CN111256700B - Edge narrowing planning method for planning operation path of automatic driving agricultural machine - Google Patents
Edge narrowing planning method for planning operation path of automatic driving agricultural machine Download PDFInfo
- Publication number
- CN111256700B CN111256700B CN202010225665.8A CN202010225665A CN111256700B CN 111256700 B CN111256700 B CN 111256700B CN 202010225665 A CN202010225665 A CN 202010225665A CN 111256700 B CN111256700 B CN 111256700B
- Authority
- CN
- China
- Prior art keywords
- farmland
- node
- planning
- path
- point
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/20—Instruments for performing navigational calculations
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Automation & Control Theory (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
- Guiding Agricultural Machines (AREA)
Abstract
The invention discloses a trimming planning method for planning an operation path of an automatic driving agricultural machine, which comprises the steps of initializing a farmland model, and acquiring farmland boundary data, a farmland entrance position and a farmland exit position; classifying the farmland models according to farmland boundary data in the farmland models; determining the main shaft direction of the agricultural machine in field operation according to the farmland boundary data and the farmland type in the farmland model, dividing the farmland boundary into an operation area and an area to be transferred; planning a full-coverage operation path in the operation area, and returning to the starting position and the ending position of the full-coverage operation path in the operation area; matching an edge folding mode according to the segmentation result in the step 3, and planning and designing an edge folding path in the area to be transferred; the invention is used for the edge folding operation of the automatic driving agricultural machine, and reduces the land blocks which are difficult to plough at the head and corners of the land.
Description
Technical Field
The invention relates to a trimming planning method, in particular to a trimming planning method for planning an operation path of an automatic driving agricultural machine.
Background
China is wide in territory, and the farmland has various existing forms, namely centralized and regular fields and scattered small fields. When the agricultural machine is operated in the field, the agricultural machine needs to turn or turn around on the ground according to operation requirements, and the agricultural machine turns or turns around when the agricultural implement descends, namely the agricultural implement enters the soil, so that the turning radius is increased, the load of the engine is increased, and the agricultural implement is easily damaged. Therefore, the agricultural machine is required to lift the agricultural implement during actual operation turning, and the situation can cause the head of the land to miss ploughing. When the agricultural machine turns, a part of farmland can not be ploughed due to the turning radius. Therefore, when the automatic driving agricultural machine works in the field, reasonable path planning is needed according to the shape of the field and the farming task, the high coverage rate of the field operation is ensured, and meanwhile, the idle running stroke of the agricultural machine is reduced, so that the utilization rate of energy and land is maximized.
When the agricultural machine carries out operation path planning, if the operation line is designed to be parallel to one long edge, the head of the land near the boundary of the farmland can be caused to miss-plough, and the head of the land needs to be ploughed at the moment, namely, the edge is collected. The edge-closing operation is that before the agricultural machine enters the farmland to operate or after the agricultural machine ploughs the long edge in the farmland, the head of the farmland and the edge of the farmland are connected, the farmland is turned around along the boundary of the farmland, and the farm tool is lifted at the corner to turn. Most of the existing automatic driving equipment does not contain a margin closing plan or only contains a limited margin closing plan, the integrity of the field operation of the automatic driving agricultural machine is difficult to ensure, and the situation of head-of-field operation vacancy is often caused.
Disclosure of Invention
In order to solve the problems in the background art, the invention provides a margin narrowing planning method for planning the operation path of the automatic driving agricultural machine, which can be used for margin narrowing operation of the automatic driving agricultural machine and reduces land blocks which are difficult to plough at the head and corners of the land.
In order to achieve the purpose, the invention provides the following technical scheme:
the edge closing planning method for planning the operation path of the automatic driving agricultural machine comprises the following specific steps:
step 1, initializing a farmland model, and acquiring farmland boundary data, a farmland entrance position and a farmland exit position;
step 2, classifying the farmland models according to farmland boundary data in the farmland models;
step 3, determining the main shaft direction of the agricultural machine in the field operation according to the farmland boundary data and the farmland type in the farmland model, dividing the farmland boundary, and dividing the farmland into an operation area and an area to be converted;
step 4, planning a full-coverage operation path in the operation area, and returning to the starting position and the ending position of the full-coverage operation path in the operation area;
and 5, matching an edge folding mode according to the segmentation result in the step 3, and planning and designing an edge folding path in the area to be converted.
The farmland model in the step 1 comprises farmland boundary data, an entrance position and an exit position. The field boundary may be described by a polygon, with the location of an inflection point in the actual field corresponding to a point on the polygon.
The farmland entrance position in the step 1, which is the position where the agricultural machinery is driven into the farmland from the farmland, is generally positioned in the farmland and is connected with the machine-ploughing channel or the field road.
The outlet position of the farmland in the step 1, namely the position where the agricultural machine completes all operation tasks and leaves the farmland, is generally positioned in the farmland and is connected with a machine-ploughing channel or a field road.
And (3) farmland classification in the step 2, namely, farmland is evaluated according to the shape of a farmland boundary, the adopted evaluation standard comprises farmland rectangle degree, and the farmland is divided into a standard farmland and a non-standard farmland according to the standard.
And 2, the boundary contour of the farmland in the farmland model in the step 2 is a convex polygon.
The main shaft direction in the step 3 is the main direction of the agricultural machinery running in the operation area, according to the farmland model input in the step 1, the optimal main shaft direction of the agricultural machinery running in the operation area can be obtained by a heuristic search method, the range of the direction is [0, pi ], through the heuristic search method, the cost in the whole range needs to be calculated because the cost function of the search is not linear, the resolution of the final calculation result is less than or equal to 5 degrees, and the calculation method of the cost function is as follows:
if the starting point of the path planning of the operation area is SP (located in the farmland), the straight line lv is perpendicular to the direction of the main shaft to be selected and passes through the SP. lv at least has two intersection points on the farmland boundary, the leftmost intersection point is ZP, the rightmost intersection point is YP, and the distance between the two points is FW, cost _ t is the overhead cornering cost, cost _ t<0. Starting from YP, a point (X) is inserted on lv at every other line spacing (specified by the user)i，Yi) Generating a job line parallel to the main axis to be selected every time when passing one point, wherein the length of the job line is LLiCost _ w is the job value, and is typically a positive value.
