Path planning method and device for autonomous driving vehicle considering uniform road compaction

Abstract

The present invention discloses a method and device for autonomous driving vehicle path planning considering uniform road rolling, which comprises: step 1, determining a road section that needs to be rolled according to mining area road information as a road section to be offset; step 2, calculating the offset l of the i-th path point according to the index information of the road section to be offset in the original reference path using formula (1) i :l i =w i ·l(1) In the formula, l is the offset of the local task, w i is the offset weight coefficient for the i-th path point. Its setting principle is as follows: the closer to the center of the road section to be offset, the closer the offset of each path point is to the preset offset value; the closer to the ends of the road section to be offset, the smaller the offset of each path point is, gradually approaching 0, thereby controlling the smooth connection between the offset path and the path before and after the offset. This invention can efficiently and effectively achieve a smooth connection between the offset path and the path before and after the offset at the breakpoint.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving technology in mines, and in particular to a method and device for path planning of an autonomous driving vehicle taking uniform road compaction into consideration. Background Art

[0002] In the field of autonomous driving in mines, the driving paths of mine trucks are mostly fixed routes collected in advance. Uniform compaction technology for autonomous driving in mines refers to the technology in which the mine autonomous driving system performs a lateral offset on the previously collected fixed driving path and ensures that the path is safe and operational. This technology can plan a path for the mining truck (short for mining truck) that is similar to but not completely overlaps with the pre-collected fixed path in each mission. The system calls the mining truck to run transportation tasks multiple times and sets different offsets for each task, ultimately achieving uniform compaction of the mining area transportation roads. This technology can prevent mining trucks from driving along fixed paths for a long time, thereby avoiding the formation of deep rutting marks. This can not only reduce the misdetection of road conditions by the vehicle-side perception system, but also maintain the soil transportation roads in the mining area, which is of great significance to ensuring the safe driving of mining trucks.

[0003] The uniform compaction technology of autonomous mining is mainly achieved by offsetting the pre-recorded transportation path to varying degrees along the normal direction of the tangent of each path point. This is also a common solution adopted by autonomous mining companies at home and abroad.

[0004] Currently, the prior art discloses a road uniform rolling solution, which has the following disadvantages:

[0005] 1. The technical solution design is relatively redundant. In the YUGONG parallel mining operating system, the remote end is responsible for issuing the original driving path, while the uniform rolling technology is implemented on the vehicle end. After receiving the driving path sent from the cloud, the vehicle end offsets the entire path (excluding the unloading area path). To address the problem of path breaks caused by path offset, the planning module will perform additional planning at the path breaks to calculate a smooth transition path.

[0006] 2. Insufficient considerations. In this solution, the vehicle management system applies an overall offset to the route sent from the cloud, but does not offset the route in the loading and unloading areas. In a real mining environment, in addition to loading and unloading areas, transport roads also include weighing sections, car wash sections, cement roads, and steep downhill sections. These sections do not require route offsets, and mining trucks must strictly follow the pre-recorded routes. Summary of the Invention

[0007] The purpose of the present invention is to provide a method and device for autonomous driving vehicle path planning that takes into account uniform road compaction, which can efficiently and in one step achieve a smooth connection between the offset path and the pre-offset path at the break point.

[0008] To achieve the above objectives, the present invention provides a path planning method for an autonomous driving vehicle taking into account uniform road compaction, which comprises:

[0009] Step 1: According to the mining area road information, determine the road section that needs to be rolled as the road section to be offset;

[0010] Step 2: Based on the index information of the road segment to be offset in the original reference path, use formula (1) to calculate the offset l of the i-th path point i , i=1,2,3,...n, n represents the total number of path points on the road segment to be offset:

[0011] l i =w i ·l (1)

[0012] Where l is the offset of the local task, w i is the offset weight coefficient of the i-th path point. Its setting principles include: the closer to the center of the road section to be offset, the closer the offset of each path point is to the preset offset value; the closer to the two ends of the road section to be offset, the smaller the offset of each path point is, gradually tending to 0, thereby controlling the smooth connection between the path after offset and the path before offset.

