driver observes the image of the wide-angle small circular mirror and the vehicle operation information during the starting operation, and transmit it to the processor 400;

[0055] The processor 400 is used to process whether the driver observes the image of the wide-angle small circular mirror and the vehicle operation information during the starting operation, obtain the number of vehicle starts and the safety score of the driver's starting operation, and transmit it to the second memory 500;

[0056] The second memory 500 is used to store the number of vehicle starts and the safety score of the driver's starting operation. The second memory 500 is provided with a USB interface, and the USB interface can be used to export the number of vehicle starts and the safety score of the driver's starting operation stored in the second memory 500.

[0057] In the above embodiment, the output end of the infrared camera 100 and the output end of the CAN acquisition module 200 are respectively connected to the input end of the first memory 300, the output end of the first memory 300 is connected to the input end of the processor 400, and the output end of the processor 400 is connected to the input end of the second memory 500. The dump truck driver starting standard operation supervision system disclosed in the present invention utilizes existing mature hardware systems such as the infrared camera 100, the CAN module 200, and the processor 400 and the memory. Its hardware composition is simple and the behavior recognition is accurate, which helps to cultivate the driver's safe and standardized starting operation habits and is suitable for wide use.

[0058] refer to Figure 2 The present invention also discloses a method for supervising the starting standard operation of a muck truck driver, comprising the following steps:

[0059] Step 1: Formulate the starting operation specifications for muck trucks;

[0060] Among them, the starting operation specifications of the muck truck are as follows:

[0061] Sub-step 1.1, start the engine of the muck truck, step on the clutch, and shift the gear lever to a low gear;

[0062] Sub-step 1.2, according to the actual situation, turn on the left turn signal or the right turn signal, honk the horn, release the parking brake operating handle, and release the clutch;

[0063] Sub-step 1.3: Hold the steering wheel and look at the wide-angle small round mirror once before pressing the accelerator pedal;

[0064] Sub-step 1.4: During the start and turn process, the driver must look at the wide-angle small round mirror once every 4 seconds, and the time for each look at the wide-angle small round mirror shall not be less than 0.5 seconds;

[0065] Step 2, obtaining a driver's head motion image, and calibrating the driver's head motion image using a convolutional neural network model to obtain a calibration result;

[0066] The method of calibrating the driver's head motion image using the convolutional neural network model is as follows:

[0067] First, a training sample set and a test sample set are obtained, and a convolutional neural network model is trained by the training sample set to obtain a trained convolutional neural network model. Then, the trained convolutional neural network model is tested by the test sample set. When the average value of the difference between the output of the convolutional neural network model and the test sample label is less than a preset error threshold, it is determined that the training of the convolutional neural network model is completed.

[0068] The training sample set and the test sample set are selected from historical data respectively. The training sample set and the test sample set are respectively composed of the driver's head direction during the starting process, the driver looking at the wide-angle small round mirror during the starting operation, and the driver not looking at the wide-angle small round mirror during the starting operation. The labels of the test samples are the driver looking at the wide-angle small round mirror during the starting operation and the driver not looking at the wide-angle small round mirror during the starting operation.

[0069] Secondly, the driver's head direction during the starting process is input into the trained convolutional neural network model, and the trained convolutional neural network model outputs that the driver looks at the wide-angle small round mirror during the starting process or the driver does not look at the wide-angle small round mirror during the starting operation;

[0070] In the above embodiment, the calibration result is specifically: the driver looks at the wide-angle small round mirror during the starting operation or the driver does not look at the wide-angle small round mirror during the starting operation;

[0071] Step 3, collecting vehicle operation information, comparing the vehicle operation information and the calibration result with the starting operation specification of the muck truck, determining whether each driver's starting operation complies with the starting operation specification of the muck truck, if so, no points are deducted; otherwise, determining the point deduction value during each driver's starting operation;

[0072] Among them, the vehicle operation information includes the ignition signal, the accelerator pedal opening and the turn signal opening status;

[0073] In the above embodiment, the deduction value during each driver's starting operation is determined as follows:

