Planning method for automatic coal raking action of excavator

Abstract

The invention discloses a planning method for automatic coal raking action of an excavator, which comprises the following steps of: acquiring length data of hydraulic rods of each mechanism of the excavator in real time through a displacement sensor arranged on the excavator, and calculating the posture of each mechanism of the excavator and the position of a current grab bucket; installing a solid-state laser radar in front of the excavator to scan environment information, identifying the form and height of a coal pile, and obtaining form data of the coal pile; the solid laser radar is adopted to calibrate the specific pose of the excavator, coordinate parameters of the grab bucket relative to the center pose of the excavator are obtained, a trajectory tracking model based on the coal pile form data is established, and the motion path of the grab bucket of the excavator relative to the coal pile form data is obtained in real time; the method comprises the following steps: acquiring path nodes of a motion path by adopting an adaptive path planning algorithm, performing global path search on the path nodes, dynamically adjusting the path by adopting a fusion algorithm, and planning a global optimal path, so that the precision and operability of path planning are ensured, and real-time automatic coal raking action is realized.

Description

Technical Field

[0001] The invention relates to the technical field of excavator motion planning, and in particular to a planning method for automatic coal raking motion of an excavator. Background Art

[0002] With the development of industrial automation and intelligence, excavators are used more and more widely in coal operation and processing. The traditional manual operation method is not only inefficient, but also has high safety risks. In order to improve operation efficiency and safety, the research and development of automated excavators has become an important research direction. However, how to realize the automatic raking action of the excavator in a complex coal pile environment is still a technical problem that needs to be solved urgently. Therefore, the development of an efficient and accurate excavator automatic raking action planning method has important practical significance and application prospects.

[0003] There are some technologies on the market for automated operation of excavators, but the following problems still exist:

[0004] (1) At present, most of the excavator automation operation technologies focus on simple excavation and handling tasks. For complex coal raking operations, there is still a problem of insufficient positioning accuracy;

[0005] (2) Existing positioning technologies have limited accuracy in complex environments, making it difficult to accurately obtain the relative position of the excavator and the coal pile;

[0006] (3) Most path planning algorithms do not fully consider the shape of the coal pile and the motion characteristics of the excavator, resulting in unreasonable planning paths that affect work efficiency;

[0007] (4) The existing system takes a long time to process data and plan paths, and cannot achieve real-time automatic coal raking. In addition, the existing technology lacks intelligent algorithm support when controlling the movement of various mechanisms of the excavator, resulting in inaccurate and unstable movements. Summary of the invention

[0008] The purpose of the present invention is to provide a planning method for the automatic coal raking action of an excavator to solve the technical problems in the prior art of the automatic control technology of the excavator, such as limited positioning accuracy, poor positioning accuracy, and imperfect path planning.

[0009] In order to solve the above technical problems, the present invention specifically provides the following technical solutions:

[0010] The present invention provides a method for planning an automatic coal raking action of an excavator, comprising the following steps:

[0011] The displacement sensor installed on the excavator can obtain the length data of the hydraulic rods of each mechanism of the excavator in real time, and calculate the posture of each mechanism of the excavator and the current position of the grab bucket;

[0012] Install a solid-state laser radar in front of the excavator to scan environmental information, obtain three-dimensional point cloud data of the coal pile, identify the shape and height of the coal pile, and obtain coal pile shape data;

[0013] The solid-state laser radar is used to calibrate the specific posture of the excavator, and the coordinate parameters of the grab bucket relative to the center posture of the excavator are obtained. In combination with the center speed and angular velocity of the excavator, a trajectory tracking model based on the coal pile morphology data is established to obtain the motion path of the excavator grab bucket relative to the coal pile morphology data in real time;

[0014] An adaptive path planning algorithm is used to collect path nodes for the motion path, a global path search is performed on the path nodes, a fusion algorithm is used to dynamically adjust the path, and a global optimal path is planned.

