CN115071686A - Parking method for unmanned mining vehicle in long and narrow area - Google Patents
Parking method for unmanned mining vehicle in long and narrow area Download PDFInfo
- Publication number
- CN115071686A CN115071686A CN202210736123.6A CN202210736123A CN115071686A CN 115071686 A CN115071686 A CN 115071686A CN 202210736123 A CN202210736123 A CN 202210736123A CN 115071686 A CN115071686 A CN 115071686A
- Authority
- CN
- China
- Prior art keywords
- path
- point
- parking
- vehicle
- loading
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units, or advanced driver assistance systems for ensuring comfort, stability and safety or drive control systems for propelling or retarding the vehicle
- B60W30/06—Automatic manoeuvring for parking
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W60/00—Drive control systems specially adapted for autonomous road vehicles
- B60W60/001—Planning or execution of driving tasks
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Abstract
The invention discloses a parking method of an unmanned mining vehicle in a long and narrow area, which comprises the following steps: acquiring preset data comprising data of a global path, a global path boundary, a loading area boundary and a loading bit; calculating and obtaining a parking path planning starting point, a middle point, a terminal point and vehicle parameters; judging whether the loading area is a long and narrow area or not according to preprocessing acquired preset data; if the parking path is judged to be a long and narrow region, obtaining an initial parking path through a Reeds Shepp curve, hybrid A and a Dubins curve; connecting the parking path planning end point and the loading position by a straight line, and calculating discrete path points between the parking path planning end point and the loading position; carrying out smoothing treatment on the initial parking path, optimizing the curvature and the curvature change rate of the path, and obtaining an optimized parking path; and converting the parking path in the inertial coordinate system into the BLH coordinate system. The parking method is higher in flexibility and adaptability, lower in calculation complexity, better in track following effect and more reliable in parking position accuracy.
Description
Technical Field
The invention belongs to the technical field of unmanned driving, and particularly relates to a parking method of an unmanned mining vehicle in a long and narrow area in an unmanned operation scene of a strip mine.
Background
With the rapid development of sensor technology, unmanned technology is gradually mature, and has been pioneered to be applied in closed scenes such as mining areas and the like, and the modern change is made to the traditional mine production operation mode. Generally, the mining area production operation mainly comprises the processes of perforation, blasting, mining loading, transportation, dumping and the like, wherein the mining loading process is to excavate mineral materials by a forklift and control a mobile mechanical arm to load the mineral materials into a box groove of the mining vehicle on the basis that the mining vehicle runs to a loading position according to a planned path, and a parking method for ensuring that the mining vehicle accurately runs to the loading position is a key part of the whole mining loading process.
However, the boundary of the loading area of the mining area can move continuously along with the change of the mining face, and the positions of the forklift and the loading position are not fixed, so that the automatic driving mining vehicle cannot drive to the loading position according to a fixed path, dynamic path planning needs to be carried out in the loading area, a reasonable path is generated, and the mining vehicle can drive to the loading position along the planned path.
At present, a parking method for an unmanned mining vehicle can only realize autonomous parking of the mining vehicle in a simple scene loading area, and a feasible parking path is difficult to obtain in a complex scene loading area.
Patent application No. CN201910797242.0 discloses a parking method for an unmanned mining vehicle, which includes calculating an intersection point of an extension line of a loading position orientation where the unmanned mining vehicle is expected to park and a global path, using the intersection point as a starting point of a parking path, and combining a fixed motion process (including a fixed front wheel rotation angle and a fixed longitudinal speed), obtaining the parking path in two steps.
Patent application No. CN202110581107.X discloses a parking method for an unmanned mining vehicle, which comprises the steps of calculating end points of an extension line segment of a loading position orientation expected to be parked by the unmanned mining vehicle, expanding the end points, obtaining a group of point sets near the end points, using the point sets as alternative sets of gear switching points of a parking path, obtaining a whole parking path in two steps (a forward path and a backward path) by solving an inscribed circle of a triangle through traversing combination of a global path and the alternative sets, wherein the adaptability of path planning is low, the requirements of vehicle dynamics are not considered, and the requirements of path planning of complex road conditions cannot be met.