[0074] First, before the accelerator pedal opening data becomes non-zero, the driver of the muck truck turns on the turn signal when starting. If yes, no points are deducted; otherwise, the deduction value a=1 point;

[0075] Secondly, before the accelerator pedal opening becomes non-zero when the muck truck driver starts, whether the driver looks at the wide-angle small round mirror once, if yes, no points will be deducted; otherwise, the deduction value b = 2 points;

[0076] Again, after the driver of the muck truck starts turning, if the driver looks at the wide-angle small round mirror once every 4 seconds, no points will be deducted, otherwise, the deduction value c = 2 points; and,

[0077] After the driver of the muck truck starts turning, if the driver looks at the wide-angle small round mirror for more than or equal to 0.5s each time, no points will be deducted. Otherwise, the deduction value is d = 2 points;

[0078] Step 4, obtaining the total deduction value of the driver's starting process within a period of time, and determining whether to stop the driver's driving work by comparing the total deduction value G of the driver's starting process within a period of time with a preset threshold value T. If so, conduct special training for the driver according to the starting operation specification of the muck truck; otherwise, the driver's starting operation process complies with the starting operation specification of the muck truck;

[0079] Here, a period of time is one day, one month or one year.

[0080] In the above embodiment, the total deduction value of the driver's starting process within a period of time is obtained, and by comparing the total deduction value of the driver's starting process within a period of time with a preset threshold, determining whether to stop the driver's driving work is specifically as follows:

[0081] If the total deduction value G of the driver's starting process within a period of time is greater than the preset threshold value T, the driver's driving work is stopped and the driver is specially trained according to the starting operation specification of the muck truck; otherwise, the driver's starting operation process complies with the starting operation specification of the muck truck;

[0082] Among them, G=a+b+c+d.

[0083] Furthermore, when the period of time is one day, the preset threshold T d is 9, if Gd>T d , stop the driver's work and conduct special training for the driver according to the starting operation specification of the muck truck, otherwise, the starting operation process of the muck truck driver complies with the starting operation specification of the muck truck;

[0084] When the period is one month, the preset threshold T m is 50, if Gd>T m, stop the driver's work and conduct special training for the driver according to the starting operation specification of the muck truck, otherwise, the starting operation process of the muck truck driver complies with the starting operation specification of the muck truck;

[0085] When the period is one year, the preset threshold T y is 480, if Gd>T y , stop the driver's work and provide special training to the driver according to the starting operation specifications of the muck truck. Otherwise, the starting operation process of the muck truck driver complies with the starting operation specifications of the muck truck.

[0086] The method for supervising the starting standard operation of a muck truck driver disclosed in the present invention manages the driver through a scoring system, scores the driving operation behaviors that do not meet the safe starting operation specifications, and retrains the drivers with higher deduction values, so as to correct the driver's unsafe driving operation at the start, and make the driver develop the habit of observing the wide-angle small circular mirror when starting, thereby improving the driver's starting safety awareness and improving the starting safety of the vehicle.