The search process is as follows:
1. costs for 0 °, 30 °, 60 °, 120 °, 150 °, 180 °, 6 directions are calculated, respectively.
2. And 3 optimal directions are selected to be placed in a queue to be selected, and the rest directions are deleted.
3. The search direction angle step is halved.
4. And respectively adding two new search directions on two sides of the three optimal directions, and calculating the cost of the new direction.
5. And repeating the step 2 and the step 3 until the resolution is less than or equal to 5 degrees.
An optimal direction is selected from the three optimal directions as a main axis direction.
And 3, dividing the farmland boundary, namely dividing the farmland boundary into an operation area and an area to be converted according to the division result of the step 2. A, B two division schemes are provided, in scheme A, dividing farmland into
、
Two areas to be transferred are provided,
an operation area; in scheme B, the farmland is divided into regions to be transferred
An operating area
。
4, the full-coverage operation path planning in the step 4 adopts a global path planning algorithm, namely, the full-coverage path planning is carried out in the operation area in a line scanning mode; when the operation of the operation area is executed after the edge is closed, the starting point of the edge closing path planning is the entrance position of the farmland, and the end point of the edge closing path planning is the starting position of the operation area planning; when planning the operation area and then closing the edge, the starting point of the edge closing path is the end position of the planning of the operation area, and the end point of the edge closing path is the outlet position of the farmland.
The edge-closing mode in the step 5 comprises a farmland configuration turn-back type edge-closing mode with two to-be-rotated areas and a configuration surrounding type edge-closing mode with a continuous to-be-rotated area.
The turn-back type edge folding mode in the step 5 generally works under the condition of two right areas to be rotated, the common edge can be reduced to the greatest extent by the turn-back type edge folding mode, namely the common edge is positioned in the operation area and is parallel to the main shaft, and the running of repeated operation is realized in the actual operation process. In the folding type edge folding mode, the edge folding planning mode is that line scanning is respectively carried out on two areas to be rotated, the base line of the line scanning is the short edge of the area to be rotated, and the paths of the two areas to be rotated are connected by a common edge.
The wrap-around edging mode in step 5 may be applied to unconventional farmlands where the agricultural machine may turn around each boundary, and therefore, the agricultural machine is required to travel around the farmland boundary to completely cover all the areas to be turned. Unlike line scanning of operation area planning and fold-back type edge closing planning, the full coverage planning adopts a broken line scanning mode, and a scanned base line is a broken line along the boundary direction.
In step 5, after the round-robin or turn-back planning is completed, the inflection point in the path needs to be smoothed.
The specific steps of the inflection point smoothing strategy are as follows:
1. setting the 1 st Node in the path point set as Start, the 1+1 st Node as Node, and the 1+2 st Node as End;
2. calculating the course difference between the Start and the Node;
3.1, if the heading difference between the Start and the Node is less than 90 degrees, setting the Test Node to be equal to the Node, and deleting the Node. And after a straight path is driven, steering by using the method shown in the figure 4, dispersing the arc line in the steering, taking a point on the arc line at an interval of 5 degrees, inserting the Start, and setting the last Node after the steering as a Node. Wherein the initial path and the steered path have a directional angle a less than 90 °;
3.2, if the difference between the Start course and the Node course is larger than 90 degrees, setting the Test Node to be equal to the Node, and deleting the Node. After a straight path is driven, steering is carried out by using a combined steering method to disperse arcs, points on one arc are removed at intervals of 5 degrees, after a Start is inserted, the last Node after steering is set as a Node;
3.3, if the heading difference between the Start and the Node is equal to 0 degree, not executing any operation, and jumping to 5;
4. and judging whether the Node is in the position relationship between the Test and the End. If Node is between Test and End, jumping to 5; if Node is on the Test and End reverse extension line, returning an error; if the Node is at the extension line of the Test and the End, a reversing route with the length equal to the distance between the Node and the End plus one meter is established along the vector (Node, End), after the starting point and the End point of the reversing route are inserted into the Node, the End point of the reversing route is set as the Node, and the Node jumps to 5;
5. setting Node to Start, End to Node, and End going forward one bit;
6. repeating the steps 2, 3, 4 and 5 until all the nodes are traversed;
the combined steering method in the step 3.2 comprises the following specific steps:
1. starting from Start, turning to the relay End node first, two arc paths are generated, namely an arc line l1 taking O as a center and an arc line l2 taking O' as a center, when the agricultural machinery reaches End and End1 along the two paths respectively, the directions are consistent with those of the target path AB, and only the path with a smaller rotation angle is reserved according to the minimum consumption principle. Taking a point on an arc every 5 degrees, inserting the Start, wherein the last point on the arc is called a relay End, and the vertical FOOT of a straight line where the relay End, the Node and the End are located is FOOT;
2. judging the position relation of FOOT, Node and End, if FOOT is between Node and End, jumping to 3, if FOOT is on the reverse extension line of Node and End, returning error, if FOOT is on the extension line of Node and End, setting relay End as Node;
3. enabling the relay End to approach a straight line where the Node and the End are located, respectively dispersing two End arcs with the o1 as the center of a circle and the o2 as the center of a circle, taking a point every 5 degrees, and sequentially inserting the points into the Start point and then setting the last point on the two segments of arcs as the Node;
through the steps, the inflection point smoothing in the waypoint sequence can be completed.
