CN114136318A - Intelligent navigation method and device for machine - Google Patents
Intelligent navigation method and device for machine Download PDFInfo
- Publication number
- CN114136318A CN114136318A CN202010813288.XA CN202010813288A CN114136318A CN 114136318 A CN114136318 A CN 114136318A CN 202010813288 A CN202010813288 A CN 202010813288A CN 114136318 A CN114136318 A CN 114136318A
- Authority
- CN
- China
- Prior art keywords
- obstacle
- newly added
- navigation path
- dynamic
- preset
- 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 45
- 230000003068 static effect Effects 0.000 claims abstract description 31
- 238000012544 monitoring process Methods 0.000 claims description 28
- 238000001514 detection method Methods 0.000 claims description 19
- 238000005516 engineering process Methods 0.000 claims description 18
- 230000008859 change Effects 0.000 claims description 13
- 238000006073 displacement reaction Methods 0.000 claims description 8
- 238000010801 machine learning Methods 0.000 claims description 7
- 239000000284 extract Substances 0.000 claims description 2
- 230000000717 retained effect Effects 0.000 claims 2
- 238000011897 real-time detection Methods 0.000 claims 1
- 230000007774 longterm Effects 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 9
- 230000008569 process Effects 0.000 description 9
- 230000004888 barrier function Effects 0.000 description 8
- 230000002093 peripheral effect Effects 0.000 description 7
- 238000012545 processing Methods 0.000 description 7
- 230000000903 blocking effect Effects 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000010408 sweeping Methods 0.000 description 2
- 230000001960 triggered effect Effects 0.000 description 2
- 208000002177 Cataract Diseases 0.000 description 1
- 241001417527 Pempheridae Species 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002085 persistent effect Effects 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/20—Instruments for performing navigational calculations
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0214—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory in accordance with safety or protection criteria, e.g. avoiding hazardous areas
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Aviation & Aerospace Engineering (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
本发明公开了一种机器智能导航方法和装置，属于智能导航技术领域。所述方法包括：当预设导航路径上出现新增障碍物时，判断所述新增障碍物为动态障碍物或静态障碍物；若为动态障碍物，则检测所述新增障碍物的运动趋势，并根据所述运动趋势重新规划导航路径；若为静态障碍物，则根据所述新增障碍物的位置信息，重新规划导航路径。采用本发明，可以避免出现自移动设备长时无效等待和路径重复规划的情况，保证了智能导航的流畅性、有效性和安全性。
The invention discloses a machine intelligent navigation method and device, belonging to the technical field of intelligent navigation. The method includes: when a newly added obstacle appears on a preset navigation path, judging that the newly added obstacle is a dynamic obstacle or a static obstacle; if it is a dynamic obstacle, detecting the movement of the newly added obstacle trend, and re-plan the navigation path according to the motion trend; if it is a static obstacle, re-plan the navigation path according to the position information of the newly added obstacle. By adopting the present invention, the situation of long-term invalid waiting and repeated path planning from the mobile device can be avoided, and the fluency, effectiveness and safety of intelligent navigation are ensured.
Description
Technical Field
The invention relates to the technical field of intelligent navigation, in particular to a machine intelligent navigation method and device.
Background
The intelligent navigation technology is a novel technology based on artificial intelligence technology and various algorithms to realize path planning and guidance, can be applied to various scenes such as families, markets, garages, outdoors and the like at present, helps a machine to determine an actual optimal route to a destination point in real time, and performs corresponding guidance according to the actual optimal route.
Generally, the smart navigation technology can be loaded on a self-moving device (i.e. a smart device with a self-moving function) in the form of a functional module, and the self-moving device can implement smart navigation through the functional module. In particular, the user may input the destination point on the self-moving device, or the self-moving device may autonomously determine the destination point based on a preset configuration. And then, the self-moving equipment can firstly determine the map coordinates corresponding to the destination point through a built-in electronic map, and then plan all routes reaching the map coordinates according to all passable paths recorded in the electronic map. Furthermore, the self-moving device can determine an actual optimal route according to a user instruction or a preset routing rule, map the actual optimal route into an actual scene, and then guide the user or automatically go to a destination.
In the process of implementing the technical scheme of the application, the inventor finds that the prior art at least has the following problems:
in the process of moving to a destination point from a mobile device, if newly added obstacles (such as pedestrians, vehicles, other temporary objects and the like) block a given route, the mobile device generally brakes to wait for the obstacles to disappear or re-plan the route to bypass. However, if the brake waiting is selected, when the newly added obstacle stays for a long time, the self-moving device will wait for a long time, which greatly reduces the navigation efficiency; if the newly-added obstacle is a moving obstacle, the newly-added obstacle may be covered by the obstacle at the next moment, so that the route is continuously re-planned by the self-moving device, which greatly affects navigation fluency.
Disclosure of Invention
In order to solve the problems in the prior art, the embodiment of the invention provides a machine intelligent navigation method and device. The technical scheme is as follows:
in a first aspect, a method for intelligent navigation of a machine is provided, the method comprising:
when a newly added obstacle appears on a preset navigation path, judging that the newly added obstacle is a dynamic obstacle or a static obstacle;
if the obstacle is a dynamic obstacle, detecting the motion trend of the newly added obstacle, and re-planning a navigation path according to the motion trend;
and if the obstacle is a static obstacle, re-planning a navigation path according to the position information of the newly added obstacle.
