CN110865642A - Path planning method based on mobile robot - Google Patents
Path planning method based on mobile robot Download PDFInfo
- Publication number
- CN110865642A CN110865642A CN201911078096.2A CN201911078096A CN110865642A CN 110865642 A CN110865642 A CN 110865642A CN 201911078096 A CN201911078096 A CN 201911078096A CN 110865642 A CN110865642 A CN 110865642A
- Authority
- CN
- China
- Prior art keywords
- path
- robot
- control point
- path planning
- obstacle
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 71
- 238000005457 optimization Methods 0.000 claims abstract description 12
- 108090000623 proteins and genes Proteins 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 abstract description 14
- 238000004422 calculation algorithm Methods 0.000 description 49
- 230000002068 genetic effect Effects 0.000 description 19
- 238000004364 calculation method Methods 0.000 description 8
- 210000000349 chromosome Anatomy 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 6
- 238000011160 research Methods 0.000 description 4
- 238000004088 simulation Methods 0.000 description 4
- 238000013459 approach Methods 0.000 description 3
- 230000004888 barrier function Effects 0.000 description 3
- 230000000875 corresponding effect Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000009499 grossing Methods 0.000 description 3
- 230000001788 irregular Effects 0.000 description 3
- 239000002245 particle Substances 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000005265 energy consumption Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 230000002829 reductive effect Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 238000003860 storage Methods 0.000 description 2
- 241000218645 Cedrus Species 0.000 description 1
- WHXSMMKQMYFTQS-UHFFFAOYSA-N Lithium Chemical compound [Li] WHXSMMKQMYFTQS-UHFFFAOYSA-N 0.000 description 1
- 230000002238 attenuated effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000003592 biomimetic effect Effects 0.000 description 1
- 230000002759 chromosomal effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000001276 controlling effect Effects 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000004387 environmental modeling Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000007429 general method Methods 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000000670 limiting effect Effects 0.000 description 1
- 229910052744 lithium Inorganic materials 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 235000001968 nicotinic acid Nutrition 0.000 description 1
- 230000010355 oscillation Effects 0.000 description 1
- 238000004064 recycling Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 230000002194 synthesizing effect Effects 0.000 description 1
- 230000000007 visual effect Effects 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
- G05D1/0214—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory in accordance with safety or protection criteria, e.g. avoiding hazardous areas
-
- 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)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
- Feedback Control In General (AREA)
Abstract
The invention discloses a path planning method based on a mobile robot, which comprises the following steps: based on a polygonal obstacle plane model, utilizing Q-IGA to search the most appropriate control point P in a collaborative mode to serve as a control point of a Bezier curve, and utilizing the control point to generate an optimal path; calculating the difference between different paths in the population at each iteration, and adding a judgment criterion for operator selection for optimization to ensure the diversity of feasible solutions in the population; and constructing a fitness function considering the distance of the obstacle, the turning angle and the size of the robot, wherein the optimization target is the maximum fitness function value. The method avoids the problem of complicated steps caused by the fact that the optimal path is obtained by utilizing the Bezier curve to directly fit and search in the robot path planning, ensures that the path planning process is efficiently completed, and has short time and higher path planning efficiency.
Description
Technical Field
The invention relates to the field of artificial intelligence, in particular to a path planning method based on a mobile robot.
Background
With the development of many interdisciplines such as computer technology, control theory, sensor technology, artificial intelligence and the like, the self-navigation robot technology is rapidly developed, and various robots are applied to many fields such as scientific research, industrial production, home life and the like[1]. The self-navigation robot technology is not only a foundation for the mobile robot to complete high-difficulty tasks, but also an important sign for the intelligent maturity of the robot[2]. Algorithm for controlling motion of self-navigation robot[3]The navigation system mainly comprises a positioning algorithm, a navigation control algorithm and a path planning algorithm. Safe obstacle avoidance in path search processThe method is a precondition for the robot to complete the task smoothly, so the quality of the path planning method has important influence on the operation of the robot. The purpose of path planning is to find a collision-free optimal path from a starting point to a target point. Currently, the main research results of path planning technology include: traditional methods, intelligent biomimetics methods, and search-based methods.