[0015] As a preferred solution of the present invention, a displacement sensor installed on the excavator is used to obtain the length data of the hydraulic rods of each mechanism of the excavator in real time, and the posture of each mechanism of the excavator and the current position of the grab bucket are calculated, including:

[0016] When the excavator drives to the coal pile, the displacement sensor is used to obtain the length data of the hydraulic rods of each mechanism during the operation of the grab bucket;

[0017] The distance that the grab bucket of the excavator extends outward relative to the excavator is calculated by using the length data, and the posture data of the hydraulic rods of each mechanism is calculated in combination with the displacement of the hydraulic rods of each mechanism;

[0018] The travel switch of the grab bucket is controlled according to the posture data to obtain the position information of the grab bucket in real time.

[0019] As a preferred solution of the present invention, a solid-state laser radar is installed in front of the excavator to scan environmental information, obtain three-dimensional point cloud data of the coal pile, identify the shape and height of the coal pile, and obtain the shape data of the coal pile, including:

[0020] The solid-state laser radar is used to scan the environmental information of the coal pile around the excavator in real time, the environmental information is rasterized, and the shape of the coal pile is determined by calibrating the grid position;

[0021] The coal pile morphology is feature extracted in the form of a grid sequence to obtain coal pile feature point data, and the coal pile feature point data is linearly fitted to identify the shape and height of the coal pile to obtain coal pile morphology data.

[0022] As a preferred solution of the present invention, the solid-state laser radar is used to calibrate the specific posture of the excavator to obtain the coordinate parameters of the grab bucket relative to the center posture of the excavator, including:

[0023] The solid-state laser radar is used to scan the environmental information of the excavator during its movement, and the environmental information is rasterized in time series to obtain environmental feature data;

[0024] Extracting environmental information feature points of the environmental feature data at the current moment and the previous moment, performing stereo matching on the environmental information feature points, and obtaining radar scanning parallax information to obtain coordinates of the environmental information feature points;

[0025] The excavator posture characteristics are estimated according to the coordinates of the environmental information feature points, and the coordinate parameters of the grab bucket relative to the center posture of the excavator are obtained.

[0026] As a preferred solution of the present invention, a trajectory tracking model based on the coal pile morphology data is established in combination with the excavator center speed and angular velocity to obtain the motion path of the excavator grab relative to the coal pile morphology data in real time, including:

[0027] The solid-state laser radar is used to feed back the moving position of the excavator in real time, and the speed sensor is used to collect the center speed and angular velocity of the excavator, and the relative position of the excavator and the coal pile is calculated according to the coordinate parameters to obtain the path coordinates between the excavator and the coal pile;

[0028] A distance constraint model is established for the path coordinates, and a safe distance h0 of the excavator during movement is set. The distance constraint model is constrained by an obstacle avoidance function, and its expression is:

[0029]

[0030] Among them, i represents the current position of the excavator relative to the coal pile, n represents the distance between the coal pile and the excavator divided into n equal parts with the rasterized length as the unit length, δ represents the safe distance between the coal pile and the excavator to avoid collision, represents the position of the excavator relative to the coal pile when it is at position i, and p0 represents the position of the coal pile;

[0031] The moving trajectory of the excavator is planned according to the safety distance, and a trajectory tracking model based on the coal pile morphology data is established.

[0032] As a preferred solution of the present invention, a trajectory tracking model based on the coal pile morphology data is established, including:

[0033] The excavator moving trajectory is used as the target trajectory, and the excavator state quantity X is obtained after the target trajectory is discretized. i ={x i ,y i ,θ i};

[0034] The host computer simulates the excavator's starting position to the target position, plans the expected driving path, and discretizes the driving path to obtain the excavator's reference posture X r ={x r ,y r ,θ r};

[0035] The current excavator state X i , reference pose X r The distance constraint model is input as input data, and the motion constraint is performed through the obstacle avoidance function to obtain the optimal control sequence U of the predicted path. r ;

[0036] For the optimal control sequence U r Establish a reward and punishment mechanism, compare each viable path with the shortest path, longest path and evaluation path in the global map, reward the pheromones of the better path, and obtain the optimal path information.

[0037] As a preferred solution of the present invention, the excavator position and posture model is established by combining the optimal path information with the excavator center speed and angular velocity, including:

[0038] Establishing a global coordinate system according to the coordinate parameters, setting the center of mass position of the grab bucket and the deflection angle of the grab bucket, and establishing the position and posture model of the excavator;

[0039] Combining the center velocity and angular velocity of the excavator, the velocity vector of the grab is integrated to obtain the position and posture of the grab;

[0040] The movement path of the grab bucket is planned by adjusting the excavator's posture and the deflection angle of the grab bucket.

