Intelligent mining electric shovel autonomous rotation bucket-opening discharging method and intelligent mining electric shovel autonomous rotation bucket-opening discharging system

Abstract

The invention discloses an intelligent mining electric shovel autonomous rotation bucket opening discharging method and system. The autonomous rotation bucket opening and discharging system comprises a positioning and attitude determining system and a control decision making system. The positioning and attitude determining system comprises a positioning module, a coordinate system unifying module and a discharging position calculating module. The control decision-making system comprises a driving data calculation module and an equipment control center; the positioning module comprises electric shovel positioning equipment and mine card positioning equipment. Therefore, according to the position and posture information of the electric shovel and the mine truck calculated by the positioning and posture determining system, the rotating angle of the upper portion of the electric shovel and the stopping position of the bucket are accurately calculated, and accurate rotating and positioning of the upper portion of the electric shovel can be achieved. And when the bucket reaches a preset position above the mine truck, the system starts the bucket opening action, so that the unloading efficiency and safety are improved.

Description

Technical Field

[0001] The invention relates to a method and system for autonomously rotating bucket opening and unloading of intelligent mining electric shovels. Background Art

[0002] At present, mining electric shovels rely on manual operation and lack autonomous capabilities. Since the national smart mine construction was proposed, higher requirements have been put forward for the intelligent and unmanned level of open-pit mining equipment, and promoting the intelligent development of electric shovels has become an inevitable trend. Electric shovels are often in extremely harsh working environments such as high temperature, extreme cold, plateau, strong vibration, high dust, and face weather such as rain, snow, wind and fog, and operate all-weather, which poses a great challenge to drivers. Long-term operation can easily lead to driver fatigue and increase the risk of accidents. Therefore, research on the intelligence of electric shovels can not only promote the autonomous operation of electric shovels, but also significantly improve production efficiency and safety. The autonomous rotation bucket unloading technology of electric shovels is the core link to improve their autonomous operation.

[0003] Although some researchers have proposed installing cameras on electric shovels to identify mining dump trucks and determine their relative positions, thereby determining the shovel's rotation angle, the current image-based relative coordinate solution method faces challenges in dynamic environments, lighting conditions, and weather changes. The uncertainty of the mining operating environment is large, which may affect the accuracy and stability of the image recognition system.

[0004] Therefore, how to provide a highly stable and high-precision method for autonomously rotating and opening the bucket to unload materials by an intelligent electric shovel is an urgent problem that needs to be solved by technicians in this field. Summary of the invention

[0005] The present invention aims to design an intelligent mining electric shovel autonomous rotation bucket unloading method and system. Based on the posture information of the electric shovel and the mine truck calculated by the positioning and posture determination system, it accurately calculates the angle at which the upper part of the electric shovel needs to rotate and the parking position of the bucket, and can achieve accurate rotation positioning of the upper part of the electric shovel. When the bucket reaches the predetermined position above the mine truck, the system starts the bucket opening action to improve the efficiency and safety of unloading. Furthermore, the present invention can also detect the bucket opening action. Specifically, the infrared camera equipped on the electric shovel will capture the state image of the bottom of the bucket, and transmit these images to the pre-trained discriminant network to determine whether the bucket opening is successful. After the material in the bucket is unloaded, the control algorithm will guide the upper part of the electric shovel to rotate to the ideal excavation position, thereby completing the entire autonomous rotation bucket unloading process.

[0006] In order to achieve the above technical objectives, the present invention will adopt the following technical solutions:

[0007] An intelligent mining electric shovel autonomous rotary bucket unloading system includes: a positioning and posture determination system and a control decision system, wherein:

[0008] The positioning and posture determination system includes a positioning module, a coordinate system module and a discharge position calculation module;

[0009] The positioning module includes an electric shovel positioning device and a mining truck positioning device; the mining truck positioning device is installed in an unobstructed position above the mining truck cab, and is used to obtain the mining truck positioning information, and transmit the obtained mining truck positioning information to the coordinate system module one; the electric shovel positioning device is installed on the top of the electric shovel cab, and is used to obtain the electric shovel positioning information, and transmit the obtained electric shovel positioning information to the coordinate system module one;

[0010] A coordinate system module 1 is used to convert the received mining truck positioning information and electric shovel positioning information into the same coordinate system and transmit it to the unloading position calculation module;

[0011] The unloading position calculation module calculates the bucket position and the unloading point position at the current moment according to the mining truck positioning information and the electric shovel positioning information transmitted from the coordinate system module 1, and transmits them to the control decision system;