[0013] Furthermore, w i Set as formula (2):

[0014]

[0015] Where σ is a preset constant, and x is a temporary variable used to ensure the symmetry between the leftmost and rightmost ends of each path. The expression is as shown in formula (3):

[0016]

[0017] Furthermore, l is set as formula (4):

[0018] l=λ·L max (4)

[0019] Where, L max is the maximum safe offset allowed at the mining site, and λ is a randomly generated parameter.

[0020] Furthermore, before step 2, the method further includes: increasing the interval between two adjacent path points of the road section to be offset, and downsampling the path points;

[0021] After step 2, the method also includes: filling the gap between two adjacent path points after the offset by an interpolation method, restoring the offset path to the preset interval distance between two adjacent path points before the offset; and detecting the curvature and safety of the offset path.

[0022] Furthermore, the road sections that need to be rolled include transportation roads, weighing and car washing sections, and reversing sections for fixed-point unloading.

[0023] The present invention also provides a path planning device for an autonomous driving vehicle taking into account uniform road compaction, comprising:

[0024] A unit for acquiring information of a road section to be shifted, which is used to determine a road section that needs to be compacted based on the road information in the mining area as the road section to be shifted;

[0025] The offset calculation unit is used to calculate the offset l of the i-th path point according to the index information of the road section to be offset in the original reference path using formula (1) i , i=1,2,3,...n, n represents the total number of path points on the road segment to be offset:

[0026] l i =w i ·l (1)

[0027] Where l is the offset of the current task of the mining card, w i is the offset weight coefficient of the i-th path point. Its setting principles include: the closer to the center of the road section to be offset, the closer the offset of each path point is to the preset offset value; the closer to the two ends of the road section to be offset, the smaller the offset of each path point is, gradually tending to 0, thereby controlling the smooth connection between the path after offset and the path before offset.

[0028] Furthermore, w i Set as formula (2):

[0029]

[0030] Where σ is a preset constant, and x is a temporary variable used to ensure the symmetry between the leftmost and rightmost ends of each path. The expression is as shown in formula (3):

[0031]

[0032] Furthermore, l is set as formula (4):

[0033] l=λ·L max (4)

[0034] Where, L max is the maximum safe offset allowed at the mining site, and λ is a randomly generated parameter.

[0035] Furthermore, the autonomous driving vehicle path planning device considering uniform road compaction further includes:

[0036] The downsampling unit is used to increase the interval between two adjacent path points before the offset of the road section to be offset, and downsample the path points.

[0037] An interpolation unit, which is used to fill the gap between two adjacent path points after the offset by an interpolation method, and restore the offset path to the preset interval distance between two adjacent path points before the offset;

[0038] The safety detection unit is used to detect the curvature and safety of the offset path.

[0039] Furthermore, the road sections that need to be rolled include transportation roads, weighing and car washing sections, and reversing sections for fixed-point unloading.

[0040] The present invention adopts the above technical solution, so it has the following advantages:

[0041] 1. Since the present invention reasonably controls the offset of each path point based on the weight function according to the driving path pre-recorded in the mining area and the actual situation on site of the mining area, it can not only efficiently and evenly compact the road, but also achieve a smooth connection between the path after offset and the path before offset at the break point in one step.

[0042] 2. Since the uniform compaction technology proposed in the present invention is achieved by lateral offsetting a pre-recorded fixed path, it can prevent the accumulation of rutting marks, thereby protecting mining roads. It also reduces the false detection of road gullies by the on-board perception system, which is crucial for the safe driving of mining trucks in autonomous driving scenarios.

[0043] 3. Since the uniform rolling scheme proposed in the present invention addresses the problem of path breakage caused by path deviation, it reasonably controls the offset of each path point based on a weight function, thereby achieving smooth connection at the expected breakage point. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is a flowchart of a method for path planning for an autonomous driving vehicle that takes uniform road compaction into consideration according to an embodiment of the present invention.

[0045] Figure 2 This is a schematic diagram of offset section selection according to an embodiment of the present invention.

[0046] Figure 3 Schematic diagram of offset point selection according to an embodiment of the present invention.

[0047] Figure 4 Schematic diagram of the offset direction of an embodiment of the present invention.

[0048] Figure 5 Schematic diagram of path breakage at a reversing point according to an embodiment of the present invention.

[0049] Figure 6 This is the weight function image of an embodiment of the present invention.

[0050] Figure 7 Schematic diagram of the overall offset effect of an embodiment of the present invention.

[0051] Figure 8 Schematic diagram of the left-end offset effect of an embodiment of the present invention, wherein the white line is the reference path and the red line is the offset path.

[0052] Figure 9 Schematic diagram of the offset effect at the right end (at the reversing point) of an embodiment of the present invention, wherein the white line is the reference path and the red line is the offset path.

[0053] Figure 10 This is a schematic diagram of the offset effect of the middle section of the embodiment of the present invention, wherein the white line is the reference path and the red line is the offset path.

[0054] Figure 11 This is a distribution diagram of path curvature after migration according to an embodiment of the present invention.

[0055] Figure 12 This is a curvature distribution diagram of a partial path after migration according to an embodiment of the present invention. DETAILED DESCRIPTION

[0056] In the accompanying drawings, the same or similar reference numerals are used to represent the same or similar elements or elements with the same or similar functions. The embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0057] In the description of the present invention, the terms "center", "longitudinal", "lateral", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like to indicate directions or positional relationships based on the directions or positional relationships shown in the accompanying drawings. They are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific direction, be constructed and operated in a specific direction. Therefore, they should not be understood as limiting the scope of protection of the present invention.

[0058] The method for autonomous driving vehicle path planning considering uniform road compaction provided by an embodiment of the present invention includes:

[0059] Step 1: Determine the road sections that need to be rolled based on the mining area road information as the sections to be offset. In this embodiment, based on the actual road properties of the mining area, only the sections that need to be evenly rolled are evenly rolled for the entire transportation route, while other sections are not evenly rolled.

[0060] like Figure 2 As shown in the diagram, conventional haul roads, weighing and car wash sections, and reverse sections for fixed-point unloading are all road sections that require compaction. The uniform compaction method of this embodiment combines the actual conditions of the mining area to efficiently achieve the goal of uniform compaction of mining area roads. The reverse path for fixed-point unloading, the car wash point, and the path near the weighing point are road sections that do not require compaction.

[0061] Step 2: Based on the index information of the road segment to be offset in the original reference path, use formula (1) to calculate the offset l of the i-th path point i , i=1,2,3,...n, n represents the total number of path points on the road segment to be offset:

[0062] l i =w i ·l (1)

[0063] Where l is the offset of the current task of the mining card, w i is the offset weight coefficient of the i-th path point. Its setting principles include: the closer to the center of the road section to be offset, the closer the offset of each path point is to the preset offset value; the closer to the two ends of the road section to be offset, the smaller the offset of each path point is, gradually tending to 0, thereby controlling the smooth connection between the path after offset and the path before offset.

[0064] This embodiment reasonably controls the offset of each path point based on a weight function, and achieves smooth connection between the path after offset and the path before offset at the break point in one step.

[0065] like Figure 3 As shown in the figure, assume that the path composed of black dots is a certain path segment on the road segment to be offset. The black dots are path points on the original reference path. This embodiment of the present invention uses a uniform sampling method, starting from the first point of each road segment to be offset, and selecting a point every 1 meter, as shown by the orange dots in the figure.

[0066] like Figure 4 As shown, taking the right offset as an example, for each original path point, the blue arrow is its tangent direction, and the orange arrow is its normal direction, which is the offset direction of the path point.

[0067] w i As a preferred embodiment, in order to solve the problem that the reference path after the shift and the reference path before the shift are broken at the junction, as shown in FIG. Figure 5 As shown, w i The preferred setting is formula (2):

[0068]

[0069] Where σ is a preset constant, and x is a temporary variable used to ensure the symmetry between the leftmost and rightmost ends of each path. The expression is as shown in formula (3):

[0070]

[0071] Among them, n represents the total number of candidate points on the current reference path, i represents the index of the candidate point, then when σ is set to 5 and n is 60, the weight function w i Images such as Figure 6 As shown in the figure, in the middle section, the function takes the value of 1, indicating that the middle waypoints are offset according to the original offset. At the ends of the path segment, the weight function gradually approaches 0, which makes the optimization weight values ​​of the corresponding indexes of the starting point and end point of the path to be offset very small. This ensures that the offset values ​​of the two end points on a path are smaller and closer to 0 as they move towards the ends. This can avoid the break between the reference path after offset and the reference path before offset at the intersection, thereby effectively maintaining the continuity of the mining truck path.