[0041] As a preferred solution of the present invention, a trajectory tracking model based on the coal pile morphology data is established through the movement path of the grab bucket, and the movement path of the excavator grab bucket relative to the coal pile morphology data is obtained in real time, including:

[0042] According to the relative motion relationship between the grab and the excavator, the kinematic equation is established, and the stable running speed of the excavator is obtained through the motion path of the grab. The expression is:

[0043]

[0044] Where v1 and v2 are the speeds of the grab and excavator, m1 and m2 are the relative masses of the grab and excavator, F is the lateral traction of the excavator, f1 is the friction of the excavator, f2 is the traction of the grab, T and T ′They represent the displacement difference between the grab bucket and the excavator during movement, θ represents the offset angle of the grab bucket relative to the excavator, and t represents the movement time;

[0045] The influence factor of parameter change on the running track of the grab bucket is determined by setting the parameter range of the stable running speed of the excavator.

[0046] As a preferred solution of the present invention, an adaptive path planning algorithm is used to collect path nodes for the motion path, and a global path search is performed on the path nodes, including:

[0047] The motion path is rasterized, multi-frame laser radar scanning is adopted, corresponding data of each unit grid is stored after addition and subtraction, a threshold is set to determine the grid state, and a safe path is planned by searching the free grid;

[0048] The safety path is discretized into different mass points, and the distance between the excavator position and the coal pile is expanded outward, and the safety distance of the excavator movement is set. The excavator is expanded outward with the safety distance, and the expansion range q is set, which is expressed as:

[0049] q=max(Δx+Lcosθ,Δy+Lsinθ)

[0050] Among them, θ represents the offset angle of the grab bucket relative to the excavator, L represents the length of the grab bucket, Δx represents the horizontal displacement of the grab bucket relative to the excavator, and Δy represents the vertical displacement of the grab bucket relative to the excavator;

[0051] The motion path is preprocessed through the expansion range q to obtain a complete grid map.

[0052] As a preferred solution of the present invention, a fusion algorithm is used to dynamically adjust the path and plan the global optimal path, including:

[0053] Select any path from the grid map, set the initial position and the target position, set the cost g(n) from the initial position to other vertices to infinity, and set a special container Closed_list to store the searched nodes;

[0054] Starting from the initial position, expand outward to traverse adjacent nodes and put them into the Open_list priority queue. If the sum of the weights from the recalculated node to the starting point is smaller, the shortest distance from the starting point to the point is updated.

[0055] Then, the node with the minimum weight in the Open_list priority queue is taken out, and the above steps are repeated until all nodes including the end point can be updated to the shortest distance, and then the shortest path is obtained by tracing back from the end point to the starting point.

[0056] Compared with the prior art, the present invention has the following beneficial effects:

[0057] The present invention obtains the length data of the hydraulic rods of each mechanism in real time through the displacement sensor installed on the excavator, calculates the posture of each mechanism of the excavator and the current position of the grab bucket, installs a solid-state laser radar in front of the excavator, obtains the three-dimensional point cloud data of the coal pile, identifies the shape and height of the coal pile, analyzes the morphological data of the coal pile through a fusion algorithm, adopts an adaptive path planning algorithm to dynamically adjust the path information, performs collaborative path planning for the excavator, considers obstacles around the mobile grab bucket during the search process, plans the global path of the excavator and the grab bucket, adds a distance optimization function and path smoothing post-processing, ensures the accuracy and operability of the path planning, and realizes real-time automatic coal raking action. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the implementation methods of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the implementation methods or the description of the prior art. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other implementation drawings can be derived from the provided drawings without creative work.

[0059] Figure 1 A flow chart of a method for planning the automatic coal raking action of an excavator provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0060] 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, not all of the embodiments. Based on the embodiments of 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.

[0061] like Figure 1 As shown, the present invention provides a method for planning the automatic coal raking action of an excavator, comprising the following steps:

[0062] The displacement sensor installed on the excavator can obtain the length data of the hydraulic rods of each mechanism of the excavator in real time, and calculate the posture of each mechanism of the excavator and the current position of the grab bucket;

[0063] In this embodiment, the rope encoder installed on the excavator is used to obtain the length data of the hydraulic rod of each mechanism in real time. Combined with the displacement sensor, the posture of each mechanism of the excavator and the current position of the grab bucket are calculated, which can provide multi-dimensional calculation of the position information of each mechanism of the excavator and improve the positioning accuracy of the excavator during movement.