[0012] The control decision system includes a drive data calculation module, a device control center and an actuator;

[0013] The driving data calculation module calculates the driving data according to the bucket position and the unloading point position at the current moment transmitted by the unloading position calculation module, and transmits the driving data to the equipment control center; the driving data includes the required rotation angle of the upper part of the electric shovel and the adjustment parameters of the bucket position;

[0014] The equipment control center controls the rotation of the upper part of the electric shovel according to the received required rotation angle of the upper part of the electric shovel until the rotation angle of the upper part of the electric shovel matches the required rotation angle of the upper part of the electric shovel calculated by the drive data calculation module; controls the bucket position according to the received adjustment parameters of the bucket position until it matches the adjustment parameters of the bucket position calculated by the drive data calculation module, and starts the bucket opening mechanism to execute the bucket opening action to complete the unloading.

[0015] Preferably, it also includes a bucket opening detection system;

[0016] The fighting detection system includes an infrared camera and a discrimination network;

[0017] The infrared camera is installed in front of the shovel cab and is used to detect the image information of the bottom of the bucket and transmit it to the discrimination network;

[0018] The discrimination network determines whether the bucket bottom has been successfully opened based on the image information fed back by the infrared camera, and transmits the judgment result to the equipment control center;

[0019] When the judgment result of the discrimination network indicates that the bottom of the bucket has been successfully opened, the equipment control center controls the upper part of the electric shovel to rotate to the designated excavation position to continue the next operation process; otherwise, the equipment control center controls the bucket opening mechanism to adjust the torque of the bucket opening motor, and then controls the bucket opening mechanism to open the bucket again, until the judgment result of the discrimination network indicates that the bottom of the bucket has been successfully opened.

[0020] Preferably, it also includes a prediction network learning update system;

[0021] The prediction network learning and updating system includes a network training database, a prediction network training module and a network weight updating module;

[0022] The network training database stores bucket bottom images captured by the infrared camera and performs classification processing on them;

[0023] The prediction network training module is constructed based on a neural network model; the input is a bucket bottom image, and the output is a prediction of the bucket opening state; the prediction network training module can use the bucket bottom image stored in the network training database to perform model training, so as to continuously optimize its own parameters and obtain optimized weight data;

[0024] The network weight updating module updates the discriminant network according to the optimized weight data obtained after training the prediction network training module.

[0025] Another technical purpose of the present invention is to provide a method for autonomously rotating and opening the bucket for unloading materials by an intelligent mining electric shovel, comprising the following steps:

[0026] Step 1: Drive the mining shovel and mining truck to the target working position respectively;

[0027] Step 2: respectively obtain the electric shovel positioning information of the mining electric shovel and the mining truck positioning information of the mining truck;

[0028] Step 3: Unify the electric shovel positioning information and the mining truck positioning information into the same coordinate system;

[0029] Step 4: Calculate the bucket position and the unloading point position at the current moment according to the electric shovel positioning information and the mining truck positioning information converted to the same coordinate system;

[0030] Step 5: Calculate the required rotation angle of the upper part of the electric shovel and the adjustment parameters of the bucket position according to the calculated bucket position and unloading point position at the current moment;

[0031] Step 6: Control the rotation of the upper part of the electric shovel according to the calculated rotation angle required for the upper part of the electric shovel; adjust the position of the bucket according to the calculated adjustment parameters of the bucket position;

[0032] Step 7: Start the bucket opening mechanism to execute the bucket opening action and complete the unloading.

[0033] Preferably, in step 2, the electric shovel positioning information is obtained by an electric shovel positioning device installed on the top of the electric shovel cab, and the mining truck positioning information is obtained by a mining truck positioning device installed at an unobstructed position above the mining truck cab;

[0034] In step three, the mine truck positioning information obtained by the mine truck positioning device is converted from the mine truck coordinate system to the electric shovel coordinate system, thereby realizing the coordinate system one of the electric shovel positioning information and the mine truck positioning information.

