A method for generating a driving track of an unmanned vehicle in an open-pit mine

Abstract

This invention relates to a method for generating the driving trajectory of unmanned vehicles in open-pit mines, belonging to the field of high-precision map acquisition technology for unmanned vehicles in mining areas. The method acquires the boundaries of roads and intersections in the mining area and generates driving trajectories for both road and intersection scenarios based on a sequential quadratic programming method and a geometric curve method. This method can meet the high-frequency changes in roads in open-pit mine scenarios while generating vehicle driving trajectories in real time, improving the generation rate of vehicle driving trajectories in mining areas. The generated vehicle driving trajectories meet vehicle dynamics requirements and are easy for unmanned vehicles in mining areas to aim and track.

Description

Technical Field

[0001] This invention belongs to the field of high-precision map acquisition technology for unmanned vehicles in mining areas, specifically involving a method for generating the driving trajectory of unmanned vehicles in open-pit mines, which can generate the driving trajectory of mining trucks based on the boundaries of the drivable area of ​​the mining area roads. Background Technology

[0002] Due to the lack of road markings on roads in mining areas, traditional visual recognition methods struggle to generate vehicle trajectories. Existing technologies for generating the trajectories of unmanned vehicles mainly include: road centerline based on road markings and high-precision satellite positioning acquisition methods.

[0003] Among these methods, road centerline based on road markings is commonly used for paved roads with clear road markings. Typically, methods such as LiDAR and aerial imagery are used to collect data on the current road conditions. Automated recognition and calculation tools are then used to analyze the road markings, and the centerlines of two road markings are extracted as the vehicle's trajectory. High-precision satellite positioning is a common method for collecting vehicle trajectories in unmanned mining areas. Its characteristic is that it does not rely on road markings; a GPS module is installed on the data collection vehicle, and the vehicle's trajectory is collected based on satellite positioning. Satellite positioning technology is used to collect trajectory points as the vehicle travels along the road. After data collection, a physical model of the data collection vehicle is established based on the location of the satellite positioning antenna mounted on the vehicle, the vehicle's orientation, vehicle width, and antenna height. The unmanned vehicle's trajectory is then constructed based on real-time data collected during the data collection process.

[0004] In mining areas, vehicles typically travel on unpaved roads without road markings, making lane centerline-based methods unsuitable. High-precision satellite positioning suffers from low data collection efficiency; reconstructing a full-area map requires extensive on-site vehicle data collection and is significantly affected by external interference factors such as weather and unstable satellite signals. Summary of the Invention

[0005] Based on the above analysis, this invention aims to provide a method for generating the driving trajectory of unmanned vehicles in open-pit mines. This method can generate vehicle driving trajectories in real time while meeting the requirements of high-frequency road changes in open-pit mine scenarios, thus improving the generation rate of vehicle driving trajectories. The generated vehicle driving trajectories meet vehicle dynamics requirements and are easy for unmanned vehicles in the mining area to anticipate and track.

[0006] The present invention provides a method for generating the driving trajectory of an unmanned vehicle in an open-pit mine, comprising the following specific steps:

[0007] Step 1: Collect the road boundaries and intersection boundaries of the driving area in the mining area;

[0008] Step 2: Obtain the road area and intersection area of ​​the driving area in the mining area;

[0009] Step 3: In each road area, extract the road centerline based on the left and right road boundaries;

[0010] Step 4: Obtain the entry and exit points of the road area based on the road centerline and vehicle driving rules;

[0011] Step 5: Obtain the center lines of the right lane and the left lane based on the road boundary and the road center line, and use the center lines of the right lane and the left lane as the initial road trajectory of the autonomous vehicle.

[0012] Step 6: Based on the length, width, and minimum turning radius of the autonomous vehicle, the road boundary is used as a constraint condition, and the initial road trajectory of the autonomous vehicle is optimized using the sequential quadratic programming method to obtain the road driving trajectory.

[0013] Step 7: Generate the driving trajectory of the autonomous vehicle in the intersection area.

[0014] Optionally, in step 1, the driving area of ​​the mining area is surveyed and collected along the boundary of the driving area of ​​the mining area to generate the road boundary and intersection boundary of the mining area.

