CN113359721B - Improved A-based AGV path planning method combined with motion control - Google Patents
Improved A-based AGV path planning method combined with motion control Download PDFInfo
- Publication number
- CN113359721B CN113359721B CN202110605550.6A CN202110605550A CN113359721B CN 113359721 B CN113359721 B CN 113359721B CN 202110605550 A CN202110605550 A CN 202110605550A CN 113359721 B CN113359721 B CN 113359721B
- Authority
- CN
- China
- Prior art keywords
- agv
- motion control
- path
- speed
- turning
- 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
- 238000000034 method Methods 0.000 title claims abstract description 46
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 61
- 238000004364 calculation method Methods 0.000 claims abstract description 19
- 238000012545 processing Methods 0.000 claims abstract description 13
- 230000008569 process Effects 0.000 claims description 16
- 230000001133 acceleration Effects 0.000 claims description 10
- 230000006872 improvement Effects 0.000 claims description 9
- 230000003247 decreasing effect Effects 0.000 claims description 8
- 230000010354 integration Effects 0.000 claims description 4
- 230000009467 reduction Effects 0.000 claims description 2
- 238000013461 design Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 230000002068 genetic effect Effects 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 238000004088 simulation Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000005265 energy consumption Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 201000004569 Blindness Diseases 0.000 description 1
- 101100012910 Plasmodium falciparum (isolate FC27 / Papua New Guinea) FIRA gene Proteins 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011065 in-situ storage Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000010355 oscillation Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0223—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving speed control of the vehicle
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0221—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving a learning process
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0276—Control of position or course in two dimensions specially adapted to land vehicles using signals provided by a source external to the vehicle
Landscapes
- Engineering & Computer Science (AREA)
- Aviation & Aerospace Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Feedback Control In General (AREA)
- Numerical Control (AREA)
Abstract
The invention relates to an AGV path planning method combining motion control and improving A, which comprises the following steps: step 1, calculating the motion control of smooth left turn and smooth right turn of the AGV according to task requirements and the AGV performance, and ensuring that the speed of the AGV before and after turning is unchanged; step 2, obtaining the minimum turning radius of the AGV according to the motion control of the step 1, rasterizing a task area, and modifying the definition of adjacent nodes in the A-star algorithm; step 3, based on the motion control of the step 1 and the grid processing of the step 2, performing path planning on the task area by using an improved A-star algorithm, and calculating a smooth path; step 4, deleting redundant path points from the path planned in the step 3; and step 5, based on the path points in the step 4, the AGV controls the movement according to the speed mark quantity of the path points to complete the operation. The invention improves the traditional A-star algorithm by combining motion control and path planning, and finishes the planning task under the constraint condition of meeting the task requirement and the calculation performance.
Description
Technical Field
The invention belongs to the field of mobile robots, and particularly relates to an AGV path planning method for improving A in combination with motion control.
Background
Path planning and motion control are both key technologies in the field of mobile robots. The path planning technology refers to that under a certain index (the index is usually the total path length, the energy consumption or the moving time of a robot and the like), an optimal or suboptimal path from a starting point to a terminal point is generated in a task area, and the path can meet the most basic obstacle avoidance requirement; the motion control refers to a motion method for controlling the mobile robot to drive according to an obstacle avoidance path drawn by a path planning rule.
With the continuous development of computer technology, artificial intelligence algorithm and sensor technology, the theoretical research suitable for AGV path planning is also continuous and deep. The current commonly used path planning algorithms include genetic algorithm, fast random search tree, artificial potential field method, A-algorithm and the like. The genetic algorithm is inspired by Darwin evolutionary theory, and the candidate solution of each generation is obtained by utilizing the cross variation and the genetic characteristics during heredity, and finally the optimal solution is obtained. The algorithm has low operation efficiency and low convergence speed, and is easy to fall into a local optimal solution; the fast random search tree is based on the idea of random sampling, is easy to fall into the problem of local minimum, and the blindness is too large during searching, so that the time consumption is long in certain environments; the environment is creatively regarded as a potential field by an artificial potential field method, the movement of the AGV is controlled by the attraction generated by a terminal point and the repulsion generated by an obstacle together, so that the AGV has a simple structure, and meanwhile, a plurality of defects are brought to the AGV, such as the phenomenon of easy oscillation, and the solved path is often not an optimal path and even stagnates in a local optimal solution; the algorithm a is a heuristic algorithm, and needs rasterization processing on an environment, and as the scale of a grid map increases, the calculation efficiency of the algorithm is obviously reduced, and the generated path has a 'stiff' broken line, which is not convenient for controlling the motion of the AGV. The algorithm a is proposed to be applied in a large amount in the actual path planning research in view of its optimality in the static environment.