In a second aspect, a machine intelligent navigation device is provided, the device comprising:
the obstacle detection module is used for judging that the newly added obstacle is a dynamic obstacle or a static obstacle when the newly added obstacle appears on a preset navigation path;
the route planning module is used for detecting the movement trend of the newly added obstacle and replanning the navigation route according to the movement trend if the newly added obstacle is a dynamic obstacle; and if the obstacle is a static obstacle, re-planning a navigation path according to the position information of the newly added obstacle.
In a third aspect, a self-moving device is provided, which comprises a processor and a memory, wherein at least one instruction, at least one program, a set of codes, or a set of instructions is stored in the memory, and the at least one instruction, the at least one program, the set of codes, or the set of instructions is loaded and executed by the processor to implement the machine intelligent navigation method according to the first aspect.
In a fourth aspect, a computer-readable storage medium is provided, in which at least one instruction, at least one program, a set of codes, or a set of instructions is stored, which is loaded and executed by a processor to implement the machine intelligent navigation method according to the first aspect.
The technical scheme provided by the embodiment of the invention has the following beneficial effects:
in the embodiment of the invention, when a newly added obstacle appears on the preset navigation path, the newly added obstacle is judged to be a dynamic obstacle or a static obstacle; if the obstacle is a dynamic obstacle, detecting the movement trend of the newly added obstacle, and re-planning the navigation path according to the movement trend; and if the obstacle is a static obstacle, replanning the navigation path according to the position information of the newly added obstacle. Therefore, when the self-moving equipment carries out intelligent navigation, for newly added obstacles, different path planning schemes can be respectively selected according to two conditions of dynamic obstacles and static obstacles, and different path planning schemes can be adaptively selected according to dynamic obstacles with different movement trends, so that the situations of long-time invalid waiting and repeated path planning of the self-moving equipment can be avoided, and the fluency, the effectiveness and the safety of the intelligent navigation are ensured.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without creative efforts.
FIG. 1 is a flow chart of a method for intelligent navigation of a machine according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a grid map based on a three-dimensional coordinate system according to an embodiment of the present invention;
FIG. 3 is a schematic diagram illustrating a principle of determining coordinates of an obstacle according to an embodiment of the present invention;
FIG. 4 is a schematic diagram illustrating a principle of determining coordinates of an obstacle according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of a method for planning a navigation path according to an embodiment of the present invention;
FIG. 6 is a schematic diagram of a method for planning a navigation path according to an embodiment of the present invention;
FIG. 7 is a schematic diagram of a method for planning a navigation path according to an embodiment of the present invention;
FIG. 8 is a schematic structural diagram of a machine-aided intelligent navigation device according to an embodiment of the present invention;
fig. 9 is a schematic structural diagram of a self-moving device according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention will be described in detail with reference to the accompanying drawings.
The embodiment of the invention provides a machine intelligent navigation method, wherein an execution main body of the method can be any self-moving equipment with an intelligent navigation function, such as self-moving robots of public service robots, sweeping robots, goods sweeping robots and the like, and at least can comprise automatic floor washers and moving chassis in a commercial environment; the road sweeper, the self-moving air purifier and the like in the household environment can also be other devices with self-moving or navigation requirements, such as a guiding robot, unmanned equipment and the like. The self-mobile equipment can be provided with a functional module of an intelligent navigation system, and the planning and the selection of a navigation path can be realized through the functional module. Meanwhile, the self-moving equipment can be further provided with a driving system and a fault detection system, wherein the driving system can be used for controlling the self-moving equipment to move according to a navigation path, and specifically can be wheels, tracks or simulated limbs; the fault detection system can utilize various technologies such as a machine learning technology, a laser ranging technology, a feature recognition technology and the like to detect all obstacles in a navigation scene and on a navigation path, and provides a detection result to the intelligent navigation system, so that the intelligent navigation system plans the navigation path according to the fault detection result.
The process flow shown in fig. 1 will be described in detail below with reference to specific embodiments, and the contents may be as follows:
In implementation, the self-moving device can be deployed in different navigation scenes such as malls, parks, garages, hotels and the like according to actual needs to execute intelligent navigation and travel processing. The self-moving device can trigger the intelligent navigation process according to the destination information input by a user or built in the device. Specifically, when a destination identifier input by a user is received or the destination identifier is obtained based on built-in logic, the self-mobile device may determine a destination point corresponding to the destination identifier by using a locally pre-stored electronic map, and then calculate all selectable paths from the current position to the destination point. Then, the self-mobile device may select an actual navigation path (i.e., a preset navigation path in the context) from all the selectable paths according to a path selection instruction input by the user or according to a built-in path selection rule, and load the actual navigation path to the driving system, so as to control the self-mobile device to travel to the destination through the driving system.
It should be noted that the electronic map pre-stored in the mobile device may be generated after the mobile device performs global scanning on the navigation scene in the initialization process, or may be directly obtained from the outside (e.g., manually input or downloaded from a network). The electronic map may be a grid map built based on a three-dimensional coordinate system, in which the X-axis and the Y-axis constitute a horizontal plane, and the Z-axis is perpendicular to the horizontal plane, as shown in fig. 2. It can be understood that, on all navigation paths generated based on the electronic map, there are no inherent obstacles in the current navigation scene, so that obstacles appearing on the navigation paths in the following process can be regarded as new obstacles.