[0087] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for supervising the starting standard operation of a muck truck driver, characterized in that: The following steps are involved: Step 1: Formulate the starting operation specifications for muck trucks; The step 1 comprises the following sub-steps: Sub-step 1.1, start the engine of the muck truck, step on the clutch, and shift the gear lever to a low gear; Sub-step 1.2, according to the actual situation, turn on the left turn signal or the right turn signal, honk the horn, release the parking brake operating handle, and release the clutch; Sub-step 1.3: Hold the steering wheel and look at the wide-angle small round mirror once before pressing the accelerator pedal; Sub-step 1.4: During the start and turn process, the driver must look at the wide-angle small round mirror once every 4 seconds, and the time for each look at the wide-angle small round mirror shall not be less than 0.5 seconds; Step 2: obtaining a driver's head motion image, and calibrating the driver's head motion image using a convolutional neural network model to obtain a calibration result. The convolutional neural network model is used to calibrate the driver's head motion image as follows: First, a training sample set and a test sample set are obtained, and a convolutional neural network model is trained by the training sample set to obtain a trained convolutional neural network model. Then, the trained convolutional neural network model is tested by the test sample set. When the average value of the difference between the output of the convolutional neural network model and the test sample label is less than a preset error threshold, it is determined that the training of the convolutional neural network model is completed. The training sample set and the test sample set are selected from historical data respectively. The training sample set and the test sample set are respectively composed of the driver's head direction during the starting process, the driver looking at the wide-angle small round mirror during the starting operation, and the driver not looking at the wide-angle small round mirror during the starting operation. The labels of the test samples are the driver looking at the wide-angle small round mirror during the starting operation and the driver not looking at the wide-angle small round mirror during the starting operation. Secondly, the driver's head direction during the start process is input into the trained convolutional neural network model, and the trained convolutional neural network model outputs whether the driver looks at the wide-angle small round mirror during the start process or the driver does not look at the wide-angle small round mirror during the start operation. The calibration result is specifically: the driver looks at the wide-angle small round mirror during the starting operation or the driver does not look at the wide-angle small round mirror during the starting operation; Step 3, collecting vehicle operation information, comparing the vehicle operation information and the calibration result with the starting operation specification of the muck truck, determining whether each driver's starting operation complies with the starting operation specification of the muck truck, if so, no points are deducted; otherwise, determining the point deduction value during each driver's starting operation; Among them, the vehicle operation information includes the ignition signal, the accelerator pedal opening and the turn signal opening status; Step 4, obtaining the total deduction value of the driver's starting process within a period of time, and determining whether to stop the driver's driving work by comparing the total deduction value G of the driver's starting process within a period of time with a preset threshold value T. If so, conduct special training for the driver according to the starting operation specification of the muck truck; otherwise, the driver's starting operation process complies with the starting operation specification of the muck truck; Here, a period of time is one day, one month or one year.

2. The method for supervising the starting standard operation of a muck truck driver according to claim 1 is characterized in that: The deduction points for each driver's starting operation are determined as follows: First, before the accelerator pedal opening data becomes non-zero, the driver of the muck truck turns on the turn signal when starting. If so, no points are deducted; otherwise, the deduction value a=1 point; Secondly, before the accelerator pedal opening becomes non-zero when the muck truck driver starts, whether the driver looks at the wide-angle small round mirror once, if so, no points will be deducted; Otherwise, the deduction value is b=2 points; Again, after the driver of the muck truck starts turning, if the driver looks at the wide-angle small round mirror once every 4 seconds, no points will be deducted, otherwise, the deduction value is c=2 points; and, After the dump truck driver starts turning, if the driver looks at the wide-angle small circular mirror for more than or equal to 0.5s each time, no points will be deducted. Otherwise, the deduction value is d=2 points.

3. The method for supervising the starting standard operation of a muck truck driver according to claim 2 is characterized in that: Obtain the total deduction points of the driver during the start-up process within a period of time, and determine whether to stop the driver's driving work by comparing the total deduction points of the driver during the start-up process within a period of time with a preset threshold value: If the total deduction value G of the driver's starting process within a period of time is greater than the preset threshold value T, the driver's driving work is stopped and the driver is specially trained according to the starting operation specification of the muck truck; otherwise, the driver's starting operation process complies with the starting operation specification of the muck truck; Among them, G=a+b+c+d.

4. The method for supervising the starting standard operation of a muck truck driver according to claim 1, characterized in that: When the period of time is one day, the preset threshold value T is 9. If G>T, the driver's work is stopped and the driver is specially trained according to the starting operation specification of the muck truck. Otherwise, the starting operation process of the muck truck driver complies with the starting operation specification of the muck truck.

5. The method for supervising the starting standard operation of a muck truck driver according to claim 1 is characterized in that: When the period of time is one month, the preset threshold T is 50. If G>T, the driver's work is stopped and the driver is specially trained according to the starting operation specification of the muck truck. Otherwise, the starting operation process of the muck truck driver complies with the starting operation specification of the muck truck.

6. The method for supervising the starting standard operation of a muck truck driver according to claim 1, characterized in that: When the period of time is one year, the preset threshold T is 480. If G>T, the driver's work is stopped and the driver is specially trained according to the starting operation specification of the muck truck. Otherwise, the starting operation process of the muck truck driver complies with the starting operation specification of the muck truck.

Images