A brief description of the a algorithm flow is required here. The traditional A-algorithm inherits and reserves the flow of Dijkstra algorithm, and adopts heuristic search on the basis of ensuring the optimal path, and the main flow is shown in figure 1. The algorithm needs to maintain both OPEN and CLOSED lists. Initially, a starting point is added into an OPEN list as a first node, a cost value is calculated by a formula f (n) = g (n) + h (n), g (n) is a distance from the starting point to a current node, h (n) is a heuristic function from the current node to an end point, an euclidean distance or a manhattan distance is usually used, obviously, g (n) of the starting point is 0, a parent node of the starting point is set to be empty, and then a loop is entered: and taking out the node with the minimum cost value in the current OPEN list, setting the node as the current node n, and putting the node into the CLOSED list. According to different cost function calculation modes, calculating and updating cost values of all adjacent feasible nodes (feasible nodes refer to the fact that the adjacent nodes can be passed by the AGV, such as non-border-crossing parts and non-obstacle parts) of the current node n and father nodes of the nodes. And after all the adjacent nodes are updated, the nodes containing the father node information and the cost value are added into an OPEN list, and then the circulation iteration is carried out. If the current node n is the terminal after a certain cycle, directly adding the current node n into a CLOSED list, and obtaining a complete path in a mode of reversely searching a father node; if the OPEN list is empty when the end point is not reached, the current environment is indicated to have no path from the starting point to the end point, and the algorithm is ended.
Fig. 2 is the result of static environment path planning in Matlab simulation using the conventional a-x algorithm. The black solid rectangle is an obstacle (including a boundary line), the semicircle represents a starting point, the triangle represents an end point, and the solid line refers to an optimal path planned by the traditional a-x algorithm. In the figure, it is clearly apparent that the optimal path is at a turn, inevitably with "stiff" polylines. This is caused by the a-algorithm being based on the discrete nature of the grid map. These fold lines can significantly reduce the operating efficiency of the AGV in practical AGVs. Because the AGV has to stop at the turning point if it works according to the route, and continue to move forward after completing the pivot rotation of the corresponding angle (if it is forced to drive and turn at the same time, there is a great risk of hitting an obstacle). It is these acceleration phases from rest to normal operation that cause the AGV to operate less efficiently. In recent years, despite improved algorithms such as Hybird A algorithm, heuristic search is performed under a continuous coordinate system, and a nonlinear optimization method is combined to further smooth the path. The method can solve the problem of broken lines of paths in the grid map to a certain extent, but the heuristic function and the optimization process of the algorithm are too complex, so that the running efficiency is reduced, and great difficulty is brought to the real-time motion control of the AGV.
Disclosure of Invention
The invention provides an improved A method which is simple in solving process, effectively solves the problem of path broken lines in a grid map by combining motion control and greatly improves the working efficiency of an AGV.
The invention is realized by the following technical scheme:
an AGV path planning method combining with improved A of motion control comprises the following steps:
step 1, according to the requirements of an operation task and the performances of an AGV mechanical structure and an electronic system, designing and calculating the motion control of the AGV, which is suitable for smooth left turning and smooth right turning of the task area, and requiring that the speed of the AGV before and after turning is unchanged;
step 2, obtaining the minimum turning radius of the AGV in the task area according to the motion control mode obtained in the step 1, carrying out rasterization processing on the task area based on the minimum turning radius, and modifying the definition of adjacent nodes in the A-x algorithm by combining the motion control method;
step 3, based on the motion control of the step 1 and the grid processing of the step 2, carrying out path planning on the task area by using an improved A-x algorithm, and finding out a smooth path from a starting point to an end point;
step 4, based on the path planned in the step 3, carrying out integration processing, and deleting redundant path points for use in motion control;
and step 5, based on the path points integrated in the step 4, the AGV controls the motion according to the speed mark quantity of the path points to complete the operation.