The self-moving equipment can continuously monitor the obstacles on the navigation path in the process of going to the destination point according to the preset navigation path. When a newly added obstacle appears on the navigation path, the self-moving device can identify the newly added obstacle by using a feature identification technology so as to judge whether the newly added obstacle is a dynamic obstacle or a static obstacle. Here, a dynamic object set may be preset on the self-moving device, all dynamic obstacles that may appear in the current navigation scene may be enumerated in the dynamic object set, and the dynamic obstacles included in the dynamic object set corresponding to different navigation scenes may be different. When the newly added obstacle does not belong to the dynamic object set corresponding to the current pilot scene, the self-moving device can judge that the newly added obstacle is a static obstacle. However, if the self-moving device detects that there is a dynamic change (including a position change, an appearance change, and the like) in the newly added obstacle through other detection techniques, the newly added obstacle may be directly determined as a dynamic obstacle, and the dynamic obstacle may be used to update the dynamic object set corresponding to the current navigation scene.
Optionally, the deviation comparison may be performed by taking the area as a unit to determine whether there is a new obstacle, and accordingly, the following processing may be performed before step 101: when the vehicle travels on a preset navigation path, detecting peripheral obstacles in real time; determining the deviation degree of the detection result and pre-stored obstacle distribution information in each preset area; and if the deviation degree of the target area is greater than a preset threshold value, determining that the newly added obstacle exists in the target area.
The pre-stored obstacle distribution information may be recorded from an electronic map built in the mobile device.
In implementation, when the self-moving device travels on a preset navigation path, peripheral obstacles in multiple directions can be detected in real time, specifically, the peripheral obstacles include obstacles outside the path in front of the self-moving device and on the path, and the obstacles behind the self-moving device do not need to be detected. After the detection result is generated from the mobile device, the detection result may be compared with the pre-stored obstacle distribution information. In the comparison process, the mobile device may perform comparison respectively with a preset region as a unit to generate a deviation degree corresponding to each region. The preset area may be an area in the obstacle distribution information, and may be an area formed by one or more grids in the grid map. When the deviation degree of the target area is larger than the preset threshold value, the self-moving equipment can determine that the newly-added obstacles exist in the target area, and for the area with the deviation degree which is smaller than or equal to the preset threshold value, the deviation can be determined to be generated due to the conventional scanning error. It can be understood that, depending on the size of the area division, one newly added obstacle may cover a plurality of areas, and one area may also have a plurality of newly added obstacles. Thus, the deviation comparison is performed by taking the area as a unit, and the newly added obstacle can be more accurately detected.
Optionally, the detection processing of the peripheral obstacle may specifically be as follows: and detecting three-dimensional coordinate information of the nearest peripheral obstacle by using a depth camera technology and/or a laser sensing technology.
In implementation, on one hand, the self-moving device may acquire a scene image in one or more directions around the self-moving device by using a depth camera such as RealSense, then determine distances between nearest obstacles in different directions and the self-moving device according to a mapping relationship between color and depth data, and then calculate three-dimensional coordinate information of the nearest obstacle according to current coordinate information of the self-moving device, which may refer to the following formula:
PointX＝robot.position.x+point.x*cos(robot.α)+point.y*sin(robot.α)；
PointY＝robot.position.x+point.x*sin(robot.α)-point.y*cos(robot.α)；
the position is the current coordinate information of the self-moving equipment and can be determined by means of an odometer and laser ranging, and the point is the coordinate information of the obstacle relative to the self-moving equipment, wherein point.x and point.y are the coordinate information of the obstacle in a coordinate system which takes the self-moving equipment as an origin and faces to an x axis from the self-moving equipment, and point x and point y are the coordinate information of the obstacle in a grid map; α is the angle from the orientation of the mobile device to the X-axis of the grid map. Reference may be made to fig. 3, wherein the Z-axis is omitted, shown in the X-Y plane.
On the other hand, the self-moving device may obtain distances between the nearest obstacle and the laser altitude horizontal plane at a plurality of angles by using a laser radar, and then calculate three-dimensional coordinate information of the nearest obstacle at each angle by combining current coordinate information of the self-moving device, which may refer to the following formula:
PointX＝robot.position.x+lds[i]*cos(robot.α-i*interval)；
PointY＝robot.position.y+lds[i]*sin(robot.α-i*interval)；
position is current coordinate information of the mobile device, i is an angle number, lds [ i ] is a laser radar ranging result corresponding to the angle i, i is an angle value corresponding to the number i, and robot is an included angle between the orientation of the mobile device and an X axis of the grid map. Reference may be made to fig. 4, wherein the Z-axis is omitted, shown in the X-Y plane.
Further, the self-moving device can jointly determine the three-dimensional coordinate information of the nearest peripheral obstacle by using a depth camera technology and a laser sensing technology. It should be added that the coordinate calculation only includes two dimensions of the X axis and the Y axis, and the coordinate information of the Z axis dimension can be directly measured by the depth camera technology, which is not described herein again.