The patent with the patent application number CN201911421601.9 discloses a path planning method, which has strong solving capability by improving a steering switching penalty term and a steering penalty term added in a traditional hybrid a-algorithm and combining with a Reeds Shepp curve heuristic algorithm, and optimizes a path after solving. Although the calculation efficiency is improved compared with the pure mixed a-x algorithm, the absolute calculation efficiency is still low, and especially in a scene with a large area and a scene with complex boundary distribution, a long calculation time and large calculation resources are required. In addition, a feasible parking path (limited by computation time and computation resources) may be set out without any rules for special scenes such as a long and narrow loading area.
In summary, the prior art has at least the following problems: (1) through the preset fixed motion process and the preset parking path gear switching point, the calculation efficiency is high, but the flexibility and the adaptability of the parking path and the parking method are low; (2) by combining hybrid A and Reeds Shepp curves, the flexibility and adaptability are improved, but the overall calculation efficiency is low; (3) the prior art can not meet the parking requirements of unmanned mining vehicles in long and narrow areas and loading areas with special shapes.
Disclosure of Invention
In view of this, in order to solve the problem that the prior art cannot complete autonomous parking of a mining vehicle in a narrow, long and small loading area, the invention provides a parking method, which is applied to an unmanned mining vehicle and aims to: (1) the flexibility and the adaptability of a parking path and a parking method are improved; (2) on the basis of ensuring the flexibility and the adaptability, the calculation efficiency is improved; (3) the parking requirement of the unmanned mining vehicle in a narrow and narrow loading area with a complex scene and a strict boundary is met.
The specific scheme of the invention is as follows:
a method of parking an unmanned mining vehicle in an elongated area, comprising the steps of:
the first step is as follows: acquiring preset data, wherein the preset data at least comprises data of a global path, a global path boundary, a loading area boundary and a loading bit;
the second step is that: under an inertial coordinate system, calculating and obtaining a parking path planning starting point, a parking path planning middle point, a parking path planning terminal point and vehicle parameters;
the third step: according to the preprocessing judgment of the acquired preset data, whether the loading area is a long and narrow area is determined;
the fourth step: if the parking path is judged to be a long and narrow region, obtaining an initial parking path through a Reeds Shepp curve, hybrid A and a Dubins curve; if the parking path is judged to be a common area, obtaining an initial parking path through the prior art;
the fifth step: connecting a parking path planning end point and a loading position by a straight line, calculating a discrete path point between the parking path planning end point and the loading position, and adding the discrete path point to the initial parking path;
and a sixth step: carrying out smoothing treatment on the initial parking path, optimizing the curvature and the curvature change rate of the path, obtaining the optimized parking path and realizing path planning;
the seventh step: and converting the parking path in the inertial coordinate system into a BLH coordinate system to obtain the parking path in the BLH coordinate system.
Further, the global path boundary and the loading area boundary are obtained through a high-precision map, wherein a road driving area is arranged in the global path boundary and is a drivable area for driving and turning the vehicle; the loading area is a loading area in the boundary, and is a drivable area for executing turning and driving to the loading position before the vehicle loading operation; the loading position is obtained through the cloud platform, and the mine vehicle is parked in the loading position to load the material.
Further, the second step of calculating and obtaining the planned starting point, the middle point, the end point and the vehicle parameters of the parking path specifically comprises:
firstly, presetting dataThe global path is included, the boundary of the loading area is converted to an inertial coordinate system which takes the loading position as an origin and takes the orientation of the loading position as a y axis; starting from the end of the global path, at a fixed length l 1 (5 m < l) 1 Less than 20 meters) as step length, and carrying out discrete sampling along the global path to the starting point direction until the starting point of the global path to obtain a set of the starting points; acquiring the middle point of two end points of the continuous boundary line as a middle point through the continuous boundary of the loading area, wherein the course of the middle point is the normal direction of the connecting line of the two end points of the boundary line; the course precision of the vehicle after being parked in the loading position is not improved, and the loading position coordinate is moved towards the direction of the loading position by the length l 2 (2 m)<l 2 <10 meters) as endpoint; and acquiring the minimum turning radius and the vehicle body size data of the vehicle from the cloud platform.