A further improvement of the invention is that in step 1 the job task requirements comprise a job time and a task area scale.
A further improvement of the invention is that in step 1, the performance of the electronic system includes a maximum speed limit, a maximum acceleration limit, a minimum turning radius, and an on-board computer calculated speed.
The invention is further improved in that in the step 1, in the turning process, the speed value of the linear speed undergoes the stage of first reduction and then increase, and the speed value of the angular speed undergoes the stage of first increase and then decrease.
A further improvement of the present invention is that, in step 3, the improved a-algorithm specifically is:
according to the requirements of the operation tasks and the performances of the AGV mechanical structure and the electronic system, the smooth left-turning and smooth right-turning motion control of the AGV, which is suitable for the task area, is designed and calculated, and the speed of the AGV before and after turning is required to be constant; in the turning process, the speed value of the linear speed is subjected to a stage of decreasing first and then increasing, and the linear speed before and after turning is kept unchanged; the speed value of the angular velocity goes through the stage of increasing first and then decreasing, and the turning process of the AGV can be smooth as much as possible while the angular velocities before and after turning are ensured to be 0; according to the motion control mode, obtaining the minimum turning radius of the AGV in the task area, performing rasterization processing on the task area based on the minimum turning radius, modifying the definition of adjacent nodes in the A-x algorithm by combining the motion control method, and controlling the motion defined in the step 1 to know that when the speed direction of the AGV at the current node is the advancing direction, the adjacent nodes are reduced into three nodes in front by eight grid nodes in the traditional A-x algorithm; the calculation of the cost value of the corresponding node is also based on the graphic approximate track length, namely, keeping the forward g (n) increment smaller than the smooth turning g (n) increment, and the values of the forward g (n) increment and the smooth turning g (n) increment are respectively represented by approximate track length difference values, wherein the forward g (n) increment is kept approximate to the straight line length, the smooth turning g (n) increment is approximate to the quarter of the arc length, and the generated final track is deviated to the straight line; carrying out path planning on the task area by using an improved A-algorithm, and finding out a smooth path from a starting point to an end point; and finally, the planned path is integrated, redundant path points are deleted for motion control, and the AGV performs motion control according to the speed mark quantity of the path points to complete the operation.
A further improvement of the invention is that the job task requirements include job time and task area dimensions.
A further improvement of the present invention is that the AGV mechanical structure and electronic system performance includes maximum speed limits, maximum acceleration limits, minimum turn radius, and on-board computer calculated speed.
In a further improvement of the present invention, in step 4, the redundant path points are path points which are on the same straight line and have the same speed direction index.
The invention has at least the following beneficial technical effects:
the invention improves the traditional A-x algorithm by combining motion control and path planning, and finishes the planning task under the constraint condition of meeting the task requirement and the calculation performance. It has the following advantages:
firstly, the method comprises the following steps: based on the discrete idea, the motion control and the path planning are combined in a simple and feasible mode, so that the path planned by the method can be directly used for controlling the motion of the AGV (the motion control can be realized by utilizing the newly-added motion direction mark quantity of each node), and the method does not need to be used for controlling the motion of the AGV by firstly carrying out the processing processes of nonlinear optimization and the like on a global path as the conventional algorithm. The structure is simple, and the use is convenient;
secondly, the method comprises the following steps: aiming at different task requirements and different calculation performances, the method can adaptively calculate the motion control corresponding to smooth left turn and smooth right turn (according to the algorithm shown in FIG. 3, the result can be obtained by combining the information of the maximum acceleration, the maximum speed and the like of the AGV). Furthermore, according to the minimum turning radius of the corresponding motion control, the resolution of the corresponding grid map can be given in a self-adaptive manner;
thirdly, the method comprises the following steps: based on the motion control of smooth left and right turns, the planned path of the present invention has smoothness that better conforms to the AGV motion than the traditional A-x algorithm, as shown in FIG. 5.
