CN115639825A - Obstacle avoidance method and system for robot - Google Patents
Obstacle avoidance method and system for robot Download PDFInfo
- Publication number
- CN115639825A CN115639825A CN202211364376.1A CN202211364376A CN115639825A CN 115639825 A CN115639825 A CN 115639825A CN 202211364376 A CN202211364376 A CN 202211364376A CN 115639825 A CN115639825 A CN 115639825A
- Authority
- CN
- China
- Prior art keywords
- obstacle
- obstacle avoidance
- type
- robot
- avoidance strategy
- 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 47
- 230000004888 barrier function Effects 0.000 claims abstract description 19
- 238000001514 detection method Methods 0.000 claims description 32
- 230000033001 locomotion Effects 0.000 claims description 14
- 230000000694 effects Effects 0.000 abstract description 5
- 230000009471 action Effects 0.000 description 7
- 238000010408 sweeping Methods 0.000 description 7
- 238000004140 cleaning Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 5
- 241000282326 Felis catus Species 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 210000003608 fece Anatomy 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 208000002177 Cataract Diseases 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 241001417527 Pempheridae Species 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 231100001261 hazardous Toxicity 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000002787 reinforcement Effects 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Images
Landscapes
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The application relates to the field of intelligent robots and discloses an obstacle avoidance method and system of a robot. The obstacle avoidance method comprises the following steps: the robot moves the position and detects the barrier, and identifies the type of the detected barrier or acquires the data of the barrier related to the type of the barrier; and determining an obstacle avoidance strategy according to the type of the obstacle or the data of the obstacle related to the type of the obstacle, and executing the determined obstacle avoidance strategy. According to the implementation mode of the application, different obstacle avoidance schemes are provided for different types of obstacles and the same type of obstacles in different environments, and the efficiency and the using effect of the whole system can be improved.
Description
Technical Field
The application relates to the field of intelligent robots, in particular to a robot and an obstacle avoidance technology thereof.
Background
At present, most intelligent robots (such as sweeping robots, unmanned automobiles and the like) have own obstacle avoidance methods, and the existing common obstacle avoidance methods scan obstacles through lasers (including single-line lasers, 2D lasers and 3D lasers) and directly avoid the obstacles after the obstacles are scanned. The slightly complex obstacle avoidance method generally incorporates various sensors, such as ultrasonic waves, millimeter wave radars and the like, which only changes the means for detecting obstacles, and avoids the obstacles directly after the obstacles are detected.
The existing obstacle avoidance scheme adopts a non-differential obstacle avoidance strategy for all obstacles and different external environments.
Disclosure of Invention
The application aims to provide an obstacle avoidance method and system for a robot, different obstacle avoidance schemes are provided for different types of obstacles and the same type of obstacles in different environments, and the efficiency and the using effect of the whole system can be improved.
The application discloses an obstacle avoidance method of a robot, which comprises the following steps:
the robot moves the position and detects an obstacle, and identifies the type of the detected obstacle or acquires data of the obstacle, which is related to the type of the obstacle;
and determining an obstacle avoidance strategy according to the type of the obstacle or the data of the obstacle related to the type of the obstacle, and executing the determined obstacle avoidance strategy.
In a preferred example, the robot performs the position movement and detects an obstacle, and identifying the type of the detected obstacle further includes: the robot moves the position and detects the barrier, and identifies the type of the detected barrier and the attribute of the current space position;
the determining an obstacle avoidance strategy according to the type of the obstacle, and executing the determined obstacle avoidance strategy further includes: and determining an obstacle avoidance strategy according to the type of the obstacle and the attribute of the current space position, and executing the determined obstacle avoidance strategy.
In a preferred embodiment, before the robot performs position movement and detects an obstacle and identifies the type of the detected obstacle, the method further includes: the robot prestores the corresponding relation between various obstacle types and obstacle avoidance strategies;
the determining an obstacle avoidance strategy according to the type of the obstacle, and executing the determined obstacle avoidance strategy further includes: and determining a corresponding obstacle avoidance strategy according to the obstacle type and the corresponding relation, and executing the determined obstacle avoidance strategy.
In a preferred embodiment, before the robot performs position movement and detects an obstacle and acquires data related to the type of the obstacle, the method further includes: taking data related to various obstacle types as input sample data, taking an obstacle avoidance strategy corresponding to each obstacle type as output sample data, and training a predetermined network model to obtain an obstacle avoidance strategy prediction model;
the robot moves the position and detects the obstacle, and data of the obstacle related to the type of the obstacle are obtained; determining an obstacle avoidance strategy according to the data of the obstacle related to the obstacle type, and executing the determined obstacle avoidance strategy, further comprising: and when the robot moves, detecting the obstacle, acquiring data of the obstacle related to the type of the obstacle, and inputting the acquired data into the obstacle avoidance strategy model to obtain the output obstacle avoidance strategy of the obstacle.
In a preferred embodiment, the different barrier types correspond to objects of different safety classes, respectively.
In a preferred example, the obstacle type includes one or more of a living object class, a movable non-living object class, a dangerous object class, a safety object class, wherein an obstacle of a type that cannot be identified belongs to the dangerous object class.
In a preferred embodiment, the obstacle avoidance strategy corresponding to the living object includes: waiting until the obstacle moves away from the target area; or avoiding the obstacle by a first preset obstacle avoidance distance;
the obstacle avoidance strategy corresponding to the movable non-living object class comprises the following steps: predicting the motion trail of the obstacle, and determining an avoidance path according to the prediction result so as to avoid the obstacle;
the obstacle avoidance strategy corresponding to the dangerous object class comprises the following steps: if the obstacle is completely detected, the obstacle is avoided by a second preset obstacle avoiding distance; if the obstacle is not completely detected, detecting the obstacle from different angles around by using a first preset detection distance, constructing an obstacle model, and avoiding the obstacle by using a second preset obstacle avoiding distance based on the obstacle model; if the obstacle model cannot be constructed or the obstacle cannot be avoided, turning back at a preset angle or the original path at a distance not less than a preset interval distance is kept;
the obstacle avoidance strategy corresponding to the safety article class comprises the following steps: if the obstacle is completely detected, avoiding the obstacle by a third preset obstacle avoidance distance, wherein the third preset obstacle avoidance distance is smaller than the second preset obstacle avoidance distance; if the obstacle is not completely detected, detecting the obstacle from different angles around by using a second preset detection distance, constructing an obstacle model, and avoiding the obstacle by using a third preset obstacle avoiding distance based on the obstacle model, wherein the second preset detection distance is smaller than the first preset detection distance.
In a preferred embodiment, the detecting obstacles from different angles around to construct the obstacle model further includes:
the robot detects the obstacles from different angles around the obstacles by using a binocular camera, and marks the coordinate information of the detected obstacles for each angle until the obstacle model is constructed.
The application also discloses a robot keep away barrier system includes:
the detection module is used for detecting the barrier when the robot moves;
the identification module is used for identifying the type of the detected obstacle, and the acquisition module is used for acquiring data of the obstacle, wherein the data are related to the type of the obstacle;
and the obstacle avoidance strategy module is used for determining an obstacle avoidance strategy according to the type of the obstacle or the data of the obstacle related to the type of the obstacle, and executing the determined obstacle avoidance strategy.
In a preferred embodiment, the obstacle avoidance system further includes a spatial position determination module, configured to determine a current spatial position attribute of the detected obstacle;
and the obstacle avoidance strategy module is also used for determining an obstacle avoidance strategy according to the type of the obstacle and the attribute of the current space position, and executing the determined obstacle avoidance strategy.
In the embodiment of the application, the types of the obstacles are considered, and obstacle avoidance schemes are automatically given for different types of obstacles, so that: 1. the obstacle avoidance can greatly improve the overall efficiency under the condition of ensuring safety, and the obstacle avoidance is not like the conventional obstacle avoidance, and can directly stop in case of emergency when the safety requirement is high to seriously influence the efficiency, and can cause some accidents when the safety is low; 2. the obstacle avoidance can be more precise, such as the closer obstacle avoidance effect can be achieved without safety accidents, the dangerous obstacles can be slightly far away instead of the traditional consistent treatment, and the efficiency and the using effect of the whole system can be improved to a certain extent. Furthermore, while considering the type of the obstacle, the influence of the environment is considered, and a scheme more suitable for the current environment is generated, for example, the overall speed and acceleration in the environment with good environment and no dangerous obstacle can be properly adjusted according to the environmental danger level, so as to achieve the maximum cleaning efficiency.
In addition, a variety of obstacle avoidance schemes are provided, and a most suitable obstacle avoidance scheme is configured for each obstacle type, and a most suitable obstacle avoidance and path planning method can be automatically selected according to various conditions, including but not limited to, bypassing obstacles, surrounding the obstacles for a circle, further detecting the obstacles, waiting for the obstacles to move away, predicting the obstacles, and the like: 1. the obstacle avoidance device can avoid obstacles, and can achieve better effect by adopting different avoidance strategies according to actual conditions; 2. the strategy for further detecting the obstacle enables a better strategy to be planned when the user encounters the obstacle next time, and a better obstacle avoidance route is planned.
The present specification describes a number of technical features distributed throughout the various technical aspects, and if all possible combinations of technical features (i.e. technical aspects) of the present specification are listed, the description is made excessively long. In order to avoid this problem, the respective technical features disclosed in the above-mentioned summary of the invention of the present application, the respective technical features disclosed in the following embodiments and examples, and the respective technical features disclosed in the drawings may be freely combined with each other to constitute various new technical solutions (all of which are considered to have been described in the present specification) unless such a combination of the technical features is technically impossible. For example, in one example, the feature a + B + C is disclosed, in another example, the feature a + B + D + E is disclosed, and the features C and D are equivalent technical means for the same purpose, and technically only one feature is used, but not simultaneously employed, and the feature E can be technically combined with the feature C, then the solution of a + B + C + D should not be considered as being described because the technology is not feasible, and the solution of a + B + C + E should be considered as being described.
Drawings
Fig. 1 is a schematic flow chart of an obstacle avoidance method of a robot according to a first embodiment of the present application;
FIG. 2 is a schematic illustration of marking obstacle information on a grid map at an initial angle according to the present application;
FIG. 3 is a schematic illustration of marking obstacle information on a grid map at another angle according to the present application;
fig. 4 is a schematic flow chart of an obstacle avoidance method of a robot according to a second embodiment of the present application;
fig. 5 is a schematic structural diagram of an obstacle avoidance system of a robot according to a third embodiment of the present application;
fig. 6 is a schematic structural diagram of an obstacle avoidance system of a robot according to a fourth embodiment of the present application;
fig. 7 is a main program frame diagram of an obstacle avoidance system of a robot according to an example of the present application;
fig. 8 is a flowchart of an obstacle avoidance process for a robot according to an example of the present application.