[0035] Preferably, in step three, the coordinate system of the electric shovel positioning information and the mining truck positioning information is set to one, which specifically includes the following steps:

[0036] Step 3.1, select at least three non-collinear common points, and use the electric shovel positioning device and the mining truck positioning device to measure the coordinate positions of these common points respectively;

[0037] Step 3.2, calculate the centroid of the common points, the formula is as follows:

[0038]

[0039] Where: C e Represents the coordinates of the mass center of the common point in the shovel coordinate system; C t Indicates the coordinates of the centroid of the common point in the mine card coordinate system; (X ei ,Y ei ,Z ei ) represents the coordinates of any common point i in the shovel coordinate system; (X ti ,Y ti ,Z ti ) represents the coordinates of any common point i in the mine card coordinate system; N represents the number of common points;

[0040] Step 3.3, calculate the central coordinates of the common point:

[0041]

[0042] Where: P ei represents the centralized coordinates of any common point i in the shovel coordinate system; P ti Represents the centralized coordinates of any common point i in the mine card coordinate system;

[0043] Step 3.4, calculate the rotation matrix R by singular value decomposition principle:

[0044] R=UV T ;

[0045] Where: U, V are the singular vectors of the matrix H, and the matrix H is defined as:

[0046]

[0047] Step 3.5: Get the coordinate transformation formula:

[0048]

[0049] In the above formula, (X te ,Y te ,Z te ) indicates the location information of the mining card (X t ,Y t ,Z t ) is converted to the coordinates of the shovel coordinate system; L represents the coordinate conversion translation vector, which is calculated by the following formula:

[0050] L=C e -RC t ;

[0051] Where: C e represents the coordinates of the center of mass of the common point in the shovel coordinate system, C t Represents the coordinates of the common point centroid in the mine card coordinate system;

[0052] The heading angle is converted to: θ te =θ t +Δθ;

[0053] In the above formula, θ te Indicates the heading angle θ of the mining truck positioning information t The heading angle converted to the shovel coordinate system, Δθ represents the coordinate conversion rotation angle.

[0054] Preferably, the calculated required rotation angle of the upper part of the electric shovel is expressed as follows:

[0055] θ=tan -1 (Y s X s )-tan -1 (Y d X d );

[0056] Where, X s , Y s Indicates the horizontal and vertical coordinates of the unloading point in the shovel coordinate system, X d , Y d Indicates the horizontal and vertical coordinates of the bucket position at the current moment in the shovel coordinate system;

[0057] Step 6 During the shovel upper rotation process, it is necessary to ensure: Z d >Z s .

[0058] Preferably, in step seven, a bucket opening detection system is triggered during the bucket opening process of the bucket opening mechanism; the bucket opening detection system includes an infrared camera and a discrimination network;

[0059] The infrared camera is installed in front of the shovel cab to detect the image information of the bucket bottom and transmit it to the discrimination network;

[0060] The discrimination network determines whether the bucket bottom has been successfully opened based on the image information fed back by the infrared camera, and transmits the judgment result to the equipment control center;

[0061] When the judgment result of the discrimination network indicates that the bottom of the bucket has been successfully opened, the equipment control center controls the upper part of the electric shovel to rotate to the designated excavation position to continue the next operation process; otherwise, the equipment control center controls the bucket opening mechanism to adjust the torque of the bucket opening motor, and then controls the bucket opening mechanism to open the bucket again, until the judgment result of the discrimination network indicates that the bottom of the bucket has been successfully opened.

[0062] Preferably, before the discrimination network discriminates the image information fed back by the infrared camera, it is necessary to first perform image restoration processing on the image information fed back by the infrared camera.

[0063] Based on the above technical objectives, the present invention has the following advantages over the prior art:

[0064] 1. Compared with image positioning, installing RTK positioning equipment on the shovel and the mining truck can obtain highly stable and high-precision positioning data, ensuring accuracy and safety during autonomous rotary unloading;

[0065] 2. The image restoration framework designed by the present invention is sufficient to cope with most production environments during the operation of the electric shovel;

[0066] 3. The method for the electric shovel to autonomously rotate and open the bucket to unload materials proposed in the present invention does not require manual construction of a complex network structure, and only requires the use of some open source network codes to achieve high-precision prediction;

[0067] 4. The method for the electric shovel to rotate and open the bucket to unload materials autonomously of the present invention ensures that the electric shovel is stable, safe, efficient and can be operated without human intervention. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 This is a block diagram of an embodiment of a method for autonomously rotating and opening the bucket to unload materials by an electric shovel according to the present invention;

[0069] Figure 2 A flowchart of image restoration proposed by the present invention;

[0070] Figure 3 This is the structure diagram of the intelligent electric shovel proposed by the present invention;

[0071] In the figure, 1-electric shovel, 2-electric shovel cab, 3-electric shovel positioning device, 4-infrared camera, 5-mining truck cab, 6-mining truck positioning device, 7-mining truck, 8-electric shovel bucket. DETAILED DESCRIPTION

[0072] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and is by no means any limitation to the present invention and its application or use. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. Unless otherwise specified, the relative arrangement, expressions and numerical values ​​of the components and steps described in these embodiments do not limit the scope of the present invention. The techniques, methods and equipment known to ordinary technicians in the relevant fields may not be discussed in detail, but in appropriate cases, the techniques, methods and equipment should be regarded as part of the specification. In all examples shown and discussed here, any specific value should be interpreted as merely exemplary, rather than as a limitation. Therefore, other examples of the exemplary embodiments may have different values.