[0064] Install a solid-state laser radar in front of the excavator to scan environmental information, obtain three-dimensional point cloud data of the coal pile, identify the shape and height of the coal pile, and obtain coal pile shape data;

[0065] In this embodiment, a solid-state laser radar is installed to scan environmental information, continuously collect and update environmental information, obtain complete three-dimensional point cloud data, and provide a data basis for the positioning of the excavator.

[0066] The solid-state laser radar is used to calibrate the specific posture of the excavator, and the coordinate parameters of the grab bucket relative to the center posture of the excavator are obtained. In combination with the center speed and angular velocity of the excavator, a trajectory tracking model based on the coal pile morphology data is established to obtain the motion path of the excavator grab bucket relative to the coal pile morphology data in real time;

[0067] An adaptive path planning algorithm is used to collect path nodes for the motion path, a global path search is performed on the path nodes, a fusion algorithm is used to dynamically adjust the path, and a global optimal path is planned.

[0068] In this embodiment, a solid-state laser radar is installed in front of the excavator to obtain three-dimensional point cloud data of the coal pile, identify the shape and height of the coal pile, analyze the morphological data of the coal pile through a fusion algorithm, calculate a reasonable grab bucket to rake the coal movement path, and discretize the path into several key nodes to ensure the accuracy and operability of the path planning, thereby effectively improving the accuracy of the path planning.

[0069] The displacement sensor installed on the excavator can obtain the length data of the hydraulic rods of each mechanism of the excavator in real time, and calculate the posture of each mechanism of the excavator and the current position of the grab bucket, including:

[0070] When the excavator drives to the coal pile, the displacement sensor is used to obtain the length data of the hydraulic rods of each mechanism during the operation of the grab bucket;

[0071] The distance that the grab bucket of the excavator extends outward relative to the excavator is calculated by using the length data, and the posture data of the hydraulic rods of each mechanism is calculated in combination with the displacement of the hydraulic rods of each mechanism;

[0072] The travel switch of the grab bucket is controlled according to the posture data to obtain the position information of the grab bucket in real time.

[0073] In this embodiment, displacement sensors are used to collect real-time length data of the hydraulic rods of various mechanisms of the excavator during the operation of the grab, and the length data is used to locate the relative position of the grab. The extension and retraction range of the grab is determined by the relative position, and a switch is used to control the travel position of the grab to obtain the grab position information in real time.

[0074] Install a solid-state laser radar in front of the excavator to scan environmental information, obtain three-dimensional point cloud data of the coal pile, identify the shape and height of the coal pile, and obtain coal pile shape data, including:

[0075] The solid-state laser radar is used to scan the environmental information of the coal pile around the excavator in real time, the environmental information is rasterized, and the shape of the coal pile is determined by calibrating the grid position;

[0076] The coal pile morphology is feature extracted in the form of a grid sequence to obtain coal pile feature point data, and the coal pile feature point data is linearly fitted to identify the shape and height of the coal pile to obtain coal pile morphology data.

[0077] In this embodiment, the environmental information scanned by the solid-state laser radar is rasterized, and a rectangular coordinate system is established to plan the position of the excavator. In the grid, 1 represents the obstacle position and 0 represents the idle position. The shape and height of the coal pile are identified by fully planning the position of grid 1, and the coal pile shape data is obtained. This can provide maximum planning of the coal pile position information, avoid obstacles for the excavator's path planning, and provide data information for the maximum telescopic path of the grab.

[0078] The solid-state laser radar is used to calibrate the specific posture of the excavator to obtain the coordinate parameters of the grab bucket relative to the center posture of the excavator, including:

[0079] The solid-state laser radar is used to scan the environmental information of the excavator during its movement, and the environmental information is rasterized in time series to obtain environmental feature data;

[0080] Extracting environmental information feature points of the environmental feature data at the current moment and the previous moment, performing stereo matching on the environmental information feature points, and obtaining radar scanning parallax information to obtain coordinates of the environmental information feature points;

[0081] The excavator posture characteristics are estimated according to the coordinates of the environmental information feature points, and the coordinate parameters of the grab bucket relative to the center posture of the excavator are obtained.