Detailed Description
In the following description, numerous technical details are set forth in order to provide a better understanding of the present application. However, it will be understood by those of ordinary skill in the art that the claimed embodiments may be practiced without these specific details and with various changes and modifications based on the following embodiments.
Interpretation of terms:
obstacle avoidance: avoiding the obstacle;
path planning: the pointer plans a path for the robot to walk for the robot, the automatic driving, the unmanned aerial vehicle and the like;
point cloud: a plurality of points with coordinate information in three-dimensional space (or other dimension space), wherein point cloud refers to a coordinate information set of points on an obstacle measured by laser or coordinate information of points on the obstacle measured or calculated by other components;
to make the objects, technical solutions and advantages of the present application more clear, embodiments of the present application will be described in further detail below with reference to the accompanying drawings.
A first embodiment of the present application relates to an obstacle avoidance method for a robot, a flow of which is shown in fig. 1, and the method includes:
and 102, determining an obstacle avoidance strategy according to the type of the obstacle, and executing the determined obstacle avoidance strategy.
In an embodiment, before the step 101, a pre-storing step "the robot pre-stores the corresponding relationship between the multiple obstacle types and the obstacle avoidance strategies", and based on the pre-storing step, the steps 101 and 102 may further be implemented as: the robot moves the position and detects the barrier, and identifies the type of the detected barrier; and determining a corresponding obstacle avoidance strategy according to the obstacle type and the corresponding relation, and executing the determined obstacle avoidance strategy. Further, for example, the robot is configured with an existing type recognition model (or classification model) trained by the target recognition network, and the robot can recognize the type of the obstacle according to the detected information of the obstacle based on the type recognition model. Wherein, the robot detects the information of the obstacle, for example, by vision, laser radar, laser sensor, mm wave radar, and the detected information of the obstacle includes but is not limited to one or any combination of vision, laser radar, laser sensor, mm wave radar, and the like.
Optionally, the plurality of obstacle types correspond to objects of different safety levels, respectively. For example, the plurality of obstacle types include one or more of a living object type (e.g., a person, a child, a dog, a cat, etc.), a movable non-living object type (e.g., another robot, such as a car, a smart home device, etc.), a dangerous object type (e.g., a cable, feces, etc.), a safety object type (e.g., a table and chair leg, etc.). The type of the obstacle that cannot be identified is also a type, such as but not limited to being attributed to the dangerous object class, or being attributed to an independent type, and the like.
Furthermore, an obstacle avoidance strategy corresponding to each obstacle type is configured in the robot in advance. Optionally, the obstacle avoidance strategy corresponding to the living object includes: waiting until the obstruction moves away from the target area (if the cataract moves away from the target area for some time); or avoid the obstacle by a first preset obstacle avoidance distance. Optionally, the obstacle avoidance strategy corresponding to the movable inactive object class includes: the motion trail of the obstacle (including the case that the motion is 0) is predicted, and an avoidance path is determined according to the prediction result so as to avoid the obstacle. Optionally, the obstacle avoidance strategy corresponding to the dangerous object class includes: if the obstacle is completely detected, the obstacle is avoided by a second preset obstacle avoidance distance; if the obstacle is not completely detected, detecting the obstacle from different angles around by using a first preset detection distance, constructing an obstacle model, and avoiding the obstacle by using a second preset obstacle avoiding distance based on the obstacle model; if the obstacle model cannot be constructed or the obstacle cannot be avoided, the road is folded back according to a preset angle or the original road by keeping the distance not less than a preset spacing distance. Optionally, the obstacle avoidance strategy corresponding to the safety article class includes: if the obstacle is completely detected, a third preset obstacle avoidance distance is used for avoiding the obstacle, and the third preset obstacle avoidance distance is smaller than the second preset obstacle avoidance distance; if the obstacle is not completely detected, detecting the obstacle from different angles around by using a second preset detection distance, constructing an obstacle model, and avoiding the obstacle by using a third preset obstacle avoiding distance based on the obstacle model, wherein the second preset detection distance is smaller than the first preset detection distance. The alternatives are merely exemplary, and not intended to limit the scope, and the obstacle avoidance strategy for each obstacle type of the present application may be implemented in other ways.
Alternatively, the aforementioned "detecting obstacles from different angles around, constructing the obstacle model" may be further implemented as: the robot detects the obstacles from different angles around the obstacle by using the binocular camera, and marks the coordinate information of the detected obstacles for each angle until the obstacle model is constructed. Specifically, as shown in fig. 2, blue represents the position of the cart, the map is a grid map, the information of the obstacles seen by the cart (representing the robot, the sweeper, etc.) through two eyes is marked on the map (as shown by green in the figure), the unseen area (the area blocked by the obstacles or the area where the whole obstacles are not detected, as shown by yellow in the figure), and the other areas are areas without obstacles seen by the cart through the camera. The trolley, seeing the unknown area behind the obstacle, will plan a route to the position where the yellow area can be seen for detection, so that the yellow area will be reduced and the outline next to the obstacle will also be displayed, as shown in fig. 3. And repeating the process continuously until the whole obstacle is seen clearly by the trolley, and constructing to obtain the obstacle model.
Optionally, step 101 and step 102 may be further implemented as: the robot moves the position and detects the barrier, and identifies the type of the detected barrier and the attribute of the current space position; and determining an obstacle avoidance strategy according to the type of the obstacle and the attribute of the current space position, and executing the determined obstacle avoidance strategy. Taking a sweeping robot as an example, the sweeping robot is further realized as follows: a, pre-storing the corresponding relation among a specific obstacle (such as a shoe of a safety article class), a spatial position attribute and an obstacle avoidance strategy of a specific obstacle type, and pre-configuring a first obstacle avoidance strategy for the shoe in a bedroom as a strategy for cleaning around the obstacle by adopting a first spacing distance; the second obstacle avoidance strategy for the shoes in the living room (or entrance room) is a turn-back strategy or a strategy of cleaning around obstacles by adopting a second spacing distance (which is greater than the first spacing distance); b, the robot moves in position and detects the obstacles, the type of the detected obstacles is identified as a safety object type, the obstacles are further identified as shoes, if a first obstacle avoidance strategy is determined according to the identified shoes, the attribute bedroom of the current spatial position and the corresponding relation, a strategy of sweeping the obstacles by adopting a first interval distance is adopted, and therefore the sweeping efficiency is higher; c, if a second obstacle avoidance strategy is determined according to the identified shoes, the attribute of the current spatial position of the living room (or entrance hall) and the corresponding relation, adopting a turn-back strategy or a strategy of cleaning around obstacles at a second interval distance to avoid touching the obstacles. Because the shoes are generally relatively clean in a relatively clean and safe environment, such as a bedroom, a close (first separation) strategy for cleaning around obstacles is preferred to achieve greater cleaning efficiency; in a relatively unsafe environment, such as a living room, where the shoes may be unclean, we would prefer to return after encountering an obstacle, trying to avoid encountering the obstacle. It is understood that the "shoes" and the corresponding bedroom and living room environments are only exemplary, and the obstacle avoidance policy determined according to the type of the obstacle and the attribute of the current spatial position is applicable to any obstacle with different safety attributes in different spatial environments.
In one embodiment, the obstacle avoidance method further includes: and determining the attribute of the current space position according to the furniture type or the combination of the furniture types of the current space. Taking the sweeping robot as an example, if a bed is in the current space, the current spatial position attribute is determined to be a bedroom, and if a combination of a sofa and a television is in the current space, the current spatial position attribute is determined to be a living room.
A second embodiment of the present application relates to an obstacle avoidance method for a robot, a flow of which is shown in fig. 4, the method including:
In one embodiment, the step 402 further includes a model training step of using data related to multiple obstacle types as input sample data, using an obstacle avoidance strategy corresponding to each obstacle type as output sample data, training a predetermined network model to obtain an obstacle avoidance strategy prediction model, and based on the obstacle avoidance strategy prediction model obtained by training, the steps 401 and 402 are further implemented as follows: and when the robot moves, detecting the obstacle, acquiring data of the obstacle related to the type of the obstacle, and inputting the acquired data into the obstacle avoidance strategy model to obtain an output obstacle avoidance strategy of the obstacle.
Optionally, the plurality of obstacle types correspond to objects of different safety levels, respectively. For example, the plurality of obstacle types include one or more of a living object type (e.g., a person, child, dog, cat, etc.), a movable non-living object type (e.g., other robots, such as a car, smart home device, etc.), a hazardous object type (e.g., cable, feces, etc.), a secure object type (e.g., table and chair legs, etc.). The type of the obstacle that cannot be identified is also a type, such as but not limited to being attributed to the dangerous object class, or being attributed to an independent type, etc.
Furthermore, an obstacle avoidance strategy corresponding to each obstacle type is configured in the robot in advance. Optionally, the obstacle avoidance strategy corresponding to the living object includes: waiting until the obstruction moves away from the target area (which may be the case if the cataract is removed for a certain amount of time); or avoid the obstacle by the first preset obstacle avoidance distance. Optionally, the obstacle avoidance strategy corresponding to the movable inactive object class includes: predicting the motion track of the obstacle (including the case that the motion is 0), and determining an avoidance path to avoid the obstacle according to the prediction result. Optionally, the obstacle avoidance strategy corresponding to the dangerous object class includes: if the obstacle is completely detected, the obstacle is avoided by a second preset obstacle avoidance distance; if the obstacle is not completely detected, detecting the obstacle from different angles around by using a first preset detection distance, constructing an obstacle model, and avoiding the obstacle by using a second preset obstacle avoiding distance based on the obstacle model; if the obstacle model cannot be constructed or the obstacle cannot be avoided, the vehicle is folded back at a preset angle or the road is folded back at a distance not less than a preset interval distance. Optionally, the obstacle avoidance policy corresponding to the class of security objects includes: if the obstacle is completely detected, a third preset obstacle avoidance distance is used for avoiding the obstacle, and the third preset obstacle avoidance distance is smaller than the second preset obstacle avoidance distance; if the obstacle is not completely detected, detecting the obstacle from different angles around by using a second preset detection distance, constructing an obstacle model, and avoiding the obstacle by using a third preset obstacle avoiding distance based on the obstacle model, wherein the second preset detection distance is smaller than the first preset detection distance. The alternatives are only exemplary and not intended to limit the scope, and the obstacle avoidance strategy for each obstacle type of the present application may be in other manners. Further, the "detecting obstacles from different angles around and constructing the obstacle model" may be further implemented as: the robot detects the obstacles from different angles around the obstacle by using the binocular cameras, and marks the detected coordinate information of the obstacles for each angle until the obstacle model is constructed.
A third embodiment of the present application relates to an obstacle avoidance system of a robot, which has a structure shown in fig. 5, and includes a detection module, an identification module, and an obstacle avoidance policy module.
The detection module is used for detecting an obstacle when the robot moves, the identification module is used for identifying the type of the detected obstacle, and the obstacle avoidance strategy module is used for determining an obstacle avoidance strategy according to the type of the obstacle and executing the determined obstacle avoidance strategy.
In one embodiment, the obstacle avoidance system of the robot further comprises a storage module, which is used for storing the corresponding relation between the types of the obstacles and the obstacle avoidance strategies in advance. In addition, in this embodiment, the detection module is configured to move the robot and detect an obstacle, the identification module is configured to identify a type of the detected obstacle, and the obstacle avoidance policy module determines a corresponding obstacle avoidance policy according to the type of the obstacle and the correspondence, and executes the determined obstacle avoidance policy.
Optionally, the plurality of obstacle types correspond to objects of different safety levels, respectively. For example, the plurality of obstacle types include one or more of a living object type (e.g., a person, a child, a dog, a cat, etc.), a movable non-living object type (e.g., another robot, such as a car, a smart home device, etc.), a dangerous object type (e.g., a cable, feces, etc.), a safety object type (e.g., a table and chair leg, etc.). The type of the obstacle that cannot be identified is also a type, such as but not limited to being attributed to the dangerous object class, or being attributed to an independent type, etc.
Further, the obstacle avoidance strategy module is pre-configured with obstacle avoidance strategies corresponding to each obstacle type. Optionally, the configuration of the obstacle avoidance policy module with the obstacle avoidance policy corresponding to the living object includes: waiting until the obstacle moves away from the target area; or avoid the obstacle by a first preset obstacle avoidance distance. Optionally, the obstacle avoidance policy module configured with an obstacle avoidance policy corresponding to the movable inactive object class includes: and predicting the motion trail of the obstacle, and determining an avoidance path according to the prediction result so as to avoid the obstacle. Optionally, the obstacle avoidance policy module configured with an obstacle avoidance policy corresponding to the dangerous object class includes: if the obstacle is completely detected, the obstacle is avoided by a second preset obstacle avoidance distance; if the obstacles are not completely detected, detecting the obstacles from different angles around by a first preset detection distance, constructing an obstacle model, and planning an avoidance path based on the obstacle model; if the obstacle can not be completely detected or can not be avoided, the obstacle is folded back according to a preset angle or the original road at a distance not less than a preset interval distance. Optionally, the configuration of the obstacle avoidance policy corresponding to the security object class in the obstacle avoidance policy module includes: if the obstacle is completely detected, a third preset obstacle avoidance distance is used for avoiding the obstacle, and the third preset obstacle avoidance distance is smaller than the second preset obstacle avoidance distance; if the obstacle is not completely detected, detecting the obstacle from different angles around by using a second preset detection distance, constructing an obstacle model, and planning an avoidance path based on the obstacle model, wherein the second preset detection distance is smaller than the first preset detection distance. The above-mentioned various alternative strategies are only exemplary, and are not intended to limit the scope, and the obstacle avoidance strategy of each obstacle type of the present application is also other ways.
Optionally, the obstacle avoidance system of the robot further includes a spatial position determination module, the spatial position determination module is configured to determine a detected attribute of a current spatial position of the obstacle, and the obstacle avoidance policy module is further configured to determine an obstacle avoidance policy according to the type of the obstacle and the attribute of the current spatial position, and execute the determined obstacle avoidance policy.
The first embodiment is a method embodiment corresponding to the present embodiment, and the technical details in the first embodiment may be applied to the present embodiment, and the technical details in the present embodiment may also be applied to the first embodiment.
A fourth embodiment of the present application relates to an obstacle avoidance system of a robot, which has a structure shown in fig. 6, and includes a detection module, an acquisition module, and an obstacle avoidance policy module. The detection module is used for detecting the barrier when the robot moves; the acquisition module is used for acquiring data of the obstacle related to the type of the obstacle; the obstacle avoidance strategy module is used for determining an obstacle avoidance strategy according to the data of the obstacle related to the obstacle type and executing the determined obstacle avoidance strategy.
In one embodiment, the obstacle avoidance strategy module further comprises an obstacle avoidance strategy prediction model, and the obstacle avoidance strategy prediction model is obtained by training a predetermined network model by taking data related to multiple obstacle types as input sample data and taking an obstacle avoidance strategy corresponding to each obstacle type as output sample data; based on the obstacle avoidance strategy prediction model, the obstacle avoidance strategy module is also used for detecting an obstacle when the robot moves, acquiring data of the obstacle related to the type of the obstacle, inputting the acquired data into the obstacle avoidance strategy model, outputting to determine an obstacle avoidance strategy of the obstacle, and executing the determined obstacle avoidance strategy. The data related to the obstacle type includes, but is not limited to, one or any combination of vision, laser, mm wave radar and the like.