Conventional path planning methods have been widely studied. Visual map method for schs army and the like[4-5]The method is used for planning the path of the robot, the robot is taken as a particle, a starting point, an end point and all vertexes of an obstacle are connected by straight lines, and then the optimal path between the starting points is found by adopting a specific searching method. This technique, although simple to implement, is not flexible enough to reconstruct a viewable view once the actual environment in which the robot travels changes; if the number of obstacles in the environment is large, the number of connecting lines between corresponding points is too large, and the path searching time is long. Grid method[6-7]When the method is used for robot path planning, an environment space is divided into continuous grids with equal size, and then an optimization algorithm is adopted for searching, so that a path consisting of a series of grids can be obtained. The grid method is simple for processing the obstacles and is convenient for computer storage, processing and analysis. Because the method expresses the environment information of the robot by using the grid, the size of the grid directly influences the environment information storage amount, the environment resolution and the planning time when the path is planned. Using artificial potential field method[8]When the path planning is carried out, the obstacle and the target point respectively generate repulsion and attraction to the robot, and the movement of the robot is guided through the resultant force of the repulsion and the attraction, so that the path planning process is efficiently completed. However, the method has certain limitations, such as the existence of local minimum points, unreachable targets, and the possibility of oscillation, which makes the path planning process unable to be performed stably.
Reference to the literature
[1]CaiP,CaiY,Chandrasekaran,I&Zheng,Parallel genetic algorithm basedautomatic path planning for crane lifting in complex environments[J].Automation in Construction,2016,62(1):133-147.
[2]G.A.Mohammed,M.Hou.Optimization ofactive muscle force-lengthmodels using least squares curve fitting[J].IEEE Trans.Biomed.Eng,2016,63(3):630–635.
[3]R.Cimurs,J.Hwang,I.H.Suh.Bezier curve-based smoothing for pathplanner with curvature constraint[C].IEEE International Conference on RoboticComputing(IRC),2017,pp.241–248.
[4] Schss, cao curio, mobile robot path planning based on visual [ J ] computer applications and software, 2011,28(3): 220-.
[5] Lemna, Zhu military Swallow, Penghang, Path planning [ J ] based on visual and A ＊ algorithms computer engineering 2014,40(3):193 and 196.
[6] Liangjiajun, Zengbi, He Yuanlie, research on a cleaning robot path planning algorithm based on an improved potential grid method [ J ]. proceedings of Guangdong university of industry, 2016,33(4):30-34.
[7] Mobile robot path planning [ J ] based on grid method environment modeling in an unstructured environment, 2016, 44 (17): 1-7.
[8]Chia-Chia Kao,Chin-Min Lin.Application ofPotential Filed Methodand Optimal Path Planning to Mobile Robot Control[C].2015IEEE InternationalConference on Automation Science and Engineering(CASE),2015:1552-1554.
[9] Liu xian feng, peng zhong ren, zhang ye, etc. unmanned aircraft path planning oriented to traffic information collection [ J ] traffic transportation system engineering and information, 2012, 12(1): 91-97.
[10]Yupei Yan,Yangnin Li.Mobile Robot Autonomous Path Planning Basedon Fuzzy Logic and Filter Smoothing in Dynamic Environment[C]201612WorldCongress on Intelligent Control and Automation(WCICA),2016:1479-1484.
[11] Elite particle swarm optimization algorithm and application thereof in robot path planning [ J ] optical precision engineering, 2013,21(12): 3160-.
[12]R.Cimurs,J.Hwang,I.H.Suh.Bezier curve-based smoothing for pathplannerwith curvature constraint[C].in:IEEE International Conference onRobotic Computing(IRC),IEEE,2017,pp.241–248.
Disclosure of Invention
The invention provides a path planning method based on a mobile robot, which avoids the complex steps caused by the optimal path obtained by directly fitting and searching by using a Bezier curve in the path planning of the robot, ensures the efficient completion of the path planning process and is described in detail as follows:
a mobile robot-based path planning method, the method comprising:
based on a polygonal obstacle plane model, utilizing Q-IGA to search the most appropriate control point P in a collaborative mode to serve as a control point of a Bezier curve, and utilizing the control point to generate an optimal path;
calculating the difference between different paths in the population at each iteration, and adding a judgment criterion for operator selection for optimization to ensure the diversity of feasible solutions in the population;
and constructing a fitness function considering the distance of the obstacle, the turning angle and the size of the robot, wherein the optimization target is the maximum fitness function value.
The step of cooperatively searching the most suitable control point P by using Q-IGA as the control point of the bezier curve specifically includes:
carrying out binarization processing on the polygonal obstacle plane model to enable the polygonal obstacle plane model to be equivalent to a grid map with a plurality of pixel points;
each grid is equivalent to a gene and has a value of 0 or 1, and when the grid is a control point of a Bezier curve, the value of the grid is 1, otherwise, the value of the grid is 0;
if the grid is covered by an obstacle, its value is set to-1, and it cannot be used as a control point.
Further, the generating the optimal path by using the control point specifically includes:
1) a line segment connecting the start points has 0 order continuity;
2) at the joint of the two line segments, the equivalent tangent line is used for ensuring first-order continuity;
3) the continuity of the Bezier curve above the third order is ensured by the curvature continuity;
and generating a smooth path by adopting a third-order Bezier curve under the condition of meeting continuity.