[0082] In this embodiment, a solid-state laser radar is used to scan environmental information in real time, and the environmental information is rasterized and sorted to obtain environmental feature data. The environmental information is then stereo matched according to the environmental features, thereby planning the excavator posture features in the environment. This not only improves the data planning accuracy of the excavator grab, but also effectively improves the positioning accuracy of the excavator.

[0083] Combined with the center speed and angular velocity of the excavator, a trajectory tracking model based on the coal pile morphology data is established to obtain the motion path of the excavator grab relative to the coal pile morphology data in real time, including:

[0084] The solid-state laser radar is used to feed back the moving position of the excavator in real time, and the speed sensor is used to collect the center speed and angular velocity of the excavator, and the relative position of the excavator and the coal pile is calculated according to the coordinate parameters to obtain the path coordinates between the excavator and the coal pile;

[0085] A distance constraint model is established for the path coordinates, and a safe distance h0 of the excavator during movement is set. The distance constraint model is constrained by an obstacle avoidance function, and its expression is:

[0086]

[0087] Among them, i represents the current position of the excavator relative to the coal pile, n represents the distance between the coal pile and the excavator divided into n equal parts with the rasterized length as the unit length, δ represents the safe distance between the coal pile and the excavator to avoid collision, represents the position of the excavator relative to the coal pile when it is at position i, and p0 represents the position of the coal pile;

[0088] The moving trajectory of the excavator is planned according to the safety distance, and a trajectory tracking model based on the coal pile morphology data is established.

[0089] In this embodiment, by establishing a distance constraint model for the path coordinates of the excavator relative to the coal pile, the movement path of the excavator can be controlled in real time in a complex coal pile environment, collision between the excavator and the coal pile can be avoided, and the relative position of the excavator and the coal pile can be accurately obtained.

[0090] Establishing a trajectory tracking model based on the coal pile morphology data, including:

[0091] The excavator moving trajectory is used as the target trajectory, and the excavator state quantity X is obtained after the target trajectory is discretized. i ={x i ,y i ,θ i};

[0092] The host computer simulates the excavator's starting position to the target position, plans the expected driving path, and discretizes the driving path to obtain the excavator's reference posture X r ={x r ,y r ,θ r};

[0093] The current excavator state X i , reference pose X r The distance constraint model is input as input data, and the motion constraint is performed through the obstacle avoidance function to obtain the optimal control sequence U of the predicted path. r ;

[0094] For the optimal control sequence U r Establish a reward and punishment mechanism, compare each viable path with the shortest path, longest path and evaluation path in the global map, reward the pheromones of the better path, and obtain the optimal path information.

[0095] In this embodiment, according to the excavator state quantity X i And the reference pose X r The position of the excavator relative to the coal pile is planned in real time, taking into account the shape of the coal pile and the motion characteristics of the excavator, and the state quantity X i And the reference pose X r The error is obtained by comparing the distance constraint model and then fed back to the model for correction to achieve the effect of closed-loop optimization control. By optimizing in a limited time domain, the impact of uncertain interference on the system is reduced.

[0096] The excavator position and posture model is established by combining the optimal path information with the excavator center speed and angular speed, including:

[0097] Establishing a global coordinate system according to the coordinate parameters, setting the center of mass position of the grab bucket and the deflection angle of the grab bucket, and establishing the position and posture model of the excavator;

[0098] Combining the center velocity and angular velocity of the excavator, the velocity vector of the grab is integrated to obtain the position and posture of the grab;

[0099] The movement path of the grab is planned by adjusting the excavator's posture and the deflection angle of the grab.

[0100] In this embodiment, the velocity vector of the grab is calculated through the center speed and angular velocity of the excavator, and the motion model of the grab in the local coordinate system is obtained. By adjusting the center speed and angular velocity of the excavator, the steering and telescopic length of the grab relative to the excavator in the local coordinate system are simulated, thereby effectively planning the motion path of the grab and improving the travel efficiency of the grab during coal transportation.