Alternatively, the dynamic obstacle may be identified by a machine learning technique, and accordingly, the process of step 101 may be as follows: and acquiring a scene image containing the newly added obstacle, performing feature recognition on the scene image based on a machine learning method, and judging whether the newly added obstacle belongs to a preset dynamic obstacle.
In implementation, if a newly added obstacle appears on the navigation path, the self-moving device may first acquire a scene image including the newly added obstacle, and then perform feature identification on the scene image by using a machine learning method to determine an object type to which the newly added obstacle belongs, and then determine whether the newly added obstacle belongs to a preset dynamic obstacle. Or, the self-moving device may further store feature information sets of all dynamic obstacles, and after feature information of a newly added obstacle in the scene image is extracted, the feature information of the newly added obstacle may be compared with the feature information sets, so as to determine whether the newly added obstacle belongs to a dynamic obstacle.
Alternatively, considering that the static obstacle moves with the dynamic obstacle, all the non-dynamic obstacles within the preset range of the dynamic obstacle may be determined as subordinate dynamic obstacles of the dynamic obstacle.
In implementation, when the self-moving device identifies a dynamic obstacle, if other non-dynamic obstacles exist in a preset range around the dynamic obstacle, the non-dynamic obstacles can be determined as subordinate dynamic obstacles of the dynamic obstacle. In a scenario, if two newly-added obstacles, namely a human body and a large carton, are detected, in general, the human body is recognized as a dynamic obstacle, the large carton is recognized as a static obstacle, but the large carton is found to be located within a preset range of the human body, it can be estimated that the human body is pushing/pulling the carton, and thus the large carton can be determined as a subordinate dynamic obstacle of the human body. Further, for the dependent dynamic obstacle, the movement trend of the dependent dynamic obstacle can be judged to be consistent with that of the dynamic obstacle. It can be understood that if there are other dynamic obstacles within the preset range of the target dynamic obstacle, the self-moving device may perform independent motion trend detection on the target dynamic obstacle and the other dynamic obstacles respectively.
And 102, if the obstacle is a dynamic obstacle, detecting the motion trend of the newly added obstacle, and re-planning the navigation path according to the motion trend.
In implementation, if the new obstacle is a dynamic obstacle, the self-moving device may further detect a movement trend of the new obstacle to determine an actual position of the new obstacle in a short period in the future. Then, the self-moving device can re-plan the original navigation path according to the movement trend of the newly added obstacle, such as deceleration, sudden stop, turning and the like. It should be noted that the planning of the navigation path in the present embodiment includes not only the determination and adjustment of the travel route, but also the determination and adjustment of the travel speed and the travel direction. In addition, the self-moving device may trigger the re-planning of the navigation path when a newly added obstacle is found, or may trigger the re-planning of the navigation path only after the distance from the newly added obstacle is detected to be smaller than a preset safe distance threshold.
Optionally, the movement trend of the newly added obstacle may be determined in a manner of monitoring states for multiple times, and the corresponding processing may be as follows: continuously monitoring the state of the newly added obstacle when the vehicle travels on the preset navigation path; determining the state change of the newly added obstacle according to the multiple state monitoring results; and judging the movement trend of the newly added obstacle based on the state change of the newly added obstacle.
In implementation, after the mobile device detects that a newly added obstacle exists on the preset navigation path, the mobile device can continue to keep moving, and the state of the newly added obstacle is continuously monitored at a certain frequency. The monitoring frequency here may be a fixed value, for example, once every 1s, or may be determined by the traveling speed of the mobile device, and the faster the traveling speed is, the higher the detection frequency is. The monitoring mode can mainly adopt image recognition, and the monitoring of the newly added obstacles is realized by shooting scene images containing the newly added obstacles. After the mobile device generates a plurality of state monitoring results, the state monitoring results can be compared and analyzed, so that the state change of the newly added obstacle is determined. And then, the self-moving equipment can judge the motion trend of the newly added obstacles in real time through the state change of the newly added obstacles.
Optionally, when there are multiple obstacles, tracking and monitoring of each obstacle may be implemented by using a feature comparison method, and corresponding processing may be as follows: extracting a target feature object in the latest state monitoring result; comparing the characteristics of the target characteristic object with at least one characteristic object in the previous state monitoring result; and determining the corresponding object of the target characteristic object in the last state monitoring result according to the characteristic comparison score.
In implementation, when the self-moving device continuously monitors the newly-added obstacles, the target feature in the latest state monitoring result can be extracted. The target feature may be any feature in the latest status monitoring result, and the feature may be an object in the status monitoring result that needs to be monitored, such as all newly added obstacles or all dynamic obstacles. And then, the mobile equipment can acquire each feature object in the previous state monitoring result, and perform feature comparison on the feature objects and the target feature object one by one to obtain a feature comparison score. Specifically, feature comparison can be performed through three angles of coordinates, width and height and a shape of the feature, the shape mainly extracts feature information through a deep learning network, and the feature similarity of the feature information and the feature information is reflected by a cosine distance, and the following formula can be referred to specifically:
motionValue＝exp(-weight*(xDiff+yDiff))；
shapeValue＝exp(-weight*(widthDiff+heightDiff))；
apperanceValue＝1.0-cosDistance；
affinityValue＝motionValue*shapeValue*apperanceValue；
wherein weight is the weight corresponding to each feature, xDiff and yDiff are the difference of coordinates, and width diff and height diff are the difference of width and height.
Next, the mobile device may compare the feature comparison score with a preset score, and when the feature comparison score of the feature a in the previous state monitoring result and the feature of the target feature is greater than the preset score, determine that the feature a matches the target feature, that is, the feature is a corresponding object of the target feature. And if the characteristic comparison scores of the plurality of characteristics A, B, C and the target characteristics are all larger than the preset score, the optimal solution can be found through the Hungarian method. Therefore, by means of characteristic comparison, the characteristic objects in the multiple monitoring results are correspondingly matched, and tracking of specific real objects can be achieved, so that corresponding movement trends can be conveniently analyzed.
Optionally, the operation trend determination mode of the newly added obstacle may specifically be as follows: calculating a displacement M value of the newly added barrier in the equipment traveling direction and a displacement N value of the newly added barrier in the traveling vertical direction in a preset historical time period; when the M value and the N value are both smaller than a preset threshold value, determining the movement trend of the newly-added obstacle as residing; when the N value is larger than a preset threshold value, determining the movement trend of the newly-added obstacles as crossing; and when the N value is smaller than the preset threshold and the M value is larger than the preset threshold, determining the movement trend of the newly added obstacle to be far away or close according to the distance between the equipment and the newly added obstacle.
In implementation, the self-moving device may determine the moving direction and the moving speed of the newly added obstacle within the historical duration by using the multiple state monitoring results. Then, the self-moving device can deduce the movement trend of the newly added obstacle in the short-term future based on the movement direction and the movement speed. The motion tendency can be mainly divided into four types, i.e., close, far, transverse, or dwell, and in detail, the displacement in the traveling direction of the mobile device can be denoted as an M value, and the displacement perpendicular to the traveling direction (i.e., the vertical traveling direction) can be denoted as an N value. In this way, if the M value and the N value are smaller than the preset threshold value in the preset historical time period, the movement trend of the newly-added barrier is considered to be resident; if the N value is larger than a preset threshold value, the movement trend of the newly-added barrier is considered to be crossing; if the value of N is smaller than the preset threshold value and the value of M is greater than the preset threshold value, the distance between the newly added obstacle and the mobile device can be further detected, if the distance between the newly added obstacle and the mobile device is increased, the movement trend of the newly added obstacle is considered to be far away, otherwise, the movement trend of the newly added obstacle is considered to be close.
Optionally, for different motion trends, corresponding different path planning schemes may exist, and the corresponding processing may be as follows: when the movement trend is close, the original navigation path is reserved, and the traveling speed of the machine is adjusted according to the distance between the machine and the newly added obstacle; when the movement trend is far away, the original navigation path is reserved, and the advancing speed of the machine is adjusted according to the movement speed of the newly-added barrier; when the movement trend is crossing, the navigation path is re-planned according to the movement direction of the newly added obstacle; and when the movement trend is resident, replanning the navigation path according to the position information of the newly added obstacle.
In implementation, when the movement trend of a new obstacle is detected to be close, the self-moving device can continue to move along the original navigation path, meanwhile, the distance between the self-moving device and the new obstacle is continuously detected, the moving speed of the self-moving device is adjusted according to the distance, the moving speed is lower when the distance is smaller, the self-moving device can stop moving to guarantee safety when the distance is smaller than a safety distance, and the self-moving device can resume moving when the new obstacle leaves the navigation path or the distance between the new obstacle and the self-moving device is larger than the safety distance. When the movement trend of the newly added obstacles is detected to be far away, the self-moving equipment can continue to move along the original navigation path, meanwhile, the movement speed of the newly added obstacles is continuously detected, and the moving speed of the self-moving equipment is adjusted according to the movement speed so as to ensure that the moving speed of the self-moving equipment is not greater than the movement speed of the newly added obstacles. When the movement trend of the new obstacle is detected to be crossing, the self-moving device can replan the navigation path in the opposite direction of the movement direction of the new obstacle. When the movement trend of the newly added obstacle is detected to be the staying, the self-mobile equipment can acquire the position information of the newly added obstacle and replan the navigation path to avoid the newly added obstacle; the step 103 may be specifically referred to for a path planning scheme of a newly-added obstacle in a residence state, and if an angle adjustment manner is adopted to avoid the newly-added obstacle, a certain early warning range needs to be reserved after a potential movement trend of the newly-added obstacle is considered, and then obstacle avoidance is performed with a minimum amplitude. It should be noted that, in order to ensure safety, no matter what kind of movement trend the newly added obstacle has, the self-moving device can adjust the traveling speed of the self-moving device in real time according to the distance between the self-moving device and the newly added obstacle.
Alternatively, if the distribution of the cataract in a certain region on the navigation path is complex, the self-moving device may bypass the whole region, and the corresponding processing may be as follows: when a plurality of dynamic obstacles with inconsistent motion trends exist on the preset navigation path, determining motion areas of the dynamic obstacles, and re-planning the navigation path according to the motion areas.
