CN114323028A - Path planning method, system, device and medium for self-adaptive map - Google Patents
Path planning method, system, device and medium for self-adaptive map Download PDFInfo
- Publication number
- CN114323028A CN114323028A CN202210256226.2A CN202210256226A CN114323028A CN 114323028 A CN114323028 A CN 114323028A CN 202210256226 A CN202210256226 A CN 202210256226A CN 114323028 A CN114323028 A CN 114323028A
- Authority
- CN
- China
- Prior art keywords
- target
- algorithm
- path
- path planning
- importing
- 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.)
- Granted
Links
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a path planning method, a system, equipment and a medium of a self-adaptive map, which belong to the technical field of data processing and specifically comprise the following steps: acquiring environmental data of a target scene and preprocessing the environmental data to obtain a target data set; importing the target data set into a matching model to generate a matching result; judging whether the matching result is of a sparse type; if yes, importing the target data into an A-algorithm for path planning and generating a target path; and if not, importing the target data into an artificial potential field-based RRT-connect algorithm, selecting a plurality of points in a target scene as root nodes to establish a random tree, and superposing a gravitational field on each root node to search nodes to generate a target path. According to the scheme, an algorithm selection mechanism is added, and the RRT random tree is guided to expand through the artificial potential field in a high-complexity environment, so that the meaningless expansion time is reduced, and the planning efficiency, the adaptability and the accuracy are improved.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a method, a system, equipment and a medium for planning a path of a self-adaptive map.
Background
Currently, path planning is a hot spot of mobile robot research. The rapid development of science and technology and the continuous improvement of automation level enable robots to be widely applied in life, and the robot path planning technology is also paid more and more attention as a core technology of robot control. Path planning is a technology for finding an optimal path from a starting point to an end point by a robot according to specific environmental conditions and task objectives. In the past decades, scholars have proposed many path planning algorithms and have achieved good results in practical applications, such as: artificial potential field method, rapid-exploration RRT random tree algorithm, genetic algorithm, A-star algorithm, reinforcement learning and the like. Path planning can be divided into global path planning and local path planning according to the degree of confidence in the environmental information. And the global path planning carries out path planning according to all map information, and the local path planning selects an optimal path reaching a certain point from the current node according to the distribution condition of the obstacles at the local position.
However, the existing path planning method cannot adapt to a complex environment or is slow in search speed, and if guidance of intermediate points is lacked in a large-scale complex environment, an optimal path may not be obtained, or even a target point cannot be reached.
Therefore, the existing path planning method has the problems of poor planning efficiency, poor adaptability and poor accuracy.
Disclosure of Invention
In view of this, the present invention provides a method, a system, a device and a medium for path planning of an adaptive map, which at least partially solve the problem in the prior art that the planning efficiency, adaptability and accuracy are poor.
In a first aspect, the present invention provides a method for planning a path of a self-adaptive map, including:
acquiring environmental data of a target scene and preprocessing the environmental data to obtain a target data set;
importing the target data set into a matching model to generate a matching result;
judging whether the matching result is of a sparse type;
if the matching result is of a sparse type, importing the target data into an A-x algorithm for path planning and generating a target path;
and if the matching result is not of a sparse type, importing the target data into an artificial potential field-based RRT-connect algorithm, selecting a plurality of points in the target scene as root nodes to establish a random tree, and superposing a gravitational field on each root node to perform node search to generate a target path.
According to a specific implementation manner of the present invention, the step of acquiring the environmental data of the target scene and preprocessing the environmental data to obtain the target data set includes:
and carrying out data noise reduction on the environmental data by adopting a grouping average and moving average algorithm and then carrying out normalization to obtain the target data set.
According to a specific implementation manner of the present invention, before the step of importing the target data set into a matching model and generating a matching result, the method further includes:
calculating the execution time of the sample path planning map under different algorithms as a first dimension;
calculating the average value of the distances between the central point of the sample path planning map and other obstacles as a second dimension;
and importing the first dimension and the second dimension into a k-means clustering algorithm for clustering, and training to obtain a matching model associated with the map and the algorithm.
According to a specific implementation manner of the present invention, the step of determining whether the matching result is a sparse type includes:
calculating the sparsity degree according to the target data set;
if the sparseness degree is larger than a threshold value, the matching result is of a non-sparse type;
and if the sparsity degree is less than or equal to the threshold value, the matching result is of a sparse type.
According to a specific implementation manner of the present invention, the step of introducing the target data into an artificial potential field-based RRT-connect algorithm, selecting a plurality of points in the target scene as root nodes to establish a random tree, and superimposing a gravitational field on each root node to perform node search, thereby generating a target path includes:
establishing a random tree by taking the starting point and the end point of the target scene as the root nodes, randomly generating a plurality of points in the non-obstacle area of the target scene as the root nodes, and establishing the random tree;
iterating one said gravitational field over each said random tree;
obtaining a new node generation formula according to a random expansion function corresponding to the random tree and a gravity function corresponding to the gravity field;
and calculating the target path according to the new node generation formula.