[0073] For ease of description, spatially relative terms such as "above", "above", "on the upper surface of", "above", etc. may be used here to describe the spatial positional relationship between a device or feature and other devices or features as shown in the figure. It should be understood that spatially relative terms are intended to include different orientations of the device in use or operation in addition to the orientation described in the figure. For example, if the device in the accompanying drawings is inverted, the device described as "above other devices or structures" or "above other devices or structures" will be positioned as "below other devices or structures" or "below other devices or structures". Thus, the exemplary term "above" can include both "above" and "below". The device can also be positioned in other different ways (rotated 90 degrees or in other orientations).

[0074] In order to realize that the electric shovel can autonomously rotate and open the bucket to unload materials after excavation, the present invention proposes a method and system for autonomously rotating the bucket and unloading materials of the electric shovel, which is used for the electric shovel to rotate above the mine truck to unload materials after the excavation action is completed. This method can improve the intelligence level, operation efficiency and safety of the electric shovel.

[0075] like Figure 1 As shown, the intelligent mining electric shovel autonomous rotary bucket unloading system of the present invention includes a positioning and posture determination system, a bucket opening detection system, a control decision system and a prediction network learning and updating system. Among them:

[0076] The positioning and posture determination system includes a positioning module, a coordinate system module and a discharge position calculation module; wherein:

[0077] The positioning module includes an electric shovel positioning device and a mine truck positioning device. The electric shovel positioning device is installed on the top of the electric shovel cab to locate the position of the electric shovel. The mine truck positioning device is installed in an unobstructed position above the mine truck cab to locate the position of the mine truck. In the present invention, the electric shovel positioning device and the mine truck positioning device both use RTK devices, each of which obtains positioning information through the RTK device, and solves the obtained data to unify the data collected by the electric shovel positioning device and the mine truck positioning device into the same coordinate system and transmit them to the control decision system.

[0078] The coordinate system module is used to unify the electric shovel coordinate system and the mining truck coordinate system, taking the electric shovel coordinate system as an example (a reference coordinate system can also be set by yourself), and the specific steps include:

[0079] Step 3.1, select at least three non-collinear common points, and use the electric shovel positioning device and the mining truck positioning device to measure the coordinate positions of these common points respectively;

[0080] Step 3.2, calculate the centroid of the common points, the formula is as follows:

[0081]

[0082] Where: C e Represents the coordinates of the mass center of the common point in the shovel coordinate system; C t Indicates the coordinates of the centroid of the common point in the mine card coordinate system; (X ei ,Y ei ,Z ei ) represents the coordinates of any common point i in the shovel coordinate system; (X ti ,Y ti ,Z ti ) represents the coordinates of any common point i in the mine card coordinate system; N represents the number of common points.

[0083] Step 3.3, calculate the central coordinates of the common point:

[0084] Based on the calculated coordinates C of the centroid of the common point in the shovel coordinate system e , coordinate C in the mine card coordinate system t , and calculate the central coordinates of each common point respectively:

[0085]

[0086] Where: P ei represents the centralized coordinates of any common point i in the shovel coordinate system; Pti Represents the centralized coordinates of any common point i in the mine card coordinate system.

[0087] Step 3.4, calculate the rotation matrix R by singular value decomposition principle:

[0088] Specifically, assuming Then the singular decomposition of the matrix H is UΣV T ; Therefore, the rotation matrix R can be expressed as:

[0089] R=UV T ;

[0090] Where U and V are the singular vectors of the matrix H respectively.

[0091] Step 3.5: Get the coordinate transformation formula:

[0092]

[0093] In the above formula, (X te ,Y te ,Z te ) indicates the location information of the mining card (X t ,Y t ,Z t ) is converted to the coordinates of the shovel coordinate system; L represents the coordinate conversion translation vector, which is calculated by the following formula:

[0094] L=C e -RC t ;

[0095] Where: C e represents the coordinates of the center of mass of the common point in the shovel coordinate system, C t Represents the coordinates of the common point centroid in the mine card coordinate system;

[0096] The heading angle is converted to: θ te =θ t +Δθ;

[0097] In the above formula, θ te Indicates the heading angle θ of the mining truck positioning information t The heading angle converted to the shovel coordinate system, Δθ represents the coordinate conversion rotation angle, which is calculated by the following formula:

[0098]

[0099] Where (x1, y1) and (x2, y2) are the coordinates of any common point in the shovel coordinate system and the mining truck coordinate system respectively.

[0100] After the coordinate system is unified, according to the shovel coordinate P in the same coordinate system e =(X e,Y e ,Z e θ, e With the mine card coordinates P te =(X te ,Y te ,Z te ,θ te ), calculate the bucket position P at the current moment based on the size of the shovel itself d =(X d ,Y d ,Z d ) and the position of the unloading point P s =(X s ,Y s ,Z s ). From this, the rotation angle Δθ = tan -1 (Y s X s )-tan -1 (Y d X d ), Z must be ensured during the rotation process d >Z s .

[0101] The control decision system includes three main components: drive data calculation 60, equipment control center 70 and actuator 80. Drive data calculation 60 receives the current bucket position information and the unloading point position information from the positioning and posture system. Drive data calculation 60 calculates the drive data based on the received current bucket position information and unloading point position information. The drive data includes the required rotation angle of the upper part of the electric shovel and the adjustment parameters of the bucket position. The drive data is then transmitted to the electric shovel equipment control center 70. After receiving the drive data, the electric shovel equipment control center 70 controls the actuator 80 according to the drive data, guides the actuator 80 to move to the predetermined position, and then controls the bucket opening mechanism to perform the bucket opening action to complete the unloading.

[0102] The bucket opening detection system consists of an infrared camera 90 and a discrimination network 100. The infrared camera 90 is installed in front of the electric shovel cab and can clearly capture the real-time image of the bucket bottom. When the actuator 80 performs the bucket opening action, the infrared camera 90 immediately captures the image data of the bucket. The discrimination network 100 is responsible for analyzing these image data to determine whether the bucket bottom has been successfully opened. The judgment result will be sent to the electric shovel equipment control center 70 for further processing. If the detection result shows that the bucket opening is successful, the electric shovel equipment control center 70 will command the actuator 80 to rotate to the designated excavation position to continue the next operation process. On the contrary, if the discrimination network 100 determines that the bucket opening is unsuccessful, the electric shovel equipment control center 70 will issue a command to adjust the torque of the bucket opening motor, prompting the actuator 80 to make a second bucket opening attempt until it is confirmed that the bucket bottom is opened correctly.

[0103] The prediction network learning and updating system includes a network training database 130, a prediction network training 120 and a network weight update 110. The network training database 130 is responsible for accumulating bucket images captured by the infrared camera 90 and classifying and processing them. These high-quality data resources provide a solid foundation for the training of the discriminant network 100. In the prediction network training 120 module, the image data in the database is used to train a network model that can effectively judge the bucket opening state. During the training process, the model will continuously adjust its own parameters to achieve the best prediction effect. Once the prediction network training 120 completes the model training task, the network weight update 110 module will take over and apply the weight data obtained from the training to the update of the discriminant network 100. At the same time, considering the dynamic environment in the mine, in order to reduce its impact on image quality, a set of image restoration architecture is integrated in the discriminant network 100. The present invention uses MPR-Net and FFA-Net to design a set of image restoration frameworks, which can effectively reduce the impact of the working environment on image quality.

[0104] Example 2

[0105] This embodiment provides a method for an intelligent mining electric shovel to autonomously rotate and open the bucket to unload materials, comprising the following steps:

[0106] Step 1: Drive the mining shovel and mining truck to the target working position respectively;

[0107] Step 2: respectively obtain the electric shovel positioning information of the mining shovel and the mining truck positioning information of the mining truck; wherein: the electric shovel positioning information is obtained by the electric shovel positioning device installed on the top of the electric shovel driver's cab, and the mining truck positioning information is obtained by the mining truck positioning device installed in an unobstructed position above the mining truck driver's cab.

[0108] Step three, unify the electric shovel positioning information and the mine truck positioning information into the same coordinate system; when implementing, convert the mine truck positioning information obtained by the mine truck positioning device from the mine truck coordinate system to the electric shovel coordinate system, thereby realizing the coordinate system one of the electric shovel positioning information and the mine truck positioning information.