[0015] Optionally, in step 3, the method for extracting the road centerline is as follows: set multiple left nodes on the left road boundary of the road area and multiple right nodes on the right road boundary; perform one-to-one matching of the left nodes on the left road boundary and the right nodes on the right road boundary according to the principle of the shortest distance between corresponding nodes on two corresponding sides; connect the one-to-one matched left nodes and right nodes to obtain the road midpoint of the line; connect all the road midpoints sequentially to obtain the road centerline.

[0016] Optionally, in step 4, the right lane and the left lane are determined by the direction of the road centerline; the right lane direction and the left lane direction corresponding to the right lane and the left lane are determined; the starting direction and the ending direction of the corresponding lane are determined by the right lane direction and the left lane direction; and the entry point position and the exit point position of the lane are determined by the starting point and the ending direction of the lane.

[0017] Optionally, in step 5, multiple center nodes are set on the center line of each road area, and the right node of the right road boundary and the center node of the road center line are matched one-to-one according to the principle of the shortest distance between the corresponding nodes of the two corresponding sides; the right node of the right road boundary and the center node of the road center line are connected, and the midpoint of the right lane of the line is taken; all the obtained lane midpoints are connected sequentially to obtain the right road center line.

[0018] Optionally, in step 5, multiple center nodes are set on the center line of each road area, and the left nodes of the left road boundary and the center nodes of the road center line are matched one-to-one according to the principle of the shortest distance between the corresponding nodes of the two corresponding sides; the left nodes of the left road boundary and the center nodes of the road center line are connected, and the midpoint of the lane is taken; all the obtained left lane midpoints are connected sequentially to obtain the left road center line.

[0019] Optionally, in step 6, the minimum radius of curvature of the optimized road area driving trajectory is ensured to not exceed the minimum turning radius of the autonomous vehicle and to avoid collision with the boundary.

[0020] Optionally, in step 7, the boundary of the intersection area and the dividing line between the intersection area and the road area are determined; multiple topological points are determined on the dividing line, and it is determined whether the multiple topological points are entry points or exit points; entry points and exit points on the boundary lines of different intersection areas are selected, and the intersection area driving trajectory is connected by the entry points and exit points on the boundary lines of different intersection areas; the trajectory points on the intersection area driving trajectory are obtained.

[0021] Compared with the prior art, the present invention has at least the following beneficial effects:

[0022] (1) The method of this invention proposes a non-contact data acquisition method that does not require on-site positioning and measurement, or on-site tracking by the data acquisition vehicle or radar scanning. When collecting vehicle travel trajectories, the data acquisition vehicle does not need to enter the unmanned driving area beforehand, avoiding the danger of collision between the data acquisition system and the unmanned mining trucks in normal operation, and improving the safety of mining operations.

[0023] (2) The method of the present invention has strong anti-interference ability. Compared with the GPS data collection scheme, this scheme is not affected by weather factors such as rain and is not constrained by road conditions such as water accumulation and slippery roads, and can be used to collect data in most mining environments.

[0024] (3) The method of the present invention makes full use of constraints such as the terrain, boundary, vehicle width, and vehicle kinematic parameters of the mining area. Compared with the GPS acquisition scheme, the generated vehicle driving trajectory has better smoothness, no data drift, uniform point spacing, and strong environmental adaptability.

[0025] (4) The method of this invention does not require repeated path data collection. Existing GPS data collection methods and road marking data collection methods require re-collection when road conditions change. During the re-collection process, they are easily affected by external interference, making it difficult to guarantee real-time updates of vehicle travel paths. The method described in this patent does not require repeated path data collection and can update vehicle travel trajectories in real time.

[0026] (5) The method of the present invention does not require repeated collection paths, thus reducing the cost of collection equipment, labor costs and operation and maintenance costs. Attached Figure Description

[0027] The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of the invention.

[0028] Figure 1 This is a flowchart illustrating the process of generating the driving trajectory of an unmanned vehicle using the method of the present invention.

[0029] Figure 2 This is the closed boundary generated by the method of the present invention for the drivable area.

[0030] Figure 3 This is a schematic diagram of the drivable area after functional segmentation and extraction of the road centerline according to the method of the present invention.

[0031] Figure 4 The method of this invention determines the starting point and ending point of the vehicle's driving trajectory and extracts a schematic diagram of the lane centerline.