According to a specific implementation manner of the present invention, before the step of obtaining the new node generation formula according to the random spreading function corresponding to the random tree and the gravity function corresponding to the gravity field, the method further includes:
and adjusting the step sizes of the random expansion function and the gravitation function according to the sparsity degree.
In a second aspect, the present invention provides a path planning system for an adaptive map, including:
the acquisition module is used for acquiring environmental data of a target scene and preprocessing the environmental data to obtain a target data set;
the matching module is used for importing the target data set into a matching model to generate a matching result;
the judging module is used for judging whether the matching result is of a sparse type;
the generating module is used for importing the target data into an A-x algorithm for path planning and generating a target path if the matching result is of a sparse type;
and if the matching result is not of a sparse type, importing the target data into an artificial potential field-based RRT-connect algorithm, selecting a plurality of points in the target scene as root nodes to establish a random tree, and superposing a gravitational field on each root node to perform node search to generate a target path.
In a third aspect, the present invention also provides an electronic device, including:
at least one processor; and the number of the first and second groups,
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of path planning for an adaptive map of the first aspect or any implementation of the first aspect.
In a fourth aspect, the present invention also provides a non-transitory computer-readable storage medium storing computer instructions for causing a computer to perform the method for path planning of an adaptive map in the first aspect or any implementation manner of the first aspect.
In a fifth aspect, the present invention also provides a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions which, when executed by a computer, cause the computer to perform the method for path planning of an adaptive map in the first aspect or any implementation form of the first aspect.
The path planning scheme of the self-adaptive map comprises the following steps: acquiring environmental data of a target scene and preprocessing the environmental data to obtain a target data set; importing the target data set into a matching model to generate a matching result; judging whether the matching result is of a sparse type; if the matching result is of a sparse type, importing the target data into an A-x algorithm for path planning and generating a target path; and if the matching result is not of a sparse type, importing the target data into an artificial potential field-based RRT-connect algorithm, selecting a plurality of points in the target scene as root nodes to establish a random tree, and superposing a gravitational field on each root node to perform node search to generate a target path.
The invention has the beneficial effects that: according to the scheme, an algorithm selection mechanism is added, a model base is established in advance to adapt to a changing environment, a proper algorithm is selected, the algorithm can automatically select the fusion of the RRT-connect and the artificial potential field in the high-complexity environment, the RRT random tree is guided to expand through the artificial potential field, the time of meaningless expansion is shortened, and the planning efficiency, the adaptability and the accuracy are improved.
Drawings
In order to more clearly illustrate the technical solution of the present invention, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic flow chart of a path planning method for an adaptive map according to the present invention;
fig. 2 is a schematic flow chart of an a-algorithm according to the present invention;
FIG. 3 is a schematic flow chart of an RRT-connect algorithm provided by the present invention;
fig. 4 is a schematic structural diagram of a path planning system of an adaptive map according to the present invention;
fig. 5 is a schematic diagram of an electronic device provided in the present invention.
Detailed Description
The present invention will be described in detail below with reference to the accompanying drawings.
The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. It is to be understood that the described embodiments are merely exemplary of the invention, and not restrictive of the full scope of the invention. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It is to be noted that the features in the following embodiments and examples may be combined with each other without conflict. 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.
It is noted that various aspects of the embodiments are described below within the scope of the appended claims. It should be apparent that the aspects described herein may be embodied in a wide variety of forms and that any specific structure and/or function described herein is merely illustrative. Based on the disclosure, one skilled in the art should appreciate that one aspect described herein may be implemented independently of any other aspects and that two or more of these aspects may be combined in various ways. For example, an apparatus may be implemented and/or a method practiced using any number of the aspects set forth herein. Additionally, such an apparatus may be implemented and/or such a method may be practiced using other structure and/or functionality in addition to one or more of the aspects set forth herein.
It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention, and the drawings only show the components related to the present invention rather than the number, shape and size of the components in practical implementation, and the type, quantity and proportion of the components in practical implementation can be changed freely, and the layout of the components can be more complicated.
In addition, in the following description, specific details are provided to facilitate a thorough understanding of the examples. However, it will be understood by those skilled in the art that the aspects may be practiced without these specific details.
Path planning is a hot spot of mobile robot research. The rapid development of science and technology and the continuous improvement of automation level enable robots to be widely applied in life, and the robot path planning technology is also paid more and more attention as a core technology of robot control. Path planning is a technology for finding an optimal path from a starting point to an end point by a robot according to specific environmental conditions and task objectives. In the past decades, scholars have proposed many path planning algorithms and have achieved good results in practical applications, such as: artificial potential field method, rapid-exploration RRT random tree algorithm, genetic algorithm, A-star algorithm, reinforcement learning and the like. Path planning can be divided into global path planning and local path planning according to the degree of confidence in the environmental information. And the global path planning carries out path planning according to all map information, and the local path planning selects an optimal path reaching a certain point from the current node according to the distribution condition of the obstacles at the local position.