Optionally, the plurality of obstacle types correspond to objects of different safety levels, respectively. For example, the plurality of obstacle types include one or more of a living object type (e.g., a person, a child, a dog, a cat, etc.), a movable non-living object type (e.g., another robot, such as a car, a smart home device, etc.), a dangerous object type (e.g., a cable, feces, etc.), a safety object type (e.g., a table and chair leg, etc.). The type of the obstacle that cannot be identified is also a type, such as but not limited to being attributed to the dangerous object class, or being attributed to an independent type, etc.
Further, the obstacle avoidance strategy module is pre-configured with obstacle avoidance strategies corresponding to each obstacle type. Optionally, the configuration of the obstacle avoidance policy module with the obstacle avoidance policy corresponding to the living object includes: waiting until the obstacle moves away from the target area; or avoid the obstacle by the first preset obstacle avoidance distance. Optionally, the obstacle avoidance policy module configured with an obstacle avoidance policy corresponding to the movable inactive object class includes: and predicting the motion trail of the obstacle, and determining an avoidance path according to the prediction result so as to avoid the obstacle. Optionally, the obstacle avoidance policy module configured with an obstacle avoidance policy corresponding to the dangerous object class includes: if the obstacle is completely detected, the obstacle is avoided by a second preset obstacle avoidance distance; if the obstacle is not completely detected, detecting the obstacle from different angles around by using a first preset detection distance, constructing an obstacle model, and planning an avoidance path based on the obstacle model; if the obstacle can not be completely detected or can not be avoided, the obstacle is folded back according to a preset angle or the original road can not be kept less than a preset spacing distance. Optionally, the configuration of the obstacle avoidance policy corresponding to the security object class in the obstacle avoidance policy module includes: if the obstacle is completely detected, a third preset obstacle avoiding distance is used for avoiding the obstacle, and the third preset obstacle avoiding distance is smaller than the second preset obstacle avoiding distance; if the obstacle is not completely detected, detecting the obstacle from different angles around by using a second preset detection distance, constructing an obstacle model, and planning an avoidance path based on the obstacle model, wherein the second preset detection distance is smaller than the first preset detection distance. The above-mentioned various alternative strategies are only exemplary, and are not intended to limit the scope, and the obstacle avoidance strategy of each obstacle type of the present application is also other ways.
Optionally, the obstacle avoidance system of the robot further includes a spatial position determination module, the spatial position determination module is configured to determine a detected attribute of a current spatial position of the obstacle, and the obstacle avoidance policy module is further configured to determine an obstacle avoidance policy according to the type of the obstacle and the attribute of the current spatial position, and execute the determined obstacle avoidance policy.
The second embodiment is a method embodiment corresponding to the present embodiment, and the technical details in the second embodiment may be applied to the present embodiment, and the technical details in the present embodiment may also be applied to the second embodiment.
It should be noted that the robot in each embodiment of the present application may be a mobile device in an intelligent home, such as a sweeping robot, or may also be a vehicle, such as an airplane and an automobile. Furthermore, the present application is not limited to the listed devices or vehicles, and may be any obstacle avoidance mobile product.
In order to better understand the technical solution of the present application, the following description is given with reference to an example, in which the listed details are mainly for the sake of understanding and are not intended to limit the scope of the present application. The main program framework of this example is shown in FIG. 7, and the flow chart is shown in FIG. 8. The data of each sensor is processed and fused in a mode including but not limited to integrating the data according to a certain method to calculate one or more results. The comprehensive judgment or determination of the obstacle avoidance strategy can be implemented by various methods, for example, one of the methods is implemented by an AI network (including but not limited to various networks such as machine learning, deep learning, reinforcement learning, and the like): input layer node is V = [ V = 1 ,v 2 ,…,v n ] T Wherein v is i Corresponding input conditions including but not limited to the size of the obstacle, the condition of the point cloud, the integrity of the detected obstacle and the like; the intermediate layer establishes a corresponding layer number and a node number according to actual needs; and the output layer outputs a corresponding obstacle avoidance strategy, including but not limited to bypassing the obstacle, surrounding the obstacle for a circle, further detecting the obstacle, waiting for the obstacle to move away, predicting the obstacle and the like. And avoiding the obstacle according to the output result of the AI network. Sources of acquisition information for the types of obstacles include, but are not limited to, vision, laser, mm-wave radar, and the like, and some combination thereof. The generated path is planned in real time (or the generated path is modified).
It should be noted that, as those skilled in the art will understand, the implementation functions of the modules shown in the above embodiments of the obstacle avoidance system of the robot may be understood by referring to the related description of the obstacle avoidance method of the obstacle avoidance system of the robot. The functions of the modules shown in the embodiments of the obstacle avoidance system for a robot described above may be implemented by a program (executable instructions) running on a processor, or may be implemented by specific logic circuits. If the obstacle avoidance system of the robot in the embodiment of the present application is implemented in the form of a software functional module and sold or used as an independent product, the obstacle avoidance system may also be stored in a computer-readable storage medium. Based on such understanding, the technical solutions of the embodiments of the present application or portions thereof that contribute to the prior art may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the methods described in the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read Only Memory (ROM), a magnetic disk, or an optical disk. Thus, embodiments of the present application are not limited to any specific combination of hardware and software.
Accordingly, the present application also provides a computer-readable storage medium, in which computer-executable instructions are stored, and when the computer-executable instructions are executed by a processor, the computer-executable instructions implement the method embodiments of the present application. Computer-readable storage media, including both permanent and non-permanent, removable and non-removable media, may implement the information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable storage medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
In addition, the embodiment of the application also provides an obstacle avoidance system of the robot, which comprises a memory for storing computer executable instructions and a processor; the processor is configured to implement the steps of the method embodiments described above when executing the computer-executable instructions in the memory. The Processor may be a Central Processing Unit (CPU), other general-purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), or the like. The memory may be a read-only memory (ROM), a Random Access Memory (RAM), a Flash memory (Flash), a hard disk or a solid state disk. The steps of the method disclosed in the embodiments of the present invention may be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules in the processor.
It is noted that, in the present patent application, relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, 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, the use of the verb "comprise a" to define an element does not exclude the presence of another, same element in a process, method, article, or apparatus that comprises the element. In the present patent application, if it is mentioned that a certain action is executed according to a certain element, it means that the action is executed according to at least the element, and two cases are included: performing the action based only on the element, and performing the action based on the element and other elements. The expression of a plurality of, a plurality of and the like includes 2, 2 and more than 2, more than 2 and more than 2.
All documents mentioned in this application are to be considered as being incorporated in their entirety into the disclosure of this application so as to be subject to modification as necessary. It should be understood that the above description is only a preferred embodiment of the present disclosure, and is not intended to limit the scope of the present disclosure. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of one or more embodiments of the present disclosure should be included in the scope of protection of one or more embodiments of the present disclosure.
Claims (10)
1. An obstacle avoidance method of a robot is characterized by comprising the following steps:
the robot moves the position and detects the obstacle, and identifies the type of the detected obstacle or acquires data of the obstacle related to the type of the obstacle;
and determining an obstacle avoidance strategy according to the type of the obstacle or the data of the obstacle related to the type of the obstacle, and executing the determined obstacle avoidance strategy.
2. An obstacle avoidance method for a robot according to claim 1, wherein the robot performs a position movement and detects an obstacle, and the identifying the type of the detected obstacle further comprises: the robot moves the position and detects the barrier, and identifies the type of the detected barrier and the attribute of the current space position of the barrier;
the determining an obstacle avoidance strategy according to the type of the obstacle, and executing the determined obstacle avoidance strategy further includes: and determining an obstacle avoidance strategy according to the type of the obstacle and the attribute of the current space position, and executing the determined obstacle avoidance strategy.
3. An obstacle avoidance method for a robot according to claim 1, wherein the robot performs a position movement and detects an obstacle, and further comprises, before identifying an obstacle type of the detected obstacle: the robot prestores the corresponding relation between various barrier types and the obstacle avoidance strategy;
the determining an obstacle avoidance strategy according to the type of the obstacle and executing the determined obstacle avoidance strategy further includes: and determining a corresponding obstacle avoidance strategy according to the obstacle type and the corresponding relation, and executing the determined obstacle avoidance strategy.
4. An obstacle avoidance method for a robot as claimed in claim 1, wherein, before the robot performs position movement and detects an obstacle and acquires data related to the type of the obstacle, the method further comprises: taking data related to various obstacle types as input sample data, taking an obstacle avoidance strategy corresponding to each obstacle type as output sample data, and training a predetermined network model to obtain an obstacle avoidance strategy prediction model;
the robot moves the position and detects the obstacle, and data of the obstacle related to the type of the obstacle are obtained; determining an obstacle avoidance strategy according to the data of the obstacle related to the obstacle type, and executing the determined obstacle avoidance strategy, further comprising: and when the robot moves, detecting the obstacle, acquiring data of the obstacle related to the type of the obstacle, and inputting the acquired data into the obstacle avoidance strategy model to obtain an output obstacle avoidance strategy of the obstacle.
5. An obstacle avoidance method for a robot according to claim 3 or 4, wherein different obstacle types correspond to objects of different safety levels, respectively.
6. A method as claimed in claim 5, wherein said obstacle types include one or more of living objects, movable non-living objects, dangerous objects, and safe objects, and wherein an obstacle of a type that cannot be identified belongs to said dangerous objects.
7. An obstacle avoidance method for a robot as claimed in claim 6, wherein the obstacle avoidance strategy corresponding to the living species comprises: waiting until the obstacle moves away from the target area; or avoiding the obstacle by a first preset obstacle avoidance distance;
the obstacle avoidance strategy corresponding to the movable inactive object class comprises the following steps: predicting the motion track of the obstacle, and determining an avoidance path according to the prediction result so as to avoid the obstacle;
the obstacle avoidance strategy corresponding to the dangerous object class comprises the following steps: if the obstacle is completely detected, the obstacle is avoided by a second preset obstacle avoidance distance; if the obstacle is not completely detected, detecting the obstacle from different angles around by using a first preset detection distance, constructing an obstacle model, and avoiding the obstacle by using a second preset obstacle avoiding distance based on the obstacle model; if the obstacle model cannot be constructed or the obstacle cannot be avoided, turning back at a preset angle or the original path at a distance not less than a preset interval distance;
the obstacle avoidance strategy corresponding to the safety object class comprises the following steps: if the obstacle is completely detected, avoiding the obstacle by a third preset obstacle avoidance distance, wherein the third preset obstacle avoidance distance is smaller than the second preset obstacle avoidance distance; if the obstacle is not completely detected, detecting the obstacle from different angles around by using a second preset detection distance, constructing an obstacle model, and avoiding the obstacle by using a third preset obstacle avoiding distance based on the obstacle model, wherein the second preset detection distance is smaller than the first preset detection distance.