Compared with the prior art, the invention has the beneficial effects that: 1. the output result of the invention can provide a full-automatic driving path for the automatic driving agricultural machine, and the result comprises the starting position and the ending position of each straight-driving road section, the starting position and the ending position of the turning road section, the turning angle and the starting position and the ending position of the reversing road section, thereby effectively improving the precision degree of the agricultural machine in field operation.
2. The invention realizes the digitalization of farmland data, and can adopt high-precision DGPS to acquire high-precision coordinate information of each position on the boundary of the farmland in the implementation process.
3. The invention analyzes the farmland model, provides a judgment standard to classify the farmland, divides the farmland boundary, divides the farmland into an operation area and a to-be-transferred area, and performs full-coverage path planning on the operation area and the to-be-transferred area, thereby improving the coverage area of the automatic driving agricultural machinery during the field operation.
Drawings
FIG. 1 is a flow chart of a planning method according to the present invention;
FIG. 2 is a schematic diagram of the division of the operation area of the planning method of the present invention;
FIG. 3 is a schematic diagram of an edge closing mode of the planning method according to the present invention;
FIG. 4 is a schematic diagram of a turning method of the planning method of the present invention;
FIG. 5 is a schematic diagram of a second method of turning according to the planning method of the present invention;
FIG. 6 is a schematic view of a third method of turning according to the planning method of the present invention;
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
According to fig. 1-6, the present invention provides a solution:
the edge closing planning method for planning the operation path of the automatic driving agricultural machine comprises the following specific steps:
step 1, initializing a farmland model, and acquiring farmland boundary data, a farmland entrance position and a farmland exit position; the farmland model comprises farmland boundary data, an inlet position and an outlet position. The farmland boundary data can be compatible with various acquisition modes, inflection point coordinates in the farmland boundary data can be extracted from data meeting requirements, and the farmland is abstracted into a polygon on a 2D plane. For example, high-precision low-altitude unmanned remote sensing images can be used for acquiring farmland boundary data, and the inlet position and the outlet position of the farmland are marked on the upper side. The entrance position is the starting point of the path planning, and the exit position is the end point of the path planning. Due to the nature of the semi-closed area with boundary limitation in the farmland, only one channel is usually arranged from a machine plowing road/field road to the farmland, if the starting point or the end point of the path planning is arranged at a place far away from an entrance, the idle running stroke of the agricultural machine is lengthened, the oil consumption is increased, the positions of the starting point and the end point of the path planning are restrained by the entrance and exit positions, and if the attribute is empty, the planning is carried out according to a default value. The entrance position of the farmland, i.e. the position where the agricultural machine drives into the farmland from the farmland, is generally positioned inside the farmland and is connected with a tractor-ploughing channel or a field road. The outlet position of the farmland is the position where the agricultural machine completes all operation tasks, and the position away from the farmland is generally positioned in the farmland and connected with a tractor-ploughing channel or a field road.
Step 2, classifying the farmland models according to farmland boundary data in the farmland models; the farmland boundary contour in the farmland model is a convex polygon. According to the topographic features of China, farmlands are divided into two types, one is a standard farmland close to a rectangle, and the other is a non-standard farmland with an irregular boundary. The standard for distinguishing the standard farmland from the non-standard farmland is the rectangle degree of the farmland outline. The squareness is calculated as follows:
wherein SRealIs the actual area of the farmland contour, SMBRIs the area of the smallest bounding rectangle. The MBR is an R tree data standard index and can be obtained by adopting a rotating caliper method. Squareness: the range of rectangular value is [0,1 ]]. The farmland with the rectangle degree larger than the critical value is a standard farmland, and the farmland with the rectangle degree smaller than the critical value is a non-standard farmland. The critical value is determined according to the actual experimental result.
And 3, determining the direction of a main shaft of operation in the operation area according to farmland boundary data and farmland types in the farmland model, dividing the farmland boundary, and dividing the farmland into the operation area and the area to be converted. And (4) planning a full-coverage operation path in a working area, wherein the turning of the agricultural machine is mainly performed in a to-be-transferred area, so that the to-be-transferred area is subjected to no-tillage. After the planning of the operation area is finished, the edge folding path planning is carried out in the area to be transferred, and the whole coverage area of field operation can be increased. And searching an optimal main shaft by adopting a heuristic search method, so that the path obtained by the full-coverage path planning in the working area has the maximum coverage area and the minimum consumption. Because the width of the farm tool is fixed, the coverage area is in direct proportion to the total length of the operation line; when the agricultural machinery is turning, often need promote the agricultural implement, the agricultural machinery is in the idle running state, and the number of turns is more, and consumption is big more. Therefore, an optimal spindle needs to be selected to meet the above requirements.
The range of the main axis direction is [0, pi ], and by a heuristic search method, the cost function of the search is not linear, so the cost in the whole range needs to be calculated, the resolution of the final calculation result is less than or equal to 5 degrees, and the calculation method of the cost function is as follows:
if the starting point of the path planning of the operation area is SP (located in the farmland), lv is vertical to the direction of the main axis to be selected, and at least two intersection points exist in the boundary of the farmland after passing through SP, lv, the leftmost intersection point is ZP, the rightmost intersection point is YP, the distance between the two points is FW, cost _ t is the overhead turning consumption, and cost _ t is the overhead turning consumption<0. Starting from YP, inserting a point on lv every other line space (the line space is designated by a user), generating a working line parallel to a main axis to be selected by each point, wherein the length of the working line is LLiCost _ w is the job value, and is typically a positive value.
The heuristic search method comprises the following search processes:
calculating the costs of 0, 30, 60, 120, 150, 180 and 6 directions respectively;
secondly, 3 optimal directions are selected to be put into a queue to be selected, and other directions are deleted;
step three, searching direction angle step length is reduced by half;
adding two new search directions on two sides of the three optimal directions respectively, and calculating the cost of the new direction;
fifthly, repeating the step 2 and the step 3 until the resolution is less than or equal to 5 degrees;
and sixthly, selecting an optimal direction from the three optimal directions as the main shaft direction.