The method for optimizing the operator selection by adding a judgment criterion specifically comprises the following steps:
in each iteration, these paths remain on the next iteration only if the similarity index of the population is below the threshold η.
The technical scheme provided by the invention has the beneficial effects that:
after the robot executes the method, a very smooth path which keeps a safe distance from the barrier can be obtained, and the requirement on the reasonability of the path is met; and the used time is short, the path planning efficiency is higher, and various requirements in practical application are met.
Drawings
FIG. 1 is a schematic diagram of a set of third order Bezier curves that satisfy continuity;
FIG. 2 is a diagram of the effect of corner angle optimization;
FIG. 3 is a flow chart of the Q-IGA algorithm;
FIG. 4 is a parametric performance graph;
wherein, (a) is population size and path length; (b) calculating population scale and calculating time; (c) the iteration times and the path length;
(d) the iteration number and the calculation time.
Fig. 5 is a diagram of simulation results in different environments.
Wherein, (a) is an irregular obstacle map; (b) a narrow-band obstacle map; (c) is a complex maze map.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention are described in further detail below.
The traditional path planning method is influenced by complexity or limited by performance, and cannot be popularized and used in the actual robot path planning. Currently, a path planning method based on the principle of intelligent bionics is commonly used, and a lot of research is availableCompared with the traditional method, the achievement proves that the method has strong self-organizing and self-learning capabilities and certain fault-tolerant capability. Literature reference[9]An improved ant colony algorithm based on a rolling window is provided, the dead angle analysis and processing capacity is enhanced, and the problem that path search is easy to get early is solved; literature reference[10]And (3) measuring environmental information in real time by using a fuzzy logic algorithm based on a sensor equipped in the robot, and making a decision through a pre-established rule to complete path planning. The method can avoid the problem that the artificial potential field method is easy to fall into local optimum, and is particularly easy to process real-time path planning. Face cedar, etc[11]An updating function and an elite selection mechanism are designed, so that the particle swarm algorithm reduces the possibility of converging to a local optimal solution, the convergence speed of the algorithm is increased, the improved algorithm is applied to robot path planning, and a better path can be obtained by the improved algorithm. The genetic algorithm can effectively process global search with probability significance because the evolutionary operator of the genetic algorithm has ergodicity, and the search process does not need to limit the inherent property of the problem, so the genetic algorithm is obvious in the aspect of solving the path planning problem. Since many paths obtained by a general method are broken-line paths, a method of smoothing a path using a bezier curve or a B-spline curve is also beginning to be widely used[12]。
Although various optimization methods currently overcome some of the disadvantages inherent in the use of genetic algorithms for path planning, such as: the exploration capacity of the method for the new space is improved so as to prevent the method from prematurely converging to a local optimal solution; complexity is reduced to prevent it from being too inefficient in computation when the number of individuals is large. However, for a robotic path navigation system to be applied in a practical production living environment, it is many times not necessarily desirable that the optimal result be the shortest path described in the above method. In contrast, a "fair" path is more desirable. Here, a reasonable path is defined as one that: the device does not penetrate through the barrier, does not closely cling to the barrier to travel, makes large-angle turning as few as possible, and makes the distance as shortest as possible on the premise of meeting the conditions.
Therefore, aiming at the problems and the requirements, the invention provides a path planning method for dynamically fitting a Bezier curve by using Q-IGA, which avoids the problem that the steps are complicated due to the fact that the optimal path is directly fitted and searched by using the Bezier curve in the path planning of the robot, and ensures that the path planning process is efficiently completed. And the diversity of solution schemes in the population is judged by utilizing the thought of a Q value inspection method in a selection operator of the genetic algorithm, the solutions with higher similarity are removed, the exploration capability of the solutions is enhanced, and the path search efficiency is improved.
Meanwhile, the invention adds other cost factors except the path length by modifying the fitness function formula, so that the generated path meets the requirement on the path rationality. Simulation results prove that compared with an improved artificial potential field method and a hybrid genetic algorithm, the Q-IGA algorithm can obtain a smooth path which keeps a certain safe distance from an obstacle and is higher in calculation efficiency. In robot applications, the method provides an effective way to avoid energy exhaustion, especially in robot applications with limited dynamic resources.
Example 1
1. Environmental modeling
The environment in which the robot travels is a two-dimensional planar space, and obstacles in the plane are arranged into any irregular polygon which is static, random and known, and the vertexes (x, y) of the polygon are represented by a circular list. Compared with the grid method, the method is easy to solve the problem of complex environmental information and occupies less resources.
2. Dynamic parallel fitting strategy
In the past research, the process of smoothing the path by using the bezier curve is an independent process, namely after the optimal path is generated, the bezier curve is used for static fitting, so that the calculation steps are increased, and the algorithm is complicated. Therefore, the best control point P is cooperatively searched by the Q-IGA to be used as the control point of the Bezier curve, and a shorter optimal path can be generated by using the selected control point.