An artificial potential field method, a common local path planning method, is a virtual force method proposed by Khatib for robot motion planning. The basic idea is to embody the influence of objects and obstacles on the movement of the robot into an artificial potential field. The potential energy at the target is low and the potential energy at the obstacle is high. The potential difference generates the attraction of the target to the robot and the repulsion of the obstacle to the robot, and the resultant force controls the robot to move towards the target point along the negative gradient direction of the potential field. The artificial potential field method is convenient to calculate, and the obtained path is safe and smooth, but a complex potential field environment may generate local minimum points outside a target point, so that the robot cannot reach the target. In order to solve the problem of local minima of the artificial potential field method, researchers have proposed various improvements. Mainly divided into two directions: one is to construct a suitable potential function to reduce or avoid the occurrence of local minima; another method is to combine other methods to make the robot leave the local minimum point after the robot meets the local minimum point. The former generally requires global map information and depends on the shape of the obstacle. It is difficult to apply when the environment is complicated. The latter method mainly utilizes methods such as a search method, a multi-potential field method and a wall-following walking method to enable the robot to leave a local minimum point. The searching method utilizes the strategies of optimal priority, simulated annealing, random searching and the like to search for points with potential field values lower than local minimum points so as to enable the robot to continue moving. The efficiency of the search method is low due to the lack of heuristic information in most of the unknown environments. The multi-potential-field method constructs potential functions with the same global minimum points and different local minimum points, and when the robot falls into a certain local minimum point, the planner switches the potential functions to enable the robot to leave the point. But such multiple potential fields are difficult to construct in unknown environments and the method may result in the robot returning to a local minimum point where it has escaped. Because the local minimum point is generated by the combined action of the repulsive potential field and the gravitational potential field of one or more barriers, the position of the local minimum point is not necessarily far away from the barriers, and the wall walking method utilizes the distance, so that the robot bypasses the barriers generating the local minimum points to continue to advance after meeting the local minimum point according to the surrounding action similar to a BUG algorithm. The method has high reliability and does not depend on prior information of the environment and the shape of the obstacle.
One of the commonly used global path planning algorithms is widely concerned due to convenience and rapidity thereof, and a rapid exploration Random Tree (RRT) is a typical representative thereof, has strong versatility, can solve the problem of space path planning of a robot in different dimensions, has a very high exploration success rate, but has very obvious defects, namely, the searching speed in a complex space is slow, and the obtained path is usually not an optimal solution thereof, and has a large deviation from the optimal path. And thirdly, the device has limitation when used in a dynamic environment. In order to overcome the defects of the traditional RRT algorithm, scholars at home and abroad respectively improve the RRT algorithm. The RRT-connect algorithm is improved by adopting a bidirectional search strategy and a self-adaptive step length method, and two random trees can be generated at a starting point and a target point simultaneously, so that the search efficiency is improved. The RRT algorithm of Karama et al finds an optimal or near optimal path when the number of iterations is large enough, but the search time is greatly increased. An inform-RRT algorithm is provided by Gamma cell and Srinivasa, an ellipse constraint sampling space is introduced for sampling optimization, the search time is greatly reduced, and the collision with obstacles is easy to occur in a dynamic environment.
The global path planning can search a global path from a starting point to an end point in a relatively short time no matter based on a sampling RRT algorithm or a heuristic search-based A algorithm, but local obstacles cannot be avoided, the adaptability to a local dynamic environment is not strong, and the requirement of avoiding the obstacles in an unknown environment cannot be met. The local path planning artificial potential field algorithm can be planned on line in real time and has good obstacle avoidance capacity, but if guidance of intermediate points is lacked in a large-scale complex environment, an optimal path can not be obtained, and even a target point can not be reached. Therefore, combining the advantages of the two methods, it is a hot spot in current research to realize the fusion of global and local algorithms for path planning.
In addition, the environment and the adaptive practice of the algorithm are also considered, when various environments are faced, a single algorithm is difficult to cover the problems, and the algorithm needs to be properly adjusted or selected according to different models in practical application.
The invention provides a path planning method of a self-adaptive map, which can be applied to a path planning process of a mobile robot scene.
Referring to fig. 1, a schematic flow chart of a path planning method for an adaptive map according to the present invention is shown. As shown in fig. 1, the method mainly comprises the following steps:
s101, collecting environmental data of a target scene and preprocessing the environmental data to obtain a target data set;
optionally, in step S101, acquiring environment data of a target scene and preprocessing the environment data to obtain a target data set, where the method includes:
and carrying out data noise reduction on the environmental data by adopting a grouping average and moving average algorithm and then carrying out normalization to obtain the target data set.
In specific implementation, considering that the nature of the K-means is a Euclidean distance-based data partitioning algorithm, the dimension with large mean and variance will have a decisive influence on the clustering of the data. Therefore, the data which is not normalized and unified in unit can not directly participate in the operation and comparison. Common data preprocessing methods are as follows: and (4) normalizing the data and standardizing the data. In addition, outliers or noisy data can have a large effect on the mean, resulting in center shifts, and therefore outlier cleaning of the data is also required. After the environmental data corresponding to the target scene is collected, considering that interference data may be included in the data, a block average and moving average algorithm may be adopted to perform data noise reduction on the environmental data to remove the interference data such as outliers and outliers, and then normalize and standardize the data to facilitate subsequent processing.