Thirdly, the steps of: based on the motion control of the invention, the AGV can avoid the phenomenon of in-situ rotation in the operation process, thereby saving the time and energy consumed in the acceleration starting stage, improving the working efficiency, reducing the energy consumption and reducing the operation time;
fourthly: and the definition of adjacent nodes in the A-algorithm is modified, so that the expanded search of the nodes which are not in accordance with the AGV motion characteristics in the planning process is reduced, and the calculated amount of the path planning algorithm is greatly reduced.
Drawings
Fig. 1 is a flow chart of a conventional a-algorithm in the prior art.
Fig. 2 is a schematic diagram of a path drawn by the conventional a-x algorithm.
Fig. 3 is a schematic diagram showing the variation of the speed of the AGV during a smooth turn by the motion control section according to the embodiment of the present invention, in which fig. 3 (a) is a linear speed variation curve during a turn, and fig. 3 (b) is an angular speed variation curve during a turn.
Fig. 4 is a schematic diagram of a modification defined by the improved a algorithm part on adjacent nodes in the embodiment of the present invention.
Fig. 5 is a schematic diagram of a path planned by the improved a-algorithm in the embodiment of the present invention.
Fig. 6 is a schematic view of a competition platform to which the present invention is applied, fig. 6 (a) is a top view, and fig. 6 (b) is an oblique view.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art. It should be noted that the embodiments and features of the embodiments may be combined with each other without conflict. The present invention will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.
The invention provides an AGV path planning method combining motion control and improving A, which comprises the following steps,
step 1, according to the requirements of an operation task (including operation time, task area size and the like) and the performances of an AGV mechanical structure and an electronic system (including maximum speed limit, maximum acceleration limit, minimum turning radius, calculation speed of an onboard computer and the like), the motion control of the AGV suitable for smooth left turning and smooth right turning of the task area is designed and calculated, and the speed of the AGV before and after turning is required to be unchanged. As shown in fig. 3, in the process of turning, the speed value of the linear velocity is subjected to a stage of decreasing first and then increasing, the linear velocity before and after turning is kept unchanged, so that the AGV can conveniently join and work, the turning radius is reduced as much as possible, and a sufficient selection space is reserved for the environment rasterization size; the speed value of the angular velocity goes through the stage of increasing first and then decreasing, and the turning process of the AGV can be smooth as far as possible while the angular velocity before and after turning is ensured to be 0.
And 2, obtaining the minimum turning radius of the AGV in the task area according to the motion control mode obtained in the step 1. Based on the turning radius, the task area is rasterized, and the definition of the adjacent nodes in the algorithm a is modified by combining the motion control method, as shown in fig. 4. As can be seen from the motion control defined in step 1, when the speed direction of the AGV at the current node is the forward direction, the adjacent nodes are reduced from the eight grid nodes in the conventional a-x algorithm to three nodes in front (obtained by smooth left turn, forward keeping and smooth right turn, respectively, the model better conforms to the actual situation of the AGV, for example, the AGV in driving is difficult to turn around suddenly to reach the node position behind the AGV), and the current speed direction is the same as the left direction and the right direction. Other nodes are no longer included in the definition of neighboring nodes for smoothness and efficiency. The calculation of the respective node cost values is also based on the illustrated approximate track length, i.e., the g (n) increments that remain in progress are smaller than the g (n) increments for smooth turns, which values can be expressed in terms of approximate track length differences, respectively (g (n) increments that remain in progress are approximately a straight length, g (n) increments for smooth turns are approximately a quarter of a circular arc length). Therefore, the generated final track is deviated to the straight movement, and the working efficiency of the AGV is improved.
And 3, planning a path of the task area by using an improved A-star algorithm based on the motion control of the step 1 and the grid processing of the step 2, and finding out a smooth path from the starting point to the end point. Compared with the traditional algorithm, the improved A-algorithm modifies the definition of adjacent nodes and the calculation of cost values, not only meets the requirement of path smoothness, but also can reduce the calculated amount on the premise of meeting the approximate optimality of the path, and simultaneously adds a mark amount in the speed direction for each node for the convenience of subsequent motion control. The path planned by the improved A-algorithm is close to the optimal path, smoothness and calculation efficiency are higher than those of the traditional A-algorithm, and motion control of the AGV is better fitted.