In implementation, if the self-moving device detects a plurality of dynamic obstacles on a preset navigation path and the movement trends of the dynamic obstacles are inconsistent, for example, a plurality of pedestrians are on the navigation path and the walking directions of the plurality of pedestrians are inconsistent, the movement areas of all the dynamic obstacles may be determined first, and then the navigation path is re-planned, so that the navigation path bypasses the movement areas. It should be noted that the determination of the motion area may be performed by accurately estimating the motion speed and the motion direction of each dynamic obstacle, or may be performed by determining a larger motion area according to the positions of a plurality of dynamic obstacles, where the motion area may be a set of all possible positions of the plurality of dynamic obstacles in a short period, and it is not necessary to detect the motion trend of each dynamic obstacle one by one. Reference may be made to fig. 5, wherein the Z-axis is omitted, shown in the X-Y plane.
And 103, if the obstacle is a static obstacle, re-planning the navigation path according to the position information of the newly added obstacle.
In implementation, if the newly added obstacle is a static obstacle, the self-moving device may obtain position information of the newly added obstacle, and then re-plan a navigation path based on the position information to avoid the newly added obstacle. In detail, the self-moving device may determine a local path where the newly added obstacle is located, delete the local path from the original navigation path, and select another shortest path that is the same as the start point and the end point of the local path, thereby completing the replanning of the navigation path. In addition, the self-moving equipment can also detect the blocking volume of the newly-added obstacle, and then adjust the advancing angle according to the blocking volume and the width of the path where the newly-added obstacle is located, so that the newly-added obstacle is avoided with the minimum amplitude, and the path does not need to be selected again. Reference may be made to fig. 6 and 7, wherein the Z-axis is omitted and the X-Y plane is shown.
Taking the self-moving device as an example of a public service robot inside a scene such as a market, a supermarket, a bank and the like, the self-moving device can have at least two moving requirements which are respectively autonomous triggered scene patrol and manually triggered destination navigation. In the scene patrol, the public service robot can patrol each area in the scene according to a preset patrol route (namely a preset navigation path), and when obstacles such as a human body, a cart and the like appear on the patrol route, the public service robot can identify the dynamic and static states and the movement trend of the obstacles so as to avoid or detour. In destination navigation, the public service robot may plan a navigation path based on a destination (such as a designated store, item, or counter) input by the user, and travel along the navigation path to guide the user to the destination. When obstacles such as a human body, a cart and the like appear on the navigation path, the public service robot can identify the dynamic and static states and the movement trend of the obstacles so as to avoid or detour.
In the embodiment of the invention, when a newly added obstacle appears on the preset navigation path, the newly added obstacle is judged to be a dynamic obstacle or a static obstacle; if the obstacle is a dynamic obstacle, detecting the movement trend of the newly added obstacle, and re-planning the navigation path according to the movement trend; and if the obstacle is a static obstacle, replanning the navigation path according to the position information of the newly added obstacle. Therefore, when the self-moving equipment carries out intelligent navigation, for newly added obstacles, different path planning schemes can be respectively selected according to two conditions of dynamic obstacles and static obstacles, and different path planning schemes can be adaptively selected according to dynamic obstacles with different movement trends, so that the situations of long-time invalid waiting and repeated path planning of the self-moving equipment can be avoided, and the fluency, the effectiveness and the safety of the intelligent navigation are ensured.
Based on the same technical concept, an embodiment of the present invention further provides a machine intelligent navigation apparatus, as shown in fig. 8, the apparatus includes:
the obstacle detection module 801 is configured to, when a newly added obstacle appears on a preset navigation path, determine that the newly added obstacle is a dynamic obstacle or a static obstacle;
a path planning module 802, configured to detect a motion trend of the newly added obstacle if the newly added obstacle is a dynamic obstacle, and re-plan a navigation path according to the motion trend; and if the obstacle is a static obstacle, re-planning the navigation path according to the position information of the newly-added obstacle.
Optionally, the obstacle detecting module 801 is further configured to:
when the vehicle travels on the preset navigation path, detecting surrounding obstacles in real time;
determining the deviation degree of the detection result and the pre-stored obstacle distribution information in each preset area;
and if the deviation degree of the target area is greater than a preset threshold value, determining that the newly added obstacle exists in the target area.
Optionally, the obstacle detecting module 801 is specifically configured to:
and detecting three-dimensional coordinate information of the nearest peripheral obstacle by using a depth camera technology and/or a laser sensing technology.
Optionally, the obstacle detecting module 801 is specifically configured to:
and acquiring a scene image containing the newly added obstacle, performing feature recognition on the scene image based on a machine learning method, and judging whether the newly added obstacle belongs to a preset dynamic obstacle.
Optionally, the obstacle detecting module 801 is further configured to:
and determining all non-dynamic obstacles within a preset range of the dynamic obstacles as subordinate dynamic obstacles of the dynamic obstacles.
Optionally, the path planning module 802 is specifically configured to:
continuously monitoring the state of the newly added obstacle when the vehicle travels on the preset navigation path;
determining the state change of the newly added obstacle according to the multiple state monitoring results;
and judging the movement trend of the newly added obstacle based on the state change of the newly added obstacle.