S102, importing the target data set into a matching model to generate a matching result;
optionally, in step S102, before importing the target data set into a matching model and generating a matching result, the method further includes:
calculating the execution time of the sample path planning map under different algorithms as a first dimension;
calculating the average value of the distances between the central point of the sample path planning map and other obstacles as a second dimension;
and importing the first dimension and the second dimension into a k-means clustering algorithm for clustering, and training to obtain a matching model associated with the map and the algorithm.
In specific implementation, at the initial stage of algorithm establishment, a representative path planning map can be selected as the sample path planning map, the representative path planning map and the execution time under different algorithms are tested, the shortest time is selected as one dimension, and the average value (called sparsity) of the distances from the center point to other obstacles is selected in the other dimension. And importing the data of the two dimensions into a training model of K-means.
The algorithm comprises the following steps: selecting initialized k samples as initial clustering centers, then importing the target data set with discarded outliers and abnormal points into Kmeans for clustering, and training a matching model of a map and an algorithm as the matching model. Meanwhile, considering that some models in the initial model matching library are not adapted any more in the subsequent algorithm execution process and a more adapted model appears in the subsequent practical application, the model library needs to be updated at this time.
Specifically, the model base can be replaced or newly added according to the accuracy rate after the model is replaced or whether the overall performance is improved; due to the dynamic change of the environment, some models in the system may fail and cannot be used any more within a long period of time, therefore, GC garbage collection should be performed on the models to improve the real-time speed of system prediction, and the model base is deleted according to whether the models are not matched for a long time or not.
S103, judging whether the matching result is of a sparse type;
further, in step S102, importing the target data set into a matching model to generate a matching result, including:
calculating the sparsity degree according to the target data set;
if the sparseness degree is larger than a threshold value, the matching result is of a non-sparse type;
and if the sparsity degree is less than or equal to the threshold value, the matching result is of a sparse type.
In specific implementation, after the matching result is obtained, the target data set may be imported into a matching model, a sparsity degree is calculated according to the target data set, if the sparsity degree is greater than a threshold value, the matching result is a non-sparse type, if the sparsity degree is less than or equal to the threshold value, the matching result is a sparse type, and after the specific type of the matching result is determined, a next operation flow is determined.
If the matching result is of a sparse type, executing step S104, importing the target data into an A-algorithm for path planning and generating a target path;
for example, when the matching result is a sparse type, the target data may be imported into an a-algorithm, and path planning is performed directly using the a-algorithm, as shown in fig. 2, the core idea of the a-algorithm is to guide the direction traversed by the algorithm by using a heuristic function. Each node of the network is evaluated using an evaluation function as follows
F (n) = g (n) + h (n), where f (n) is an evaluation cost from an originating node to a target node, g (n) is an evaluation cost from a specific originating node to a specific target node, and h (n) is a predicted evaluation cost from a current node to a target node. For H (n), a Manhattan estimation algorithm is adopted
H (n) = D (ab (xn-xgoal) + ab (yn-ygoal)), where D is manhattan distance, ab is grid square area, xn and yn are the horizontal and vertical positions of the current square, respectively, and xgoal and ygoal are the horizontal and vertical positions of the target square, respectively. The working principle of the A-algorithm is shown in the figure. The a-algorithm implementation requires two lists, OpenList and CloseList, for storing nodes on the path. OpenList is used to store the nodes to be checked, and when it is initialized, it only contains one start node, and CloseList is used to store the checked nodes, and the nodes in CloseList are not checked for the 2 nd time, and their initial values are null. The algorithm a performs path planning starting from the starting node s, and the nodes generated in the planning are stored in the two tables above. Each node n along the path has a pointer to its parent node, and each node n has the value of its associated valuation function f (n). Each loop of the a-algorithm selects the node n with the smallest f (n) value from the OpenList, and if n is the target node, the loop ends and the best path is found. If the node n is not the target node, the node n is taken out of the OpenList and stored in the CloseList. Meanwhile, all nodes adjacent to the node are checked, except the nodes in the CloseList, if the nodes are not in the OpenList, the nodes are stored in the OpenList, and the nodes just stored in the CloseList are taken as parent nodes of the newly stored nodes. By expanding the nodes in this way, finally, there will be a target node on the OpenList, the algorithm is terminated, and then the target node is moved from the end point to the parent node through the pointer back to the start point, which is the optimal path planned by the a-algorithm, that is, the target path.
And if the matching result is not of a sparse type, executing step S105, importing the target data into an artificial potential field-based RRT-connect algorithm, selecting a plurality of points in the target scene as root nodes to establish a random tree, and superposing a gravitational field on each root node to perform node search to generate a target path.
On the basis of the foregoing embodiment, step S105, importing the target data into an artificial potential field-based RRT-connect algorithm, selecting multiple points in the target scene as root nodes to build a random tree, and superimposing a gravitational field on each root node to perform node search to generate a target path, where the method includes:
establishing a random tree by taking the starting point and the end point of the target scene as the root nodes, randomly generating a plurality of points in the non-obstacle area of the target scene as the root nodes, and establishing the random tree;
iterating one said gravitational field over each said random tree;
obtaining a new node generation formula according to a random expansion function corresponding to the random tree and a gravity function corresponding to the gravity field;
and calculating the target path according to the new node generation formula.