And 4, performing integration processing based on the path planned in the step 3, deleting redundant path points (the redundancy refers to the fact that the paths are on the same straight line and the speed direction mark quantity is the same) and using the path points for motion control.
And step 5, based on the path points integrated in the step 4, the AGV controls the motion according to the speed mark quantity of the path points to complete the operation.
Preferably, the improved a algorithm is specifically:
according to the requirements of a work task (including work time, a scale of a task area and the like) and the performances of a mechanical structure and an electronic system of the AGV (including maximum speed limit, maximum acceleration limit, minimum turning radius, calculation speed of an on-board computer and the like), the motion control of the AGV suitable for smooth left turning and smooth right turning of the task area is designed and calculated, and the speed of the AGV before and after turning is required to be unchanged. As shown in fig. 3, in the process of turning, the speed value of the linear velocity is subjected to a stage of decreasing first and then increasing, the linear velocity before and after turning is kept unchanged, so that the AGV can conveniently join and work, the turning radius is reduced as much as possible, and a sufficient selection space is reserved for the environment rasterization size; the speed value of the angular velocity goes through the stages of increasing first and then decreasing, and the turning process of the AGV can be smooth as far as possible while the angular velocity before and after turning is ensured to be 0. And according to the motion control mode, obtaining the minimum turning radius of the AGV in the task area. Based on the turning radius, the task area is rasterized, and the definition of the adjacent nodes in the algorithm a is modified by combining the motion control method, as shown in fig. 4. As can be seen from the motion control defined in step 1, when the speed direction of the AGV at the current node is the forward direction, the adjacent nodes are reduced from the eight grid nodes in the conventional a-x algorithm to three nodes in front (obtained by smooth left turn, forward keeping and smooth right turn, respectively, the model better conforms to the actual situation of the AGV, for example, the AGV in driving is difficult to turn around suddenly to reach the node position behind the AGV), and the current speed direction is the same as the left direction and the right direction. Other nodes are no longer included in the definition of neighboring nodes for smoothness and efficiency. The calculation of the cost values of the respective nodes is also based on the illustrated approximate trajectory lengths, i.e., g (n) increments that remain forward are smaller than g (n) increments for smooth turns, which values can be represented by approximate trajectory length differences, respectively (g (n) increments that remain forward are approximately a straight length and g (n) increments for smooth turns are approximately a quarter of a circular arc length). Therefore, the generated final track is deviated to straight movement, and the working efficiency of the AGV is improved. And planning the path of the task area by using an improved A-x algorithm, and finding out a smooth path from the starting point to the end point. Compared with the traditional algorithm, the improved A-algorithm modifies the definition of adjacent nodes and the calculation of cost values, meets the requirement of path smoothness and reduces the calculation amount on the premise of meeting the approximate optimality of the path, and simultaneously adds a mark amount in the speed direction for each node for the convenience of the following motion control. The path planned by the improved A-algorithm is close to the optimal path, smoothness and calculation efficiency are higher than those of the traditional A-algorithm, and motion control of the AGV is better fitted. And finally, performing integration processing on the planned path, and deleting redundant path points (the redundancy means that the paths are on the same straight line and the mark quantity of the speed direction is the same) for motion control. The AGV can perform motion control according to the speed mark quantity of the path point to complete the operation.
Examples
The AGV path planning combining with the improved A of the motion control mainly comprises two parts of a motion control design and a path planning, and each part is as follows:
1. and (3) motion control design:
as shown in FIG. 4, based on the present invention, the AGV movement between adjacent nodes includes only three modes, smooth left turn, keep forward and smooth right turn, as previously described. The designer needs to adjust the variation law of linear and angular velocities corresponding to smooth left and right turns according to the task requirements (including operation time, task area size, etc.) of operation in practical applications and the performance of the AGV mechanical structure and electronic system (including maximum speed limit, maximum acceleration limit, minimum turning radius, and on-board computer calculated speed, etc.), as shown in fig. 3. For example, in the first fili group-simulation group obstacle avoidance challenge (hereinafter referred to as "obstacle avoidance challenge") project of Robocup chinese robot race 2020, based on the performance of the turtlebot robot, the parameters are set as follows:
v 1 ＝v 2 ＝0.6m/s
w 1 ＝0rad/s，w 2 ≈7.54rad/s
the uniform acceleration and uniform deceleration process in fig. 3 is approximated by a high-frequency message mechanism of the ROS system, that is, a continuous change process of the speed is simulated by sending a uniformly increased or decreased speed value message to the AGV high frequency, the distance error can be reduced to a millimeter level by reasonable parameter setting, and the angle error is controlled within 0.5 degree.
