CN110908374B - Mountain orchard obstacle avoidance system and method based on ROS platform - Google Patents
Mountain orchard obstacle avoidance system and method based on ROS platform Download PDFInfo
- Publication number
- CN110908374B CN110908374B CN201911111146.2A CN201911111146A CN110908374B CN 110908374 B CN110908374 B CN 110908374B CN 201911111146 A CN201911111146 A CN 201911111146A CN 110908374 B CN110908374 B CN 110908374B
- Authority
- CN
- China
- Prior art keywords
- obstacle
- data
- algorithm
- clustering
- distance
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 20
- 239000002420 orchard Substances 0.000 title claims abstract description 17
- 230000003068 static effect Effects 0.000 claims abstract description 10
- 238000004891 communication Methods 0.000 claims abstract description 7
- 238000001914 filtration Methods 0.000 claims description 21
- 238000009826 distribution Methods 0.000 claims description 14
- 238000010586 diagram Methods 0.000 claims description 13
- 238000012545 processing Methods 0.000 claims description 7
- 230000004888 barrier function Effects 0.000 claims description 6
- 238000007781 pre-processing Methods 0.000 claims description 4
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 238000011161 development Methods 0.000 description 17
- 230000000694 effects Effects 0.000 description 6
- 238000011160 research Methods 0.000 description 6
- 238000013461 design Methods 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 3
- 238000001514 detection method Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0214—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
- G05D1/0231—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means
- G05D1/0238—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using obstacle or wall sensors
- G05D1/024—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using obstacle or wall sensors in combination with a laser
-
- 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
- G05D1/0257—Control of position or course in two dimensions specially adapted to land vehicles using a radar
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Aviation & Aerospace Engineering (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Optics & Photonics (AREA)
- Electromagnetism (AREA)
- Traffic Control Systems (AREA)
- Optical Radar Systems And Details Thereof (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention discloses a hillside orchard obstacle avoidance system and method based on a ROS platform, wherein the system comprises a controller, a motor, a steering system, a communication module and a laser radar, the laser radar acquires static obstacles in the motion direction in real time and transmits acquired radar cloud point data to the controller; the controller preprocesses radar cloud point data, and combines an obstacle cloud point information clustering algorithm, a two-dimensional cloud point clustering data convex hull algorithm and a rotary hull-jamming algorithm to position and identify an obstacle, so as to obtain position and angle information of the obstacle, determine the convex hull diameter of the obstacle, establish an obstacle dangerous area by taking a clustering center point as a circle center, call an obstacle avoidance algorithm to establish an obstacle avoidance path planning and course control mode, generate a lowest-cost path for obstacle avoidance, and control a motor and a steering system. The invention realizes and optimizes the obstacle avoidance function by analyzing the obstacle avoidance data and combining a plurality of groups of algorithms.
Description
Technical Field
The invention relates to a laser radar and vision guide technology, in particular to a hillside orchard obstacle avoidance system and method based on a ROS platform.
Background
The recent research results of foreign research on lidar, such as the 2016 university of columbia robot system research institute, include processing algorithms for obstacle detection of stationary and moving objects under driving conditions using 3D lidar based on three-dimensional pixels and multiple regions. An example of applying a gating strategy on cloud points, efficient ground estimation and recognition using the piecewise plane fitting algorithm RANSAC analyzes three-dimensional network models of static and dynamic obstacles. On 6/10/2016, Google formally sourced the SLAM algorithm, which was downloaded in Google searches.
The introduction of the laser radar technology into agricultural transportation is relatively late in domestic colleges and universities. However, many robot development companies for indoor two-dimensional modeling research have also introduced new two-dimensional modeling robots in the market. Among them, the real-time closed-loop algorithm is applied to a portable laser range finder (also called LIDAR), synchronous positioning and mapping (SLAM) to obtain a building plan, and a floor plan is generated and displayed in real time to help an operator to evaluate the quality and coverage of captured data.
Disclosure of Invention
Aiming at the condition of the prior art, the invention provides a hillside orchard obstacle avoidance system and method based on a ROS platform.
The mountain orchard obstacle avoidance system based on the ROS platform comprises a controller, a motor, a steering system, a communication module and a laser radar, wherein the laser radar is connected with the controller through the communication module, collects static obstacles in the movement direction in real time and transmits collected radar cloud point data to the controller; the controller preprocesses radar cloud point data, combines an obstacle cloud point information clustering algorithm, a two-dimensional cloud point clustering data convex hull algorithm and a rotary hull-jamming algorithm to position and identify an obstacle, obtains position and angle information of the obstacle, determines the convex hull diameter of the obstacle, establishes an obstacle dangerous area by taking a clustering center point as a circle center, calls an obstacle avoidance algorithm, formulates an obstacle avoidance path planning and course control mode, generates a lowest-cost path for obstacle avoidance, and controls a motor and a steering system.
The invention relates to a hillside orchard obstacle avoidance method based on a ROS platform, which comprises the following steps:
s1, identifying the environment through a laser radar, collecting static obstacles in the motion direction in real time, and preprocessing the collected radar cloud point data;
s2, analyzing and processing the preprocessed radar cloud point data, and positioning and describing the obstacle through a recognition algorithm to obtain position and angle information of the obstacle; clustering the preprocessed radar cloud point data, and determining a clustering center; determining the convex hull diameter of the barrier by a convex hull algorithm and a rotary hull-jamming algorithm, and establishing a barrier dangerous area by taking a clustering center as a circle center;
and S3, calling an obstacle avoidance algorithm according to the position and angle information of the obstacle and the danger area of the obstacle, formulating an obstacle avoidance path planning and course control mode, and generating a lowest-cost obstacle avoidance path.
In a preferred embodiment, in step S2, the preprocessed radar cloud point data are clustered by using a density-based clustering algorithm, so as to obtain a distribution map of the obstacle points.
Further, in step S2, a convex polygon corresponding to the obstacle point distribution map is obtained by using a Graham' S Scan algorithm; after the clustering center is determined, the maximum diameter of a convex polygon corresponding to the obstacle point distribution diagram is obtained through a convex hull algorithm and a rotary hull clamping algorithm and is used as the convex hull diameter of the obstacle, and then the clustering center is used as the circle center to establish an obstacle dangerous area.
Compared with the prior art, the invention has the following advantages and beneficial effects:
1. the invention carries out integral design on the obstacle avoidance of the ROS platform, and combines various algorithms to realize and optimize the obstacle avoidance function. In the aspect of hardware, the two-dimensional laser radar which is more popular in price and meets the design requirement in performance is used as a hardware platform for obstacle avoidance research, the two-dimensional laser radar and the ROS platform are calibrated and corrected, and an accurate data acquisition basis is provided for subsequent experiments. In the aspect of software, firstly, a Rodriguez A fast clustering algorithm is used for finding out a clustering center, then, a convex hull algorithm and a rotary hull algorithm are combined to obtain the maximum diameter of a convex hull, then, an obstacle danger area is established by taking the clustering center point as the center of a circle, and then, an obstacle avoidance path planning and course control mode is established according to the obstacle danger area.
2. The invention relates to an obstacle avoidance system designed for a transport vehicle, which adopts a two-dimensional laser radar to perform environment modeling on an external environment so as to realize transportation system environment acquisition, in particular to perform real-time acquisition on static obstacles in a movement direction and analyze the edge characteristics of the obstacles. And researching a low-cost obstacle avoidance algorithm according to the edge characteristics of the obstacle. The improved median filtering algorithm is adopted for filtering processing, and the problem that the previous median filtering effect depends on filtering window selection is improved.
3. Through ROS platform test experiments, the obstacle avoidance algorithm can basically achieve the obstacle avoidance function on the ROS platform, and the feasibility of two-dimensional laser radar obstacle avoidance based on the ROS platform is preliminarily verified.
Drawings
Fig. 1 is a structural block diagram of an obstacle avoidance system of the present invention;
FIG. 2 is a schematic representation of a ROS development platform coordinate system;
FIG. 3 is a schematic diagram of coordinate transformation;
FIG. 4 is a comparison graph of the effect before and after filtering, in which (a) and (b) represent a median before-filtering rectangular coordinate graph and a median after-filtering rectangular coordinate graph, respectively;
FIG. 5 is a diagram of clustering effect, in which (a) and (b) represent a data decision diagram and a clustered point distribution diagram, respectively;
FIG. 6 is a cloud point layout after convex hull algorithm processing;
fig. 7 is a schematic diagram of obstacle avoidance angle determination.
Detailed Description
The present invention will be described in further detail with reference to examples and drawings, but the present invention is not limited thereto.
Examples
The mountain orchard obstacle avoidance system based on the ROS platform is mainly used for collecting static obstacles in the moving direction in real time through the two-dimensional laser radar, analyzing the edge characteristics of the obstacles and researching an obstacle avoidance algorithm according to the characteristics of the obstacles to perform obstacle avoidance operation.
As shown in fig. 1, the hillside orchard obstacle avoidance system based on the ROS platform of the present invention comprises a controller, a motor, a steering system, a communication module and a two-dimensional laser radar, wherein the two-dimensional laser radar is connected with the controller via the communication module, and the two-dimensional laser radar collects static obstacles in the movement direction in real time and transmits the collected cloud point data of the radar to the controller; the controller carries out preprocessing such as filtering on acquired radar cloud point data through a median filtering algorithm and the like, and carries out positioning and identification on the obstacle by combining an obstacle cloud point information clustering algorithm, a two-dimensional cloud point clustering data convex hull algorithm and a rotary hull algorithm to obtain position and angle information of the obstacle, determine the convex hull diameter of the obstacle, establish an obstacle dangerous area by taking a clustering center point as a circle center, then call an obstacle avoidance algorithm, formulate an obstacle avoidance path planning and course control mode, generate a lowest cost path for avoiding the obstacle, control a motor and a steering system, and achieve basic obstacle avoidance requirements of an ROS platform.
The invention discloses the overall design of ROS platform obstacle avoidance, and realizes and optimizes the obstacle avoidance function by combining multiple groups of algorithms. In the aspect of hardware, the two-dimensional laser radar and the ROS platform are calibrated and corrected, and an accurate data acquisition basis is provided for subsequent experiments. In the aspect of software, firstly, a Rodriguez A fast clustering algorithm is used for finding out a clustering center, then, a convex hull algorithm and a rotary hull algorithm are combined to obtain the maximum diameter of a convex hull, then, an obstacle dangerous area is established by taking the clustering center as the center of a circle, and then, an obstacle avoidance algorithm is called to establish an obstacle avoidance path planning and course control mode. The invention relates to a hillside orchard obstacle avoidance method based on an ROS platform, which comprises the following steps:
and S1, identifying the environment through the two-dimensional laser radar, collecting static obstacles in the motion direction in real time, and preprocessing the collected radar cloud point data through coordinate conversion, filtering and the like.
Because the environment data collected by the laser radar is relative to the form of a laser radar polar coordinate system, if the laser radar is additionally arranged on the ROS development platform, the coordinate system takes the ROS development platform as a standard. Therefore, the relationship between the laser radar coordinate system and the ROS development platform coordinate system needs to be calibrated again in the step, and the global coordinate system, the ROS development platform coordinate system and the laser radar coordinate system are respectively adopted for defining.
The ROS development platform positioning system is provided with three coordinate definitions which are a global coordinate system, an ROS development platform coordinate system and a laser radar coordinate system respectively. The relationship between the development platform and the lidar coordinates is determined by the relative position of the lidar to the development platform. Hence the global coordinate system XwOwYwAnd a vehicle coordinate system XvOvYvAs a main research coordinate system, the lidar coordinate system is XsOsYsAs in fig. 2. The pose of the platform in the global coordinate system is denoted as (X, Y, theta). In this coordinate system, (X, Y) represents the ROS platform position and θ represents the heading angle.
FIG. 3 shows the relationship between the ROS development platform coordinate system and the global coordinate system of the point P, as shown in the following formula (1), and the development platform pose coordinate is (x)v,yv,θ)。
The laser radar adopts a polar coordinate form as a coordinate system, and is determined by taking a development platform as a standard after being added to an ROS platform. And determining the relation between the laser radar coordinate system and the ROS development platform through laser radar calibration. Referring to FIG. 3, the lidar coordinate system is XsOsYs. According to FIG. 2, the lidar coordinates are (x)s,ys) In polar form (L, β). Then, the conversion relationship between the rectangular coordinate and the polar coordinate of the laser radar is as follows:
the position and orientation coordinates of the laser radar are (X)s,YsPhi) where phi is the ROS platform coordinate system axis OvXvTo the axis OsXsThe angle formed. Let the coordinate of the lidar coordinate point s under the development platform be (x)vs,yvs) Then the rotational translation of the coordinate system is as in equation (3):
and carrying out calibration test according to the calibration method. Firstly, determining the pose of the ROS platform under the global coordinate as (X)s,YsPhi). The coordinate of the auxiliary calibration object placed under the global coordinate system is (x)w,yw). The relation data of the global coordinate system and the laser radar coordinate system can be obtained according to the relation of the coordinate systems in the figures 2 and 3:
the center of the auxiliary calibration object under the laser radar coordinate system is calculated as (x) by the least square methods,ys). Calculating the position and orientation coordinates (X) of the laser radar by using an M-estimation method through a formula (4)s,Ys,Ф)。
The final calibration result is: xs=2cm;Ys=7cm;φ=3°。
Because the laser radar data collection can produce the interference cloud point signal, interference noise can make the detection effect of barrier unstable. Therefore, the invention also needs to filter the laser radar collected data.
In this embodiment, a median filtering method is adopted to perform filtering processing on radar cloud point data acquired by a laser radar. Since the median filtering effect depends on the choice of the filtering window, the present embodiment adopts an improved median filtering method. The median filtering method firstly judges whether the extreme value of the data point is in the filtering window range, if so, the data point can be filtered, otherwise, the filtering operation is cancelled. In fig. 4, b is the improved median filtering algorithm effect, and the filtering algorithm is as formula (5):
wherein: x is the number ofmin=min{xi-s,…,xi-1,xi,xi+1,…,xi+s},xmax=max{xi-s,…,xi-1,xi,xi+1,…,xi+s};
And S2, analyzing and processing the preprocessed radar cloud point data, positioning and describing the obstacle through a recognition algorithm, and finally obtaining the environmental distance information of the obstacle, wherein the environmental distance information comprises position and angle information.
The method adopts a density-based clustering algorithm (such as Rodriguez A fast clustering algorithm) to fast and accurately cluster the preprocessed obstacle radar cloud point data, further determines a clustering center point, and positions the obstacle radar cloud point data center according to the clustering center. In the traditional K-means algorithm, K needs to be given in advance before the program is executed, and the selection of the K value cannot be guaranteed to be constant; the DBSCAN clustering algorithm must assign a density threshold to cluster arbitrary shape distributions to remove noise below the density threshold. In addition, the density-based clustering algorithm can obtain non-spherical clustering results, can well describe data distribution, and is lower than a K-means algorithm in algorithm complexity.
The density-based clustering algorithm is first defined conceptually, and for each data point i, two quantities need to be calculated: local density ρiAnd a minimum distance from the density above data point i. The local density is defined as follows:
ρi=∑jχ(dij-dc) (6)
wherein can be defined as:
where d iscIs a truncation distance. Definition of dcComprises the following steps: and arranging the mutual distances of all the points from small to large, and taking the distance value of the first 2% position as a truncation distance. The distance between data points is defined as follows:
the above equation (8) is used to find the minimum distance between the data points with density greater than the data point i and the data point i. The distance δ between the data pointsiThe larger the value of (d), the longer the distance of the data point i from the high density point is, the more likely the data point i becomes the cluster center. For the global maximum density of data points, its distance to all data points is maximum. The clustering algorithm is shown in table 1 below.
TABLE 1 clustering Algorithm
Fig. 5 is a diagram of a clustered obstacle point distribution diagram, which is a data decision diagram of a clustering algorithm and a result of the clustering algorithm. In fig. 5, (a) shows a data decision map, and (b) shows a clustered obstacle point distribution map.
A convex hull is a concept in computer geometry, and in a real vector space V, the intersection S of all convex sets containing X is called the convex hull ch (X) of X for a given set X. The convex hull CH (X) of X may use all points (X) in set X1,x2,…,xn) Are constructed from linear combinations of (a). The invention adopts Graham's Scan algorithm to calculate the convex polygon corresponding to the obstacle clustering data (namely the obstacle point distribution diagram) after the radar cloud point data clustering. After the clustering center is determined, the maximum diameter of the convex polygon corresponding to the obstacle point distribution diagram is obtained through a convex hull algorithm and a rotary hull clamping algorithm (namely the convex hull of the obstacle is determined)Diameter) and establishing an obstacle dangerous area by taking the clustering center as a circle center so as to establish an obstacle avoidance path planning and course control mode in the following.
The basic idea of the Graham's Scan algorithm is described below: the convex hull problem is solved by setting a stack S on the candidate points. Every data point in set X is pushed onto the stack once, and the data points of the vertices in non-CH (X) are finally popped off the stack; when the computation is terminated, only the vertices CH (X) in the stack S are arranged in the counterclockwise direction with respect to the boundary where the vertices appear. Table 2 shows the Graham's Scan algorithm.
Let P0Is the data point in the set X with the smallest Y coordinate, and if there are multiple such data points, the leftmost data point is taken as P0And | X | ≧ 3, the call function TOP (S) returns the data point at the top of the stack S, and the call function NEXT-TO-TOP (S) returns the data point below the top of the stack, but does not change the structure of the stack.
TABLE 2 Graham's Scan Algorithm
The diameter of the convex hull is obtained by adopting a rotating hull clamping algorithm, two points which are the farthest points on the convex hull exist, and a pair of parallel lines can be drawn through the two points respectively. By rotating the pair of parallel lines, the parallel lines and one edge of the convex hull are coincided. According to the characteristics of the convex hull, the farthest point can be prevented from being repeatedly calculated by enumerating the edges anticlockwise, and therefore the maximum butt point of the diameter of the convex hull is obtained. The rotating jam algorithm is shown in table 3 below.
The maximum diameter d of the convex polygon in FIG. 6 is processed by a rotating and clamping shell algorithmmax5.65cm, 15.55cm, 12.65cm, 9.43cm, 8.94cm and 12.04cm, respectively. Finally, the maximum diameter d of the convex polygon is used as the centermaxThe area being a circle as the danger area for obstacles, i.e. avoidanceAnd (4) barrier range.
And S3, calling an obstacle avoidance algorithm according to the position and angle information of the obstacle and the danger area of the obstacle, formulating an obstacle avoidance path planning and course control mode, and generating a lowest-cost obstacle avoidance path.
The basic obstacle avoidance requirement of the ROS platform is met through an improved VFH + path planning method, the passing direction interval is determined according to the distribution of surrounding environment obstacles on a guide track of the ROS development platform and a certain angle threshold, and then the passing direction and speed are determined. And finally, establishing an obstacle avoidance coordinate system by taking the ROS platform as a center, as shown in fig. 7.
And on the basis of the positioning of the obstacle obtained in the previous step, expanding the radius of the obstacle avoidance range to be half of the width of the ROS platform through clustering center circle expansion. Setting the diameter d at two sides of the development platform through dynamic constraint1And d2The no-pass area of (1); the ROS platform will be limited during cornering and set the kinetic constraint angle to 30. And the distance from the obstacle to the two-dimensional laser radar on the ROS platform can be known from the polar coordinates.
And after the threshold value is determined, acquiring a passing direction interval. Defined according to the threshold h (i) as follows:
when H (i) is 0, the direction i is passable, otherwise, the direction i is impassable.
The above embodiments are preferred embodiments of the present invention, but the present invention is not limited to the above embodiments, and any other changes, modifications, substitutions, combinations, and simplifications which do not depart from the spirit and principle of the present invention should be construed as equivalents thereof, and all such changes, modifications, substitutions, combinations, and simplifications are intended to be included in the scope of the present invention.
Claims (7)
1. A hillside orchard obstacle avoidance system based on a ROS platform is characterized by comprising a controller, a motor, a steering system, a communication module and a laser radar, wherein the laser radar is connected with the controller through the communication module, collects static obstacles in the motion direction in real time and transmits collected radar cloud point data to the controller; the controller preprocesses radar cloud point data, and performs positioning and identification on an obstacle by combining an obstacle cloud point information clustering algorithm, a two-dimensional cloud point clustering data convex hull algorithm and a rotary hull-jamming algorithm to obtain position and angle information of the obstacle, determine the convex hull diameter of the obstacle, establish an obstacle dangerous area by taking a clustering center point as a circle center, then call an obstacle avoidance algorithm, formulate an obstacle avoidance path planning and course control mode, generate a lowest-cost path for obstacle avoidance, and control a motor and a steering system;
the controller accurately clusters the preprocessed obstacle radar cloud point data by adopting a density-based clustering algorithm to obtain an aspheric clustering result, determines a clustering center point, and positions the obstacle radar cloud point data center according to the clustering center; in a density-based clustering algorithm, the local density ρ needs to be calculated for each data point iiAnd a minimum distance from the density above the data point i, the local density being defined as:
ρi=∑jχ(dij-dc)
wherein d iscThe distance is a truncation distance, the mutual distances of all points are arranged from small to large, and the distance value of the first 2% of positions is taken as the truncation distance; the distance between data points is defined as follows:
the distance δ between the above data pointsiFor obtainingTaking the minimum distance value between the data points with the density larger than the data points i and the data points i; distance δ between data pointsiThe larger the value of (d), the longer the distance of the data point i from the high-density point is, the more likely the data point i becomes the clustering center; for the global maximum density of data points, its distance from all data points is maximum.
2. The hillside orchard obstacle avoidance system of claim 1, wherein the lidar is a two-dimensional lidar.
3. The hillside orchard obstacle avoidance system of claim 1, wherein the controller preprocesses the radar cloud point data including coordinate conversion and filtering.
4. A hillside orchard obstacle avoidance method based on a ROS platform is characterized by comprising the following steps:
s1, identifying the environment through a laser radar, collecting static obstacles in the motion direction in real time, and preprocessing the collected radar cloud point data;
s2, analyzing and processing the preprocessed radar cloud point data, and positioning and describing the obstacle through a recognition algorithm to obtain position and angle information of the obstacle; clustering the preprocessed radar cloud point data, and determining a clustering center; determining the convex hull diameter of the barrier by a convex hull algorithm and a rotary hull-jamming algorithm, and establishing a barrier dangerous area by taking a clustering center as a circle center;
accurately clustering the preprocessed obstacle radar cloud point data by adopting a density-based clustering algorithm to obtain an aspheric clustering result, determining a clustering center point, and positioning the obstacle radar cloud point data center according to the clustering center; in a density-based clustering algorithm, the local density ρ needs to be calculated for each data point iiAnd a minimum distance from the density above the data point i, the local density being defined as:
ρi=∑jχ(dij-dc)
wherein d iscThe distance is a truncation distance, the mutual distances of all points are arranged from small to large, and the distance value of the first 2% of positions is taken as the truncation distance; the distance between data points is defined as follows:
the distance δ between the above data pointsiThe distance minimum value between the data points with the density larger than the data points i and the data points i is obtained; distance δ between data pointsiThe larger the value of (d), the longer the distance of the data point i from the high-density point is, the more likely the data point i becomes the clustering center; for the data point with the maximum global density, the distance between the data point and all the data points is maximum;
and S3, calling an obstacle avoidance algorithm according to the position and angle information of the obstacle and the danger area of the obstacle, formulating an obstacle avoidance path planning and course control mode, and generating a lowest-cost obstacle avoidance path.
5. The hillside orchard obstacle avoidance method according to claim 4, wherein in step S2, the preprocessed radar cloud point data are clustered by a density-based clustering algorithm to obtain an obstacle point distribution map.
6. The obstacle avoidance method for the hillside orchard according to claim 5, wherein in step S2, a Graham' S Scan algorithm is adopted to find a convex polygon corresponding to the obstacle point distribution map; after the clustering center is determined, the maximum diameter of a convex polygon corresponding to the obstacle point distribution diagram is obtained through a convex hull algorithm and a rotary hull clamping algorithm and is used as the convex hull diameter of the obstacle, and then the clustering center is used as the circle center to establish an obstacle dangerous area.
7. The obstacle avoidance method for hillside orchards according to claim 4, wherein step S3 is implemented by using an improved VFH + path planning method to make an obstacle avoidance path plan and a heading control manner.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911111146.2A CN110908374B (en) | 2019-11-14 | 2019-11-14 | Mountain orchard obstacle avoidance system and method based on ROS platform |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911111146.2A CN110908374B (en) | 2019-11-14 | 2019-11-14 | Mountain orchard obstacle avoidance system and method based on ROS platform |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110908374A CN110908374A (en) | 2020-03-24 |
CN110908374B true CN110908374B (en) | 2021-04-23 |
Family
ID=69817584
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201911111146.2A Active CN110908374B (en) | 2019-11-14 | 2019-11-14 | Mountain orchard obstacle avoidance system and method based on ROS platform |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110908374B (en) |
Families Citing this family (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111427355B (en) * | 2020-04-13 | 2023-05-02 | 京东科技信息技术有限公司 | Obstacle data processing method, device, equipment and storage medium |
CN112130569B (en) * | 2020-09-27 | 2024-03-26 | 上海木木聚枞机器人科技有限公司 | Ultrasonic wave range setting method and system |
CN112529874A (en) * | 2020-12-14 | 2021-03-19 | 上海智蕙林医疗科技有限公司 | Obstacle detection method, device, medium and robot based on three-dimensional radar |
CN114654471A (en) * | 2022-04-29 | 2022-06-24 | 中国煤炭科工集团太原研究院有限公司 | Anchor protection mechanical arm obstacle avoidance system and method based on laser scanner |
CN115903853B (en) * | 2023-01-06 | 2023-05-30 | 北京理工大学 | Safe feasible domain generation method and system based on effective barrier |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107272680A (en) * | 2017-06-16 | 2017-10-20 | 华南理工大学 | A kind of automatic follower method of robot based on ROS robot operating systems |
CN107390681A (en) * | 2017-06-21 | 2017-11-24 | 华南理工大学 | A kind of mobile robot real-time location method based on laser radar and map match |
CN107450571A (en) * | 2017-09-30 | 2017-12-08 | 江西洪都航空工业集团有限责任公司 | A kind of AGV dolly Laser navigation systems based on ROS |
CN109001756A (en) * | 2018-05-04 | 2018-12-14 | 上海交通大学 | Multi-line laser radar obstacle detection system and method based on embedded device |
CN109032162A (en) * | 2018-08-20 | 2018-12-18 | 辽宁壮龙无人机科技有限公司 | A kind of unmanned plane obstacle avoidance system and control method based on laser radar |
CN109633676A (en) * | 2018-11-22 | 2019-04-16 | 浙江中车电车有限公司 | A kind of method and system based on the laser radar obstruction detection direction of motion |
CN109828589A (en) * | 2019-03-08 | 2019-05-31 | 北京大成高科机器人技术有限公司 | Modularization farm machinery platform and control method |
CN110147106A (en) * | 2019-05-29 | 2019-08-20 | 福建(泉州)哈工大工程技术研究院 | Has the intelligent Mobile Service robot of laser and vision fusion obstacle avoidance system |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6525751B2 (en) * | 2001-05-25 | 2003-02-25 | Xerox Corporation | Raster output scanner fraction-of-scan polygon rephasing and algorithm |
-
2019
- 2019-11-14 CN CN201911111146.2A patent/CN110908374B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107272680A (en) * | 2017-06-16 | 2017-10-20 | 华南理工大学 | A kind of automatic follower method of robot based on ROS robot operating systems |
CN107390681A (en) * | 2017-06-21 | 2017-11-24 | 华南理工大学 | A kind of mobile robot real-time location method based on laser radar and map match |
CN107450571A (en) * | 2017-09-30 | 2017-12-08 | 江西洪都航空工业集团有限责任公司 | A kind of AGV dolly Laser navigation systems based on ROS |
CN109001756A (en) * | 2018-05-04 | 2018-12-14 | 上海交通大学 | Multi-line laser radar obstacle detection system and method based on embedded device |
CN109032162A (en) * | 2018-08-20 | 2018-12-18 | 辽宁壮龙无人机科技有限公司 | A kind of unmanned plane obstacle avoidance system and control method based on laser radar |
CN109633676A (en) * | 2018-11-22 | 2019-04-16 | 浙江中车电车有限公司 | A kind of method and system based on the laser radar obstruction detection direction of motion |
CN109828589A (en) * | 2019-03-08 | 2019-05-31 | 北京大成高科机器人技术有限公司 | Modularization farm machinery platform and control method |
CN110147106A (en) * | 2019-05-29 | 2019-08-20 | 福建(泉州)哈工大工程技术研究院 | Has the intelligent Mobile Service robot of laser and vision fusion obstacle avoidance system |
Non-Patent Citations (1)
Title |
---|
移动机器人前景障碍物检测及避障系统设计与实现;李晓飞;《中国优秀硕士学位论文全文数据库信息科技辑》;20190215(第2期);第I138-1412页 * |
Also Published As
Publication number | Publication date |
---|---|
CN110908374A (en) | 2020-03-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110908374B (en) | Mountain orchard obstacle avoidance system and method based on ROS platform | |
US10665115B2 (en) | Controlling unmanned aerial vehicles to avoid obstacle collision | |
CN108647646B (en) | Low-beam radar-based short obstacle optimized detection method and device | |
Kang et al. | Automatic targetless camera–lidar calibration by aligning edge with gaussian mixture model | |
CN110703268B (en) | Air route planning method and device for autonomous positioning navigation | |
McDaniel et al. | Ground plane identification using LIDAR in forested environments | |
Andreasson et al. | Mini-SLAM: Minimalistic visual SLAM in large-scale environments based on a new interpretation of image similarity | |
Shinzato et al. | Road estimation with sparse 3D points from stereo data | |
Milella et al. | 3D reconstruction and classification of natural environments by an autonomous vehicle using multi-baseline stereo | |
Liu et al. | Autonomous road detection and modeling for UGVs using vision-laser data fusion | |
Guo et al. | E-LOAM: LiDAR odometry and mapping with expanded local structural information | |
CN112393735B (en) | Positioning method and device, storage medium and electronic device | |
Hoang et al. | Motion estimation based on two corresponding points and angular deviation optimization | |
WO2023131203A1 (en) | Semantic map updating method, path planning method, and related apparatuses | |
Xu et al. | Real-time road detection and description for robot navigation in an unstructured campus environment | |
CN115330861A (en) | Repositioning algorithm based on object plane common representation and semantic descriptor matching | |
Vatavu et al. | Environment perception using dynamic polylines and particle based occupancy grids | |
Ren et al. | Towards efficient and robust LiDAR-based 3D mapping in urban environments | |
Dang et al. | LiDAR-based online navigation algorithm for an autonomous agricultural robot | |
Thanh et al. | LiDAR-Based Online Navigation Algorithm For An Autonomous Agricultural Robot⁑ | |
Lv et al. | Planar scan matching using incident angle | |
Hroob et al. | Learned Long-Term Stability Scan Filtering for Robust Robot Localisation in Continuously Changing Environments | |
CN116382308B (en) | Intelligent mobile machinery autonomous path finding and obstacle avoiding method, device, equipment and medium | |
Li et al. | L-LO: Enhancing Pose Estimation Precision via a Landmark-Based LiDAR Odometry | |
KR102618951B1 (en) | Method for visual mapping, and computer program recorded on record-medium for executing method therefor |
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 |