Optionally, the path planning module 802 is specifically configured to:
extracting a target feature object in the latest state monitoring result;
comparing the characteristics of the target characteristic object with the characteristics of each characteristic object in the previous state monitoring result;
and determining a corresponding object of the target characteristic object in the last state monitoring result according to the characteristic comparison score.
Optionally, the path planning module 802 is specifically configured to:
calculating a displacement M value of the newly-added barrier in the equipment traveling direction and a displacement N value of the newly-added barrier in the traveling vertical direction within a preset historical time period;
when the M value and the N value are both smaller than a preset threshold value, determining the movement trend of the newly-added obstacle as residing;
when the N value is larger than a preset threshold value, determining the movement trend of the new obstacle as crossing;
and when the N value is smaller than a preset threshold value and the M value is larger than the preset threshold value, determining the movement trend of the newly added obstacle to be far away or close according to the distance between the equipment and the newly added obstacle.
Optionally, the path planning module 802 is specifically configured to:
when the movement trend is close, the original navigation path is reserved, and the traveling speed of the machine is adjusted according to the distance between the machine and the newly added obstacle;
when the movement trend is far away, the original navigation path is reserved, and the advancing speed of the machine is adjusted according to the movement speed of the newly added obstacle;
when the movement trend is crossing, replanning a navigation path according to the movement direction of the newly-added barrier;
and when the movement trend is resident, re-planning a navigation path according to the position information of the newly added obstacle.
Optionally, the path planning module 802 is further configured to:
when a plurality of dynamic obstacles with inconsistent motion trends exist on the preset navigation path, determining motion areas of the dynamic obstacles, and replanning the navigation path according to the motion areas.
In the embodiment of the invention, when a newly added obstacle appears on the preset navigation path, the newly added obstacle is judged to be a dynamic obstacle or a static obstacle; if the obstacle is a dynamic obstacle, detecting the movement trend of the newly added obstacle, and re-planning the navigation path according to the movement trend; and if the obstacle is a static obstacle, replanning the navigation path according to the position information of the newly added obstacle. Therefore, when the self-moving equipment carries out intelligent navigation, for newly added obstacles, different path planning schemes can be respectively selected according to two conditions of dynamic obstacles and static obstacles, and different path planning schemes can be adaptively selected according to dynamic obstacles with different movement trends, so that the situations of long-time invalid waiting and repeated path planning of the self-moving equipment can be avoided, and the fluency, the effectiveness and the safety of the intelligent navigation are ensured.
Fig. 9 is a schematic structural diagram of a self-moving device according to an embodiment of the present invention. The mobile device 900 may vary widely in configuration or performance and may include one or more central processors 922 (e.g., one or more processors) and memory 932, one or more storage media 930 (e.g., one or more mass storage devices) storing applications 942 or data 944. Memory 932 and storage media 930 can be, among other things, transient storage or persistent storage. The program stored in the storage medium 930 may include one or more modules (not shown), each of which may include a sequence of instructions that operate on the mobile device 900. Still further, central processor 922 may be configured to communicate with storage medium 930 to execute a sequence of instruction operations in storage medium 930 on mobile device 900.
The self-mobile device 900 may also include one or more power supplies 929, one or more wired or wireless network interfaces 950, one or more input-output interfaces 958, one or more keyboards 956, and/or one or more operating systems 941, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc.
The self-moving apparatus 900 may include memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs including instructions for performing the machine intelligent navigation described above.
It will be understood by those skilled in the art that all or part of the steps for implementing the above embodiments may be implemented by hardware, or may be implemented by a program instructing relevant hardware, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a magnetic disk, an optical disk, or the like.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (15)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010813288.XA CN114136318A (en) | 2020-08-13 | 2020-08-13 | Intelligent navigation method and device for machine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010813288.XA CN114136318A (en) | 2020-08-13 | 2020-08-13 | Intelligent navigation method and device for machine |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114136318A true CN114136318A (en) | 2022-03-04 |
Family
ID=80438001
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010813288.XA Pending CN114136318A (en) | 2020-08-13 | 2020-08-13 | Intelligent navigation method and device for machine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114136318A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114750159A (en) * | 2022-05-07 | 2022-07-15 | 哈尔滨工业大学（深圳） | Control method of catering robot and catering robot |