[0101] By using the movement path of the grab bucket, a trajectory tracking model based on the coal pile shape data is established to obtain the movement path of the excavator grab bucket relative to the coal pile shape data in real time, including:

[0102] According to the relative motion relationship between the grab and the excavator, the kinematic equation is established, and the stable running speed of the excavator is obtained through the motion path of the grab. The expression is:

[0103]

[0104] Where v1 and v2 are the speeds of the grab and excavator, m1 and m2 are the relative masses of the grab and excavator, F is the lateral traction of the excavator, f1 is the friction of the excavator, f2 is the traction of the grab, T and T ′ They represent the displacement difference between the grab bucket and the excavator during movement, θ represents the offset angle of the grab bucket relative to the excavator, and t represents the movement time;

[0105] The influence factor of parameter change on the running track of the grab bucket is determined by setting the parameter range of the stable running speed of the excavator.

[0106] The motion path is subjected to the adaptive path planning algorithm to collect path nodes, and the path nodes are subjected to a global path search, including:

[0107] The motion path is rasterized, multi-frame laser radar scanning is adopted, corresponding data of each unit grid is stored after addition and subtraction, a threshold is set to determine the grid state, and a safe path is planned by searching the free grid;

[0108] The safety path is discretized into different mass points, and the distance between the excavator position and the coal pile is expanded outward, and the safety distance of the excavator movement is set. The excavator is expanded outward with the safety distance, and the expansion range q is set, which is expressed as:

[0109] q=max(Δx+Lcosθ,Δy+Lsinθ)

[0110] Among them, θ represents the offset angle of the grab bucket relative to the excavator, L represents the length of the grab bucket, Δx represents the horizontal displacement of the grab bucket relative to the excavator, and Δy represents the vertical displacement of the grab bucket relative to the excavator;

[0111] The motion path is preprocessed through the expansion range q to obtain a complete grid map.

[0112] In this embodiment, in order to facilitate the excavator to be assumed as a particle during path planning, it is necessary to expand outward around the obstacle based on the maximum value of the excavator's shape, so as to ensure that the excavator does not collide with the obstacle and remains at a relatively safe distance. By setting the expansion range q, the excavator's moving path is constrained, thereby controlling the distance between the excavator and the coal pile.

[0113] Use fusion algorithms to dynamically adjust the path and plan the global optimal path, including:

[0114] Select any path from the grid map, set the initial position and the target position, set the cost g(n) from the initial position to other vertices to infinity, and set a special container Closed_list to store the searched nodes;

[0115] Starting from the initial position, expand outward to traverse adjacent nodes and put them into the Open_list priority queue. If the sum of the weights from the recalculated node to the starting point is smaller, the shortest distance from the starting point to the point is updated.

[0116] Then, the node with the minimum weight in the Open_list priority queue is taken out, and the above steps are repeated until all nodes including the end point can be updated to the shortest distance, and then the shortest path is obtained by tracing back from the end point to the starting point.

[0117] In this embodiment, collaborative path planning is performed on the excavator. During the search process, obstacles around the mobile grab are taken into consideration, and a global path for the excavator and the grab is planned. A distance optimization function and path smoothing post-processing are added, and finally a smooth and collision-free global path for the excavator can be planned within 1 second, thus realizing the autonomous path-finding function in complex environments.

[0118] The present invention obtains the length data of the hydraulic rods of each mechanism in real time through the displacement sensor installed on the excavator, calculates the posture of each mechanism of the excavator and the current position of the grab bucket, installs a solid-state laser radar in front of the excavator, obtains the three-dimensional point cloud data of the coal pile, identifies the shape and height of the coal pile, analyzes the morphological data of the coal pile through a fusion algorithm, adopts an adaptive path planning algorithm to dynamically adjust the path information, performs collaborative path planning for the excavator, considers obstacles around the mobile grab bucket during the search process, plans the global path of the excavator and the grab bucket, adds a distance optimization function and path smoothing post-processing, ensures the accuracy and operability of the path planning, and realizes real-time automatic coal raking action.

[0119] The above embodiments are only exemplary embodiments of the present application and are not intended to limit the present application. The protection scope of the present application is defined by the claims. Those skilled in the art may make various modifications or equivalent substitutions to the present application within the essence and protection scope of the present application, and such modifications or equivalent substitutions shall also be deemed to fall within the protection scope of the present application.