[0072] Of course, w can also be changed according to the road attributes of the road section to be offset. i The value of is set to a fixed value. You can also add or reduce the "1.25" in the denominator of formula (2) to obtain w based on formula (2). i We can even transform the entire denominator of formula (2) to obtain w i .

[0073] In one embodiment, based on L max To determine the specific offset l for each task, set l to formula (4):

[0074] l=λ·L max (4)

[0075] Where, L max The maximum safety offset allowed at the mining site is read in advance from the configuration file. λ is a randomly generated parameter, for example: λ∈{-1,-0.75,-0.5,-0.25,0,0.25,0.5,0.75,1.0}.

[0076] Of course, the value of the offset l can also be set according to the road attributes of the road section to be offset. It is also possible to transform the formula (4) based on the formula (4) to obtain l.

[0077] In one embodiment, Figure 1As shown, the autonomous driving vehicle path planning method considering uniform road compaction further includes, before step 2, increasing the distance between adjacent path points on the road section to be offset and downsampling the path points. For example, the downsampling of the road section to be offset is performed. In this example, the distance between two adjacent path points on the original reference path is 10 cm, and the distance between two adjacent path points after downsampling is 1 meter.

[0078] After step 2, it also includes: filling the gaps between two adjacent path points after the offset by an interpolation method, and restoring the offset path to the preset interval distance between two adjacent path points before the offset. Because the path points after the offset are too sparse and do not meet the requirements of vehicle operation, this embodiment uses an interpolation method based on a cubic spline curve to fill the gaps between adjacent points, and restores the offset path to an interval of 10 cm between adjacent points. The cubic parameter spline interpolation scheme adopted in this example is a specific parameter selected as the distance s along the path, i.e., the x(s) cubic spline interpolation and the y(s) cubic spline interpolation. This scheme can ensure the continuity of the curvature curve.

[0079] After step 2, it also includes: in order to ensure the safety of the offset path, it is necessary to test the curvature and safety of the offset path. Safety detection involves collision detection between each offset path point and the map boundary, and checking whether the curvature of each path point meets the maximum steering limit of the vehicle. Among them, the collision detection is carried out by building a box with each path point as the center, combining the vehicle's geometric model and safety threshold, and checking whether each map boundary point intersects with the box one by one. Curvature detection mainly determines whether the curvature of each point is greater than a preset threshold, which is usually determined by the minimum turning radius of the vehicle. Safety detection determines whether the offset path exceeds the map boundary.

[0080] In one embodiment, the road sections that need to be compacted include transport roads, weighing and car washing sections, and reverse sections for fixed-point unloading.

[0081] An embodiment of the present invention further provides a path planning device for an autonomous driving vehicle that takes into account uniform road compaction, characterized in that it includes a unit for acquiring information about a road section to be offset and a unit for calculating an offset, wherein:

[0082] The road section information acquisition unit to be offset is used to determine the road section that needs to be rolled according to the mining area road information as the road section to be offset.

[0083] The offset calculation unit is used to calculate the offset l of the i-th path point according to the index information of the road section to be offset in the original reference path using formula (1): i .

[0084] In one embodiment, the autonomous driving vehicle path planning device considering uniform road compaction further includes a downsampling unit, an interpolation unit, and / or a safety detection unit, wherein:

[0085] The downsampling unit is used to increase the interval between two adjacent path points before the offset of the road section to be offset, and downsample the path points.

[0086] The interpolation unit is used to fill the gap between two adjacent path points after the offset by using an interpolation method, and restore the offset path to the preset interval distance between two adjacent path points before the offset.

[0087] The safety detection unit is used to detect the curvature and safety of the offset path.

[0088] The following is the simulation effect of the above embodiment:

[0089] Figures 7 to 10 This is the effect diagram of the planned path being offset 0.19m to the left compared to the reference path, where the grid resolution behind it is 1m / pix. Figure 11 and Figure 12 The corresponding curvature distribution diagram is shown in Figure 2. It can be seen that the curvature of the interpolated points changes continuously, and the maximum curvature does not exceed the maximum turning limit of the vehicle. In this example, the minimum turning radius of the vehicle is 12m.