[0032] Figure 5 This is a schematic diagram illustrating the method of the present invention for generating and connecting vehicle road travel trajectories and intersection travel trajectories.

[0033] Figure 6 This is a schematic diagram illustrating the method for generating vehicle trajectories at intersections according to the present invention. Detailed Implementation

[0034] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0035] A specific embodiment of the present invention, such as Figure 1-6 A method for generating the driving trajectory of an unmanned vehicle in an open-pit mine is disclosed, including the following steps:

[0036] Step 1, as follows Figure 2 As shown, the boundaries of the working area, roads, and intersections of the entire mining area's driving area are collected.

[0037] The entire drivable area of ​​the mining area is mapped using lidar. A data collection vehicle is driven along the boundaries of the drivable area to collect data, generating the working area boundaries, road boundaries, and intersection boundaries. These boundaries constitute the closed boundaries of the entire drivable area of ​​the mining area. Multiple working area boundaries, road boundaries, and intersection boundaries are included.

[0038] Step 2, as follows Figure 3 As shown, the work area, road area, and intersection area of ​​the entire drivable area of ​​the mining area are obtained;

[0039] The work area includes loading area, unloading area and parking area; no driving trajectory is generated within the work area; multiple work areas, road areas and intersection areas are set up.

[0040] Step 3, as follows Figure 3 As shown, in each road region, the road centerline is extracted based on the left and right road boundaries of that road;

[0041] The method for extracting the road centerline is as follows:

[0042] Multiple left nodes are set on the left road boundary of each road area, and multiple right nodes are set on the right road boundary. The left nodes on the left road boundary and the right nodes on the right road boundary are matched one-to-one according to the principle of the shortest distance between the corresponding nodes on the two corresponding sides. The one-to-one matched left and right nodes are connected by a line, and the midpoint of the line is taken. All the obtained midpoints of the road are connected in sequence to obtain the road centerline.

[0043] Step 4, as follows Figure 4 As shown, the entry and exit points of the road area are obtained based on the road centerline and vehicle driving rules.

[0044] The right and left lanes are determined by the direction of the road centerline;

[0045] According to vehicle driving rules (such as right-hand drive rules), the right-hand and left-hand lanes are designated with corresponding right-hand and left-hand lane directions. The method for determining lane direction under right-hand drive rules is: the right-hand lane's direction is the same as the road center line, and the left-hand lane's direction is opposite to the road center line. The method for determining lane direction under left-hand drive rules is: the right-hand lane's direction is opposite to the road center line, and the left-hand lane's direction is the same as the road center line.

[0046] The starting and ending directions of the corresponding lanes are determined by the direction of the right lane and the direction of the left lane;

[0047] The entry and exit points of a lane are determined by the starting and ending directions of that lane. When the road width allows for simultaneous two-way traffic, the entry and exit points are located at 1 / 4 of the distance from the boundary line of the corresponding lane. When the road width does not allow for simultaneous two-way traffic, the entry and exit points are located at 1 / 2 of the distance from the boundary line of the corresponding lane.

[0048] Step 5, as follows Figure 4 As shown, the center lines of the right lane and the left lane are obtained based on the road boundary and the road center line, and the center lines of the right lane and the left lane are used as the initial road trajectory of the autonomous vehicle.

[0049] The specific method is as follows: Multiple center nodes are set along the centerline of each road area. A one-to-one match is performed between the right node of the right road boundary and the center node of the road centerline, based on the principle of finding the shortest distance between corresponding nodes on two corresponding sides. The matched right nodes of the right road boundary and the center node of the road centerline are then connected, and the midpoint of the lane along the connecting line is taken. All the obtained lane midpoints are then sequentially connected to form a line to obtain the right road centerline. The method for generating the left road centerline is the same as described above.

[0050] Step 6, as follows Figure 5 As shown, based on vehicle parameters such as the length, width, and minimum turning radius of the autonomous vehicle, the road boundary is used as a constraint. The Sequential Quadratic Programming (SQP) method is used to optimize the initial road trajectory of the autonomous vehicle, so that the minimum curvature radius of the optimized trajectory does not exceed the minimum turning radius of the vehicle and does not collide with the road boundary to obtain the driving trajectory in the road area.

[0051] Step 7, as follows Figure 5 As shown, the driving trajectory of the autonomous vehicle in the intersection area is generated.