In the segmentation method described in step 3, as shown in scheme a in fig. 1, the direction of the straight line where the main axis is located can be selected as the main axis direction according to the main axis selection rule of the standard farmland.
In the embodiment shown in FIG. 2, A, B two division schemes are provided, in scheme A, dividing the field into two
、
Two areas to be transferred are provided,
an operation area; in scheme B, the farmland is divided into regions to be transferred
An operating area
。
Step 4, designing a full-coverage operation path in the operation area, and returning to the starting position and the ending position of the full-coverage operation path in the operation area; and (3) planning the full-coverage operation path of the operation area, wherein during the implementation process, the full-coverage operation can be planned in a line scanning mode. When the agricultural machine is used for carrying out soil preparation operation and sowing operation, the operation of an operation area needs to be carried out firstly, and then the operation of a to-be-transferred area needs to be carried out. Therefore, the starting position of the edge closing plan is the end point of the operation in the operation area, and the ending position of the edge closing plan is the exit position of the farmland; when the agricultural machine executes harvesting operation, the starting position of the edge closing plan is the entrance position of the farmland, and the end position of the edge closing plan is the exit position of the farmland.
And 5, matching an edge folding mode according to the segmentation result in the step 3, and planning and designing an edge folding path, wherein the edge folding mode in the step 5 comprises a farmland configuration turning-back type edge folding mode with two areas to be rotated and a configuration surrounding type edge folding mode with one continuous area to be rotated. The turn-back type edge folding mode generally works under the condition of two right areas to be rotated, the common edge can be reduced to the greatest extent by the turn-back type edge folding mode, namely the common edge is positioned in the operation area and is parallel to the main shaft, and the operation belongs to the repeated operation in the actual operation process. In the folding type edge folding mode, the edge folding planning mode is that line scanning is respectively carried out on two areas to be rotated, the base line of the line scanning is the short edge of the area to be rotated, and the paths of the two areas to be rotated are connected by a common edge. The wrap-around edge-closing mode can be applied to unconventional farmlands in which agricultural machinery may turn on each boundary, and therefore, the agricultural machinery is required to travel around the farmland boundary to completely cover all the areas to be turned. Unlike line scanning of operation area planning and fold-back type edge closing planning, the full coverage planning adopts a broken line scanning mode, and a scanned base line is a broken line along the boundary direction. After the surrounding type or turning-back type planning is completed, the inflection point in the path needs to be smoothed.
Taking the turn-back type as an example, the edge closing planning strategy for planning the operation path of the automatic driving agricultural machine is implemented by the following specific implementation method:
as shown in the scheme a in fig. 3, the foldback edge closing planning strategy needs to perform full coverage planning on two areas to be transferred in a line scanning manner. The two regions to be switched are then connected by a transition line DL p0p1 as shown in scheme a in fig. 3. In actual work, p0p1 needs to be translated by a half line width in the direction pointing to the center of the farmland, and the line width is specified by a user.
If the operation of the operation area is executed first and then the edge is closed. The transfer waiting area near the work end position of the work area is T1, and the other transfer waiting area is T2. And coding straight lines of the T1 area and the T2 area from inside to outside, sequentially setting the straight lines into a road section one, a road section two and a road section three … …, and sequentially putting the straight lines into a road section queue.
If the operation of the operation area is executed first and then the edge is closed. The region to be transferred near the entrance of the farmland is taken as T1, and the other region to be transferred is taken as T2. And coding straight lines in a T1 area and a T2 area from outside to inside, sequentially setting the straight lines as a road section one, a road section two and a road section three … …, and sequentially putting the straight lines into a road section queue.
The starting point of DL is the droop of the end point of the full-coverage plan in the T1 area on DL, and the end point of DL is the droop of the starting point of the full-coverage plan in the T2 area on DL.
The processing process of the foldback edge closing algorithm comprises the following steps:
firstly, creating a path point set E;
secondly, setting the starting position of the edge closing planning as a road node and placing the road node into E;
thirdly, a straight line starting from a T1 area along the direction of the current road node intersects at a point P, the point P is set as a road node, and the road node is placed in E;
fourthly, setting an end point on the first road section far away from the point P as a road node, and putting the road node into the E;
fifthly, taking the distance between the current road node and two end points of the current road section as an evaluation standard, successively setting the end points of the next road point as road nodes according to a principle of first approach and then distance, and placing the road nodes into E;
sixthly, repeating the fifth step until the last road section is reached, setting the end point which is closer to the current road node as a road node, and placing the road node into E;
step seven, sequentially setting the starting point and the end point of the transition line as road nodes, putting the road nodes into an E, and turning to a T2 area to prepare for traversing a T2 area;
and step eight, repeating the step 5 until the traversal of all the road segments in the T2 area is completed.
And ninthly, ending the traversal, wherein the load node set in the step E is the node set of the edge closing path.
In step 5, taking a surrounding type as an example, the edge closing planning strategy for planning the operation path of the automatic planning driving agricultural machinery is implemented, and the specific implementation method is as follows:
as shown in fig. 3, scenario B, there is a pending zone within the field. If the operation of the operation area is executed firstly and then the edge is closed, a loop line (an edge parallel to the main axis and a plurality of loop lines share a path) is established from the boundary of the operation area along the parallel direction of the boundary, the loop line is sequentially expanded by taking p1 ' p2 ', p2 ' p3 ', p3 ' p4 ', p4 ' p0 ' and p0 ' p1 as base lines, is parallel to each base line respectively and is far away from the operation area, wherein the p0 ' p1 ' is a common edge, and the line width is translated by half to the direction pointing to the center of the farmland during actual operation. The distance between two adjacent loop lines is one line width, and the line width is input by a user. And when the distance between the loop line and the boundary of the farmland is less than one line width, the expansion is stopped. The loop closest to the work area is referred to as a loop, which in turn is referred to as a two-loop, three-loop … ….