[0109] Step 4: Calculate the bucket position and the unloading point position at the current moment according to the electric shovel positioning information and the mining truck positioning information converted to the same coordinate system;

[0110] Step 5: Calculate the required rotation angle of the upper part of the electric shovel and the adjustment parameters of the bucket position according to the calculated bucket position and unloading point position at the current moment; the calculated rotation angle required for the upper part of the electric shovel is expressed as follows:

[0111] θ=tan -1 (Y s X s )-tan -1 (Y d X d );

[0112] Where, X s , Y s Indicates the horizontal and vertical coordinates of the unloading point in the shovel coordinate system, X d , Y d Indicates the horizontal and vertical coordinates of the bucket position at the current moment in the shovel coordinate system.

[0113] Step 6: Control the rotation of the upper part of the shovel according to the calculated rotation angle required by the upper part of the shovel. During the rotation of the upper part of the shovel, it is necessary to ensure that: Z d >Z s ; Adjust the bucket position according to the calculated bucket position adjustment parameters;

[0114] Step 7: Start the bucket opening mechanism to execute the bucket opening action and complete the unloading.

[0115] The bucket opening detection system is triggered during the bucket opening process of the bucket opening mechanism; the bucket opening detection system includes an infrared camera and a discrimination network. The infrared camera is installed in front of the electric shovel cab to detect the image information of the bucket bottom and transmit it to the discrimination network;

[0116] The discrimination network determines whether the bottom of the bucket has been successfully opened based on the image information fed back by the infrared camera, and transmits the judgment result to the equipment control center; when the judgment result of the discrimination network shows that the bottom of the bucket has been successfully opened, the equipment control center controls the upper part of the electric shovel to rotate to the designated excavation position to continue the next operation process; otherwise, the equipment control center controls the bucket opening mechanism to adjust the torque of the bucket opening motor, and then controls the bucket opening mechanism to open the bucket again, until the judgment result of the discrimination network shows that the bottom of the bucket has been successfully opened.

[0117] The method for autonomously rotating and opening the bucket to unload materials by a mining electric shovel according to the present invention mainly comprises:

[0118] S1. Coordinate system 1 for the electric shovel positioning device and the mining truck positioning device: by selecting at least three common points to measure the position information of the common points in the two sets of equipment, and then calculating the rotation matrix and translation vector between the two coordinate systems, the measurement data of the two devices can be unified into the same coordinate system;

[0119] S2, obtain the current posture information of the shovel through measurement (X e ,Y e ,Z e ,θ e ) and the position information of the mining card (X t ,Y t ,Z t ,θ t Based on this information, the ideal position of the shovel bucket after the excavation is completed (X d ,Y d ,Z d ,θ d ) and the required rotation angle θ of the electric shovel. To ensure the safety of the operation, the electric shovel must maintain its bucket height Z during the rotation process. d >Z s , thereby preventing possible collisions between the bucket and the mining truck bucket;

[0120] S4, prediction result processing 50 sends the processed data to drive data calculation 60. Drive data calculation 60 calculates the required precise drive instructions based on these data. The calculated drive data is then transmitted to the electric shovel equipment control center 70, which accurately commands the actuator based on these data to move the bucket to the predetermined target position, thereby completing the expected bucket opening and unloading task;

[0121] S5. After the execution structure executes the bucket opening and unloading, the infrared camera 90 in the bucket opening detection system will capture the image information of the bucket bottom state, and the discrimination network 100 receives the captured image information to determine whether the bucket opening is successful, and sends the judgment result to the electric shovel equipment control center 70. If the bucket opening is judged to be successful, the control center 70 will instruct the actuator to rotate back to the original excavation position; if the bucket opening is unsuccessful, the control center 70 will command the bucket opening motor to increase the torque to try to open the bucket for the second time;

[0122] The image restoration in the discriminant network 100 in step S4 specifically includes the following steps:

[0123] S5.1. For a single influencing factor, the processing method is very easy. For example, the processing methods for rainy days, foggy days, blur, dust, etc. are to simply perform operations such as removing rain, removing fog (dust), and removing noise.

[0124] S5.2, in the case of rainy days and blurred superposition, the deraining operation is first performed followed by the deblurring operation, both using the MPR-Net network;

[0125] S5.3. When rain and fog (dust) overlap, first use the MPR-Net network to perform rain removal, and then use FFA-Net to remove fog (dust);

[0126] S5.4, fog (dust) and blur superposition, first perform deblurring, then use FFA-Net to remove fog (dust):

[0127] S5.5. When rain, fog (dust) and blur are superimposed, first perform rain removal, then deblurring, and finally defogging (dust). Defogging (dust) uses FFA-Net, and the rest of the operations use MPR-Net.