After the control parameters of smooth left turn and smooth right turn are set, the minimum turning radius required by turning based on the parameters can be further obtained, and the minimum turning radius is an important parameter of the next rasterization map. For example, in the obstacle avoidance challenge game described above, the turtlebot minimum turning radius is 0.15m.
2. Path planning:
after the motion control design of the corresponding AGV is completed, the path planning initially requires rasterization of the task area. The resolution of the grid map, i.e. the side length of each grid, is an important parameter that affects the computational efficiency and planning results. The designer in the motion control design can obtain the minimum turning radius of the AGV, and the resolution ratio of the grid map cannot be smaller than the value. In order to plan a smooth path closer to the optimal path, it is proposed to choose the minimum turning radius as the resolution of the grid map. For example, in the obstacle avoidance challenge race described above, the resolution of the grid map is 0.15m.
As shown in fig. 4, the definition of the neighboring nodes is modified to calculate the trajectory according to the conventional a-algorithm flow. The specific programming implementation flow is shown in the pseudo code of fig. 6: two lists, OPEN and CLOSED, are also maintained. Initially, adding a starting point serving as a first node into an OPEN list, calculating a cost value, setting a father node to be null, setting an initial speed direction flag quantity to be 'forward', and entering a loop: and taking out the node with the minimum cost value in the current OPEN list, setting the node as the current node N, and putting the node into the CLOSED list. According to the cost value calculation method (based on the approximate track length), cost values of three adjacent nodes of the current node N and father nodes of the current node N are calculated and updated, and the updating method is that the node with the lower cost value of the node is selected as the father node of the node. And after all the adjacent nodes are updated, the nodes including the father node information, the cost value and the speed direction mark amount are added into an OPEN list, and then the loop iteration is carried out. If the current node N is the terminal after a certain cycle, directly adding the current node N into a CLOSED list, and obtaining a complete Path point set Path in a mode of reversely searching a father node; if the OPEN list is empty when the end point is not reached, the current environment is indicated to have no path from the starting point to the end point, and the algorithm is ended.
After obtaining the set of path points and deleting redundant path points (if the speed direction flag quantities of some path points are the same and are in the same straight line, only the head and the tail path points are reserved), the AGV can perform operation according to the speed direction flag quantities of each path point. For example, if the speed direction flag of the current node N is "forward" and the speed direction flag of the next node M is "left turn", the AGV may perform a smooth left turn according to the motion control method described above. Because the motion control command can be debugged in advance, the AGV can quickly respond, and the operating efficiency is improved.
As shown in FIG. 5, the method and the device effectively solve the problem of path broken lines in the grid map by designing specific motion control, so that the generated path is smoother and is beneficial to the driving of the AGV.
The invention has been applied to the obstacle avoidance challenge game project of FIRA mini-simulation group of Robocup China robot major game in 2020, and the national champion is obtained. The primary scoring criteria for this project is speed, i.e., the time the AGV takes from the start to the end. Therefore, the invention can fully exert the advantages in the project: by means of the combination of motion control and path planning, the calculation amount is greatly reduced; the path is smoothed by a discrete idea, the slow starting stage of the AGV is avoided, and the running speed of the AGV is greatly improved; the redefinition of the adjacent nodes reduces the time spent on node searching and further improves the calculation speed. After testing on the platform of the obstacle avoidance challenge race, as shown in the map shown in fig. 6, the conventional a-star algorithm is used to drive the turnlebot robot to travel from the starting point to the end point, which requires 9.21 seconds on average, whereas with the present invention, it requires only 6.45 seconds on average, which is about 30% higher.