If the operation of the operation area is executed after the edge is taken away, a loop line (a plurality of loop lines share a path on the edge parallel to the main axis) is established along the parallel direction of the boundary from the boundary of the farmland, the loop line sequentially takes p1p2, p2p3, p3p4, p4p0 and p0p1 as base lines and expands towards the direction close to the operation area, wherein p0p1 is a common edge, and the loop line is translated by half line width towards the direction pointing to the center of the farmland in actual work. The distance between two adjacent loop lines is one line width, and the line width is input by a user. And when the distance from the loop line to the boundary of the operation area is less than one line width, the expansion is stopped. The loop closest to the field boundary is set to one loop and in turn set inward to two loops, three loops … ….
The path points in the edge closing path are called road nodes.
When the operation of the operation area is executed after the edge is closed, the end point of the outermost loop line is the end position of the edge closing planning and the foot of the loop line.
When the operation of the operation area is executed firstly and then the edge is closed, the end point of the outermost side loop is the drop foot of the loop at which the farmland outlet is positioned.
The surrounding type edge closing algorithm processing process comprises the following steps:
firstly, creating a route point set E of an edge closing path;
step two, setting the starting position of the edge closing plan as a road node and placing the road node into E;
thirdly, starting along the direction of the current road node, intersecting with a ring at a point P, setting the coordinate of the point P as a road node, and placing the road node into E;
fourthly, calculating the positions of the outlets and the vertical feet of each loop line, and calling the positions as the end points of the current loop lines;
fifthly, calculating the distance from the road node to the current loop line terminal along the loop line, and setting the direction of the route with longer detour distance as the detour direction;
and sixthly, winding along the loop, sequentially setting the turning points on the loop along the way as road nodes according to the sequence of successive arrival, and placing the road nodes into E. Setting the current loop line end point as a road node, and placing the road node into E;
step seven, entering the next loop line, detouring along the same direction as the previous detouring route, and repeating the operation of the step six;
eighthly, repeating the operation of the seventh step until all the loop lines are traversed;
and ninthly, ending the traversal, wherein the load node set in the step E is the node set of the edge closing path.
And 5, an inflection point smoothing strategy, aiming at smoothing corners appearing in the path. The agricultural machinery is an Ackerman steering structure driven by front steering and rear steering, the pivot steering cannot be realized, and the turning radius exists in the process of turning the agricultural machinery. The global path realized in the scheme is actually a series of waypoint sets E, and the inflection point smoothing strategy is a comprehensive path combining a forward straight line and a backward straight line with an arc line, so that the differential characteristics of scattered points are continuous. Based on fitting and interpolation of quadratic and cubic spline lines, the dragon lattice effect is easy to occur. The model matching mode provided by the application uses the circular arc tangent to the linear end point to connect each waypoint, and is safer and more reliable.
Course of the waypoint: taking the waypoint i as the start position of the vector v, the waypoint i +1 as the end position of the vector v, and the azimuth angle of the vector v in space
I.e. the heading of waypoint i.
Course difference: starting from a waypoint, the angle through which it rotates along one heading (clockwise/counterclockwise) to another.
The specific implementation of the knee smoothing strategy is as follows:
1. setting the 1 st Node in the path point set as Start, the 1+1 st Node as Node, and the 1+2 st Node as End;
2. calculating the course difference between the Start and the Node;
3.1, if the heading difference between the Start and the Node is less than 90 degrees, setting the Test Node to be equal to the Node, and deleting the Node. And after a straight path is driven, steering by using the method shown in the figure 4, dispersing the arc line in the steering, taking a point on the arc line at an interval of 5 degrees, inserting the Start, and setting the last Node after the steering as a Node. Wherein the initial path and the steered path have a directional angle a less than 90 °;
3.2, if the difference between the Start course and the Node course is larger than 90 degrees, setting the Test Node to be equal to the Node, and deleting the Node. After a straight path is driven, steering is carried out by using a combined steering method to disperse arcs, points on one arc are removed at intervals of 5 degrees, after a Start is inserted, the last Node after steering is set as a Node;
3.3, if the heading difference between the Start and the Node is equal to 0 degree, not executing any operation, and jumping to 5;
4. and judging whether the Node is in the position relationship between the Test and the End. If Node is between Test and End, jumping to 5; if Node is on the Test and End reverse extension line, returning an error; if the Node is at the extension line of the Test and the End, a reversing route with the length equal to the distance between the Node and the End plus one meter is established along the vector (Node, End), after the starting point and the End point of the reversing route are inserted into the Node, the End point of the reversing route is set as the Node, and the Node jumps to 5;
5. setting Node to Start, End to Node, and End going forward one bit;
6. repeating the steps 2, 3, 4 and 5 until all the nodes are traversed;
the specific implementation manner of the combined steering in the step 3.2 is as follows:
first, starting from Start, the method shown in fig. 5 is first switched to the relay End node, and in this way, two circular arc paths are generated, namely an arc line l1 centered at O and an arc line l2 centered at O', and the agricultural machinery respectively reaches End and End1 along the two circular arc paths. The direction is consistent with that of the target path AB, and only the path with a smaller rotation angle is reserved according to the minimum consumption principle. Taking a point on an arc every 5 degrees, inserting the Start, wherein the last point on the arc is called a relay End, and the vertical FOOT of a straight line where the relay End, the Node and the End are located is FOOT;
secondly, judging the position relation of FOOT, Node and End, if FOOT is positioned between Node and End, jumping to 3, if FOOT is positioned on the reverse extension line of Node and End, returning an error, if FOOT is positioned on the extension line of Node and End, setting relay End as Node;
thirdly, using the method shown in the picture 6 to make the relay End approach the straight line where the Node and the End are located, respectively dispersing the two End arcs taking o1 as the center of circle and o2 as the center of circle, taking a point every 5 degrees, and sequentially inserting the point after the Start point to set the last point on the two arcs as the Node.
