Unmanned loader shoveling point selection method based on laser radar and system and equipment thereof

Abstract

The invention discloses an unmanned loader shoveling point selection method based on a laser radar and a system and equipment thereof, relates to the technical field of unmanned engineering machinery, and aims to solve the problems of low efficiency, high cost, potential safety hazards and the like of a traditional manual shoveling method of a loader in the prior art. The method comprises the steps of obtaining point cloud data of a construction area environment; preprocessing the point cloud data of the construction area environment to obtain preprocessed point cloud data; constructing a three-dimensional environment model according to the preprocessed point cloud data; generating a plurality of candidate spading points by using the three-dimensional environment model, and calculating and recording spading parameters of the candidate spading points; and evaluating the candidate spading points according to the spading parameters of the candidate spading points to obtain an optimal spading point. According to the method, the spading points can be automatically generated in a complex construction environment, and efficient and accurate spading operation is realized through path planning.

Description

Technical Field

[0001] The invention relates to a laser radar-based unmanned loader excavation point selection method and a system and equipment thereof, belonging to the technical field of unmanned engineering machinery. Background Art

[0002] Loaders are earthwork machines widely used in construction projects such as roads, railways, buildings, hydropower, ports, and mines. They are mainly used to shovel bulk materials such as soil, sand, lime, and coal, and can also do light shoveling of ore and hard soil. By replacing different auxiliary working devices, they can also do bulldozing, lifting, and loading and unloading of other materials such as wood.

[0003] In road construction, especially in high-grade highway construction, loaders are used for filling and excavation of roadbed engineering, collection and loading of asphalt mixture and cement concrete material fields, etc. In addition, they can also carry out operations such as pushing soil, leveling the ground, and pulling other machinery. Because loaders have the advantages of fast operation speed, high efficiency, good maneuverability, and easy operation, they have become one of the main machines for earthwork construction in engineering construction.

[0004] As the demand for construction and earthwork continues to increase, the traditional manual shoveling method of loaders faces problems such as low efficiency, high cost and safety hazards. The emergence of unmanned loaders provides a new solution for earthwork construction. Summary of the invention

[0005] The purpose of the present invention is to overcome the deficiencies in the prior art and to provide a laser radar-based unmanned loader excavation point selection method and system and equipment, which can automatically generate excavation points in a complex construction environment and achieve efficient and accurate excavation operations through path planning.

[0006] To achieve the above object, the present invention is implemented by adopting the following technical solutions: In a first aspect, the present invention provides a method for selecting a digging point for an unmanned loader based on a laser radar, comprising: Obtain point cloud data of the construction area environment; Preprocessing the point cloud data of the construction area environment to obtain preprocessed point cloud data; Construct a three-dimensional environment model based on the preprocessed point cloud data; Generate multiple candidate excavation points using the three-dimensional environment model, and calculate and record excavation parameters of the candidate excavation points; The candidate digging points are evaluated according to their digging parameters to obtain the optimal digging point.

[0007] Furthermore, the point cloud data of the construction area environment is obtained by performing a 360° scan of the construction area using a laser radar.

[0008] Furthermore, the point cloud data of the construction area environment is preprocessed to obtain the preprocessed point cloud data, including: Through-through filtering is used to remove point cloud data outside the construction area, and statistical filtering is used to remove noise points to obtain preprocessed environmental point cloud data.

[0009] Furthermore, constructing a three-dimensional environment model according to the pre-processed environment point cloud data includes: The pre-processed environmental point cloud data is segmented using a ground segmentation algorithm to obtain ground point cloud data and non-ground point cloud data; Use the ground detection algorithm to identify the ground point cloud data to obtain the average height of the ground point cloud data, generate a height map, and identify the non-ground point cloud data to obtain the three-dimensional contour and position of the obstacle; A three-dimensional environment model is constructed based on the height map and the three-dimensional contours and positions of obstacles.

[0010] Furthermore, the method of generating a plurality of candidate excavation points by using the three-dimensional environment model, and calculating and recording excavation parameters of the candidate excavation points includes: Uniformly generate multiple candidate excavation points on the height map of the three-dimensional environment model, wherein all of the candidate excavation points must be located within the excavation area; Calculate and record the location coordinates, digging depth and digging efficiency of the candidate digging points.