The environment model established in the method is a polygonal obstacle plane model, so that the environment model is firstly subjected to binarization processing to be equivalent to a grid map with a plurality of pixel points. Each grid corresponds to a gene and may have a value of 0 or 1, and when the grid is a control point of a bezier curve, the grid takes a value of 1, and when the grid is not a control point of a bezier curve, the grid takes a value of 0. If the grid is covered by an obstacle, its value is set to-1, and it cannot be used as a control point.
Therefore, each chromosome in the initial population of the genetic algorithm is coded in a binary mode, and the gene value is 0 or 1, namely, whether the path point is the candidate control point or not is represented. If a path passes through an obstacle, i.e., passes through a grid point with a value of-1, the path point is removed by an obstacle avoidance operation in a subsequent genetic operation. It is important to ensure the continuity of the path in the path planning process, and the low-order continuity criterion is defined as follows:
1) a line segment connecting the start points has continuity of order 0.
2) And at the joint of the two line segments, the equivalent tangent is used for ensuring the first-order continuity.
3) The continuity of the Bezier curve above the third order is ensured by the curvature continuity.
In consideration of computational complexity, the lower the order of the bezier curve is, the better the continuity is satisfied. The invention adopts a third-order Bezier curve to generate a smooth path, and the third-order Bezier curve is defined as follows:
Q(t)＝B0(1-t)3+3B1(1-t)2t+3B2(1-t)2+B3t3
wherein, B0Etc. 4 control point coordinates required to obtain a third order bezier curve.
3. Increasing population adaptability by optimizing selection operator
In Q-IGA, a population is first randomly generated, based on the basic steps of the genetic algorithm, comprising S chromosomes, each of which is a path planning scheme, the structure of which can be represented as xs＝{u1，u2，....，uP}. One of the problems of the optimization algorithm is that the solution process has strong randomness, which results in that the iteration result is vulnerable to diversity loss.
In order to improve the diversity of solutions generated by genetic algorithms, the method optimizes the selection operator. By calculating the difference between different path schemes in the population at each iteration, a judgment criterion is added for the selection operator, and the diversity of feasible solutions in the population is ensured. The algorithm sets the judgment criterion as follows:
in each iteration, these schemes can only be retained for the next iteration if the similarity indicator of the population is below the threshold η. the initial value of η is a random value of (0, 1.) during the iteration, the value of η drops linearly from the initial value to 0. in the algorithm model, the attenuation factor w is 0.99 to represent the attenuation rate, i.e., the attenuation formula of η is:
ηt+1＝ωηt
η approaches 0 as t approaches infinity.
In each iteration η is reduced, meaning that the path solutions iteratively obtained are more and more similar as the solution approaches the optimal solution later in the search, i.e., a balance is struck between the beginning and the end of the search.
The method judges the diversity of the population by referring to the thought of a Q value detection method. From the path model, two solutions x1 are defined, with q values between x2 being:
wherein N is11Representing the number of gene values of 1 at corresponding positions in both chromosomes, N00Represents the number of 0 gene values at the corresponding positions in both chromosomes; n is a radical of01Represents x1The position of the chromosomal gene value 0 corresponds to x2Number of gene value 1; n is a radical of10Represents x1Position of chromosome gene value 1 corresponds to x2The number of genes was 0.
If two chromosomes x1＝[1 1 0 0 1 1 0 0 1 0]，x2＝[1 0 1 0 1 0 1 0 1 1]And then:
the population similarity Q is the calculated average of Q values between all individuals, that is:
4. optimizing fitness function
Another important step of genetic algorithms is the calculation of individual fitness values. From the common knowledge, in a similar motion environment, the motion energy consumption of the robot is positively correlated with the motion distance, i.e. the longer the motion distance is, the more the energy consumption is. The traditional genetic algorithm judges the length of the path through a fitness function, and can ensure that the movement distance of the robot is shortened as far as possible on the premise that the robot finishes a moving target. However, the path thus obtained may be too close to the obstacle or may have a large-angle turn that consumes a large amount of energy. Therefore, in the method, the fitness function takes the turning angle and the robot volume as evaluation factors, and the optimization goal is to maximize the redefined fitness function value.
(1) Distance factor
After the genetic algorithm completes the control point search, the path length D can be expressed as:
where n is the number of control points, d (p)i,pi+1) Is the distance between two adjacent control points.
The path length may be calculated by the integral of the bezier curve:
(2) volume of robot:
in an ideal environment, a mobile robot is treated as a particle and does not consider the effect of its volume on the travel process. However, in practical applications, the robot should not only avoid the obstacle during the movement, but also keep a certain safe distance from the obstacle. In a classical path planning algorithm, in order to ensure that an optimal path is a shortest path, a restriction situation that a robot is close to an obstacle or a wall often occurs.