Claims

1. An intelligent mining electric shovel autonomous rotary bucket unloading system, characterized in that: include: Positioning and attitude determination system and control decision system, including: The positioning and posture determination system includes a positioning module, a coordinate system module and a discharge position calculation module; The positioning module includes an electric shovel positioning device and a mining truck positioning device; the mining truck positioning device is installed in an unobstructed position above the mining truck cab, and is used to obtain the mining truck positioning information, and transmit the obtained mining truck positioning information to the coordinate system module one; the electric shovel positioning device is installed on the top of the electric shovel cab, and is used to obtain the electric shovel positioning information, and transmit the obtained electric shovel positioning information to the coordinate system module one; A coordinate system module 1 is used to convert the received mining truck positioning information and electric shovel positioning information into the same coordinate system and transmit it to the unloading position calculation module; The unloading position calculation module calculates the bucket position and the unloading point position at the current moment according to the mining truck positioning information and the electric shovel positioning information transmitted from the coordinate system module 1, and transmits them to the control decision system; The control decision system includes a drive data calculation module, a device control center and an actuator; The driving data calculation module calculates the driving data according to the bucket position and the unloading point position at the current moment transmitted by the unloading position calculation module, and transmits the driving data to the equipment control center; the driving data includes the required rotation angle of the upper part of the electric shovel and the adjustment parameters of the bucket position; The equipment control center controls the rotation of the upper part of the electric shovel according to the received required rotation angle of the upper part of the electric shovel until the rotation angle of the upper part of the electric shovel matches the required rotation angle of the upper part of the electric shovel calculated by the drive data calculation module; controls the bucket position according to the received adjustment parameters of the bucket position until it matches the adjustment parameters of the bucket position calculated by the drive data calculation module, and starts the bucket opening mechanism to execute the bucket opening action to complete the unloading.

2. The intelligent mining electric shovel autonomous rotary bucket unloading system according to claim 1 is characterized in that: It also includes a bucket opening detection system; The fighting detection system includes an infrared camera and a discrimination network; The infrared camera is installed in front of the shovel cab and is used to detect the image information of the bottom of the bucket and transmit it to the discrimination network; The discrimination network determines whether the bucket bottom has been successfully opened based on the image information fed back by the infrared camera, and transmits the judgment result to the equipment control center; When the judgment result of the discrimination network indicates that the bottom of the bucket has been successfully opened, the equipment control center controls the upper part of the electric shovel to rotate to the designated excavation position to continue the next operation process; otherwise, the equipment control center controls the bucket opening mechanism to adjust the torque of the bucket opening motor, and then controls the bucket opening mechanism to open the bucket again, until the judgment result of the discrimination network indicates that the bottom of the bucket has been successfully opened.

3. The intelligent mining electric shovel autonomous rotary bucket unloading system according to claim 2 is characterized in that: It also includes a predictive network learning update system; The prediction network learning and updating system includes a network training database, a prediction network training module and a network weight updating module; the network training database stores bucket bottom images captured by an infrared camera and performs classification processing on the images; The prediction network training module is constructed based on a neural network model; the input is a bucket bottom image, and the output is a prediction of the bucket opening state; the prediction network training module can use the bucket bottom image stored in the network training database to perform model training, so as to continuously optimize its own parameters and obtain optimized weight data; The network weight updating module updates the discriminant network according to the optimized weight data obtained after training the prediction network training module.

4. A method for autonomously rotating bucket opening and unloading of intelligent mining electric shovel, characterized in that: The steps include: Step 1: Drive the mining shovel and mining truck to the target working position respectively; Step 2: respectively obtain the electric shovel positioning information of the mining electric shovel and the mining truck positioning information of the mining truck; Step 3: Unify the electric shovel positioning information and the mining truck positioning information into the same coordinate system; Step 4: Calculate the bucket position and the unloading point position at the current moment according to the electric shovel positioning information and the mining truck positioning information converted to the same coordinate system; Step 5: Calculate the required rotation angle of the upper part of the electric shovel and the adjustment parameters of the bucket position according to the calculated bucket position and unloading point position at the current moment; Step 6: Control the upper part of the electric shovel to rotate according to the calculated rotation angle required for the upper part of the electric shovel; adjust the position of the bucket according to the calculated adjustment parameters of the bucket position; Step 7: Start the bucket opening mechanism to execute the bucket opening action and complete the unloading.