| CN115752481A (en) * | 2022-12-09 | 2023-03-07 | 广东车卫士信息科技有限公司 | AR navigation method, AR glasses, medium and equipment based on image recognition |
| CN115993830A (en) * | 2023-03-21 | 2023-04-21 | 佛山隆深机器人有限公司 | Path planning method and device based on obstacle avoidance and robot |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102160006A (en) * | 2008-07-15 | 2011-08-17 | 空中侦察辨识和避免技术有限责任公司 | System and method for preventing a collis |
| CN105629989A (en) * | 2015-12-28 | 2016-06-01 | 电子科技大学 | Obstacle region division method based on minimum enclosing circle and maximum inscribed circle |
| CN106599832A (en) * | 2016-12-09 | 2017-04-26 | 重庆邮电大学 | Method for detecting and recognizing various types of obstacles based on convolution neural network |
| CN108762264A (en) * | 2018-05-22 | 2018-11-06 | 重庆邮电大学 | The dynamic obstacle avoidance method of robot based on Artificial Potential Field and rolling window |
| CN109725650A (en) * | 2019-03-08 | 2019-05-07 | 哈尔滨工程大学 | An AUV Obstacle Avoidance Method in Dense Obstacle Environment |
| CN109782763A (en) * | 2019-01-18 | 2019-05-21 | 中国电子科技集团公司信息科学研究院 | A kind of method for planning path for mobile robot under dynamic environment |
| CN111319615A (en) * | 2020-03-16 | 2020-06-23 | 湖北亿咖通科技有限公司 | Intelligent passenger-replacing parking method, computer-readable storage medium and electronic device |
| CN111426326A (en) * | 2020-01-17 | 2020-07-17 | 深圳市镭神智能系统有限公司 | Navigation method, device, equipment, system and storage medium |
-
2020
- 2020-08-13 CN CN202010813288.XA patent/CN114136318A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102160006A (en) * | 2008-07-15 | 2011-08-17 | 空中侦察辨识和避免技术有限责任公司 | System and method for preventing a collis |
| CN105629989A (en) * | 2015-12-28 | 2016-06-01 | 电子科技大学 | Obstacle region division method based on minimum enclosing circle and maximum inscribed circle |
| CN106599832A (en) * | 2016-12-09 | 2017-04-26 | 重庆邮电大学 | Method for detecting and recognizing various types of obstacles based on convolution neural network |
| CN108762264A (en) * | 2018-05-22 | 2018-11-06 | 重庆邮电大学 | The dynamic obstacle avoidance method of robot based on Artificial Potential Field and rolling window |
| CN109782763A (en) * | 2019-01-18 | 2019-05-21 | 中国电子科技集团公司信息科学研究院 | A kind of method for planning path for mobile robot under dynamic environment |
| CN109725650A (en) * | 2019-03-08 | 2019-05-07 | 哈尔滨工程大学 | An AUV Obstacle Avoidance Method in Dense Obstacle Environment |
| CN111426326A (en) * | 2020-01-17 | 2020-07-17 | 深圳市镭神智能系统有限公司 | Navigation method, device, equipment, system and storage medium |
| CN111319615A (en) * | 2020-03-16 | 2020-06-23 | 湖北亿咖通科技有限公司 | Intelligent passenger-replacing parking method, computer-readable storage medium and electronic device |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114750159A (en) * | 2022-05-07 | 2022-07-15 | 哈尔滨工业大学（深圳） | Control method of catering robot and catering robot |
| CN114750159B (en) * | 2022-05-07 | 2023-08-22 | 哈尔滨工业大学（深圳） | Control method of catering robot and catering robot |
| CN115752481A (en) * | 2022-12-09 | 2023-03-07 | 广东车卫士信息科技有限公司 | AR navigation method, AR glasses, medium and equipment based on image recognition |
| CN115752481B (en) * | 2022-12-09 | 2023-09-01 | 广东车卫士信息科技有限公司 | AR navigation method, AR glasses, medium and equipment based on image recognition |
| CN115993830A (en) * | 2023-03-21 | 2023-04-21 | 佛山隆深机器人有限公司 | Path planning method and device based on obstacle avoidance and robot |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10705528B2 (en) | Autonomous visual navigation | |
| US11454973B2 (en) | Mapped driving paths for autonomous vehicle | |
| US10551850B2 (en) | Low quality pose | |
| CN111788102B (en) | Odometer system and method for tracking traffic lights | |
| Kümmerle et al. | A navigation system for robots operating in crowded urban environments | |
| US11829147B2 (en) | Path planning | |
| JP2022542807A (en) | intermediate waypoint generator | |
| CN113110457A (en) | Autonomous coverage inspection method for intelligent robot in indoor complex dynamic environment | |
| CN114136318A (en) | Intelligent navigation method and device for machine | |
| Lidoris et al. | The autonomous city explorer (ACE) project—mobile robot navigation in highly populated urban environments | |
| CN117629205A (en) | A navigation method, chip and robot for a robot to pass through a narrow passage | |
| WO2022049918A1 (en) | Information processing device, information processing method, and program | |
| JP6809913B2 (en) | Robots, robot control methods, and map generation methods | |
| Gao et al. | Design and implementation of an autonomous driving delivery robot | |
| Molinos et al. | Dynamic obstacle avoidance based on curvature arcs | |
| JP2023063149A (en) | Processing device, mobile robot, movement control system, processing method and program | |
| CN116954265B (en) | Method and device for rescheduling local motion trail and electronic equipment | |
| Lunenburg et al. | A representation method based on the probability of collision for safe robot navigation in domestic environments | |
| Matsuura et al. | Model predictive obstacle avoidance control for an electric wheelchair in indoor environments using artificial potential field method | |
| Ayomoh et al. | Robot navigation model in a multi-target domain amidst static and dynamic obstacles | |
| Priyasad et al. | Point cloud based autonomous area exploration algorithm | |
| Karakaya et al. | A novel local motion planner: Navibug | |
| Nordström et al. | DTAA: A Detect, Track and Avoid Architecture for navigation in spaces with Multiple Velocity Objects | |
| KR102647612B1 (en) | Robot for guiding an evacuation route for persons in the space in the event of emergency and method for controlling the same | |
| CN118225105B (en) | Textile workshop logistics AGV navigation method based on multi-source information perception |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220304 |