Further, the third step of determining whether the loading area is a long and narrow area by performing preprocessing judgment on the acquired preset data specifically includes:
collision detection is carried out through the boundary of the loading area and an OBB bounding box of a space required by vehicle turning, and whether the loading area is a long and narrow area or not is judged; establishing a grid map with the resolution of (R + Lr) meters on the basis of an AABB bounding box formed by the boundary of the loading area, traversing the OBB bounding box with the boundary of the loading area and the loading position to perform collision detection, if a collision-free bounding box exists, judging that the loading area is not a long and narrow area, and if the collision-free bounding box does not exist, judging that the loading area is the long and narrow area; the length of the bounding box is 2 x (R + Lr), the width is R + L + Lf + W/2, R is the running radius of the corresponding rear wheel outside the vehicle when the front wheel steering angle of the vehicle reaches the maximum, L is the vehicle wheel base, Lf is the front overhang length of the vehicle, Lr is the rear overhang length of the vehicle, and W is the vehicle width.
Further, in the fourth step, if the loading area is a normal area, the initial parking path is obtained by connecting the starting point and the end point through a Reeds Shepp curve.
Further, in the fourth step, if the loading region is an elongated region, the method includes the following three steps:
(1) traversing the starting points in the starting point set by using a Reeds Shepp curve, connecting the starting points with the intermediate points by making the Reeds Shepp curve, ending traversing the starting point set until the starting point i in the starting point set is connected to the intermediate points by the Reeds Shepp curve, expecting to obtain a path connecting the starting points with the intermediate points, and after the vehicle can finish running out from the starting points on the path, turning around to reach the intermediate points;
(2) establishing a grid map with the resolution of 1m on the basis of an AABB bounding box formed by the boundary of a loading area, searching hybrid A, and expecting to obtain a path connecting an intermediate point to an end point until the path reaches a return path near the end point, or the search times reach a preset value and return failure, wherein the node expansion mode of hybrid A is only forward expansion;
(3) in the hybrid A searching process, node expansion is performed every n (1< n <20), 1 time of connecting the current node and the terminal of the hybrid A with a Dubins curve is tried, and if the hybrid A and the terminal of the hybrid A are successfully connected, the hybrid A and the Dubins path are searched and output, and the initial parking path is obtained.
Further, in the sixth step, the curvature and the curvature change rate of the path are constrained according to constraint conditions of the vehicle parameters in the sequence quadratic programming, and the objective function and the constraint conditions are specifically as follows:
an objective function: cost is cost 1 +cost 2 Wherein cost 1 Cost for smoothness 2 In order to offset the cost with respect to the original point,
in the formula, x i ,x i-1 ,x i+1 Respectively the abscissa of the ith optimization path point, the abscissa of the (i-1) th optimization path point, the abscissa of the (i + 1) th optimization path point, and x ref,i Is the abscissa, y, of the ith path point of the original path i ,y i-1 ,y i+1 Respectively is the ordinate of the ith optimization path point, the ith-1 optimization pathOrdinate of the point, ordinate of the i +1 th optimized Path Point, y ref,i The vertical coordinate of the ith path point of the original path is shown, and n is the number of the path points of the original path;
constraint conditions are as follows:
(1) displacement constraint from origin
In the formula, x l As a minimum value of a constraint on the lateral offset from the original path point, x u Is a maximum value, y, of a lateral offset constraint with respect to the original path point l Is a minimum value, y, of a longitudinal offset constraint with respect to the original path point u Is the maximum value of the longitudinal offset constraint relative to the original path point;
(2) maximum curvature constraint
(x i-1 +x i+1 -2×x i ) 2 +(y i-1 +y i+1 -2×y i ) 2 ≤(Δs 2 ×cur) 2
i＝0,1,2,……,n-1；
In the formula, Δ s is the distance between path points of the original path, and cur is the maximum curvature constraint;
(3) maximum curvature rate constraint
In the formula, x i+2 For the abscissa, y, of the i +2 th optimized path point i+2 For the ordinate of the i +2 th optimized path point, cur _ rate is the maximum curvature rate constraint.