[0011] Furthermore, the calculation expression of the excavation depth is: B = H target -H ground in, B Indicates the digging depth. H target Indicates the target digging height, H ground Indicates the average height of ground point cloud data; The factors affecting the shoveling efficiency include the distance between the candidate shoveling point and the obstacle, the accessibility of the shoveling point and the ground soil type, and the expression is as follows: E = k(R × S) / D in, E Indicates the digging efficiency, k represents the empirical coefficient, D represents the distance between the candidate excavation point and the nearest obstacle, R represents the accessibility coefficient of the candidate shovel point, S Represents the ground soil type coefficient.

[0012] Further, the candidate digging points are evaluated according to the digging parameters of the candidate digging points to obtain the optimal digging point, including: Preset objective function; The parameter vector is used as input and the optimal position coordinates are used as output. The particle swarm optimization algorithm is used to iteratively optimize according to the objective function until the maximum number of iterations is reached, and the optimal shoveling point is output; The parameter vector is { i, x, y, z, E, D, P, H},in, i is the index number of the candidate shovel point, x, y, z is the position coordinate of the candidate shovel point, E For excavation efficiency, D is the distance between the candidate excavation point and the nearest obstacle, P is the straight-line distance from the candidate shovel point to the current position of the loader, H The candidate excavation point is the center point K × K The average height of the material within the range, K is the bucket width.

[0013] Furthermore, the optimization algorithm is one of a genetic algorithm, a particle swarm optimization algorithm and a simulated annealing algorithm, and the calculation expression of the objective function is: f(x, y, z) = w 1 ×Cost(x, y, z) + w 2 ×R(x, y, z) + w 3 ×Risk(x, y, z) in, f(x, y, z) represents the objective function, w 1 , w 2 , w 3 represents the weight coefficient, Cost(x, y) is the cost function, R (x, y, z) is the resource maximization objective function, Risk(x, y, z) is the risk function; The expression of the cost function is: Cost(x, y) = C fixed +C variable ×P(x, y) in, C fixed is a fixed cost, C variable is the unit cost, P (x, y) is the distance from the candidate shovel point to the current position of the loader; The expression of the resource maximization objective function is: R ( x , y, z )=1 / H(x, y, z) in, H(x, y, z) Represents the candidate excavation point as the center K × K The average height of the material within the range, K is the bucket width; The expression of the risk function is: Risk ( x , y, z )=1 / E in, E Indicates the digging efficiency of the candidate digging point.

[0014] In a second aspect, the present invention further provides an unmanned loader digging point selection system based on laser radar, comprising: LiDAR, used to obtain environmental point cloud data of the construction area; A data processing module is used to pre-process the environmental point cloud data of the construction area to obtain pre-processed environmental point cloud data; A model building module, used to build a three-dimensional environment model based on pre-processed environment point cloud data; A candidate shoveling point generation module is used to generate multiple candidate shoveling points using a three-dimensional environment model, and calculate and record shoveling parameters of the candidate shoveling points; The optimal shoveling point selection module is used to evaluate the candidate shoveling points according to the shoveling parameters of the candidate shoveling points to obtain the optimal shoveling point.

[0015] In a third aspect, the present invention further provides a device, comprising: A memory for storing instructions; A processor is used to execute the instructions so that the device performs operations to implement the laser radar-based unmanned loader excavation point selection method as described in any of the above items.

[0016] Compared with the prior art, the present invention has the following beneficial effects: The present invention is combined with lidar technology, which can realize real-time perception and intelligent decision-making of the construction environment. It can automatically generate multiple candidate excavation points in a complex construction environment, and obtain the optimal excavation point after iterative optimization by a particle swarm algorithm. Finally, through path planning, the unmanned loader can implement efficient and accurate excavation operations at the optimal excavation point, thereby greatly improving construction efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic diagram of a flow chart of a method for selecting a digging point for an unmanned loader based on a laser radar in an embodiment of the present invention; Figure 2 Schematic diagram of a three-dimensional environment model of a method for selecting excavation points for an unmanned loader based on laser radar in an embodiment of the present invention, wherein a is a schematic diagram of preprocessed point cloud data, and b is a schematic diagram of a three-dimensional environment model; Figure 3 The figure is a schematic diagram of the hardware structure of an unmanned loader excavation point selection system based on laser radar in an embodiment of the present invention. DETAILED DESCRIPTION

[0018] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of the present invention.

[0019] Example 1

[0020] like Figure 1 As shown, an embodiment of the present invention provides a method for selecting a digging point for an unmanned loader based on a laser radar, comprising the following steps: The point cloud data of the construction area environment is obtained. In this embodiment, a laser radar is used to perform a 360° scan of the surrounding environment of the construction area to obtain high-precision point cloud data in the area.