Further, before the step of obtaining a new node generation formula according to the random spreading function corresponding to the random tree and the gravitational function corresponding to the gravitational field, the method further includes:
and adjusting the step sizes of the random expansion function and the gravitation function according to the sparsity degree.
The RRT-connect algorithm is a path planning algorithm based on random sampling and is evolved from the RRT algorithm. The algorithm takes the starting point and the target point of a path as root nodes of a random tree and opposite random expansion nodes, and leaf nodes are expanded by combining methods such as step length limitation, collision detection and the like in the node expansion process. And when the distance between two leaf nodes in the two random trees is smaller than a set fixed threshold, stopping the expansion of the random trees, and connecting each node to finish searching the path. The RRT-connect algorithm alternately expands the random tree from two nodes, so that the planning time can be shortened, and the search efficiency can be improved. The algorithm firstly initializes two random trees T1 and T2 by taking a starting point and an end point of a path as root nodes, randomly expands a 1 st tree T1 to obtain a node Xrand, continuously expands the nodes by adopting a greedy strategy-based method by taking the Xrand node as a direction of a 2 nd tree T2, and continuously detects collision in the expansion process until the node collides with an obstacle or the expansion is finished when the distance between the node and a certain node of the 1 st tree is smaller than a fixed threshold value.
If the distance between the node and the 1 st tree is smaller than a fixed threshold, the path is successfully generated, otherwise, the path and the 1 st tree continue to be alternately expanded. The flow chart of the RRT-connect algorithm is shown in FIG. 3. Compared with the RRT algorithm, the RRT-connect algorithm adds the 2 nd expansion node, and the 2 nd expansion node adopts a greedy-based expansion strategy, so that the path search time is reduced, and the path planning efficiency is improved. Although the RRT-connect algorithm is obviously improved in planning efficiency compared with the RRT algorithm, the execution efficiency of the RRT-connect algorithm is slightly insufficient for industries with high reaction sensitivity, and the RRT-connect algorithm guided based on the gravitational field is provided for solving the problem.
Compared with the RRT-connect algorithm, the algorithm is mainly improved in the following two aspects that 1) the nth random tree expansion node is added, and the original two random trees are changed into n +1 random trees, so that the node searching efficiency is improved. 2) And a gravitational field is respectively superposed on each root node to guide the generation direction of random nodes, so that the search range of an invalid space is reduced, and the adaptability to narrow channels is improved. And n is the sparsity rating of the matching model, the more sparse the model needs fewer additional points, and the more dense the model needs more points generated in the graph, so as to achieve the purposes of quickly exploring the full space and finding out a path in a narrow space.
Selecting rules of the middle extra points;
randomly generating points in a non-obstacle area;
superposing gravitational fields;
the randomness of the RRT-connect algorithm in node searching causes the nodes to expand a large amount of invalid space, so that the concept of a gravitational field in an artificial potential field is introduced. And respectively superposing a gravitational field on each expansion node to guide the generation direction of the random node. But the expansion node only takes care of the gravitational field generated by the next target node closest to its location. Equivalently, an extended node is added, the improved algorithm can be regarded as that two bidirectional extended random trees are introduced into the same configuration space essentially, the core idea of the algorithm after the gravity field is superimposed is that a target gravity function G (n) is added to each node n in any bidirectional extended random tree, the n node represents the nth new node extended outwards from the root node, and the extension mode is as follows
F(n)=R(n)+G(n)
F (n) represents the expansion function of the new node to the target point, R (n) is the random expansion function of the new node, and G (n) is the attraction function of the target point to the nearest node in the random tree. Random spread function R (n) of new node
U and sparsity
The method is in direct proportion, is an expansion step length, and is less expanded by a small step length due to the obstacle when sparse, so that a better path is searched and the obstacle is not easy to collide; and expanding large step length when the density is dense so as to achieve the purpose of escaping from the local minimum value.
The target gravity function g (n) is expressed as follows.
Wherein epsilon is a gravity coefficient, | xgoal-xnear | 2 represents the square of the Euclidean distance from the target point to the nearest point in the nearest random tree, thus constructing a gravity field with larger gravity as the distance from the target point is farther and avoiding the complexity of calculation of the gravity gradient caused by considering a plurality of points at the same time when the sparsity is too large.
The two formulas are superposed to obtain
Finally, the new node generation formula obtained after the substitution is as follows
And expanding each node of the new algorithm after the gravity field is superposed according to the method of the formula, so that the plurality of bidirectional random trees are expanded towards the direction of the target point under the guidance of respective gravity components.
The number of initial nodes is determined according to sparsity, meanwhile, the nodes are randomly generated in an interval between a starting point and an ending point, each node occupies one (xmax/initial node number, ymax) space, and the nodes are randomly generated in the space.
Meanwhile, if the local minimum value is difficult to sample to an external point for a long time (the iteration number is greater than or equal to 50, and the next spanning tree is not replaced), the sampling is randomly performed once.