Although the invention has been described in detail hereinabove with respect to a general description and specific embodiments thereof, it will be apparent to those skilled in the art that modifications or improvements may be made thereto based on the invention. Accordingly, such modifications and improvements are intended to be within the scope of the invention as claimed.
Claims (4)
1. An AGV path planning method combining with improved A of motion control is characterized by comprising the following steps:
step 1, according to the requirements of an operation task and the performances of an AGV mechanical structure and an electronic system, designing and calculating the motion control of the AGV, which is suitable for smooth left turning and smooth right turning of a task area, and requiring that the speed of the AGV before and after turning is unchanged; in the turning process, the speed value of the linear speed is subjected to a first reduction and then increase stage, and the linear speed before and after turning is kept unchanged; the speed value of the angular velocity goes through the stage of increasing first and then decreasing, and the turning process of the AGV can be smooth as much as possible while the angular velocities before and after turning are ensured to be 0;
step 2, obtaining the minimum turning radius of the AGV in the task area according to the motion control mode obtained in the step 1, carrying out rasterization processing on the task area based on the minimum turning radius, and modifying the definition of adjacent nodes in the A-x algorithm by combining the motion control method; according to the motion control defined in the step 1, when the speed direction of the AGV at the current node is the advancing direction, the adjacent nodes are reduced to three front nodes from eight grid nodes in the traditional A-star algorithm; the calculation of the cost value of the corresponding node is based on the length of the approximate motion track and keeps advancingg(n) With increments smaller than smooth turnsg(n) Increment, the values of which are respectively expressed by the difference of the approximate motion track lengths, wherein the advance is keptg(n) With increments of straight length, smoothly-turningg(n) The increment is one fourth of the arc length, and the generated final track is deflected to go straight;
step 3, based on the motion control in the step 1 and the rasterization processing in the step 2, carrying out path planning on the task area by using an improved A-x algorithm, and finding out a smooth path from a starting point to an end point;
step 4, based on the path planned in the step 3, carrying out integration processing, and deleting redundant path points for use in motion control;
and step 5, based on the path points integrated in the step 4, the AGV controls the movement according to the speed mark quantity of the path points to complete the operation.
2. The AGV path planning method according to claim 1 with improved a in combination with motion control, wherein in step 1, the job task requirements include a job time and a task area size.
3. The AGV path planning method in combination with motion control improvement a of claim 1, wherein the performance of the electronic system includes maximum speed limit, maximum acceleration limit, minimum turning radius and on-board computer calculated speed in step 1.
4. The AGV path planning method according to claim 1, wherein the redundant path points in step 4 are path points that are on the same straight line and have the same speed index.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110605550.6A CN113359721B (en) | 2021-05-31 | 2021-05-31 | Improved A-based AGV path planning method combined with motion control |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110605550.6A CN113359721B (en) | 2021-05-31 | 2021-05-31 | Improved A-based AGV path planning method combined with motion control |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113359721A CN113359721A (en) | 2021-09-07 |
| CN113359721B true CN113359721B (en) | 2022-10-25 |
Family
ID=77530611
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110605550.6A Active CN113359721B (en) | 2021-05-31 | 2021-05-31 | Improved A-based AGV path planning method combined with motion control |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113359721B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113848985B (en) * | 2021-11-01 | 2022-09-09 | 西南交通大学 | Multi-carrier-based unmanned aerial vehicle path planning method based on Hybird A-star algorithm |