Compared with the prior art, the invention has the following beneficial effects:
the method has the advantages that: the invention has higher flexibility and adaptability. (1) There are a total of 9 classes of 48 combinations of Reeds shepp curves, as shown in table 1; (2) there are 6 combinations of Dubins curves: LSL, RSR, RSL, LSR, RLR, LRL, where L represents motion from an arc in a counter-clockwise direction, R represents motion in an arc in a clockwise direction, and L represents motion in a straight line. (3) hybrid a is more diverse based on the number and manner of node extensions set. Compared with a planning method by presetting one or a group of motion processes, the Reeds Shepp curve, hybrid A curve and Dubins curve have higher flexibility and adaptability, and the comprehensive three modes have higher flexibility and adaptability.
TABLE 1 combination of Reeds shepp curves
The advantages are two: the invention has lower computational complexity. (1) The node expansion mode of hybrid A is modified from forward and backward expansion to forward direction expansion only, so that unnecessary backward expansion is reduced; (2) in the Hybrid A expansion process, the Dubins curve is used for trying to connect the current point and the planning terminal point, and compared with the RSP trial connection, the formula calculation times are reduced.
The advantages are three: the invention has better track following effect and more reliable parking position parking precision. (1) The curvature smoothing is carried out on the initial path by using the sequence quadratic programming, so that the curvature mutation condition is avoided; (2) the distance of the translation L1 of the loading position is used as a planning terminal point, so that the vehicle can run in a straight line as far as possible when approaching the loading position, and the position and the heading accuracy of parking and parking are improved.
Drawings
FIG. 1 is a flow chart of a parking method;
FIG. 2 is a schematic view of a loading zone; wherein 2(a) is a common loading region and 2(b) is an elongated loading region;
FIG. 3 is an OBB enclosure of the space required for vehicle turnaround;
FIG. 4 is a loading area crash detection with a vehicle turnaround bounding box;
FIG. 5 is a node expansion mode of Hybrid A @; wherein, 5(a) is a conventional hybrid A node expansion mode, and 5(b) is the hybrid A node expansion mode in the invention.
Detailed Description
The following description will be made in detail by way of a specific embodiment of a parking method for an unmanned mining vehicle in an elongated area according to the present invention.
The first step is as follows: and acquiring preset data, wherein the preset data comprises data of a global path, a global path boundary, a loading area boundary and a loading bit. The global path, the global path boundary and the loading area boundary are obtained through a high-precision map, wherein a road driving area is arranged in the global path boundary and is a drivable area which can be used for driving and turning around vehicles; the boundary of the loading area is a loading area which is a drivable area for executing turning and driving to the loading position before vehicle loading operation; the loading position is obtained through the cloud platform, and the mine vehicle can be loaded after being parked into the loading position.
The second step is that: and calculating and obtaining a planned starting point, a middle point, a terminal point and vehicle parameters of the parking path. Firstly, converting preset data including a global path and a loading area boundary into an inertial coordinate system with a loading position as an origin and a loading position orientation as a y axis; starting from the end point of the global path, carrying out discrete sampling along the global path in the reverse direction of 5m with a fixed length until the starting point of the global path is reached, and obtaining a set of the starting points; acquiring the middle point of two end points of the continuous boundary line as a middle point through the continuous boundary of the loading area, wherein the course of the middle point is the normal direction of the connecting line of the two end points of the boundary line; the course precision of the vehicle after being parked in the loading position is not improved, and the coordinate of the loading position is moved by a length of 10m along the direction of the loading position to be used as a terminal point; and acquiring the minimum turning radius and the vehicle body size data of the vehicle from the cloud platform.
The third step: and judging whether the loading area is a long and narrow area or not according to the preprocessing of the acquired preset data. Collision detection is carried out through the boundary of the loading area and an OBB bounding box of a space required by vehicle turning, and whether the loading area is a long and narrow area is judged; and establishing a grid map with the resolution of (R + Lr) m on the basis of an AABB bounding box formed by the boundary of the loading area, traversing the OBB bounding box through the grid map, the boundary of the loading area and the loading position for collision detection, judging that the loading area is not a long and narrow area if a bounding box without collision exists, and judging that the loading area is the long and narrow area if the bounding box without collision does not exist. The length of the bounding box is 2 x (R + Lr), the width is R + L + Lf + W/2, R is the running radius of the corresponding rear wheel outside the vehicle when the front wheel intersection of the vehicle reaches the maximum, L is the vehicle wheel base, Lf is the front overhang length of the vehicle, Lr is the rear overhang length of the vehicle, and W is the vehicle width. The pretreatment result comprises the following steps: the loading region is an elongated region.
The fourth step: and planning the parking path between the acquired starting point and the acquired end point. If the loading area is a common area, connecting a starting point and an end point through a Reeds Shepp curve to obtain an initial parking path; if the loading region is a narrow region, the method comprises the following three steps: (1) traversing the starting points in the starting point set by using a Reeds Shepp curve, connecting the starting points with the intermediate points by making the Reeds Shepp curve, ending traversing the starting point set until the starting point i in the starting point set is connected to the intermediate points by the Reeds Shepp curve, expecting to obtain a path connecting the starting points with the intermediate points, and after the vehicle can finish running out from the starting points on the path, turning around to reach the intermediate points; (2) establishing a grid map with the resolution of 1m on the basis of an AABB bounding box formed by a loading region boundary, searching hybrid A, expecting to obtain a path connecting an intermediate point to an end point until the path reaches a return path near the end point, or the searching times reach a preset value and the return fails, wherein the node expansion mode of hybrid A is forward expansion only; (3) in the hybrid A searching process, trying to connect the current node and the terminal of the hybrid A with a Dubins curve for 1 time every 5 times of node expansion, and if the hybrid A and the terminal are successful, ending the search and outputting the hybrid A and the Dubins path to obtain an initial parking path; it should be understood by those skilled in the art that the specific process of hybrid a search is not described in detail, and only the improvement part will be described.
The fifth step: connecting a parking path planning end point and a loading position by a straight line, dispersing a line segment between the parking path planning end point and the loading position at fixed length intervals, obtaining a dispersed path point, and adding the dispersed path point to an initial parking path;
and a sixth step: carrying out smoothing processing on the initial parking path, constraining curvature constraint and curvature change rate according to constraint conditions of vehicle parameters in sequence quadratic programming, optimizing the initial parking path, obtaining an optimized path, and realizing path planning; through the processing of discrete points of the parking path, the parking path is optimized, and those skilled in the art should understand that the specific process of the sequential quadratic programming is not described again, and only the objective function and the constraint condition are described:
an objective function: cost is defined as cost 1 +cost 2 Wherein cost 1 Cost for smoothness 2 Is the cost of offsetting from the original point.
In the formula, x i ,x i-1 ,x i+1 Respectively the abscissa of the ith optimization path point, the abscissa of the (i-1) th optimization path point, the abscissa of the (i + 1) th optimization path point, and x ref,i Is the abscissa, y, of the ith path point of the original path i ,y i-1 ,y i+1 Respectively the ordinate of the ith optimization path point, the ordinate of the (i-1) th optimization path point, the ordinate of the (i + 1) th optimization path point, y ref,i Is the ordinate of the ith path point of the original path, and n is the number of path points of the original path.
Constraint conditions are as follows:
(1) displacement constraint from origin
In the formula, x l As a minimum value of a constraint on the lateral offset from the original path point, x u Is a maximum value, y, of a lateral offset constraint with respect to the original path point l As a minimum value, y, of a longitudinal offset constraint with respect to the original waypoint u Is a relative original pathThe maximum value of the point longitudinal offset constraint.