Through the steps, the inflection point smoothing in the waypoint sequence can be completed.
The output result of the method can provide a full-automatic running path for the automatic agricultural machinery, and the result comprises the starting position and the ending position of each road section which runs straight, the starting position and the ending position of the road section which turns, the turning angle and the starting position and the ending position of the road section which backs a car. Compared with experience estimation in the process of driving by people, the method can effectively improve the precision degree of the agricultural machine in the field operation.
In the step 1 of the application, the farmland data is digitalized, and high-precision DGPS can be adopted to obtain high-precision coordinate information of each position on the farmland boundary in the implementation process.
In step 2 of this application, carry out the analysis to the farmland model to provide the judgement standard and classify to the farmland, then cut apart the farmland border according to the end of step 2 in step 3, cut apart the farmland into operation district and wait to change the district, step 4 and step 5 carry out the path planning of full coverage respectively in operation district and waiting to change the district, can improve the coverage area of autopilot agricultural machinery when the field operation.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (9)
1. A margin narrowing planning method for planning an operation path of an automatic driving agricultural machine is characterized by comprising the following steps: the method comprises the following specific steps:
step 1, initializing a farmland model, and acquiring farmland boundary data, a farmland entrance position and a farmland exit position;
step 2, classifying the farmland models according to farmland boundary data in the farmland models;
step 3, determining the main shaft direction of the agricultural machine in the field operation according to the farmland boundary data and the farmland type in the farmland model, dividing the farmland boundary, and dividing the farmland into an operation area and an area to be converted; the main shaft direction is the main direction of the agricultural machinery running in the operation area, the optimal main shaft direction of the running in the operation area can be obtained by a heuristic search method according to the farmland model input in the step 1, the range of the direction is [0, pi ], the cost in the whole range is calculated by the heuristic search method, the resolution of the final calculation result is less than or equal to 5 degrees, and the calculation method of the cost function is as follows:
if the starting point of the path planning of the operation area is SP, a straight line lv is perpendicular to the direction of the main axis to be selected and passes through SP, lv has at least two intersection points on the boundary of the farmland, the leftmost intersection point is ZP, the rightmost intersection point is YP, the distance between the two points is FW, cost _ t is the overhead turning consumption, and cost _ t is<0, starting from YP, every other line spacing, specified by the user, inserting a dot on lv
Generating a job line parallel to the main axis to be selected every time when passing one point, wherein the length of the job line is LLiCost _ w is the operation value and is a positive value;
the heuristic search method comprises the following search processes:
respectively calculating the costs of 0 degrees, 30 degrees, 60 degrees, 120 degrees, 150 degrees, 180 degrees and 6 directions;
selecting 3 optimal directions to be placed in a queue to be selected, and deleting the other directions;
step three, the step length of the angle of the search direction is reduced by half;
respectively adding two new search directions on two sides of the three optimal directions, and calculating the cost of the new direction;
(V) repeating the step 2 and the step 3 until the resolution is less than or equal to 5 degrees;
(VI), selecting an optimal direction from the three optimal directions as a main shaft direction;
step 4, planning a full-coverage operation path in the operation area, and returning to the starting position and the ending position of the full-coverage operation path in the operation area;
and 5, matching an edge folding mode according to the segmentation result in the step 3, and planning and designing an edge folding path in the area to be converted.
2. The edging method for planning a work path of an autonomous agricultural machine according to claim 1, characterized in that: the farmland model in the step 1 comprises farmland boundary data, an entrance position and an exit position; the field boundary can be described by a polygon, and the location of an inflection point in the actual field corresponds to a point on the polygon.
3. The edging method for planning a work path of an autonomous agricultural machine according to claim 1, characterized in that: and 2, classifying the farmland in step 2, namely evaluating the farmland according to the shape of the boundary of the farmland, wherein the adopted evaluation standard comprises the farmland rectangle degree, and classifying the farmland into a standard farmland and a non-standard farmland according to the standard.
4. The edging method for planning a work path of an autonomous agricultural machine according to claim 1, characterized in that: and 2, the boundary contour of the farmland in the farmland model in the step 2 is a convex polygon.
5. According toThe edging planning method for planning a work path of an autonomous agricultural machine according to claim 1, characterized in that: dividing the farmland boundary in the step 3, namely dividing the farmland boundary into an operation area and a to-be-transferred area according to the division result in the step 2, wherein the method comprises A, B two division schemes, and in the scheme A, the farmland is divided into two paths
、
Two areas to be transferred are provided,
an operation area; in scheme B, the farmland is divided into regions to be transferred
An operating area
。
6. The edging method for planning a work path of an autonomous agricultural machine according to claim 1, characterized in that: 4, the full-coverage operation path planning in the step 4 adopts a global path planning algorithm, namely, the full-coverage path planning is carried out in the operation area in a line scanning mode; when the operation of the operation area is executed after the edge is closed, the starting point of the edge closing path planning is the entrance position of the farmland, and the end point of the edge closing path planning is the starting position of the operation area planning; when planning the operation area and then closing the edge, the starting point of the edge closing path is the end position of the planning of the operation area, and the end point of the edge closing path is the outlet position of the farmland.
7. The edging method for planning a work path of an autonomous agricultural machine according to claim 1, characterized in that: the edge-closing mode in the step 5 comprises a farmland configuration turn-back type edge-closing mode with two to-be-rotated areas and a configuration surrounding type edge-closing mode with one continuous to-be-rotated area.