Therefore, the method optimizes the calculation method of the classical fitness function. For an obstacle that cannot pass through, its cost is described by + ∞. Thus, the cost of a point near an obstacle to the robot's motion is defined as:
wherein k is1D is the distance between the center of the robot and the obstacle, d is a normalized coefficient0Is the outer dimension of the robot, and σ is the width of the gaussian function.
It can be seen that when the distance between the robot and the obstacle is smaller than the size of the robot, the cost is + ∞, and the fitness function value is 0 at the moment, that is, the robot is considered to collide with the obstacle under the condition; when the distance between the robot and the obstacle is larger than the size of the robot, the cost value of the robot is attenuated by a Gaussian function, so that the robot is ensured to be far away from the obstacle as far as possible in a certain range.
(3) Turning angle of robot
Too large or too frequent path turning angle not only affects the working efficiency of the robot, but also consumes more energy under the condition of the same moving distance. Although the endurance time of the robot is longer and longer with the rapid development and maturity of rechargeable battery technologies such as lithium batteries in recent years, the recycling efficiency of the robot is still limited by the defects of long charging time and the like. Therefore, the invention introduces the turning angle factor into the fitness function, and the cost of each turning point caused by turning is defined as:
wherein α is the turning angle, k2The optimized path has fewer turns and smoother corners for normalizing the coefficients.
Synthesizing the path length factor, the robot volume and the cost brought by the robot turning angle, and finally correcting the fitness function as follows:
wherein D is the path length, R1(n) is the robot volume penalty, R2And (n) is the turning angle cost of the robot, and the iteration target of the algorithm is that the newly defined fitness function value reaches the maximum.
5. Path search
After the genetic algorithm selection operator and the fitness function are improved, the optimized Q-IGA algorithm can be used for searching path points and dynamically searching and determining control points of a Bezier curve, so that the path searching and path smoothing processes are executed in parallel.
Example 2
Before carrying out a simulation experiment, reasonable parameters are required to be set to enable an algorithm to obtain an optimal solution. In fig. 4, (a), (b), (c), and (d) are relationships between the population size and the path length, the calculation time, and the number of iterations and the path length, and the calculation time, respectively. The 4 performance graphs are synthesized to know that the execution time of the algorithm is continuously increased along with the increase of the population scale and the iteration times, the linear rising trend is basically presented, and the direct proportion relation between the operation time and the algorithm complexity is met; and as the population size and the iteration number increase, the path length tends to be stable and basically unchanged after a certain node after a period of time reduction. As can be seen from fig. (a) and (c), when the population size is 110, the number of iterations is 150 and then the path length is stabilized, and therefore, for the first map, the population size is set to 110 and the number of iterations is set to 160 under the condition that the path length is kept as short as possible and the calculation time is also short. The parameter selection method of the other two maps is the same as the principle of the first map, and is not described herein again.
The invention sets three map environments for simulation experiments, namely an irregular obstacle environment (with the size of 20 x 20), a narrow-band obstacle environment (with the size of 30 x 30) and a complex maze pattern environment (with the size of 45 x 45). The initial environment is represented by a circular coordinate system, and the shaded portions are obstacles. An improved artificial potential field method and a hybrid genetic algorithm are used as comparison algorithms, and the performance of the three algorithms is verified in each map environment. Fig. 5 shows the path effects generated by the three algorithms, all of which can obtain a path that can avoid an obstacle from a starting point to a target point, and the paths generated by the improved potential field method and the hybrid genetic algorithm are the shortest in some environments, but have larger turning angles, and some path points are too close to the obstacle because the influence caused by the volume of the robot is not considered, so that the risk of friction between the robot and the obstacle is increased. The path generated by the method of the invention is not the shortest path, but meets the requirement on the reasonability of the path planning of the robot, can finish obstacle avoidance and keep a certain distance with the obstacle according to the volume of the robot, and meanwhile, the path is smoother and more conforms to the requirement of the robot on the path in the actual engineering application.
The method provided by the invention aims to obtain a reasonable path and improve the searching efficiency of the algorithm, so that the time performance of the hybrid genetic algorithm and the Q-IGA (method) algorithm is investigated. As can be seen from table 1, the Q-IGA algorithm is shorter in time than the hybrid genetic algorithm, and the time efficiency of the algorithm is effectively improved while the optimal path is obtained.
TABLE 1 time Performance of each algorithm
In the embodiment of the present invention, except for the specific description of the model of each device, the model of other devices is not limited, as long as the device can perform the above functions.