| CN114578808A (en) * | 2022-01-10 | 2022-06-03 | 美的集团(上海)有限公司 | Path planning method, electronic device, computer program product, and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108563243A (en) * | 2018-06-28 | 2018-09-21 | 西北工业大学 | A kind of unmanned aerial vehicle flight path planing method based on improvement RRT algorithms |
| CN110320933A (en) * | 2019-07-29 | 2019-10-11 | 南京航空航天大学 | Unmanned plane avoidance motion planning method under a kind of cruise task |
| CN110487279A (en) * | 2019-08-27 | 2019-11-22 | 东南大学 | A kind of paths planning method based on improvement A* algorithm |
| CN111452868A (en) * | 2019-01-22 | 2020-07-28 | 北京京东尚科信息技术有限公司 | Curve turning control method and device, storage medium and vehicle |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102008050951A1 (en) * | 2008-10-10 | 2010-04-22 | Eads Deutschland Gmbh | Computer time optimized route planning for aircraft |
| CN104075717A (en) * | 2014-01-21 | 2014-10-01 | 武汉吉嘉伟业科技发展有限公司 | Unmanned plane airline routing algorithm based on improved A* algorithm |
| CN104121903B (en) * | 2014-07-04 | 2017-06-30 | 沈阳航空航天大学 | A kind of rolling Route planner based on boundary value problem |
| CN109542098A (en) * | 2018-11-06 | 2019-03-29 | 上海威瞳视觉技术有限公司 | A kind of AGV paths planning method based on minimum turning cost |
| CN112327927B (en) * | 2020-11-25 | 2021-11-26 | 北京理工大学 | Multi-angle strike track planning method for formation unmanned aerial vehicles based on grid planning algorithm |
-
2021
- 2021-05-31 CN CN202110605550.6A patent/CN113359721B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108563243A (en) * | 2018-06-28 | 2018-09-21 | 西北工业大学 | A kind of unmanned aerial vehicle flight path planing method based on improvement RRT algorithms |
| CN111452868A (en) * | 2019-01-22 | 2020-07-28 | 北京京东尚科信息技术有限公司 | Curve turning control method and device, storage medium and vehicle |
| CN110320933A (en) * | 2019-07-29 | 2019-10-11 | 南京航空航天大学 | Unmanned plane avoidance motion planning method under a kind of cruise task |
| CN110487279A (en) * | 2019-08-27 | 2019-11-22 | 东南大学 | A kind of paths planning method based on improvement A* algorithm |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113359721A (en) | 2021-09-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109945873B (en) | Hybrid path planning method for indoor mobile robot motion control | |
| CN113359721B (en) | Improved A-based AGV path planning method combined with motion control | |
| CN110609547B (en) | Mobile robot planning method based on visual map guidance | |
| Lai et al. | Enhanced center constraint weighted A* algorithm for path planning of petrochemical inspection robot | |
| CN110006429A (en) | A kind of unmanned boat path planning method based on depth optimization | |
| CN113359718B (en) | Method and equipment for fusing global path planning and local path planning of mobile robot | |
| CN113296520B (en) | Routing inspection robot path planning method integrating A and improved gray wolf algorithm | |
| Yang et al. | LF-ACO: An effective formation path planning for multi-mobile robot | |
| Klančar et al. | Drivable path planning using hybrid search algorithm based on E* and Bernstein–Bézier motion primitives | |
| CN112683290A (en) | Vehicle track planning method, electronic equipment and computer readable storage medium | |
| CN111664851B (en) | Robot state planning method and device based on sequence optimization and storage medium | |
| CN116203965A (en) | AGV-based path planning algorithm for improving ant colony algorithm | |
| CN115167398A (en) | Unmanned ship path planning method based on improved A star algorithm | |
| Zhang et al. | Robot navigation with reinforcement learned path generation and fine-tuned motion control | |
| CN117032231A (en) | Multi-agent path planning method based on improved RRT | |
| Zhao et al. | A compound path planning algorithm for mobile robots | |
| Li et al. | Research on AUV path planning based on improved ant colony algorithm | |
| Broderick et al. | Maximizing coverage for mobile robots while conserving energy | |
| Junyi et al. | Path planning for USV with FG-DA-RRT algorithm | |
| Wang et al. | OGBPS: orientation and gradient based path smoothing algorithm for various robot path planners | |
| Wu et al. | Smooth path planning method of agricultural vehicles based on improved Hybrid A | |
| Yu et al. | Robot path planning based on improved A* algorithm | |
| Bao et al. | Improved a-star algorithm for mobile robot path planning based on sixteen-direction search | |
| Chen et al. | Research on Improved A* Algorithm of Fusion Auction Mechanism | |
| Aine et al. | Integrating planning and control for efficient Path planning in the presence of environmental disturbances |
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 |