The path planning method for the adaptive map provided by the embodiment adapts to a changing environment by adding an algorithm selection mechanism and establishing a model base in advance, and selects a proper algorithm; meanwhile, the algorithm model can be automatically updated to adapt to a new change environment according to the change of the utilization rate and the accuracy rate in the later period. In a high-complexity environment, the algorithm can automatically select the fusion of the RRT-connect and the artificial potential field, and the RRT random tree is guided to expand through the artificial potential field, so that the time of meaningless expansion is shortened. When the iteration times are excessive, a safety node is randomly selected to jump out of a deadlock area. In addition, only three nodes (namely one extra point) are selected in the original document; points of phase proportion are generated according to the complexity of the model so as to rapidly expand the trees. The improvement can improve the adaptability of the algorithm under different environments, a simpler algorithm is used when a sparse map is faced, the path can be generated quickly while the memory loss is reduced, the calculation amount is used for getting the speed in time when a complex map is faced, the actual demand is met, and the planning efficiency, the adaptability and the accuracy are improved.
In correspondence with the above method embodiment, referring to fig. 4, the present invention further provides a path planning system 40 for adaptive map, including:
the acquisition module 401 is configured to acquire environmental data of a target scene and perform preprocessing to obtain a target data set;
a matching module 402, configured to import the target data set into a matching model, and generate a matching result;
a judging module 403, configured to judge whether the matching result is a sparse type;
a generating module 404, configured to import the target data into an a-x algorithm for path planning and generate a target path if the matching result is a sparse type;
and if the matching result is not of a sparse type, importing the target data into an artificial potential field-based RRT-connect algorithm, selecting a plurality of points in the target scene as root nodes to establish a random tree, and superposing a gravitational field on each root node to perform node search to generate a target path.
The system shown in fig. 4 can correspondingly execute the content in the above method embodiment, and details of the part not described in detail in this embodiment refer to the content described in the above method embodiment, which is not described again here.
Referring to fig. 5, the present invention also provides an electronic device 50, including: at least one processor and a memory communicatively coupled to the at least one processor. The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method for path planning of the adaptive map in the foregoing method embodiments.
The present invention also provides a non-transitory computer-readable storage medium storing computer instructions for causing a computer to perform the method of path planning for an adaptive map in the aforementioned method embodiments.
The invention also provides a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions which, when executed by a computer, cause the computer to perform the method of path planning for an adaptive map in the aforementioned method embodiments.
Referring now to FIG. 5, there is shown a schematic diagram of an electronic device 50 suitable for use in implementing the present invention. The electronic device in the present invention may include, but is not limited to, a mobile terminal such as a mobile phone, a notebook computer, a digital broadcast receiver, a PDA (personal digital assistant), a PAD (tablet computer), a PMP (portable multimedia player), a vehicle-mounted terminal (e.g., a car navigation terminal), etc., and a stationary terminal such as a digital TV, a desktop computer, etc. The electronic device shown in fig. 5 is only an example, and should not bring any limitation to the function and the scope of use of the present invention.
As shown in fig. 5, electronic device 50 may include a processing means (e.g., central processing unit, graphics processor, etc.) 501 that may perform various appropriate actions and processes in accordance with a program stored in a Read Only Memory (ROM) 502 or a program loaded from a storage means 508 into a Random Access Memory (RAM) 503. In the RAM 503, various programs and data necessary for the operation of the electronic apparatus 50 are also stored. The processing device 501, the ROM 502, and the RAM 503 are connected to each other through a bus 504. An input/output (I/O) interface 505 is also connected to bus 504.
Generally, the following devices may be connected to the I/O interface 505: input devices 506 including, for example, a touch screen, touch pad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; output devices 507 including, for example, a Liquid Crystal Display (LCD), speakers, vibrators, and the like; storage devices 508 including, for example, magnetic tape, hard disk, etc.; and a communication device 509. The communication means 509 may allow the electronic device 50 to communicate with other devices wirelessly or by wire to exchange data. While the figures illustrate an electronic device 50 having various means, it is to be understood that not all illustrated means are required to be implemented or provided. More or fewer devices may alternatively be implemented or provided.
In particular, according to an embodiment of the present invention, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the invention include a computer program product comprising a computer program embodied on a computer-readable medium, the computer program comprising program code for performing the method illustrated in the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network via the communication means 509, or installed from the storage means 508, or installed from the ROM 502. The computer program performs the above-mentioned functions defined in the method of the present invention when executed by the processing means 501.
It should be noted that the computer readable medium of the present invention can be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present invention, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In the present invention, however, a computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, either in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, optical cables, RF (radio frequency), etc., or any suitable combination of the foregoing.
The computer readable medium may be embodied in the electronic device; or may exist separately without being assembled into the electronic device.
The computer readable medium carries one or more programs which, when executed by the electronic device, cause the electronic device to perform the steps associated with the method embodiments.
Alternatively, the computer readable medium carries one or more programs which, when executed by the electronic device, enable the electronic device to perform the steps associated with the method embodiments.
Computer program code for carrying out operations for aspects of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The units described in the present invention can be implemented by software or hardware.
It should be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof.