[0052] For the driving trajectory in the intersection area, Figure 6 The following is an example to illustrate the method:

[0053] First, determine the boundaries of the intersection area and its dividing line with the road area. Then, on the dividing line between the intersection area and the road area, determine multiple topological points based on the entry and exit points obtained in step 4, and determine whether these topological points are entry or exit points according to the right-hand traffic rule. Next, select entry and exit points on the boundary lines of different intersection areas, and use a reed-sheep curve to fit the intersection area driving trajectory connecting these entry and exit points. Obtain the trajectory points on the intersection area driving trajectory. Finally, according to the separating axis theorem, perform boundary collision detection on the obtained trajectory point sequence. If no collision occurs, the initial intersection area driving trajectory is a collision-free trajectory. If a collision occurs, select control points and use a B-spline curve to fit the intersection area driving trajectory to obtain a collision-free trajectory.

[0054] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for generating the driving trajectory of an unmanned vehicle in an open-pit mine, characterized in that, The specific steps include the following: Step 1: Collect the road boundaries and intersection boundaries of the driving area in the mining area; Step 2: Obtain the road area and intersection area of ​​the driving area in the mining area; Step 3: In each road area, extract the road centerline based on the left and right road boundaries. The specific steps are as follows: The method for extracting the road centerline is as follows: set multiple left nodes on the left road boundary and multiple right nodes on the right road boundary of the road area; perform one-to-one matching of the left nodes on the left road boundary and the right nodes on the right road boundary according to the principle of the shortest distance between corresponding nodes on two corresponding sides; connect the one-to-one matched left nodes and right nodes to obtain the road midpoint of the line; connect all the road midpoints sequentially to obtain the road centerline. Step 4: Based on the road centerline and vehicle driving rules, obtain the entry and exit points of the road area, and determine the right and left lanes according to the direction of the road centerline. The specific steps are as follows: Determine the right lane direction and left lane direction corresponding to the right lane and left lane; determine the start direction and end direction of the corresponding lane based on the right lane direction and left lane direction; determine the entry point and exit point of the lane based on the start and end direction of the lane. Step 5: Obtain the center lines of the right and left lanes based on the road boundaries and center lines. Use these center lines as the initial road trajectory for the autonomous vehicle. The specific steps are as follows: Multiple center nodes are set on the centerline of each road area. The right nodes of the right road boundary and the center nodes of the road centerline are matched one-to-one according to the principle of the shortest distance between the corresponding nodes of the two corresponding sides. The right nodes of the right road boundary and the center nodes of the road centerline are connected by a line, and the midpoint of the right lane of the line is taken. All the obtained lane midpoints are connected in sequence to obtain the right road centerline. Multiple center nodes are set on the centerline of each road area. The left nodes of the left road boundary and the center nodes of the road centerline are matched one-to-one according to the principle of the shortest distance between the corresponding nodes of the two corresponding sides. The left nodes of the left road boundary and the center nodes of the road centerline are connected by a line, and the midpoint of the lane is taken. All the obtained left lane midpoints are connected in sequence to obtain the left road centerline. Step 6: Based on the length, width, and minimum turning radius of the autonomous vehicle, the road boundary is used as a constraint condition, and the initial road trajectory of the autonomous vehicle is optimized using the sequential quadratic programming method to obtain the road driving trajectory. The optimized road area driving trajectory has a minimum curvature radius that does not exceed the minimum turning radius of the autonomous vehicle and does not collide with the boundary. Step 7: Generate the driving trajectory of the autonomous vehicle in the intersection area.

2. The method for generating a driving trajectory according to claim 1, characterized in that, In step 1, the driving area of ​​the mining area is surveyed and collected along the boundary of the driving area of ​​the mining area to generate the road boundary and intersection boundary of the mining area.

3. The method for generating a driving trajectory according to claim 1, characterized in that, In step 7, the boundary of the intersection area and the dividing line between the intersection area and the road area are determined; multiple topological points are determined on the dividing line, and it is determined whether the multiple topological points are entry points or exit points; Select entry and exit points on the boundary lines of different intersection areas, connect the entry and exit points on the boundary lines of different intersection areas to obtain the intersection area driving trajectory; obtain the trajectory points on the intersection area driving trajectory.

Images