[0021] Preprocess the point cloud data of the construction area environment. The preprocessing specifically includes: First, use the straight-through filter to remove the point cloud data in the range of no interest, which mainly refers to the area outside the construction area. This is because the laser radar scanning process will scan some areas outside the construction area. Then use the statistical filter to remove noise points, reduce the amount of calculation, and ensure the accuracy of the point cloud data, so as to obtain the preprocessed point cloud data.

[0022] Construct a 3D environment model based on preprocessed point cloud data, including: The pre-processed environment point cloud data is segmented using a ground segmentation algorithm to obtain ground point cloud data and non-ground point cloud data.

[0023] The ground detection algorithm is used to identify the ground point cloud data to obtain the average height of the ground point cloud data, generate a height map, and identify the non-ground point cloud data to obtain the three-dimensional contour and position of the obstacle, such as Figure 2 As shown, Figure 2 a is the pre-processed point cloud data, Figure 2b is a three-dimensional environment model, in which the ground (gray area), obstacles (black area) and materials (purple area) are marked. In this embodiment, the ground segmentation algorithm and the ground detection algorithm use RANSAC.

[0024] According to the three-dimensional environmental model, multiple candidate excavation points are evenly generated on the height map. It should be noted that these candidate excavation points should all be located in the excavable area, which is within 3 meters from the front of the pile in this embodiment. Multiple candidate excavation points are evenly generated with a spacing of half the bucket width, and the bucket width is 3 meters.

[0025] At the same time, the digging parameters of the candidate digging points are calculated and recorded.

[0026] In this embodiment, the excavation parameters include the position coordinates of the candidate excavation points, the excavation depth and the excavation efficiency, but are not limited thereto. Other conditions that need to be considered in the actual construction process should also be acceptable. Among them, the calculation expression of the excavation depth is: B = H target -H ground in, B Indicates the digging depth. H target Indicates the target digging height, H ground Indicates the average height of the ground point cloud data.

[0027] The factors affecting the shoveling efficiency include but are not limited to the distance between the candidate shoveling point and the obstacle, the accessibility of the shoveling point and the ground soil type. The expression is as follows: E = k(R × S) / D in, E Indicates the digging efficiency, k It represents the empirical coefficient, which can be adjusted according to the specific project conditions. D Indicates the distance between the candidate excavation point and the nearest obstacle. The longer the distance, the lower the efficiency. R It represents the accessibility coefficient of the shovel point. The accessibility of the candidate shovel point can be evaluated according to the site conditions (such as road, slope, etc.). The better the accessibility of the candidate shovel point, the larger the coefficient. Its value range is 0~1. S It indicates the ground soil type coefficient, which represents the difficulty of digging different soil types. The larger the coefficient value, the easier the soil is to dig. Its value range is 0~1.

[0028] The candidate digging points are evaluated according to their digging parameters to obtain the optimal digging point. In this embodiment, the optimization algorithm selected is the particle swarm algorithm, but not limited to this. Other algorithms such as genetic algorithm (by simulating natural selection, gradually iterating to find the optimal solution) and simulated annealing algorithm (by simulating the physical annealing process to avoid falling into the local optimum and thus obtain the optimal solution) should all be possible.

[0029] The evaluation process is as follows: a. First, to clarify the selection target of the optimal excavation point, it is necessary to comprehensively consider minimizing costs, maximizing efficiency or optimizing resource utilization, etc. In this embodiment, the expression of the preset objective function is as follows: f(x, y, z) = w 1 ×Cost(x, y, z) + w 2 ×R(x, y, z) + w 3 ×Risk(x, y, z) in, f(x, y, z) represents the objective function, w 1 , w 2 , w 3 represents the weight coefficient, Cost(x, y) is the cost function, R (x, y, z) is the resource maximization objective function, i.e., the estimated value of excavable material at the candidate excavation point, Risk(x, y, z) is the risk function.

[0030] The expression of the cost function is: Cost(x, y) = C fixed +C variable ×P(x, y) in, C fixed is a fixed cost, C variable is the unit cost, P (x, y) is the distance from the candidate digging point to the current position of the loader.

[0031] The expression of the resource maximization objective function is: R ( x , y, z )=1 / H(x, y, z) in, H(x, y, z) Represents the candidate excavation point as the center K × K The average height of the material within the range, K is the bucket width.

[0032] The expression of the risk function is: Risk ( x , y, z )=1 / E in, E Indicates the digging efficiency of the candidate digging point.