The above description is only for the specific embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (8)
1. A path planning method of an adaptive map is characterized by comprising the following steps:
acquiring environmental data of a target scene and preprocessing the environmental data to obtain a target data set;
importing the target data set into a matching model to generate a matching result;
calculating the sparsity degree according to the target data set and comparing the sparsity degree with a threshold value to obtain a matching result, wherein the matching result is any one of a sparse type and a non-sparse type;
if the matching result is of a sparse type, importing the target data into an A-x algorithm for path planning and generating a target path;
and if the matching result is of a non-sparse type, importing the target data into an artificial potential field-based RRT-connect algorithm, selecting a plurality of points in the target scene as root nodes to establish a random tree, and superposing a gravitational field on each root node to perform node search to generate a target path.
2. The method of claim 1, wherein the step of acquiring environmental data of the target scene and preprocessing the environmental data to obtain the target data set comprises:
and carrying out data noise reduction on the environmental data by adopting a grouping average and moving average algorithm and then carrying out normalization to obtain the target data set.
3. The method of claim 1, wherein prior to the step of importing the target dataset into a matching model to generate a matching result, the method further comprises:
calculating the execution time of the sample path planning map under different algorithms as a first dimension;
calculating the average value of the distances between the central point of the sample path planning map and other obstacles as a second dimension;
and importing the first dimension and the second dimension into a k-means clustering algorithm for clustering, and training to obtain a matching model associated with the map and the algorithm.
4. The method of claim 1, wherein the step of introducing the target data into an artificial potential field-based RRT-connect algorithm, selecting a plurality of points in the target scene as root nodes to build a random tree, and superimposing a gravitational field on each root node to perform node search to generate a target path comprises:
establishing a random tree by taking the starting point and the end point of the target scene as the root nodes, randomly generating a plurality of points in the non-obstacle area of the target scene as the root nodes, and establishing the random tree;
iterating one said gravitational field over each said random tree;
obtaining a new node generation formula according to a random expansion function corresponding to the random tree and a gravity function corresponding to the gravity field;
and calculating the target path according to the new node generation formula.
5. The method of claim 4, wherein before the step of deriving the new node generation formula according to the random spreading function corresponding to the random tree and the gravity function corresponding to the gravity field, the method further comprises:
and adjusting the step sizes of the random expansion function and the gravitation function according to the sparsity degree.
6. A system for adaptive map path planning, the system comprising:
the acquisition module is used for acquiring environmental data of a target scene and preprocessing the environmental data to obtain a target data set;
the matching module is used for importing the target data set into a matching model to generate a matching result;
the judging module is used for judging whether the matching result is of a sparse type;
the generating module is used for importing the target data into an A-x algorithm for path planning and generating a target path if the matching result is of a sparse type;
and if the matching result is not of a sparse type, importing the target data into an artificial potential field-based RRT-connect algorithm, selecting a plurality of points in the target scene as root nodes to establish a random tree, and superposing a gravitational field on each root node to perform node search to generate a target path.
7. An electronic device, characterized in that the electronic device comprises:
at least one processor; and the number of the first and second groups,
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of path planning for an adaptive map of any of the preceding claims 1-5.
8. A non-transitory computer-readable storage medium storing computer instructions for causing a computer to perform the method of path planning for an adaptive map of any of the preceding claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210256226.2A CN114323028B (en) | 2022-03-16 | 2022-03-16 | Path planning method, system, device and medium for self-adaptive map |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210256226.2A CN114323028B (en) | 2022-03-16 | 2022-03-16 | Path planning method, system, device and medium for self-adaptive map |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114323028A true CN114323028A (en) | 2022-04-12 |
| CN114323028B CN114323028B (en) | 2022-06-07 |
Family
ID=81033755
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210256226.2A Active CN114323028B (en) | 2022-03-16 | 2022-03-16 | Path planning method, system, device and medium for self-adaptive map |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114323028B (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20110015833A (en) * | 2009-08-10 | 2011-02-17 | 삼성전자주식회사 | Method and apparatus of path planing for a robot |
| US20140180526A1 (en) * | 2012-12-21 | 2014-06-26 | Toyota Jidosha Kabushiki Kaisha | Autonomous Navigation Through Obstacles |