Those skilled in the art will appreciate that the drawings are only schematic illustrations of preferred embodiments, and the above-described embodiments of the present invention are merely provided for description and do not represent the merits of the embodiments.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (4)
1. A path planning method based on a mobile robot is characterized by comprising the following steps:
based on a polygonal obstacle plane model, utilizing Q-IGA to search the most appropriate control point P in a collaborative mode to serve as a control point of a Bezier curve, and utilizing the control point to generate an optimal path;
calculating the difference between different paths in the population at each iteration, and adding a judgment criterion for operator selection for optimization to ensure the diversity of feasible solutions in the population;
and constructing a fitness function considering the distance of the obstacle, the turning angle and the size of the robot, wherein the optimization target is the maximum fitness function value.
2. The method according to claim 1, wherein the step of searching the most suitable control point P by using the Q-IGA cooperation as the control point of the bezier curve specifically comprises:
carrying out binarization processing on the polygonal obstacle plane model to enable the polygonal obstacle plane model to be equivalent to a grid map with a plurality of pixel points;
each grid is equivalent to a gene and has a value of 0 or 1, and when the grid is a control point of a Bezier curve, the value of the grid is 1, otherwise, the value of the grid is 0;
if the grid is covered by an obstacle, its value is set to-1, and it cannot be used as a control point.
3. The method according to claim 1, wherein the generating of the optimal path using the control points specifically comprises:
1) a line segment connecting the start points has 0 order continuity;
2) at the joint of the two line segments, the equivalent tangent line is used for ensuring first-order continuity;
3) the continuity of the Bezier curve above the third order is ensured by the curvature continuity;
and generating a smooth path by adopting a third-order Bezier curve under the condition of meeting continuity.
4. The mobile robot-based path planning method according to claim 1, wherein the optimization by adding a judgment criterion to the selection operator specifically comprises:
in each iteration, these paths remain on the next iteration only if the similarity index of the population is below the threshold η.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911078096.2A CN110865642A (en) | 2019-11-06 | 2019-11-06 | Path planning method based on mobile robot |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911078096.2A CN110865642A (en) | 2019-11-06 | 2019-11-06 | Path planning method based on mobile robot |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110865642A true CN110865642A (en) | 2020-03-06 |
Family
ID=69653772
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911078096.2A Pending CN110865642A (en) | 2019-11-06 | 2019-11-06 | Path planning method based on mobile robot |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110865642A (en) |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111481933A (en) * | 2020-05-18 | 2020-08-04 | 浙江工业大学 | Game path planning method based on improved potential field grid method |
| CN111766875A (en) * | 2020-06-18 | 2020-10-13 | 珠海格力智能装备有限公司 | Obstacle avoidance method and device for dust collection equipment and electronic equipment |
| CN112099493A (en) * | 2020-08-31 | 2020-12-18 | 西安交通大学 | Autonomous mobile robot trajectory planning method, system and equipment |
| CN112214031A (en) * | 2020-09-25 | 2021-01-12 | 北京理工大学 | Multi-node collaborative landing position planning method based on genetic particle swarm optimization |
| CN112256023A (en) * | 2020-09-28 | 2021-01-22 | 南京理工大学 | Bezier curve-based airport border patrol robot local path planning method and system |
| CN112947490A (en) * | 2021-04-09 | 2021-06-11 | 京东数科海益信息科技有限公司 | Path smoothing method, device, equipment, storage medium and product |
| CN113110422A (en) * | 2021-03-25 | 2021-07-13 | 贵州电网有限责任公司 | Path planning method based on Q-IGA dynamic fitting Bezier curve |
| CN113341876A (en) * | 2021-06-24 | 2021-09-03 | 合肥工业大学 | Five-axis curved surface machining track planning method based on differential vector optimization |
| CN113390412A (en) * | 2020-03-11 | 2021-09-14 | 宁波方太厨具有限公司 | Full-coverage path planning method and system for robot, electronic equipment and medium |