[0033] b. Determine the particle swarm parameters, such as: Number of particles: Select the number of particles, usually 20~50; Dimension: Determine the dimension based on the complexity of the problem. For example, the coordinates of the excavation point are (x, y) because the excavation point needs to be on the ground. Speed ​​Range: Set the speed range of particles to control the search range.

[0034] c. Process the particle swarm, including position initialization and velocity initialization. Position initialization is to randomly generate the initial position of each particle, usually within the defined search space. Velocity initialization is to randomly generate the initial velocity of each particle.

[0035] d. Input parameter vector, the parameter vector is { i, x, y, z, E, D, P, H},in, i is the index number of the candidate shovel point, x, y, z is the position coordinate of the candidate shovel point, E For excavation efficiency, D is the distance between the candidate excavation point and the nearest obstacle, P is the straight-line distance from the candidate shovel point to the current position of the loader, H The candidate excavation point is the center point K × K The average height of the material within the range, K is the bucket width.

[0036] e. Calculate the objective function value of each particle's current position based on the objective function. The smaller the objective function value, the better the particle's current position.

[0037] f. Update particle information: Individual best position (pBest): For each particle, record its best historical position; Global optimal position (gBest): the best position among the individual optimal positions of all particles.

[0038] g. Update the positions and velocities of all particles. Their expressions are: ;

[0039] ; in, For the i The updated velocity of each particle is For the i The last speed of a particle, For the i The individual historical optimal solution of each particle, is the global historical optimal solution, For the i The current vector of the particle, For the i The last position of a particle, After the update i The position of a particle, and is the learning factor, also called the acceleration constant, which usually takes the same value and is in the interval [0,4]. and is a random number between (0,1).

[0040] Repeat steps e~g until the maximum number of iterations is reached, and output the global optimal position, which is the optimal digging point.

[0041] After obtaining the optimal shoveling point, the A* algorithm or Dijkstra algorithm is used for path planning to ensure that the loader can smoothly reach the optimal shoveling point from the current position. During the path planning process, the surrounding environment is monitored in real time and the path is adjusted to avoid dynamic obstacles.

[0042] Embodiment 2: Based on Example 1, this embodiment provides an unmanned loader digging point selection system based on laser radar, including: LiDAR, used to obtain environmental point cloud data of the construction area; A data processing module is used to pre-process the environmental point cloud data of the construction area to obtain pre-processed environmental point cloud data; A model building module, used to build a three-dimensional environment model based on pre-processed environment point cloud data; A candidate shoveling point generation module is used to generate multiple candidate shoveling points using a three-dimensional environment model, and calculate and record shoveling parameters of the candidate shoveling points; The optimal shoveling point selection module is used to evaluate the candidate shoveling points according to the shoveling parameters of the candidate shoveling points to obtain the optimal shoveling point.

[0043] Among them, the data processing module, model building module, candidate digging point generation module and optimal digging point selection module are all integrated on the industrial computer in the loader cab. The laser radar is connected to the industrial computer by electrical signals, which makes it convenient for the staff to control the laser radar through the industrial computer.

[0044] Embodiment 3: Based on Embodiment 1, this embodiment provides a device, including: A memory for storing instructions; The processor is used to execute the instructions so that the device performs the operations of the unmanned loader excavation point selection method based on laser radar as described in Example 1.

[0045] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for selecting digging points for an unmanned loader based on laser radar, characterized in that: include: Obtain point cloud data of the construction area environment; Preprocessing the point cloud data of the construction area environment to obtain preprocessed point cloud data; Construct a three-dimensional environment model based on the preprocessed point cloud data; Generate multiple candidate excavation points using the three-dimensional environment model, and calculate and record excavation parameters of the candidate excavation points; The candidate digging points are evaluated according to their digging parameters to obtain the optimal digging point.

2. The method for selecting digging points for unmanned loaders based on laser radar according to claim 1 is characterized in that: The point cloud data of the construction area environment is obtained by performing a 360° scan of the construction area by a laser radar.

3. The method for selecting digging points for unmanned loaders based on laser radar according to claim 1 is characterized in that: The step of preprocessing the point cloud data of the construction area environment to obtain the preprocessed point cloud data includes: Through-through filtering is used to remove point cloud data outside the construction area, and statistical filtering is used to remove noise points to obtain preprocessed environmental point cloud data.