Claims

1. A method for planning the automatic coal raking action of an excavator, characterized in that: The following steps are involved: The displacement sensor installed on the excavator can obtain the length data of the hydraulic rods of each mechanism of the excavator in real time, and calculate the posture of each mechanism of the excavator and the current position of the grab bucket; Install a solid-state laser radar in front of the excavator to scan environmental information, obtain three-dimensional point cloud data of the coal pile, identify the shape and height of the coal pile, and obtain coal pile shape data; The solid-state laser radar is used to calibrate the specific posture of the excavator, and the coordinate parameters of the grab bucket relative to the center posture of the excavator are obtained. In combination with the center speed and angular velocity of the excavator, a trajectory tracking model based on the coal pile morphology data is established to obtain the motion path of the excavator grab bucket relative to the coal pile morphology data in real time; An adaptive path planning algorithm is used to collect path nodes for the motion path, a global path search is performed on the path nodes, a fusion algorithm is used to dynamically adjust the path, and a global optimal path is planned.

2. The method for planning the automatic coal raking action of an excavator according to claim 1, characterized in that: The displacement sensor installed on the excavator can obtain the length data of the hydraulic rods of each mechanism of the excavator in real time, and calculate the posture of each mechanism of the excavator and the current position of the grab bucket, including: When the excavator drives to the coal pile, the displacement sensor is used to obtain the length data of the hydraulic rods of each mechanism during the operation of the grab bucket; The distance that the grab bucket of the excavator extends outward relative to the excavator is calculated by using the length data, and the posture data of the hydraulic rods of each mechanism is calculated in combination with the displacement of the hydraulic rods of each mechanism; The travel switch of the grab bucket is controlled according to the posture data to obtain the position information of the grab bucket in real time.

3. The method for planning the automatic coal raking action of an excavator according to claim 2, characterized in that: Install a solid-state laser radar in front of the excavator to scan environmental information, obtain three-dimensional point cloud data of the coal pile, identify the shape and height of the coal pile, and obtain coal pile shape data, including: The solid-state laser radar is used to scan the environmental information of the coal pile around the excavator in real time, the environmental information is rasterized, and the shape of the coal pile is determined by calibrating the grid position; The coal pile morphology is feature extracted in the form of a grid sequence to obtain coal pile feature point data, and the coal pile feature point data is linearly fitted to identify the shape and height of the coal pile to obtain coal pile morphology data.

4. The method for planning the automatic coal raking action of an excavator according to claim 3, characterized in that: The solid-state laser radar is used to calibrate the specific posture of the excavator to obtain the coordinate parameters of the grab bucket relative to the center posture of the excavator, including: The solid-state laser radar is used to scan the environmental information of the excavator during its movement, and the environmental information is rasterized in time series to obtain environmental feature data; Extracting environmental information feature points of the environmental feature data at the current moment and the previous moment, performing stereo matching on the environmental information feature points, and obtaining radar scanning parallax information to obtain coordinates of the environmental information feature points; The excavator posture characteristics are estimated according to the coordinates of the environmental information feature points, and the coordinate parameters of the grab bucket relative to the center posture of the excavator are obtained.

5. The method for planning the automatic coal raking action of an excavator according to claim 4, characterized in that: Combined with the center speed and angular velocity of the excavator, a trajectory tracking model based on the coal pile morphology data is established to obtain the motion path of the excavator grab relative to the coal pile morphology data in real time, including: The solid-state laser radar is used to feed back the moving position of the excavator in real time, and the speed sensor is used to collect the center speed and angular velocity of the excavator, and the relative position of the excavator and the coal pile is calculated according to the coordinate parameters to obtain the path coordinates of the excavator relative to the coal pile; A distance constraint model is established for the path coordinates, and a safe distance h0 of the excavator during movement is set. The distance constraint model is constrained by an obstacle avoidance function, and its expression is: Among them, i represents the current position of the excavator relative to the coal pile, n represents the distance between the coal pile and the excavator divided into n equal parts with the rasterized length as the unit length, δ represents the safe distance between the coal pile and the excavator to avoid collision, represents the position of the excavator relative to the coal pile when it is at position i, and p0 represents the position of the coal pile; The moving trajectory of the excavator is planned according to the safety distance, and a trajectory tracking model based on the coal pile morphology data is established.