5. The method for autonomously rotating bucket opening and unloading of intelligent mining electric shovel according to claim 4 is characterized in that: In step 2, the electric shovel positioning information is obtained by an electric shovel positioning device installed on the top of the electric shovel cab, and the mining truck positioning information is obtained by a mining truck positioning device installed in an unobstructed position above the mining truck cab; In step three, the mine truck positioning information obtained by the mine truck positioning device is converted from the mine truck coordinate system to the electric shovel coordinate system, thereby realizing the coordinate system one of the electric shovel positioning information and the mine truck positioning information.

6. The method for autonomously rotating bucket opening and unloading of intelligent mining electric shovel according to claim 4 is characterized in that: In step 3, the coordinate system of the electric shovel positioning information and the mining truck positioning information is set to one, which specifically includes the following steps: Step 3.1, select at least three non-collinear common points, and use the electric shovel positioning device and the mining truck positioning device to measure the coordinate positions of these common points respectively; Step 3.2, calculate the centroid of the common points, the formula is as follows: Where: C e Represents the coordinates of the mass center of the common point in the shovel coordinate system; C t Indicates the coordinates of the centroid of the common point in the mine card coordinate system; (X ei ,Y ei ,Z ei ) represents the coordinates of any common point i in the shovel coordinate system; (X ti ,Y ti ,Z ti ) represents the coordinates of any common point i in the mine card coordinate system; N represents the number of common points; Step 3.3, calculate the central coordinates of the common point: Where: P ei represents the centralized coordinates of any common point i in the shovel coordinate system; P ti Represents the centralized coordinates of any common point i in the mine card coordinate system; Step 3.4, calculate the rotation matrix R by singular value decomposition principle: R=UV T ; Where: U, V are the singular vectors of the matrix H, and the matrix H is defined as: Step 3.5: Get the coordinate transformation formula: In the above formula, (X te ,Y te ,Z te ) indicates the location information of the mining card (X t ,Y t ,Z t ) is converted to the coordinates of the shovel coordinate system; L represents the coordinate conversion translation vector, which is calculated by the following formula: L=C e -R t ; Where: C e represents the coordinates of the center of mass of the common point in the shovel coordinate system, C t Represents the coordinates of the common point centroid in the mine card coordinate system; The heading angle is converted to: θ te =θ t +Δθ; In the above formula, θ te Indicates the heading angle θ of the mining truck positioning information t The heading angle converted to the shovel coordinate system, Δθ represents the coordinate conversion rotation angle.

7. The method for autonomously rotating bucket opening and unloading of intelligent mining electric shovel according to claim 6 is characterized in that: The calculated required rotation angle of the upper part of the shovel is expressed as follows: θ=tan -1 (AND s X s )-so -1 (AND d X d ); Where, X s , Y s Indicates the horizontal and vertical coordinates of the unloading point in the shovel coordinate system, X d , Y d Indicates the horizontal and vertical coordinates of the bucket position at the current moment in the shovel coordinate system; Step 6 During the shovel upper rotation process, it is necessary to ensure: Z d >Z s .

8. The method for autonomously rotating bucket opening and unloading of intelligent mining electric shovel according to claim 4 is characterized in that: Step 7: triggering a bucket opening detection system during the bucket opening process of the bucket opening mechanism; the bucket opening detection system includes an infrared camera and a discrimination network; The infrared camera is installed in front of the shovel cab to detect the image information of the bucket bottom and transmit it to the discrimination network; The discrimination network determines whether the bucket bottom has been successfully opened based on the image information fed back by the infrared camera, and transmits the judgment result to the equipment control center; When the judgment result of the discrimination network indicates that the bottom of the bucket has been successfully opened, the equipment control center controls the upper part of the electric shovel to rotate to the designated excavation position to continue the next operation process; otherwise, the equipment control center controls the bucket opening mechanism to adjust the torque of the bucket opening motor, and then controls the bucket opening mechanism to open the bucket again, until the judgment result of the discrimination network indicates that the bottom of the bucket has been successfully opened.

9. The method for autonomously rotating bucket opening and unloading of intelligent mining electric shovel according to claim 8 is characterized in that: Before the discriminant network discriminates the image information fed back by the infrared camera, it is necessary to first perform image restoration processing on the image information fed back by the infrared camera.

Images