8. An obstacle avoidance method for a robot according to claim 7, wherein the detecting obstacles from different angles around the robot to construct an obstacle model, further comprises:
the robot detects obstacles from different angles around the obstacle by using binocular cameras, and marks the detected coordinate information of the obstacles at each angle until the obstacle model is constructed.
9. An obstacle avoidance system for a robot, the obstacle avoidance system comprising:
the detection module is used for detecting the barrier when the robot moves;
the identification module is used for identifying the type of the detected obstacle of the obstacle, and the acquisition module is used for acquiring data of the obstacle related to the type of the obstacle;
and the obstacle avoidance strategy module is used for determining an obstacle avoidance strategy according to the type of the obstacle or the data of the obstacle related to the type of the obstacle, and executing the determined obstacle avoidance strategy.
10. An obstacle avoidance system according to claim 9, further comprising a spatial position determination module for determining a spatial position attribute of the detected obstacle at which the robot is currently located;
and the obstacle avoidance strategy module is also used for determining an obstacle avoidance strategy according to the type of the obstacle and the attribute of the current space position, and executing the determined obstacle avoidance strategy.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211364376.1A CN115639825A (en) | 2022-11-02 | 2022-11-02 | Obstacle avoidance method and system for robot |
| PCT/CN2023/128357 WO2024093989A1 (en) | 2022-11-02 | 2023-10-31 | Obstacle avoidance method and system for robot |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211364376.1A CN115639825A (en) | 2022-11-02 | 2022-11-02 | Obstacle avoidance method and system for robot |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115639825A true CN115639825A (en) | 2023-01-24 |
Family
ID=84947478
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211364376.1A Pending CN115639825A (en) | 2022-11-02 | 2022-11-02 | Obstacle avoidance method and system for robot |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN115639825A (en) |
| WO (1) | WO2024093989A1 (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109508019A (en) * | 2018-12-28 | 2019-03-22 | 北京猎户星空科技有限公司 | A kind of motion planning and robot control method, apparatus and storage medium |
| CN110502014A (en) * | 2019-08-22 | 2019-11-26 | 深圳乐动机器人有限公司 | A kind of method and robot of robot obstacle-avoiding |
| CN111930127A (en) * | 2020-09-02 | 2020-11-13 | 广州赛特智能科技有限公司 | Robot obstacle identification and obstacle avoidance method |
-
2022
- 2022-11-02 CN CN202211364376.1A patent/CN115639825A/en active Pending
-
2023
- 2023-10-31 WO PCT/CN2023/128357 patent/WO2024093989A1/en unknown
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109508019A (en) * | 2018-12-28 | 2019-03-22 | 北京猎户星空科技有限公司 | A kind of motion planning and robot control method, apparatus and storage medium |
| CN110502014A (en) * | 2019-08-22 | 2019-11-26 | 深圳乐动机器人有限公司 | A kind of method and robot of robot obstacle-avoiding |
| CN111930127A (en) * | 2020-09-02 | 2020-11-13 | 广州赛特智能科技有限公司 | Robot obstacle identification and obstacle avoidance method |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2024093989A1 (en) | 2024-05-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110632921B (en) | Robot path planning method and device, electronic equipment and storage medium | |
| US11709058B2 (en) | Path planning method and device and mobile device | |
| WO2019128933A1 (en) | Map construction and navigation method, and device and system | |
| US11487297B2 (en) | Method of travel control, device and storage medium | |
| KR102577785B1 (en) | Cleaning robot and Method of performing task thereof | |
| CN112526993B (en) | Grid map updating method, device, robot and storage medium | |
| JP4256812B2 (en) | Obstacle avoidance method for moving body and moving body | |
| EP3719538B1 (en) | Method and device for drawing region outline and computer readable storage medium | |
| CN108984741B (en) | Map generation method and device, robot and computer-readable storage medium | |
| CN110562221A (en) | automobile and brake control method, device and system thereof | |
| CN115494834A (en) | Robot path planning method and device and robot | |
| CN114911228A (en) | Robot path planning method and device and robot | |
| CN115639825A (en) | Obstacle avoidance method and system for robot | |
| CN112578798B (en) | Robot map acquisition method and device, processor and electronic device | |
| CN116310743A (en) | Method, device, mobile device and storage medium for determining expansion strategy | |
| CN112667924B (en) | Robot map acquisition method and device, processor and electronic device | |
| Vatavu et al. | Environment perception using dynamic polylines and particle based occupancy grids | |
| KR20210065837A (en) | Apparatus and method for generating occupancy grid map | |
| Qin et al. | Road detection and mapping using 3d rolling window | |
| García et al. | Context aided fusion procedure for road safety application | |
| CN116185002A (en) | Obstacle avoidance method and device based on robot multi-vision module and robot | |
| US20220289237A1 (en) | Map-free generic obstacle detection for collision avoidance systems | |
| WO2024009756A1 (en) | Object identification device and object identification method | |
| CN115562305A (en) | Self-walking equipment movement control method and device and self-walking equipment | |
| CN117367447A (en) | Navigation method, navigation device, electronic equipment and storage medium |
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 |