4. The method for selecting digging points for unmanned loaders based on laser radar according to claim 1, characterized in that: The method of constructing a three-dimensional environment model according to the pre-processed environment point cloud data includes: The pre-processed environmental point cloud data is segmented using a ground segmentation algorithm to obtain ground point cloud data and non-ground point cloud data; Use the ground detection algorithm to identify the ground point cloud data to obtain the average height of the ground point cloud data, generate a height map, and identify the non-ground point cloud data to obtain the three-dimensional contour and position of the obstacle; A three-dimensional environment model is constructed based on the height map and the three-dimensional contours and positions of obstacles.

5. The method for selecting digging points for unmanned loaders based on laser radar according to claim 1, characterized in that: The method of generating a plurality of candidate excavation points by using the three-dimensional environment model and calculating and recording the excavation parameters of the candidate excavation points includes: Uniformly generate multiple candidate excavation points on the height map of the three-dimensional environment model, wherein all of the candidate excavation points must be located within the excavation area; Calculate and record the location coordinates, digging depth and digging efficiency of the candidate digging points.

6. The method for selecting digging points for unmanned loaders based on laser radar according to claim 5 is characterized in that: The calculation expression of the excavation depth is: D = H target -H ground in, D Indicates the digging depth. H target Indicates the target digging height, H ground Indicates the average height of ground point cloud data; The factors affecting the shoveling efficiency include the distance to the obstacle, the accessibility of the shoveling point and the ground soil type, and the expression is as follows: E=k(R × S) / D in, E Indicates the digging efficiency, k represents the empirical coefficient, D represents the distance between the candidate excavation point and the nearest obstacle, R represents the accessibility coefficient of the candidate shovel point, S Represents the ground soil type coefficient.

7. The method for selecting digging points for an unmanned loader based on laser radar according to claim 1, characterized in that: The step of evaluating the candidate digging points according to the digging parameters of the candidate digging points to obtain the optimal digging point includes: Select the optimization algorithm and preset the objective function; The parameter vector is used as input and the optimal position coordinates are used as output. The particle swarm optimization algorithm is used to iteratively optimize according to the objective function until the maximum number of iterations is reached, and the optimal shoveling point is output; The parameter vector is { i,x,y,z,E,D,P,H },in, i is the index number of the candidate shovel point, x,y,z is the position coordinate of the candidate shovel point, E is the digging efficiency of the candidate digging point, D is the distance between the candidate excavation point and the nearest obstacle, P is the straight-line distance from the candidate shovel point to the current position of the loader, H The candidate excavation point is the center point K × K The average height of the material within the range, K is the bucket width.

8. The method for selecting digging points for an unmanned loader based on laser radar according to claim 1, characterized in that: The optimization algorithm is one of a genetic algorithm, a particle swarm optimization algorithm and a simulated annealing algorithm, and the calculation expression of the objective function is: f(x,y,z)=w 1 ×Cost(x,y,z)+w 2 ×R(x,y,z)+w 3 ×Risk(x,y,z) in, f(x,y,z) represents the objective function, w 1 , w 2 , w 3 represents the weight coefficient, Cost(x,y) is the cost function, R(x,y, z) is the resource maximization objective function, Risk(x,y,z) is the risk function; The expression of the cost function is: Cost(x,y)=C fixed +C variable ×P(x,y) in, C fixed is a fixed cost, C variable is the unit cost, P (x,y) is the distance from the candidate shovel point to the current position of the loader; The expression of the resource maximization objective function is: R ( x , y,z )=1 / H(x,y,z) in, H(x,y,z) represents the candidate shovel point as the center K × K The average height of the material within the range, K is the bucket width; The expression of the risk function is: Risk ( x , y,z )=1 / E in, E Indicates the digging efficiency of the candidate digging point.

9. An unmanned loader excavation point selection system based on laser radar, characterized in that: include: LiDAR, used to obtain environmental point cloud data of the construction area; A data processing module is used to pre-process the environmental point cloud data of the construction area to obtain pre-processed environmental point cloud data; A model building module, used to build a three-dimensional environment model based on pre-processed environment point cloud data; A candidate shoveling point generation module is used to generate multiple candidate shoveling points using a three-dimensional environment model, and calculate and record shoveling parameters of the candidate shoveling points; The optimal shoveling point selection module is used to evaluate the candidate shoveling points according to the shoveling parameters of the candidate shoveling points to obtain the optimal shoveling point.

10. A device, characterized in that: include: A memory for storing instructions; A processor is used to execute the instruction so that the device performs the operation of implementing the unmanned loader excavation point selection method based on laser radar as described in any one of claims 1 to 8.

Images