6. The method for planning the automatic coal raking action of an excavator according to claim 5, characterized in that: Establishing a trajectory tracking model based on the coal pile morphology data, including: The excavator moving trajectory is used as the target trajectory, and the excavator state quantity X is obtained after the target trajectory is discretized. i ={x i ,y i ,θ i }; The host computer simulates the excavator's starting position to the target position, plans the expected driving path, and discretizes the driving path to obtain the excavator's reference posture X r ={x r ,y r ,θ r }; The current excavator state X i , reference pose X r The distance constraint model is input as input data, and the motion constraint is performed through the obstacle avoidance function to obtain the optimal control sequence U of the predicted path. r ; For the optimal control sequence U r Establish a reward and punishment mechanism, compare each viable path with the shortest path, longest path and evaluation path in the global map, reward the pheromones of the better path, and obtain the optimal path information.

7. The method for planning the automatic coal raking action of an excavator according to claim 6, characterized in that: The excavator position and posture model is established by combining the optimal path information with the excavator center speed and angular speed, including: Establishing a global coordinate system according to the coordinate parameters, setting the center of mass position of the grab bucket and the deflection angle of the grab bucket, and establishing the position and posture model of the excavator; Combining the center velocity and angular velocity of the excavator, the velocity vector of the grab is integrated to obtain the position and posture of the grab; The movement path of the grab bucket is planned by adjusting the excavator's posture and the grab bucket's deflection angle.

8. The method for planning the automatic coal raking action of an excavator according to claim 7, characterized in that: By using the movement path of the grab bucket, a trajectory tracking model based on the coal pile shape data is established to obtain the movement path of the excavator grab bucket relative to the coal pile shape data in real time, including: According to the relative motion relationship between the grab and the excavator, the kinematic equation is established, and the stable running speed of the excavator is obtained through the motion path of the grab. The expression is: Where v1 and v2 are the speeds of the grab and excavator, m1 and m2 are the relative masses of the grab and excavator, F is the lateral traction of the excavator, f1 is the friction of the excavator, f2 is the traction of the grab, T and T ′ They represent the displacement difference between the grab bucket and the excavator during movement, θ represents the offset angle of the grab bucket relative to the excavator, and t represents the movement time; The speed setting parameter range of the stable operation of the excavator is used to determine the influence factor of the parameter change on the operation trajectory of the grab bucket.

9. The method for planning the automatic coal raking action of an excavator according to claim 8, characterized in that: The motion path is subjected to the adaptive path planning algorithm to collect path nodes, and the path nodes are subjected to a global path search, including: The motion path is rasterized, multi-frame laser radar scanning is adopted, corresponding data of each unit grid is stored after addition and subtraction, a threshold is set to determine the grid state, and a safe path is planned by searching the free grid; The safety path is discretized into different mass points, and the distance between the excavator position and the coal pile is expanded outward, and the safety distance of the excavator movement is set. The excavator is expanded outward with the safety distance, and the expansion range q is set, which is expressed as: q=max(Δx+Lcosθ,Δy+Lsinθ) Among them, θ represents the offset angle of the grab bucket relative to the excavator, L represents the length of the grab bucket, Δx represents the horizontal displacement of the grab bucket relative to the excavator, and Δy represents the vertical displacement of the grab bucket relative to the excavator; The motion path is preprocessed through the expansion range q to obtain a complete grid map.

10. The method for planning the automatic coal raking action of an excavator according to claim 9, characterized in that: Use fusion algorithms to dynamically adjust the path and plan the global optimal path, including: Select any path from the grid map, set the initial position and the target position, set the cost g(n) from the initial position to other vertices to infinity, and set a special container Closed_list to store the searched nodes; Starting from the initial position, expand outward to traverse adjacent nodes and put them into the Open_list priority queue. If the sum of the weights from the recalculated node to the starting point is smaller, the shortest distance from the starting point to the point is updated. Then, the node with the minimum weight in the Open_list priority queue is taken out, and the above steps are repeated until all nodes including the end point can be updated to the shortest distance, and then the shortest path is obtained by tracing back from the end point to the starting point.

Images