(2) Maximum curvature constraint
(x i-1 +x i+1 -2×x i ) 2 +(y i-1 +y i+1 -2×y i ) 2 ≤(Δs 2 ×cur) 2
i＝0,1,2,……,n-1；
Where Δ s is the path point spacing of the original path and cur is the maximum curvature constraint.
(3) Maximum curvature rate constraint
In the formula, x i+2 For the abscissa, y, of the i +2 th optimized path point i+2 For the ordinate of the i +2 th optimized path point, cur _ rate is the maximum curvature rate constraint.
The seventh step: the path is converted to BLH (earth coordinate system) coordinate system. And converting the parking path points in the inertial coordinate system into a BLH coordinate system one by one, and outputting the BLH coordinate system as a planning result.
The above description is only for the purpose of illustrating the technical solutions of the present invention, but not for the purpose of limiting the same, and the scope of the present invention is not limited thereto, although the present invention is described in detail with reference to the foregoing embodiment examples. It should be noted that, for those skilled in the art, it is possible to make various improvements and modifications to the invention without departing from the principle of the invention, and those improvements and modifications also fall within the scope of the claims of the invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (7)
1. A method of parking an unmanned mining vehicle in an elongated area, comprising the steps of:
the first step is as follows: acquiring preset data, wherein the preset data at least comprises data of a global path, a global path boundary, a loading area boundary and a loading bit;
the second step: under an inertial coordinate system, calculating and obtaining a parking path planning starting point, a parking path planning middle point, a parking path planning end point and vehicle parameters;
the third step: according to the preprocessing judgment of the acquired preset data, whether the loading area is a long and narrow area is determined;
the fourth step: if the parking path is judged to be a long and narrow region, obtaining an initial parking path through a Reeds Shepp curve, hybrid A and a Dubins curve; if the parking path is judged to be a common area, obtaining an initial parking path through the prior art;
the fifth step: connecting a parking path planning end point and a loading position by a straight line, calculating a discrete path point between the parking path planning end point and the loading position, and adding the discrete path point to the initial parking path;
and a sixth step: carrying out smoothing treatment on the initial parking path, optimizing the curvature and the curvature change rate of the path, obtaining the optimized parking path and realizing path planning;
the seventh step: and converting the parking path in the inertial coordinate system into a BLH coordinate system to obtain the parking path in the BLH coordinate system.
2. The parking method for the unmanned mining vehicle in the long and narrow area as claimed in claim 1, wherein the global path, the global path boundary and the loading area boundary are obtained through a high-precision map, wherein a road driving area is arranged in the global path boundary and is a drivable area for driving and turning around the vehicle; the loading area is a loading area in the boundary, and is a drivable area for executing turning and driving to the loading position before the vehicle loading operation; the loading position is obtained through the cloud platform, and the mine vehicle is parked in the loading position to load the material.
3. The method for parking an unmanned mining vehicle in an elongated area according to claim 1, wherein in the second step, the calculating and obtaining of the planned starting point, the intermediate point, the end point of the parking path and the vehicle parameters specifically comprises:
first, preset data is included into the wholeThe path and the boundary of the loading area are converted to an inertial coordinate system which takes the loading position as an origin and takes the orientation of the loading position as a y axis; starting from the end of the global path, at a fixed length l 1 Step length, 5m < l 1 If the distance is less than 20 meters, carrying out discrete sampling along the global path to the starting point direction until the starting point of the global path is reached to obtain a set of the starting points; acquiring the middle point of two end points of the continuous boundary line as a middle point through the continuous boundary of the loading area, wherein the course of the middle point is the normal direction of the connecting line of the two end points of the boundary line; the course precision of the vehicle after being parked in the loading position is not improved, and the loading position coordinate is moved towards the direction of the loading position by the length l 2 2 m<l 2 <10 meters as end point; and acquiring the minimum turning radius and the vehicle body size data of the vehicle from the cloud platform.
4. The parking method for the unmanned mining vehicle in the long and narrow area according to claim 1, wherein in the third step, according to the preprocessing judgment of the acquired preset data, the determination of whether the loading area is the long and narrow area is specifically:
collision detection is carried out through the boundary of the loading area and an OBB bounding box of a space required by vehicle turning, and whether the loading area is a long and narrow area is judged; establishing a grid map with the resolution of (R + Lr) meters on the basis of an AABB bounding box formed by the boundary of the loading area, traversing the OBB bounding box with the boundary of the loading area and the loading position to perform collision detection, if a collision-free bounding box exists, judging that the loading area is not a long and narrow area, and if the collision-free bounding box does not exist, judging that the loading area is the long and narrow area; the length of the bounding box is 2 x (R + Lr), the width is R + L + Lf + W/2, R is the running radius of the corresponding rear wheel outside the vehicle when the front wheel steering angle of the vehicle reaches the maximum, L is the vehicle wheel base, Lf is the front overhang length of the vehicle, Lr is the rear overhang length of the vehicle, and W is the vehicle width.
5. The method for parking an unmanned mining vehicle in an elongated region, according to claim 1, wherein in the fourth step, if the loading region is a normal region, an initial parking path is obtained by connecting a starting point and an end point through a Reeds Shepp curve.
6. A method of parking an elongated area of unmanned mining vehicle, according to claim 1, wherein said fourth step, if the loading area is an elongated area, is performed by the following three steps:
(1) traversing the starting point in the starting point set by using a Reeds Shepp curve, connecting the starting point with the middle point by using the Reeds Shepp curve until the starting point i in the starting point set is connected to the middle point by using the Reeds Shepp curve, finishing traversing the starting point set, expecting to obtain a path connecting the starting point with the middle point, and turning around to reach the middle point after the vehicle can finish driving out from the starting point on the path;
(2) establishing a grid map with the resolution of 1m on the basis of an AABB bounding box formed by a loading region boundary, searching hybrid A, expecting to obtain a path connecting an intermediate point to an end point until the path reaches a return path near the end point, or the searching times reach a preset value and the return fails, wherein the node expansion mode of hybrid A is forward expansion only;
(3) in the hybrid A searching process, every n times of node expansion is carried out, 1< n <20, 1 time of connecting the current node and the terminal of the hybrid A with a Dubins curve is tried, if the hybrid A and the terminal are successful, the hybrid A and the Dubins path is searched and output, and the initial parking path is obtained.
7. The method for parking an unmanned mining vehicle in an elongated region according to claim 1, wherein in the sixth step, the curvature and the curvature change rate of the path are constrained according to constraint conditions of vehicle parameters in sequence quadratic programming, and the objective function and constraint conditions are specifically as follows:
an objective function: cost is defined as cost 1 +cost 2 Wherein cost 1 Cost for smoothness 2 In order to offset the cost with respect to the original point,
in the formula, x i ，x i-1 ，x i+1 Respectively are the abscissa of the ith optimization path point, the abscissa of the (i-1) th optimization path point, the abscissa of the (i + 1) th optimization path point, and x ref，i Is the abscissa, y, of the ith path point of the original path i ，y i-1 ，y i+1 Respectively the ordinate of the ith optimized path point, the ordinate of the (i-1) th optimized path point, the ordinate of the (i + 1) th optimized path point, y ref，i The vertical coordinate of the ith path point of the original path, and n is the number of the path points of the original path;
constraint conditions are as follows:
(1) displacement constraint from origin
In the formula, x l As a minimum value, x, of a lateral offset constraint with respect to the original waypoint u Is a maximum value, y, of a lateral offset constraint with respect to the original path point l Is a minimum value, y, of a longitudinal offset constraint with respect to the original path point u Is the maximum value of the longitudinal offset constraint relative to the original path point;
(2) maximum curvature constraint
(x i-1 +x i+1 -2×x i ) 2 +(y i-1 +y i+1 -2×y i ) 2 ≤(Δs 2 ×cur) 2