| CN113442140A (en) * | 2021-06-30 | 2021-09-28 | 同济人工智能研究院(苏州)有限公司 | Bezier optimization-based Cartesian space obstacle avoidance planning method |
| CN113932818A (en) * | 2020-12-30 | 2022-01-14 | 浙江德源智能科技股份有限公司 | Robot walking route planning method, program and storage medium |
| CN114131591A (en) * | 2021-12-03 | 2022-03-04 | 山东大学 | Semi-physical simulation method and system for operation strategy of outer limb robot |
| WO2022252869A1 (en) * | 2021-06-02 | 2022-12-08 | 北京迈格威科技有限公司 | Obstacle bypassing method for mobile device, and mobile device and storage medium |
| CN116408793A (en) * | 2023-02-16 | 2023-07-11 | 广州数控设备有限公司 | Industrial robot path fairing method and system with continuous curvature |
| CN116460830A (en) * | 2023-03-17 | 2023-07-21 | 北京信息科技大学 | Robot intelligent control system and control method based on artificial intelligence |
| EP4180894A4 (en) * | 2020-07-20 | 2023-12-27 | Huawei Digital Power Technologies Co., Ltd. | Method and device for planning obstacle avoidance path for traveling device |
| WO2024077716A1 (en) * | 2022-10-11 | 2024-04-18 | 劢微机器人科技(深圳)有限公司 | Local path planning method for amr |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20120098152A (en) * | 2011-02-28 | 2012-09-05 | 한국과학기술연구원 | Path planning system for mobile robot |
| CN106681335A (en) * | 2017-01-22 | 2017-05-17 | 无锡卡尔曼导航技术有限公司 | Obstacle-avoiding route planning and control method for unmanned agricultural machine driving |
| CN106709601A (en) * | 2016-12-19 | 2017-05-24 | 广西师范大学 | Cellular quantum genetic algorithm and PH curve combination-based unmanned aerial vehicle three-dimensional route planning method |
| CN107168305A (en) * | 2017-04-01 | 2017-09-15 | 西安交通大学 | Unmanned vehicle method for planning track based on Bezier and VFH under the scene of crossing |
| CN107702723A (en) * | 2017-11-27 | 2018-02-16 | 安徽工程大学 | A kind of robot path planning method, storage medium and equipment |
| CN109799822A (en) * | 2019-01-30 | 2019-05-24 | 中国石油大学(华东) | Mobile robot global smooth paths planing method |
| CN110347151A (en) * | 2019-05-31 | 2019-10-18 | 河南科技大学 | A kind of robot path planning method merging Bezier optimized Genetic Algorithm |
-
2019
- 2019-11-06 CN CN201911078096.2A patent/CN110865642A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20120098152A (en) * | 2011-02-28 | 2012-09-05 | 한국과학기술연구원 | Path planning system for mobile robot |
| CN106709601A (en) * | 2016-12-19 | 2017-05-24 | 广西师范大学 | Cellular quantum genetic algorithm and PH curve combination-based unmanned aerial vehicle three-dimensional route planning method |
| CN106681335A (en) * | 2017-01-22 | 2017-05-17 | 无锡卡尔曼导航技术有限公司 | Obstacle-avoiding route planning and control method for unmanned agricultural machine driving |
| CN107168305A (en) * | 2017-04-01 | 2017-09-15 | 西安交通大学 | Unmanned vehicle method for planning track based on Bezier and VFH under the scene of crossing |
| CN107702723A (en) * | 2017-11-27 | 2018-02-16 | 安徽工程大学 | A kind of robot path planning method, storage medium and equipment |
| CN109799822A (en) * | 2019-01-30 | 2019-05-24 | 中国石油大学(华东) | Mobile robot global smooth paths planing method |
| CN110347151A (en) * | 2019-05-31 | 2019-10-18 | 河南科技大学 | A kind of robot path planning method merging Bezier optimized Genetic Algorithm |
Non-Patent Citations (2)
| Title |
|---|
| 徐林: "餐厅服务机器人路径规划研究", 《万方数据知识服务平台》 * |
| 柯鹏: "基于多目标进化算法的车辆路径问题的研究", 《中国优秀博硕士学位论文全文数据库(博士) 信息科技辑》 * |
Cited By (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113390412A (en) * | 2020-03-11 | 2021-09-14 | 宁波方太厨具有限公司 | Full-coverage path planning method and system for robot, electronic equipment and medium |
| CN111481933A (en) * | 2020-05-18 | 2020-08-04 | 浙江工业大学 | Game path planning method based on improved potential field grid method |
| CN111766875A (en) * | 2020-06-18 | 2020-10-13 | 珠海格力智能装备有限公司 | Obstacle avoidance method and device for dust collection equipment and electronic equipment |
| EP4180894A4 (en) * | 2020-07-20 | 2023-12-27 | Huawei Digital Power Technologies Co., Ltd. | Method and device for planning obstacle avoidance path for traveling device |
| CN112099493A (en) * | 2020-08-31 | 2020-12-18 | 西安交通大学 | Autonomous mobile robot trajectory planning method, system and equipment |
| CN112214031A (en) * | 2020-09-25 | 2021-01-12 | 北京理工大学 | Multi-node collaborative landing position planning method based on genetic particle swarm optimization |
| CN112214031B (en) * | 2020-09-25 | 2021-08-20 | 北京理工大学 | Multi-node collaborative landing position planning method based on genetic particle swarm optimization |
| CN112256023A (en) * | 2020-09-28 | 2021-01-22 | 南京理工大学 | Bezier curve-based airport border patrol robot local path planning method and system |