[0090] Finally, it should be noted that the above embodiments are intended only to illustrate the technical solutions of the present invention and are not intended to limit them. Those skilled in the art will appreciate that the technical solutions described in the aforementioned embodiments may be modified, or some of the technical features thereof may be replaced with equivalents; such modifications or replacements do not deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A path planning method for an autonomous driving vehicle considering uniform road compaction, characterized in that: include: Step 1: According to the mining area road information, determine the road section that needs to be rolled as the road section to be offset; Step 2: Based on the index information of the road segment to be offset in the original reference path, use formula (1) to calculate the offset l of the i-th path point i , i=1,2,3,...n, n represents the total number of path points on the road segment to be offset: l i ＝w i ·l(1) Where l is the offset of the local task, w i is the offset weight coefficient of the i-th path point. Its setting principle is as follows: the closer to the center of the road section to be offset, the closer the offset of each path point is to the preset offset value; the closer to the ends of the road section to be offset, the smaller the offset of each path point is, gradually tending to 0, thereby ensuring a smooth connection between the path after offset and the path before offset; Will w i Set as formula (2): Where σ is a preset constant, and x is a temporary variable used to ensure the symmetry between the leftmost and rightmost ends of each path. The expression is as shown in formula (3):

2. The method for autonomous driving vehicle path planning considering uniform road compaction according to claim 1, wherein: Set l as formula (4): l=λ·L max (4) Where, L max is the maximum safe offset allowed at the mining site, and λ is a randomly generated parameter.

3. The method for autonomous driving vehicle path planning considering uniform road compaction according to claim 1 or 2, characterized in that: Before step 2, the method further includes: increasing the interval between two adjacent path points of the road section to be offset and downsampling the path points; After step 2, the method also includes: filling the gap between two adjacent path points after the offset by an interpolation method, restoring the offset path to the preset interval distance between two adjacent path points before the offset; and detecting the curvature and safety of the offset path.

4. The method for autonomous driving vehicle path planning considering uniform road compaction according to claim 3, wherein: The road sections that need to be compacted include transportation roads, weighing and car washing sections, and reversing sections for fixed-point unloading.

5. A path planning device for an autonomous driving vehicle that takes into account uniform road compaction, characterized in that: include: A unit for acquiring information of a road section to be shifted, which is used to determine a road section that needs to be compacted based on the road information in the mining area as the road section to be shifted; The offset calculation unit is used to calculate the offset l of the i-th path point according to the index information of the road section to be offset in the original reference path using formula (1) i , i=1,2,3,...n, n represents the total number of path points on the road segment to be offset: l i ＝w i ·l(1) Where l is the offset of the current task of the mining card, w i is the offset weight coefficient of the i-th path point. Its setting principle is as follows: the closer to the center of the road section to be offset, the closer the offset of each path point is to the preset offset value; the closer to the ends of the road section to be offset, the smaller the offset of each path point is, gradually tending to 0, thereby ensuring a smooth connection between the path after offset and the path before offset; Will w i Set as formula (2): Where σ is a preset constant, and x is a temporary variable used to ensure the symmetry between the leftmost and rightmost ends of each path. The expression is as shown in formula (3):

6. The autonomous driving vehicle path planning device considering uniform road compaction according to claim 5, characterized in that: Set l as formula (4): l=λ·L max (4) Where, L max is the maximum safe offset allowed at the mining site, and λ is a randomly generated parameter.

7. The autonomous driving vehicle path planning device considering uniform road compaction according to claim 5 or 6, characterized in that: Also includes: The downsampling unit is used to increase the interval between two adjacent path points before the offset of the road section to be offset, and downsample the path points. An interpolation unit, which is used to fill the gap between two adjacent path points after the offset by an interpolation method, and restore the offset path to the preset interval distance between two adjacent path points before the offset; The safety detection unit is used to detect the curvature and safety of the offset path.

8. The autonomous driving vehicle path planning device considering uniform road compaction according to claim 7, characterized in that: The road sections that need to be compacted include transportation roads, weighing and car washing sections, and reversing sections for fixed-point unloading.

Images