8. The edging method for planning a work path of an autonomous agricultural machine according to claim 1, characterized in that: in step 5, after the edge-closing path planning design is completed, the inflection point in the path needs to be smoothed, and the path node generated in the above step is processed by adopting an inflection point smoothing strategy.
9. The edging method for planning a work path of an autonomous agricultural machine according to claim 8, characterized in that: the specific steps of the inflection point smoothing strategy are as follows:
firstly, setting a 1 st Node in a path point set as Start, setting a 1+1 st Node as Node, and setting a 1+2 st Node as End;
secondly, calculating the course difference between the Start and the Node;
thirdly, if the course difference between the Start and the Node is less than 90 degrees, setting a Test Node to be equal to the Node, deleting the Node, steering by using the method shown in FIG. 4 after a section of straight path is driven, dispersing an arc line in steering, taking a point on the arc line every 5 degrees, inserting the Start, and setting the last Node after steering as the Node, wherein the direction angle a between the initial path and the steered path is less than 90 degrees;
if the course difference between the Start and the Node is larger than 90 degrees, setting a Test Node to be equal to the Node, deleting the Node, driving a section of straight path, then using a combined steering method to steer to disperse the arc line, removing a point on the arc line at an interval of 5 degrees, inserting the Start, and setting the last Node after steering as the Node;
thirdly, if the heading difference between the Start and the Node is equal to 0 degree, no operation is executed, and the step jumps to 5;
judging whether the Node is in the position relationship between the Test and the End, and if the Node is between the Test and the End, jumping to 5; if Node is on the Test and End reverse extension line, returning an error; if the Node is at the extension line of the Test and the End, a reversing route with the length equal to the distance between the Node and the End plus one meter is established along the vector (Node, End), after the starting point and the End point of the reversing route are inserted into the Node, the End point of the reversing route is set as the Node, and the Node jumps to 5;
fifthly, setting Node as Start, setting End as Node, and advancing End by one bit;
sixthly, repeating the second, third, fourth and fifth steps until all the nodes are traversed;
the combined steering method in the third step comprises the following specific steps:
through the steps, the inflection point smoothing in the waypoint sequence can be completed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010225665.8A CN111256700B (en) | 2020-03-26 | 2020-03-26 | Edge narrowing planning method for planning operation path of automatic driving agricultural machine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010225665.8A CN111256700B (en) | 2020-03-26 | 2020-03-26 | Edge narrowing planning method for planning operation path of automatic driving agricultural machine |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111256700A CN111256700A (en) | 2020-06-09 |
| CN111256700B true CN111256700B (en) | 2021-11-02 |
Family
ID=70953443
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010225665.8A Active CN111256700B (en) | 2020-03-26 | 2020-03-26 | Edge narrowing planning method for planning operation path of automatic driving agricultural machine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111256700B (en) |
Families Citing this family (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112197775B (en) * | 2020-11-12 | 2024-02-09 | 扬州大学 | Agricultural machinery multi-machine collaborative operation path planning method |
| CN112434386B (en) * | 2020-11-17 | 2024-02-20 | 北京农业智能装备技术研究中心 | Operation control method and system of agricultural machine unit |
| CN112703834B (en) * | 2020-12-03 | 2022-08-16 | 广州极飞科技股份有限公司 | Sweeping path planning method, land block leveling operation path determining method and device |
| CN113375674B (en) * | 2021-06-16 | 2024-02-27 | 上海联适导航技术股份有限公司 | Curve path generation method, device, equipment and readable storage medium |
| CN113686344B (en) * | 2021-08-30 | 2023-08-18 | 中车大连电力牵引研发中心有限公司 | Agricultural machinery coverage path planning method |
| CN114091896B (en) * | 2021-11-22 | 2022-04-29 | 中化现代农业有限公司 | Agricultural machinery running state analysis method |
| CN114137974A (en) * | 2021-11-26 | 2022-03-04 | 广州极飞科技股份有限公司 | Intertillage control method, device and system and electronic equipment |
| CN114554308B (en) * | 2021-11-30 | 2023-12-22 | 潍坊中科晶上智能装备研究院有限公司 | Intelligent agricultural machinery remote control system |
| CN114326717A (en) * | 2021-12-13 | 2022-04-12 | 中国科学院计算技术研究所 | Intelligent agricultural machinery transition and operation fusion path planning method and system |
| CN114485610B (en) * | 2021-12-27 | 2023-07-18 | 广州极飞科技股份有限公司 | Vector map construction method and device and path planning method and device |
| CN115250720A (en) * | 2022-07-12 | 2022-11-01 | 松灵机器人(深圳)有限公司 | Mowing method, mowing device, mowing robot and storage medium |
| CN115226458B (en) * | 2022-07-29 | 2024-05-03 | 驻马店志恒信息科技有限公司 | Unmanned agricultural machinery fertilization control method |
| CN115200589B (en) * | 2022-09-16 | 2022-11-29 | 北京博创联动科技有限公司 | Agricultural machinery edge folding track planning method and device, computer terminal and storage medium |
| CN115469664A (en) * | 2022-09-16 | 2022-12-13 | 北京博创联动科技有限公司 | Agricultural machine edge folding method and device, agricultural machine and storage medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102167038A (en) * | 2010-12-03 | 2011-08-31 | 北京农业信息技术研究中心 | Method and device for generating all-region-covering optimal working path for farmland plot |