| CN112256023B (en) * | 2020-09-28 | 2022-08-19 | 南京理工大学 | Bezier curve-based airport border patrol robot local path planning method and system |
| CN113932818B (en) * | 2020-12-30 | 2022-03-29 | 浙江德源智能科技股份有限公司 | Robot walking route planning method, program and storage medium |
| CN113932818A (en) * | 2020-12-30 | 2022-01-14 | 浙江德源智能科技股份有限公司 | Robot walking route planning method, program and storage medium |
| CN113110422A (en) * | 2021-03-25 | 2021-07-13 | 贵州电网有限责任公司 | Path planning method based on Q-IGA dynamic fitting Bezier curve |
| CN112947490A (en) * | 2021-04-09 | 2021-06-11 | 京东数科海益信息科技有限公司 | Path smoothing method, device, equipment, storage medium and product |
| WO2022252869A1 (en) * | 2021-06-02 | 2022-12-08 | 北京迈格威科技有限公司 | Obstacle bypassing method for mobile device, and mobile device and storage medium |
| CN113341876A (en) * | 2021-06-24 | 2021-09-03 | 合肥工业大学 | Five-axis curved surface machining track planning method based on differential vector optimization |
| CN113442140B (en) * | 2021-06-30 | 2022-05-24 | 同济人工智能研究院(苏州)有限公司 | Cartesian space obstacle avoidance planning method based on Bezier optimization |
| CN113442140A (en) * | 2021-06-30 | 2021-09-28 | 同济人工智能研究院(苏州)有限公司 | Bezier optimization-based Cartesian space obstacle avoidance planning method |
| CN114131591A (en) * | 2021-12-03 | 2022-03-04 | 山东大学 | Semi-physical simulation method and system for operation strategy of outer limb robot |
| WO2024077716A1 (en) * | 2022-10-11 | 2024-04-18 | 劢微机器人科技(深圳)有限公司 | Local path planning method for amr |
| CN116408793A (en) * | 2023-02-16 | 2023-07-11 | 广州数控设备有限公司 | Industrial robot path fairing method and system with continuous curvature |
| CN116408793B (en) * | 2023-02-16 | 2023-11-14 | 广州数控设备有限公司 | Industrial robot path fairing method and system with continuous curvature |
| CN116460830A (en) * | 2023-03-17 | 2023-07-21 | 北京信息科技大学 | Robot intelligent control system and control method based on artificial intelligence |
| CN116460830B (en) * | 2023-03-17 | 2023-10-20 | 北京信息科技大学 | Robot intelligent control system and control method based on artificial intelligence |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110865642A (en) | Path planning method based on mobile robot | |
| Wang et al. | Dynamic path planning for unmanned surface vehicle in complex offshore areas based on hybrid algorithm | |
| Zhang et al. | Mobile robot path planning based on improved localized particle swarm optimization | |
| CN108664022B (en) | Robot path planning method and system based on topological map | |
| CN110989352B (en) | Group robot collaborative search method based on Monte Carlo tree search algorithm | |
| Su et al. | Robot path planning based on random coding particle swarm optimization | |
| Tian | Research on robot optimal path planning method based on improved ant colony algorithm | |
| Huang et al. | APSO: An A*-PSO hybrid algorithm for mobile robot path planning | |
| Gan et al. | Dp-a*: For path planing of ugv and contactless delivery | |
| CN113110422A (en) | Path planning method based on Q-IGA dynamic fitting Bezier curve | |
| Tang et al. | An improved PSO for path planning of mobile robots and its parameters discussion | |
| Yu et al. | SOF-RRT*: An improved path planning algorithm using spatial offset sampling | |
| Xiao et al. | Multiobjective path optimization of an indoor AGV based on an improved ACO-DWA | |
| Yin et al. | Reinforcement learning path planning based on step batch Q-learning algorithm | |
| Morelli et al. | Integrated mapping and path planning for very large-scale robotic (vlsr) systems | |
| Liu et al. | UUV path planning method based on QPSO | |
| Masuzawa et al. | Observation planning for environment information summarization with deadlines | |
| CN115097814A (en) | Mobile robot path planning method, system and application based on improved PSO algorithm | |
| Neeraja et al. | A multi objective hybrid collision-free optimal path finder for autonomous robots in known static environments | |
| Mo et al. | Bio-geography based differential evolution for robot path planning | |
| Pang et al. | Task allocation for multi-AUV system under ocean current environment | |
| Zou et al. | Particle swarm optimization based topology control mechanism for holonomic unmanned vehicles operating in three-dimensional space | |
| Fang et al. | Research and application of a new lion swarm algorithm | |
| Xi et al. | Robust ga based global path planning for iop oriented mobil robot | |
| Liu et al. | A path optimization algorithm for motion planning with the moving target |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200306 |