i＝0，1，2，……，n-1；
In the formula, Δ s is the distance between path points of the original path, and cur is the maximum curvature constraint;
(3) maximum curvature rate constraint
In the formula, x i+2 For the abscissa, y, of the i +2 th optimized path point i+2 For the ordinate of the i +2 th optimized path point, cur _ rate is the maximum curvature rate constraint.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210736123.6A CN115071686A (en) | 2022-06-15 | 2022-06-15 | Parking method for unmanned mining vehicle in long and narrow area |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210736123.6A CN115071686A (en) | 2022-06-15 | 2022-06-15 | Parking method for unmanned mining vehicle in long and narrow area |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115071686A true CN115071686A (en) | 2022-09-20 |
Family
ID=83254774
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210736123.6A Pending CN115071686A (en) | 2022-06-15 | 2022-06-15 | Parking method for unmanned mining vehicle in long and narrow area |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115071686A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116380086A (en) * | 2023-03-28 | 2023-07-04 | 安徽海博智能科技有限责任公司 | Unmanned mining card track planning method based on drivable area |
| CN116442992A (en) * | 2023-06-15 | 2023-07-18 | 广汽埃安新能源汽车股份有限公司 | Parking control method and device |
| CN117826825A (en) * | 2024-02-29 | 2024-04-05 | 苏州观瑞汽车技术有限公司 | Unmanned mining card local path planning method and system based on artificial potential field algorithm |
-
2022
- 2022-06-15 CN CN202210736123.6A patent/CN115071686A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116380086A (en) * | 2023-03-28 | 2023-07-04 | 安徽海博智能科技有限责任公司 | Unmanned mining card track planning method based on drivable area |
| CN116442992A (en) * | 2023-06-15 | 2023-07-18 | 广汽埃安新能源汽车股份有限公司 | Parking control method and device |
| CN116442992B (en) * | 2023-06-15 | 2023-09-05 | 广汽埃安新能源汽车股份有限公司 | Parking control method and device |
| CN117826825A (en) * | 2024-02-29 | 2024-04-05 | 苏州观瑞汽车技术有限公司 | Unmanned mining card local path planning method and system based on artificial potential field algorithm |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115071686A (en) | Parking method for unmanned mining vehicle in long and narrow area | |
| US11460311B2 (en) | Path planning method, system and device for autonomous driving | |
| CN104977933A (en) | Regional path tracking control method for autonomous land vehicle | |
| CN111026133A (en) | Path planning method, vehicle and computer readable medium | |
| CN112606830B (en) | Two-section type autonomous parking path planning method based on mixed A-algorithm | |
| CN113721637B (en) | Intelligent vehicle dynamic obstacle avoidance path continuous planning method and system and storage medium | |
| CN113978452B (en) | Automatic parallel parking path planning method | |
| CN111538331A (en) | Reaction type navigation method of underground unmanned articulated vehicle | |
| CN111006667A (en) | Automatic driving track generation system under high-speed scene | |
| CN116185014A (en) | Intelligent vehicle global optimal track planning method and system based on dynamic planning | |
| CN113895463B (en) | Path planning method suitable for turning around of automatic driving vehicle | |
| CN113670305A (en) | Parking trajectory generation method and device, computer equipment and storage medium | |
| CN114162113A (en) | Parallel parking path planning method combined with multi-section curve | |
| CN115077553A (en) | Method, system, automobile, equipment and medium for planning track based on grid search | |
| CN112706760B (en) | Unmanned parking path planning method for special road scene | |
| CN111857148B (en) | Unstructured road vehicle path planning method | |
| CN114763133A (en) | Vehicle parking planning method, device, equipment and computer storage medium | |
| CN115840454B (en) | Multi-vehicle track collaborative planning method and device for unstructured road conflict area | |
| CN113914406B (en) | Electric loader path tracking system and control method | |
| CN115837904A (en) | Parking planning method for four-wheel steering vehicle in mining area | |
| CN113741463B (en) | Fixed-point and directional parking control method and system for unmanned ground maneuvering platform | |
| CN116360420A (en) | CLF-CBF optimized unmanned vehicle motion planning control method | |
| CN114877904A (en) | Road curvature estimation method, road curvature measurement device and computer equipment | |
| Chen et al. | Research and application of improved pure pursuit algorithm in low-speed driverless vehicle system | |
| Qin et al. | A feasible parking algorithm in form of path planning and following |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |