CN112506196A - Robot obstacle avoidance method and system based on priori knowledge - Google Patents
Robot obstacle avoidance method and system based on priori knowledge Download PDFInfo
- Publication number
- CN112506196A CN112506196A CN202011416636.6A CN202011416636A CN112506196A CN 112506196 A CN112506196 A CN 112506196A CN 202011416636 A CN202011416636 A CN 202011416636A CN 112506196 A CN112506196 A CN 112506196A
- Authority
- CN
- China
- Prior art keywords
- obstacle
- robot
- image
- obstacle avoidance
- attribute information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 64
- 239000013598 vector Substances 0.000 claims abstract description 24
- 230000008569 process Effects 0.000 claims abstract description 14
- 238000013507 mapping Methods 0.000 claims abstract description 13
- 238000012549 training Methods 0.000 claims description 18
- 238000010276 construction Methods 0.000 claims description 3
- 238000005259 measurement Methods 0.000 claims description 3
- 230000004888 barrier function Effects 0.000 claims description 2
- 239000011159 matrix material Substances 0.000 description 7
- 238000012360 testing method Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 230000003068 static effect Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000005266 casting Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 230000006872 improvement Effects 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
- 238000006467 substitution reaction Methods 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
- G05D1/0231—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
-
- 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
- G05D1/0246—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using a video camera in combination with image processing means
-
- 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)
- Electromagnetism (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention provides a robot obstacle avoidance method and system based on prior knowledge, and relates to the technical field of robot obstacle avoidance. According to the embodiment of the invention, the image of the obstacle is obtained, the prior knowledge is utilized to obtain the obstacle category and attribute information, the vector mapping method is adopted to establish the environment grid map, the angle range formed by the robot as the center and the obstacle is obtained, and finally the obstacle avoidance strategy is generated. By introducing prior knowledge, the obstacle avoidance method can more accurately obtain obstacle information in the obstacle avoidance process, and a more appropriate obstacle avoidance strategy is selected according to the type and attribute information of the obstacle, so that the obstacle avoidance efficiency is improved.
Description
Technical Field
The invention relates to the technical field of robot obstacle avoidance, in particular to a robot obstacle avoidance method and system based on priori knowledge.
Background
The obstacle avoidance problem is a hot problem of research in recent years, and a mobile robot needs to identify and avoid obstacles during traveling. Many sensors are available in the art for identifying obstacles, such as lidar, sonar, infrared and vision sensors, etc.
The traditional obstacle avoidance algorithm comprises an artificial potential field method, a vector field histogram method (VFH), a visual graph method and the like; the intelligent obstacle avoidance algorithm comprises an ant colony algorithm, a genetic algorithm, a neural network, fuzzy control and the like.
In the research directions of the research, obstacles are avoided by improving the accuracy of the algorithm, so that the path is optimized. Although the success rate and efficiency of obstacle avoidance are improved through algorithm improvement, a large amount of useless exploration and random exploration always exist in the training process, the more complex the environment is, the longer the time required for training is, the obstacle avoidance efficiency is gradually reduced, obstacles are not distinguished, and the obstacle avoidance strategies are the same.
Disclosure of Invention
Technical problem to be solved
Aiming at the defects of the prior art, the invention provides a robot obstacle avoidance method and system based on prior knowledge, and solves the technical problems that the existing obstacle avoidance technology does not distinguish obstacles, and the obstacle avoidance strategies are the same.
(II) technical scheme
In order to achieve the purpose, the invention is realized by the following technical scheme:
in a first aspect, a robot obstacle avoidance method based on prior knowledge includes:
determining the distance from the robot to an obstacle, and acquiring the initial position of the robot;
acquiring an image of an obstacle, and obtaining the type and attribute information of the obstacle by using priori knowledge according to the image;
establishing an environment grid map by adopting a vector mapping method, and acquiring an angle range formed by a robot as a center and an obstacle;
and generating an obstacle avoidance strategy by combining the distance from the robot to the obstacle, the initial position of the robot, the obstacle category and attribute information and the angle range.
Preferably, the determining the distance from the robot to the obstacle specifically includes:
d＝0.5(c*t)
where c is the speed of light and t is the time interval from the emission of laser light to the reception of laser light by the laser rangefinder.
Preferably, the obtaining the image of the obstacle and obtaining the obstacle category information by using the priori knowledge according to the image specifically includes:
acquiring an image of the obstacle;
and inputting the obstacle image into a pre-trained class recognition model to obtain the obstacle class information.
Preferably, the construction process of the category identification model includes:
acquiring a plurality of obstacle images and determining the corresponding category information of the plurality of obstacle images;
taking the multiple obstacle images as training samples, taking corresponding category information as output labels, and constructing a training database;
and based on the training data, performing model training by adopting a nearest neighbor classifier under the Euclidean distance to obtain the class identification model.
Preferably, the obtaining the image of the obstacle and obtaining the attribute information of the obstacle according to the image by using the prior knowledge specifically includes:
the robot inquires the attribute information of the obstacles in a knowrob knowledge base, and the knowrob knowledge base writes a plurality of obstacles and corresponding attribute information thereof in advance through an entity description language.
Preferably, the x-axis of the environment grid map indicates an angle between the robot center and the obstacle, and the y-axis indicates a probability p of the obstacle existing in the angular direction.
Preferably, the angle range between the center of the robot and the obstacle is determined according to whether the obstacle is located right in front of the robot or in front of the robot.
In a second aspect, a robot obstacle avoidance system based on prior knowledge includes:
the distance measurement module is used for determining the distance from the robot to the obstacle and acquiring the initial position of the robot;
the image module is used for acquiring an image of the obstacle and obtaining the type and attribute information of the obstacle by using the prior knowledge according to the image;
the map module is used for establishing an environment grid map by adopting a vector mapping method and acquiring an angle range formed by a robot as a center and an obstacle;
and the obstacle avoidance module is used for generating an obstacle avoidance strategy by combining the distance from the robot to the obstacle, the initial position of the robot, the obstacle category and attribute information and the angle range.
(III) advantageous effects
The invention provides a robot obstacle avoidance method and system based on prior knowledge. Compared with the prior art, the method has the following beneficial effects:
according to the method, the image of the obstacle is obtained, the prior knowledge is utilized to obtain the obstacle category and attribute information, the environment grid map is established by adopting a vector mapping method, the angle range formed by the center of the robot and the obstacle is obtained, and finally the obstacle avoidance strategy is generated. By introducing prior knowledge, the obstacle avoidance method can more accurately obtain obstacle information in the obstacle avoidance process, and a more appropriate obstacle avoidance strategy is selected according to the type and attribute information of the obstacle, so that the obstacle avoidance efficiency is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic flow chart of a robot obstacle avoidance method based on prior knowledge according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a polar histogram generated by the VFH algorithm provided by an embodiment of the present invention;
FIG. 3 is a schematic view of an embodiment of the present invention with an obstacle positioned directly in front of the robot;
FIG. 4 is a schematic view of an embodiment of the present invention when an obstacle is positioned laterally forward of the robot;
fig. 5 is a structural block diagram of a robot obstacle avoidance system based on prior knowledge according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention are clearly and completely described, and it is obvious that the described embodiments are a part of the embodiments of the present invention, but not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The embodiment of the application provides the robot obstacle avoidance method and system based on the priori knowledge, solves the technical problems that the existing obstacle avoidance technology does not distinguish obstacles and the obstacle avoidance strategies are the same, and achieves the beneficial effect of improving the obstacle avoidance efficiency.
In order to solve the technical problems, the general idea of the embodiment of the application is as follows:
according to the embodiment of the invention, the image of the obstacle is obtained, the prior knowledge is utilized to obtain the obstacle category and attribute information, the vector mapping method is adopted to establish the environment grid map, the angle range formed by the robot as the center and the obstacle is obtained, and finally the obstacle avoidance strategy is generated. By introducing prior knowledge, the obstacle avoidance method can more accurately obtain obstacle information in the obstacle avoidance process, and a more appropriate obstacle avoidance strategy is selected according to the type and attribute information of the obstacle, so that the obstacle avoidance efficiency is improved.
In order to better understand the technical solution, the technical solution will be described in detail with reference to the drawings and the specific embodiments.
As shown in fig. 1, an embodiment of the present invention provides a robot obstacle avoidance method based on prior knowledge, including:
scanning the surrounding environment by using a laser range finder, determining the distance from the robot to an obstacle, and acquiring the initial position of the robot;
acquiring an image of an obstacle, and obtaining the type and attribute information of the obstacle by using priori knowledge according to the image;
establishing an environment grid map by adopting a vector mapping method, and acquiring an angle range formed by a robot as a center and an obstacle;
and generating an obstacle avoidance strategy by combining the distance from the robot to the obstacle, the initial position of the robot, the obstacle category and attribute information and the angle range.
According to the embodiment of the invention, the image of the obstacle is obtained, the prior knowledge is utilized to obtain the obstacle category and attribute information, the vector mapping method is adopted to establish the environment grid map, the angle range formed by the robot as the center and the obstacle is obtained, and finally the obstacle avoidance strategy is generated. By introducing prior knowledge, the obstacle avoidance method can more accurately obtain obstacle information in the obstacle avoidance process, and a more appropriate obstacle avoidance strategy is selected according to the type and attribute information of the obstacle, so that the obstacle avoidance efficiency is improved.
Example 1:
in a first aspect, as shown in fig. 1, an embodiment of the present invention provides a robot obstacle avoidance method based on prior knowledge, which specifically includes:
and S1, scanning the surrounding environment by using the laser range finder, determining the distance from the robot to the obstacle, and acquiring the initial position (x, y) of the robot.
The laser range finder comprises a transmitter and a receiver, wherein the transmitter irradiates a target with laser, and the receiver receives the backward light wave. Mechanical lidar comprises a mechanical mechanism with a mirror that rotates so that the beam covers a plane and distance information to a plane can be measured. The robot measures the distance to the obstacle by measuring the flight time of the laser in the obstacle avoidance process:
d＝0.5(c*t)
where c is the speed of light and t is the time interval from the emission of laser light to the reception of laser light by the laser rangefinder.
And S2, acquiring an image of the obstacle, and obtaining the obstacle category and attribute information by using the priori knowledge according to the image.
The obtaining of the image of the obstacle and the obtaining of the obstacle category information by using the priori knowledge according to the image specifically include:
acquiring an image of the obstacle;
and inputting the obstacle image into a pre-trained class recognition model to obtain the obstacle class information.
The construction process of the category identification model comprises the following steps:
acquiring a plurality of obstacle images and determining the corresponding category information of the plurality of obstacle images;
taking the multiple obstacle images as training samples, taking corresponding category information as output labels, and constructing a training database;
and based on the training data, performing model training by adopting a nearest neighbor classifier under the Euclidean distance to obtain the class identification model. The method specifically comprises the following steps:
(1) the obstacle picture sample set is as follows 4: the scale of 1 is divided into a training set and a test set, assuming that the training set has n images (100 x 100). Each pixel of the image is defined as a one-dimensional vector with a size of n, and for a 100 × 100 normal image, the n samples belong to c classes, that is, c obstacles, which can be abstracted as a 100 × 100-10000 high-dimensional vector. x is the number ofiFor the ith sample, which is a column vector of 10000 dimensions, the image is PCA-transformed to reduce the dimension k to 99 dimensions.
(2) Calculating a covariance matrix of the training sample:
the mean u of all samples is obtained by averaging each column of the high-dimensional vectors (n) corresponding to all images, which is also a 10000-dimensional column vector.
(3) Solving the covariance matrix StAnd the feature root λ and the feature vector w:
Stw＝λw
(4) obtaining a projection matrix W:
if the projection matrix is projected to a k-dimensional space, only the eigenvectors corresponding to the k largest eigenvalues need to be selected to form a projection matrix W:
W＝[w1,w2,…wk]
w is 10000 xk.
(5) Projecting each sample x in the training samples to a low-dimensional space through a transformation matrix W, wherein a projected vector y is as follows:
y＝WT(x-u)
and y is k 1.
(6) Projecting the test set samples to a low-dimensional space through a transformation matrix W:
y＝WT(x-u)
classifying and identifying by using a nearest neighbor classifier under an Euclidean distance to obtain a low-dimensional vector omega, and continuously solving the following objective function:
argmin‖Ωt-Ω‖
wherein omegatAnd (4) solving the minimum value for the low-dimensional vector of the c-type test sample by argmin, and finally determining the class information of the test sample.
In addition, the types of obstacles can be defined, such as tables, chairs and the like belonging to static obstacles, and vehicles, pedestrians and the like belonging to dynamic obstacles; the robot can directly avoid the obstacle facing the static obstacle without considering the change of the obstacle and other attributes, such as the length m and the width n of the size.
The obtaining of the image of the obstacle and the obtaining of the attribute information of the obstacle by using the priori knowledge according to the image specifically include:
the robot inquires the attribute information of the obstacles in a knowrob knowledge base, and the knowrob knowledge base writes a plurality of obstacles and corresponding attribute information thereof in advance through an entity description language.
Writing a plurality of obstacles and corresponding attribute information thereof in a knowrob knowledge base through an entity description language in advance refers to: the obstacle category and its attribute are written into the knowrob knowledge base platform by using the entity description language description, for example, if we train a cup picture as an obstacle, then a cup can be described as follows by using the entity description language description:
entity(Cup,[an,object,[type,cup],[shape,cylinder],[color,orange]])
entity represents entity, cup is entity name, type is its type, shape is its shape, color is its color.
The robot queries obstacle information in a knowrob knowledge base, such as a classic Query in knowrob for grabbing cups:
it retrieves a casting operation Tsk, source refers to the action object, which has a sub-operation to grab the source container, called Spt-Tsk. Then, the grabbing gestures G and PG and the corresponding grabbing Force of the arriving movement are inquired, and the movement parameters are displayed through the inquiry result Answer:
Answer
Tsk＝’Pouring_0’,Spt-Tsk＝’Grasping_6’,PG＝’Pose_788’,G＝’Pose_685’,
Force＝50Nm.
that is, PG is Pose _788, G is Pose _685, and Force is 50Nm
If the obstacle information is not inquired, the correlation between the two objects can be calculated according to wup similarity measurement, and the type with the maximum correlation is selected to select a proper obstacle avoidance strategy
Wherein S is C1And C2D (S) is the (lowest) depth of concept C in the body, dS(C) Is the (lowest) depth of concept C in the ontology when the path of super concept S through C.
The following table shows some examples of objects in knowrob and their similarities to cups, cookware and cutlery:
s3, environment information grid: and establishing an environment grid map by adopting a vector mapping method, and acquiring an angle range formed by the center of the robot and the barrier. The method specifically comprises the following steps:
in the obstacle avoidance process, a local map based on polar coordinate representation is created by adopting a vector mapping method aiming at the current surrounding environment of the mobile robot, and the local map uses a grid map representation method. The VFH algorithm (vector field histogram method) produces a polar histogram as shown in fig. 2.
The x-axis of the environment grid map represents an angle between the robot center and the obstacle, and the y-axis represents the probability p of the obstacle existing in the angular direction.
The angle range formed by the center of the robot and the obstacle is respectively determined according to whether the obstacle is positioned right in front of the robot or in front of the side of the robot.
And S4, generating an obstacle avoidance strategy by combining the distance from the robot to the obstacle, the initial position of the robot, the obstacle category and attribute information and the angle range.
As shown in fig. 3, when the obstacle is positioned right in front of the robot, the angle (α) at which the obstacle exists in front of the robot is estimated from the probability p1，α2) Then, the robot can know the self position (x, y) determined by the laser range finder, the obstacle category information (such as static obstacle) and the attribute information (such as size, length m, width n), and the distance d from the obstacle:
the length of the obstacle to the left of the robot is | d tan α1The length of the obstacle to the right of the robot is | d tan α2The robot can get the rough distribution of obstacles as follows:
the obstacle is located on an abscissa from x-d tan alpha1From | to x + | d tan α2The range of | the ordinate ranges from y + d to y + d + n. The robot can select left-turn straight line | d tan α1L + L (distance L between multiple lines to prevent collision, L can be freely set according to the size of the robot) and then the distance d + n is rotated to the right and the distance d tan alpha is rotated to the right1The distance of | + L, bypass the obstacle and turn left back to the original direction.
As shown in fig. 4, when the obstacle is positioned in front of the robot side, the angle is(α3，α4) If alpha is present0∈(α3，α4) So that d is α0<And L, obstacle avoidance is needed, and the robot can select the distance of the left-turn straight line L or the right-turn straight line L and then turn right to straight line or turn left to straight line to return to the original direction, so that the obstacle avoidance is completed.
In a second aspect, as shown in fig. 5, an embodiment of the present invention further provides a robot obstacle avoidance system based on prior knowledge, including:
the distance measuring module is used for scanning the surrounding environment by using the laser distance measuring device, determining the distance from the robot to the obstacle and acquiring the initial position of the robot;
the image module is used for acquiring an image of the obstacle and obtaining the type and attribute information of the obstacle by using the prior knowledge according to the image;
the map module is used for establishing an environment grid map by adopting a vector mapping method and acquiring an angle range formed by a robot as a center and an obstacle;
and the obstacle avoidance module is used for generating an obstacle avoidance strategy by combining the distance from the robot to the obstacle, the initial position of the robot, the obstacle category and attribute information and the angle range.
It can be understood that other contents of the robot obstacle avoidance system based on the prior knowledge provided in the embodiment of the present invention correspond to the robot obstacle avoidance method based on the prior knowledge one to one, and are not described again here.
In summary, compared with the prior art, the method has the following beneficial effects:
according to the embodiment of the invention, the image of the obstacle is obtained, the prior knowledge is utilized to obtain the obstacle category and attribute information, the vector mapping method is adopted to establish the environment grid map, the angle range formed by the robot as the center and the obstacle is obtained, and finally the obstacle avoidance strategy is generated. By introducing prior knowledge, the obstacle avoidance method can more accurately obtain obstacle information in the obstacle avoidance process, and a more appropriate obstacle avoidance strategy is selected according to the type and attribute information of the obstacle, so that the obstacle avoidance efficiency is improved.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (8)
1. A robot obstacle avoidance method based on prior knowledge is characterized by comprising the following steps:
determining the distance from the robot to an obstacle, and acquiring the initial position of the robot;
acquiring an image of an obstacle, and obtaining the type and attribute information of the obstacle by using priori knowledge according to the image;
establishing an environment grid map by adopting a vector mapping method, and acquiring an angle range formed by a robot as a center and an obstacle;
and generating an obstacle avoidance strategy by combining the distance from the robot to the obstacle, the initial position of the robot, the obstacle category and attribute information and the angle range.
2. The robot obstacle avoidance method of claim 1, wherein the determining the distance from the robot to the obstacle specifically comprises:
d＝0.5(c*t)
where c is the speed of light and t is the time interval from the emission of laser light to the reception of laser light by the laser rangefinder.
3. The robot obstacle avoidance method of claim 1, wherein the obtaining of the image of the obstacle and the obtaining of the obstacle category information according to the image by using the prior knowledge specifically comprises:
acquiring an image of the obstacle;
and inputting the obstacle image into a pre-trained class recognition model to obtain the obstacle class information.
4. A robot obstacle avoidance method according to claim 3, wherein the construction process of the category identification model includes:
acquiring a plurality of obstacle images and determining the corresponding category information of the plurality of obstacle images;
taking the multiple obstacle images as training samples, taking corresponding category information as output labels, and constructing a training database;
and based on the training data, performing model training by adopting a nearest neighbor classifier under the Euclidean distance to obtain the class identification model.
5. The robot obstacle avoidance method of claim 1, wherein the obtaining of the image of the obstacle and the obtaining of the obstacle attribute information according to the image by using the prior knowledge specifically comprises:
the robot inquires the attribute information of the obstacles in a knowrob knowledge base, and the knowrob knowledge base writes a plurality of obstacles and corresponding attribute information thereof in advance through an entity description language.
6. A robot obstacle avoidance method according to claim 1, wherein an x-axis in the environment grid map represents an angle between a robot center and an obstacle, and a y-axis represents a probability p of the obstacle existing in the angular direction.
7. A robot obstacle avoidance method according to claim 1, wherein the angular range between the center of the robot and the obstacle is determined according to whether the obstacle is located right in front of the robot or laterally in front of the robot.
8. The utility model provides a robot keeps away barrier system based on priori knowledge which characterized in that includes:
the distance measurement module is used for determining the distance from the robot to the obstacle and acquiring the initial position of the robot;
the image module is used for acquiring an image of the obstacle and obtaining the type and attribute information of the obstacle by using the prior knowledge according to the image;
the map module is used for establishing an environment grid map by adopting a vector mapping method and acquiring an angle range formed by a robot as a center and an obstacle;
and the obstacle avoidance module is used for generating an obstacle avoidance strategy by combining the distance from the robot to the obstacle, the initial position of the robot, the obstacle category and attribute information and the angle range.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011416636.6A CN112506196B (en) | 2020-12-07 | 2020-12-07 | Robot obstacle avoidance method and system based on priori knowledge |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011416636.6A CN112506196B (en) | 2020-12-07 | 2020-12-07 | Robot obstacle avoidance method and system based on priori knowledge |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112506196A true CN112506196A (en) | 2021-03-16 |
| CN112506196B CN112506196B (en) | 2022-09-20 |
Family
ID=74970725
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011416636.6A Active CN112506196B (en) | 2020-12-07 | 2020-12-07 | Robot obstacle avoidance method and system based on priori knowledge |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112506196B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113778119A (en) * | 2021-09-26 | 2021-12-10 | 江西理工大学 | Unmanned aerial vehicle control path optimization method |
| WO2024093989A1 (en) * | 2022-11-02 | 2024-05-10 | 神顶科技（南京）有限公司 | Obstacle avoidance method and system for robot |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002345703A (en) * | 2002-03-15 | 2002-12-03 | Matsushita Electric Ind Co Ltd | Movable working robot |
| CN102854880A (en) * | 2012-10-08 | 2013-01-02 | 中国矿业大学 | Robot whole-situation path planning method facing uncertain environment of mixed terrain and region |
| CN107976999A (en) * | 2017-11-21 | 2018-05-01 | 深圳普思英察科技有限公司 | A kind of mobile robot and its avoidance and paths planning method and system |
| US10043076B1 (en) * | 2016-08-29 | 2018-08-07 | PerceptIn, Inc. | Visual-inertial positional awareness for autonomous and non-autonomous tracking |
| CN109765901A (en) * | 2019-02-18 | 2019-05-17 | 华南理工大学 | Dynamic cost digital map navigation method based on line laser and binocular vision |
| CN110705545A (en) * | 2019-09-09 | 2020-01-17 | 云南电网有限责任公司曲靖供电局 | Obstacle identification method and device, storage medium and inspection robot |
| CN110968083A (en) * | 2018-09-30 | 2020-04-07 | 科沃斯机器人股份有限公司 | Method for constructing grid map, method, device and medium for avoiding obstacles |
| CN111976718A (en) * | 2020-07-13 | 2020-11-24 | 浙江大华汽车技术有限公司 | Automatic parking control method and system |
-
2020
- 2020-12-07 CN CN202011416636.6A patent/CN112506196B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002345703A (en) * | 2002-03-15 | 2002-12-03 | Matsushita Electric Ind Co Ltd | Movable working robot |
| CN102854880A (en) * | 2012-10-08 | 2013-01-02 | 中国矿业大学 | Robot whole-situation path planning method facing uncertain environment of mixed terrain and region |
| US10043076B1 (en) * | 2016-08-29 | 2018-08-07 | PerceptIn, Inc. | Visual-inertial positional awareness for autonomous and non-autonomous tracking |
| CN107976999A (en) * | 2017-11-21 | 2018-05-01 | 深圳普思英察科技有限公司 | A kind of mobile robot and its avoidance and paths planning method and system |
| CN110968083A (en) * | 2018-09-30 | 2020-04-07 | 科沃斯机器人股份有限公司 | Method for constructing grid map, method, device and medium for avoiding obstacles |
| CN109765901A (en) * | 2019-02-18 | 2019-05-17 | 华南理工大学 | Dynamic cost digital map navigation method based on line laser and binocular vision |
| CN110705545A (en) * | 2019-09-09 | 2020-01-17 | 云南电网有限责任公司曲靖供电局 | Obstacle identification method and device, storage medium and inspection robot |
| CN111976718A (en) * | 2020-07-13 | 2020-11-24 | 浙江大华汽车技术有限公司 | Automatic parking control method and system |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113778119A (en) * | 2021-09-26 | 2021-12-10 | 江西理工大学 | Unmanned aerial vehicle control path optimization method |
| CN113778119B (en) * | 2021-09-26 | 2022-10-21 | 江西理工大学 | Unmanned aerial vehicle control path optimization method |
| WO2024093989A1 (en) * | 2022-11-02 | 2024-05-10 | 神顶科技（南京）有限公司 | Obstacle avoidance method and system for robot |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112506196B (en) | 2022-09-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11720799B2 (en) | Object detection neural networks | |
| KR102109941B1 (en) | Method and Apparatus for Vehicle Detection Using Lidar Sensor and Camera | |
| CN112506196B (en) | Robot obstacle avoidance method and system based on priori knowledge | |
| CN110969064B (en) | Image detection method and device based on monocular vision and storage equipment | |
| JP7224682B1 (en) | 3D multiple object detection device and method for autonomous driving | |
| CN113706480A (en) | Point cloud 3D target detection method based on key point multi-scale feature fusion | |
| Zelener et al. | Cnn-based object segmentation in urban lidar with missing points | |
| CN109633686B (en) | Method and system for detecting ground obstacle based on laser radar | |
| Park et al. | I-LOAM: intensity enhanced LiDAR odometry and mapping | |
| Lee et al. | 3D reconstruction using a sparse laser scanner and a single camera for outdoor autonomous vehicle | |
| Hussain et al. | Drivable region estimation for self-driving vehicles using radar | |
| Chiang et al. | 3D point cloud classification for autonomous driving via dense-residual fusion network | |
| CN113255779B (en) | Multi-source perception data fusion identification method, system and computer readable storage medium | |
| US11941875B2 (en) | Processing perspective view range images using neural networks | |
| Carneiro et al. | Mapping road lanes using laser remission and deep neural networks | |
| CN109635692B (en) | Scene re-identification method based on ultrasonic sensor | |
| US20240151855A1 (en) | Lidar-based object tracking | |
| Hoy et al. | Bayesian tracking of multiple objects with vision and radar | |
| Zhang et al. | Depth Monocular Estimation with Attention-based Encoder-Decoder Network from Single Image | |
| Nakashima et al. | Learning geometric and photometric features from panoramic LiDAR scans for outdoor place categorization | |
| Shan et al. | Recognitions of collision avoidance for unmanned surface vessels with visible light camera using broad learning system | |
| Fan et al. | Three-dimensional real-time object perception based on a 16-beam lidar for an autonomous driving car | |
| CN112329678B (en) | Monocular pedestrian 3D positioning method based on information fusion | |
| US20220351412A1 (en) | Method and Device for Passive Ranging by Image Processing and Use of Three-Dimensional Models | |
| Weber et al. | Leveraging Saliency-Aware Gaze Heatmaps for Multiperspective Teaching of Unknown Objects |
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 |