| KR20150126482A (en) * | 2014-05-02 | 2015-11-12 | 한화테크윈 주식회사 | Path planning apparatus of mobile robot |
| CN110083165A (en) * | 2019-05-21 | 2019-08-02 | 大连大学 | A kind of robot paths planning method under complicated narrow environment |
| US20200159216A1 (en) * | 2018-11-16 | 2020-05-21 | Great Wall Motor Company Limited | Motion Planning Methods And Systems For Autonomous Vehicle |
| CN112013846A (en) * | 2020-08-18 | 2020-12-01 | 南京信息工程大学 | Path planning method combining dynamic step RRT algorithm and potential field method |
| CN112902971A (en) * | 2021-03-22 | 2021-06-04 | 中国长江三峡集团有限公司 | Fast extended random tree algorithm based on Gaussian sampling and target deviation guidance, electronic equipment and storage medium |
| CN112923944A (en) * | 2021-01-29 | 2021-06-08 | 的卢技术有限公司 | Automatic driving path planning method and system and computer readable storage medium |
| CN112975961A (en) * | 2021-02-20 | 2021-06-18 | 华南理工大学 | Picking mechanical arm motion planning method based on CTB-RRT algorithm |
| US20210341309A1 (en) * | 2019-01-14 | 2021-11-04 | Zhejiang Huaray Technology Co., Ltd. | Systems and methods for route planning |
| CN114115362A (en) * | 2021-11-30 | 2022-03-01 | 沈阳航空航天大学 | Unmanned aerial vehicle flight path planning method based on bidirectional APF-RRT algorithm |
-
2022
- 2022-03-16 CN CN202210256226.2A patent/CN114323028B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20110015833A (en) * | 2009-08-10 | 2011-02-17 | 삼성전자주식회사 | Method and apparatus of path planing for a robot |
| US20140180526A1 (en) * | 2012-12-21 | 2014-06-26 | Toyota Jidosha Kabushiki Kaisha | Autonomous Navigation Through Obstacles |
| KR20150126482A (en) * | 2014-05-02 | 2015-11-12 | 한화테크윈 주식회사 | Path planning apparatus of mobile robot |
| US20200159216A1 (en) * | 2018-11-16 | 2020-05-21 | Great Wall Motor Company Limited | Motion Planning Methods And Systems For Autonomous Vehicle |
| US20210341309A1 (en) * | 2019-01-14 | 2021-11-04 | Zhejiang Huaray Technology Co., Ltd. | Systems and methods for route planning |
| CN110083165A (en) * | 2019-05-21 | 2019-08-02 | 大连大学 | A kind of robot paths planning method under complicated narrow environment |
| CN112013846A (en) * | 2020-08-18 | 2020-12-01 | 南京信息工程大学 | Path planning method combining dynamic step RRT algorithm and potential field method |
| CN112923944A (en) * | 2021-01-29 | 2021-06-08 | 的卢技术有限公司 | Automatic driving path planning method and system and computer readable storage medium |
| CN112975961A (en) * | 2021-02-20 | 2021-06-18 | 华南理工大学 | Picking mechanical arm motion planning method based on CTB-RRT algorithm |
| CN112902971A (en) * | 2021-03-22 | 2021-06-04 | 中国长江三峡集团有限公司 | Fast extended random tree algorithm based on Gaussian sampling and target deviation guidance, electronic equipment and storage medium |
| CN114115362A (en) * | 2021-11-30 | 2022-03-01 | 沈阳航空航天大学 | Unmanned aerial vehicle flight path planning method based on bidirectional APF-RRT algorithm |
Non-Patent Citations (2)
| Title |
|---|
| OLZHAS ADIYATOV 等: "Sparse tree heuristics for RRT family motion planners", 《2017 IEEE INTERNATIONAL CONFERENCE ON ADVANCED INTELLIGENT MECHATRONICS (AIM)》 * |
| 黄文刚 等: "变步长稀疏A*算法的无人机航路规划", 《计算机工程与应用》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114323028B (en) | 2022-06-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10593110B2 (en) | Method and device for computing a path in a game scene | |
| Dickinson et al. | Indoor positioning of shoppers using a network of Bluetooth Low Energy beacons | |
| US20110137833A1 (en) | Data processing apparatus, data processing method and program | |
| CN106714110A (en) | Auto building method and system of Wi-Fi position fingerprint map | |
| JP2012008659A (en) | Data processing device, data processing method, and program | |
| US20110137831A1 (en) | Learning apparatus, learning method and program | |
| KR20190017454A (en) | Device and method for generating location estimation model | |
| CN113110605B (en) | Multi-unmanned aerial vehicle collaborative search method and device in urban environment | |
| CN114089752A (en) | Autonomous exploration method for robot, and computer-readable storage medium | |
| Sarmiento et al. | An efficient strategy for rapidly finding an object in a polygonal world | |
| Chen et al. | A wifi indoor localization method based on dilated cnn and support vector regression | |
| CN114323028B (en) | Path planning method, system, device and medium for self-adaptive map | |
| Chand et al. | A two-tiered global path planning strategy for limited memory mobile robots | |
| CA2894863A1 (en) | Indoor localization using crowdsourced data | |
| US20220258344A1 (en) | Information processing device, control method, and storage medium | |
| JP5845604B2 (en) | Information processing apparatus, information processing method, and program | |
| CN115933638A (en) | Multi-robot collaborative exploration method and device, electronic equipment and medium | |
| CN114048333B (en) | Multisource fusion voice interactive indoor positioning method, terminal and storage medium | |
| Lawrance et al. | Fast marching adaptive sampling | |
| CN113790729B (en) | Unmanned overhead traveling crane path planning method and device based on reinforcement learning algorithm | |
| KR20190042540A (en) | Device and method for generating geomagnetic sensor based location estimation model using artificial neural networks | |
| CN115200584A (en) | Path planning method, device and equipment and readable storage medium | |
| Opoku et al. | The Ar-Star (Ar) Pathfinder | |
| Chung et al. | Robot motion planning in dynamic uncertain environments | |
| CN112985397B (en) | Robot track planning method and device, storage medium and electronic equipment |
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 |