| CN105955194A (en) * | 2016-05-10 | 2016-09-21 | 大连理工大学 | Method for smoothing inflection point of discrete processing path |
| CN106845371A (en) * | 2016-12-31 | 2017-06-13 | 中国科学技术大学 | A kind of city road network automotive emission remote sensing monitoring system |
| JP2018033382A (en) * | 2016-08-31 | 2018-03-08 | 三菱マヒンドラ農機株式会社 | Mulch rice transplanter |
| CN107990903A (en) * | 2017-12-29 | 2018-05-04 | 东南大学 | A kind of indoor AGV paths planning methods based on improvement A* algorithms |
| CN108334068A (en) * | 2017-01-20 | 2018-07-27 | 株式会社久保田 | Driving path generating means |
| CN108955695A (en) * | 2018-08-22 | 2018-12-07 | 洛阳中科龙网创新科技有限公司 | A kind of global path planning method for farmland robot |
| CN109964190A (en) * | 2016-12-19 | 2019-07-02 | 株式会社久保田 | Operation Van's automatic running system |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR100263982B1 (en) * | 1996-04-28 | 2000-08-16 | 모리 하루오 | Navigation apparatus |
| US6810315B2 (en) * | 2001-11-21 | 2004-10-26 | Pgi International, Ltd. | Agricultural vehicle dispenser regulator and method |
| CN203219779U (en) * | 2013-03-29 | 2013-10-02 | 北京德邦大为科技有限公司 | Novel water supply system with large-scale lateral moving sprinkling machine |
| CN104596531B (en) * | 2014-05-28 | 2018-12-28 | 腾讯科技（深圳）有限公司 | A kind of generation method of navigation routine, device and server |
| US9826674B2 (en) * | 2014-07-25 | 2017-11-28 | Agco Corporation | Vehicle guidance system |
| CN106989748A (en) * | 2017-05-16 | 2017-07-28 | 南京农业大学 | A kind of Agriculture Mobile Robot man-computer cooperation paths planning method based on cloud model |
-
2020
- 2020-03-26 CN CN202010225665.8A patent/CN111256700B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102167038A (en) * | 2010-12-03 | 2011-08-31 | 北京农业信息技术研究中心 | Method and device for generating all-region-covering optimal working path for farmland plot |
| CN105955194A (en) * | 2016-05-10 | 2016-09-21 | 大连理工大学 | Method for smoothing inflection point of discrete processing path |
| JP2018033382A (en) * | 2016-08-31 | 2018-03-08 | 三菱マヒンドラ農機株式会社 | Mulch rice transplanter |
| CN109964190A (en) * | 2016-12-19 | 2019-07-02 | 株式会社久保田 | Operation Van's automatic running system |
| CN106845371A (en) * | 2016-12-31 | 2017-06-13 | 中国科学技术大学 | A kind of city road network automotive emission remote sensing monitoring system |
| CN108334068A (en) * | 2017-01-20 | 2018-07-27 | 株式会社久保田 | Driving path generating means |
| CN107990903A (en) * | 2017-12-29 | 2018-05-04 | 东南大学 | A kind of indoor AGV paths planning methods based on improvement A* algorithms |
| CN108955695A (en) * | 2018-08-22 | 2018-12-07 | 洛阳中科龙网创新科技有限公司 | A kind of global path planning method for farmland robot |
Non-Patent Citations (3)
| Title |
|---|
| Smooth turning path generation for agricultural vehicles in headlands;Juha Backman 等;《Biosystems Engineering》;20150830;第139卷;第76-86页 * |
| 农田作业机械路径优化方法;孟志军 等;《农业机械学报》;20120630;第43卷(第6期);第147-152页 * |
| 无人驾驶电动拖拉机制动控制系统设计;杨柯 等;《农业开发与装备》;20191231(第5期);第114-117页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111256700A (en) | 2020-06-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111256700B (en) | Edge narrowing planning method for planning operation path of automatic driving agricultural machine | |
| US11044842B2 (en) | Path planning for area coverage | |
| AU2019278049B2 (en) | Route planning method for mobile vehicle | |
| Jin et al. | Optimal coverage path planning for arable farming on 2D surfaces | |
| CN111189444A (en) | Automatic driving agricultural machinery field operation path planning system and planning method | |
| CN108362285B (en) | Plant protection unmanned aerial vehicle accurate coverage track planning method and system | |
| CN108955695B (en) | Global path planning method for farmland robot | |
| CN106643719B (en) | Path planning algorithm of intelligent mowing vehicle | |
| CN109238298A (en) | A kind of unmanned paths planning method with avoidance | |
| CN107924188A (en) | Flight path planning, control method and the system of a kind of unmanned plane | |
| CN107608372A (en) | It is a kind of that path planning method is cooperateed with improving the multiple no-manned plane that PH curves are combined based on improvement RRT algorithms | |
| CN107860387A (en) | The unmanned machine operation flight course planning method of plant protection and plant protection unmanned plane | |
| CN106813667B (en) | A kind of Route planner and device based on no-fly zone constraint | |
| CN113064407B (en) | Cleaning method and device for full-area coverage, cleaning robot and storage device | |
| CN115164907B (en) | Forest operation robot path planning method based on A-algorithm of dynamic weight | |
| CN106774434B (en) | Automatic obstacle avoidance method and system applied to unmanned aerial vehicle plant protection | |
| CN109214596B (en) | Method for planning shortest path of grid with direction constraint and barrier limitation | |
| CN114035572B (en) | Obstacle avoidance tour method and system for mowing robot | |
| Nehme et al. | Lidar-based structure tracking for agricultural robots: Application to autonomous navigation in vineyards | |
| CN112462763A (en) | Mowing robot path planning method based on grid map | |
| CN113359780A (en) | Unmanned ship cleaning path planning method and device, computer equipment and storage medium | |
| CN109753075A (en) | A kind of agricultural garden robot navigation method of view-based access control model | |
| CN116880497A (en) | Full-coverage path planning method, device and equipment for automatic agricultural machine | |
| CN106910219B (en) | A method of agricultural machinery work area is counted based on geometric ways | |
| EP3518649B1 (en) | A method for determining placement of new obstacles in an agricultural field |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |