CN112847370A - Mobile robot path planning method based on deep learning - Google Patents
Mobile robot path planning method based on deep learning Download PDFInfo
- Publication number
- CN112847370A CN112847370A CN202110027874.6A CN202110027874A CN112847370A CN 112847370 A CN112847370 A CN 112847370A CN 202110027874 A CN202110027874 A CN 202110027874A CN 112847370 A CN112847370 A CN 112847370A
- Authority
- CN
- China
- Prior art keywords
- operated
- point
- time value
- specific time
- moving
- 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.)
- Withdrawn
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1656—Programme controls characterised by programming, planning systems for manipulators
- B25J9/1664—Programme controls characterised by programming, planning systems for manipulators characterised by motion, path, trajectory planning
Landscapes
- Engineering & Computer Science (AREA)
- Robotics (AREA)
- Mechanical Engineering (AREA)
- Manipulator (AREA)
Abstract
The invention provides a mobile robot path planning method based on deep learning, which comprises the steps that a master controller calculates a moving time stamp of a first point to be operated according to an expected speed, and if the moving time stamp is larger than a first specific time value and smaller than a second specific time value, the controller controls a driver to drive a motor shaft to act; if the time value is less than the first specific time value, the controller controls the driver to maintain the original working state; if the time value is larger than the second specific time value, the controller controls the driver to interpolate N second to-be-operated points behind the first to-be-operated point to serve as the first to-be-operated point in the next path planning process, comparison of moving time stamps is carried out, whether interpolation of a new operation point is carried out or operation is carried out or the original state is maintained, so that a path curve of the next to-be-operated point is corrected again.
Description
Technical Field
The invention relates to the technical field of industrial robots, in particular to a mobile robot path planning method based on deep learning.
Background
At present, automation technology is rapidly developed, in the field, a robot system is an automatic operation system composed of a robot, peripheral equipment and tools, the robot, particularly an industrial robot, is generally a multi-axis mechanical arm, an axis of the robot is mainly composed of a gear box and a servo motor, the robot system also comprises a plurality of parts which move and rotate independently, and a robot control system is mainly composed of a main controller and a servo drive controller.
At present, leading robot control systems at home and abroad, particularly industrial robot control systems, are developing towards the directions of further improving control precision, having better safety, being more convenient for input and output, and the like. In order to achieve the purpose that the robot completes the designated operation, the robot needs to be established with the operation of how to move. The traditional mode is through the teaching box, under the teaching mode, control every joint motion, the continuous recording operation point. After recording a plurality of work points of the entire job, the type of movement between the work points, the movement speed, and the like may be set. The hand teaching technology developed at present is more and more popular, the principle is that a force sensor is added at the tail end of a robot, when the hand teaching function is started, a worker pulls the tail end of the robot by using a hand, the force sensor collects data and then sends the data to a robot controller, the controller calculates the direction and the size of a tension vector and drives the robot to move, and the tail end of the robot moves (including rotates) in the tension direction, namely follows the hand movement of the worker. In the process, the controller records the whole motion path, and after the teaching is finished, the path recorded by the controller is the teaching task.
In general, teaching tasks are executed in two ways: one is reproduction of the hand-grip teaching, i.e. the controller will record
The time interval between the points of the path is increased or decreased and then sent out as an execution job. When the time interval is not changed, the operation executed by the robot is completely reproduced when the hand is taught, and after the interval is increased or decreased, the operation is slowly released or quickly released when the hand is taught; the other is that the path of the teaching task is not changed, but the speed in the whole path is re-planned, and the teaching is not referred to when the hand pulls fast and when the hand pulls slow. In some special application environments, such as welding, material filling and the like, the tail end of the robot needs to be at a constant speed to ensure the operation effect, and the hand-held teaching operation in the special application needs to be executed according to a second mode, namely, the speed in the whole path is re-planned.
In the second implementation, the type of movement and speed of movement between the work points needs to be re-planned. A simple way is to pick distinctive points in the path, such as inflection points, and use these points to plan a path. The method can discard a plurality of collected path points and also has an influence on the restoration of the path. The other mode is that a path curve is calculated by utilizing the collected path points, a new operation point is calculated according to the expected speed, and then the operation point is reversely solved into a joint value. The disadvantage of this method is that the amount of calculation is too large, which affects the work efficiency.
Disclosure of Invention
The invention provides a mobile robot path planning method and system based on deep learning, aiming at the problems that the existing operation point re-planning technology is complex to realize, the calculated amount is too large, the working efficiency is influenced and the like, and solving the problems that the existing operation point re-planning technology is complex to realize, the calculated amount is too large, the communication time is long, the working efficiency is influenced and the like.
On one hand, the invention provides a mobile robot path planning method based on deep learning, which comprises the steps that a master controller calculates a moving timestamp of a first point to be operated;
if the moving timestamp is greater than a first specific time value and less than a second specific time value, the master controller controls a driver corresponding to the first to-be-operated point to drive a motor shaft to act; the second specific time value is greater than the first specific time value;
if the moving timestamp is smaller than the first specific time value, the master controller controls the driver corresponding to the first to-be-operated point to maintain the original working state;
if the moving timestamp is greater than the second specific time value, the master controller controls a driver corresponding to the first point to be operated to interpolate N second points to be operated after the first point to be operated so as to take the N second points to be operated as the first point to be operated in the next path planning process to re-plan the operation movement path of the robot, wherein N is an integer greater than zero.
As an alternative embodiment, the general controller calculates the moving time stamp of the first point-to-be-operated, and includes:
detecting through a sensor, and recording spatial position points of the M first points to be operated in a sampling teaching process; wherein M is an integer greater than zero;
calculating the position distance between two first points to be operated corresponding to the adjacent operation time;
obtaining a set expected speed in unit time;
obtaining the movement timestamp from the position separation and the desired velocity.
In another aspect, the present invention further provides a system for planning a motion path of a robot operation, including:
the master controller is used for calculating the moving timestamp of the first point to be operated;
the total controller is further used for judging whether the moving timestamp reaches the second specific time value from the first specific time value
Within two specific time value ranges; if the moving timestamp is greater than the first specific time value and less than the second specific time value, the master controller is further used for controlling a driver corresponding to the first to-be-operated point to drive a motor shaft to act;
the driver is controlled by the controller to drive a motor shaft to act;
if the moving timestamp is smaller than the first specific time value, the master controller is further used for controlling the driver corresponding to the first to-be-operated point to maintain the original working state;
the moving timestamp is greater than the second specific time value, and the controller is further configured to control the driver corresponding to the first point to be operated to interpolate N second points to be operated after the first point to be operated, so that the N second points to be operated are used as the first points to be operated in the next path planning process, and the operation motion path of the robot is re-planned;
and the driver is further configured to interpolate, by the controller, N second points to be operated after the first point to be operated according to a result that the moving timestamp determined by the controller is greater than the second specific time value, so as to take the N second points to be operated as the first point to be operated in a next path planning process, and to re-plan an operation movement path of the robot, where N is an integer greater than zero.
As an optional implementation manner, the system further includes a sensor, configured to perform detection and record spatial position points of the M first to-be-operated points in a sampling teaching process; wherein M is an integer greater than zero;
the master controller is further configured to calculate a position distance between two first to-be-operated points corresponding to adjacent operation time, obtain an expected speed in a set unit time, and obtain the moving timestamp according to the position distance and the expected speed.
The invention provides a mobile robot path planning method and a system based on deep learning, wherein a master controller is used for calculating a moving timestamp of a first point to be operated; (ii) a If the moving timestamp is greater than a first specific time value and less than a second specific time value, controlling the driver to drive a motor shaft to act; the second specific time value is greater than the first specific time value; if the moving timestamp is smaller than the first specific time value, controlling the driver to maintain the original working state; if the moving timestamp is greater than the second specific time value, controlling the driver to interpolate N second points to be operated after the first point to be operated so as to use the N second points to be operated as the first points to be operated in the next path planning process to re-plan the operation movement path of the robot, comparing the calculated moving timestamp with the specific time, selecting to interpolate a new point to maintain the original state to re-correct the path curve of the next point to be operated, and having the advantages of simple steps and convenient calculation.
Drawings
Fig. 1 is a flowchart of a method for planning a path of a mobile robot based on deep learning according to an embodiment of the present invention;
fig. 2 is a flowchart of another method for planning a path of a mobile robot based on deep learning according to an embodiment of the present invention.
Detailed Description
In order that the objects and advantages of the invention will be more clearly understood, the invention is further described below with reference to examples; it should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only for explaining the technical principle of the present invention, and do not limit the scope of the present invention.
It should be noted that in the description of the present invention, the terms of direction or positional relationship indicated by the terms "upper", "lower", "left", "right", "inner", "outer", etc. are based on the directions or positional relationships shown in the drawings, which are only for convenience of description, and do not indicate or imply that the device or element must have a specific orientation, be constructed in a specific orientation, and be operated, and thus, should not be construed as limiting the present invention.
Furthermore, it should be noted that, in the description of the present invention, unless otherwise explicitly specified or limited, the terms "mounted," "connected," and "connected" are to be construed broadly, and may be, for example, fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood by those skilled in the art according to specific situations.
Fig. 1 is a flowchart of a mobile robot path planning method based on deep learning according to an embodiment of the present invention. As shown in fig. 1, the method for planning a path of a mobile robot based on deep learning provided in this embodiment includes:
110. the master controller calculates a moving timestamp t of a first point to be operated;
120. judging the sizes of the moving timestamp T and the first and second specific time values T1 and T2;
if the moving timestamp T is greater than the first specific time value T1 and less than the second specific time value T2, the following is performed
130. controlling the driver to drive a motor shaft to act; the second specific time value T2 is greater than the first specific time value T1;
if the moving timestamp T is less than the first specific time value T1, performing the following step 140;
140. controlling a driver to maintain the original working state;
if the moving timestamp T is greater than a second specific time value T2, performing the following step 150;
150. the driver is controlled to interpolate N second points to be operated after the first point to be operated so as to take the N second points to be operated as the first points to be operated in the next path planning process and plan the operation movement path of the robot again; wherein N is an integer greater than zero. In this embodiment, the controller, while controlling the driver, issues a key value of the standby operation point so that the driver performs a key action according to the key value.
As an alternative implementation, referring to fig. 2, fig. 2 is a flowchart of another method for planning a path of a mobile robot based on deep learning according to an embodiment of the present invention, as shown in fig. 2, based on step 110 shown in fig. 1, the method further includes the following steps, that is, the main controller calculates a moving time stamp t of the first point to be operated, including:
111. detecting through a sensor, and recording spatial position points of M first points to be operated in a sampling teaching process; wherein M is an integer greater than zero;
112. calculating the position distance between two first points to be operated corresponding to the adjacent operation time;
113. obtaining a set expected speed in unit time;
114. the movement time stamp t is obtained from the position separation and the desired velocity.
On the other hand, the embodiment of the invention also provides a robot operation motion path planning system, please refer to.
The invention provides a structure diagram of a robot operation motion path planning system. As shown, the robot working motion path planning system 100 provided by the present embodiment includes an overall controller 310 and a driver 320, where the overall controller 310 is configured to calculate a moving time stamp t of a first point to be worked; the general controller 310 is configured to determine whether the moving timestamp T is within a range from a first specific time value T1 to a second specific time value T2; the overall controller 310 is further configured to determine whether the moving timestamp is within a range from the first specific time value to the second specific time value; if the moving timestamp is greater than the first specific time value and less than the second specific time value, the overall controller 310 is further configured to control the driver 320 corresponding to the first to-be-operated point to drive a motor shaft to act; the driver 320 is also used for driving a motor shaft to act if the moving time stamp T is greater than the first specific time value T1 and less than the second specific time value T2 according to the judgment of the overall controller 310; if the moving timestamp is less than the first specific time value, the general controller 310 is further configured to control the driver 320 corresponding to the first to-be-operated point to maintain an original operating state; the driver 320 is further configured to maintain the original working state according to the result that the moving timestamp T judged by the overall controller 310 is smaller than the first specific time value T1; if the moving timestamp is greater than the second specific time value, the overall controller 310 is further configured to control the driver 320 corresponding to the first point to be operated to interpolate N second points to be operated after the first point to be operated, so as to use the N second points to be operated as the first point to be operated in the next path planning process, so as to re-plan the operation motion path of the robot; the driver 320 is further configured to interpolate N second points to be operated after the first point to be operated according to the result that the moving timestamp T determined by the overall controller 310 is greater than the second specific time value T2, so as to use the N second points to be operated as the first points to be operated in the next path planning process, so as to re-plan the operation movement path of the robot, where N is an integer greater than zero.
As an alternative embodiment, the present embodiment provides a robot working motion path planning system 100
The system also comprises a sensor, a signal processing module and a signal processing module, wherein the sensor is used for detecting and recording spatial position points of M first to-be-operated points in the sampling teaching process; wherein M is an integer greater than zero; the general controller 310 is further configured to calculate a position distance between two first to-be-operated points corresponding to adjacent operation times, obtain a desired speed in a set unit time, and obtain the moving timestamp t according to the position distance and the desired speed.
In summary, in the mobile robot path planning method and system based on deep learning provided by the embodiment of the present invention, the master controller calculates the moving timestamp of the first point to be operated; (ii) a If the moving timestamp is greater than a first specific time value and less than a second specific time value, controlling the driver to drive a motor shaft to act; the second specific time value is greater than the first specific time value; if the moving timestamp is smaller than the first specific time value, controlling the driver to maintain the original working state; if the moving timestamp is greater than the second specific time value, controlling the driver to interpolate N second points to be operated after the first point to be operated so as to use the N second points to be operated as the first points to be operated in the next path planning process to re-plan the operation movement path of the robot, comparing the calculated moving timestamp with the specific time, selecting to interpolate a new point to maintain the original state to re-correct the path curve of the next point to be operated, and having the advantages of simple steps and convenient calculation.
So far, the technical solutions of the present invention have been described in connection with the preferred embodiments shown in the drawings, but it is easily understood by those skilled in the art that the scope of the present invention is obviously not limited to these specific embodiments. Equivalent changes or substitutions of related technical features can be made by those skilled in the art without departing from the principle of the invention, and the technical scheme after the changes or substitutions can fall into the protection scope of the invention.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention; various modifications and alterations to this invention will become apparent to those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (4)
1. A mobile robot path planning method based on deep learning is characterized by comprising the following steps: the master controller calculates a moving timestamp of a first point to be operated;
if the moving timestamp is greater than a first specific time value and less than a second specific time value, the master controller controls a driver corresponding to the first to-be-operated point to drive a motor shaft to act; the second specific time value is greater than the first specific time value;
if the moving timestamp is smaller than the first specific time value, the master controller controls the driver corresponding to the first to-be-operated point to maintain the original working state;
if the moving timestamp is greater than the second specific time value, the master controller controls a driver corresponding to the first point to be operated to interpolate N second points to be operated after the first point to be operated so as to take the N second points to be operated as the first point to be operated in the next path planning process to re-plan the operation movement path of the robot, wherein N is an integer greater than zero.
2. The deep learning-based mobile robot path planning method according to claim 1, wherein the overall controller calculates a moving time stamp of the first point to be operated, including:
detecting through a sensor, and recording spatial position points of the M first points to be operated in a sampling teaching process; wherein M is an integer greater than zero;
calculating the position distance between two first points to be operated corresponding to the adjacent operation time; obtaining a set expected speed in unit time;
obtaining the movement timestamp from the position separation and the desired velocity.
3. A robot work motion path planning system, comprising: the master controller is used for calculating the moving timestamp of the first point to be operated;
the master controller is further configured to determine whether the moving timestamp is within a range from the first specific time value to the second specific time value; if the moving timestamp is greater than the first specific time value and less than the second specific time value, the master controller is further configured to control a driver corresponding to the first to-be-operated point to drive a motor shaft to act;
the driver is controlled by the controller to drive a motor shaft to act;
if the moving timestamp is smaller than the first specific time value, the master controller is further used for controlling the driver corresponding to the first to-be-operated point to maintain the original working state;
if the moving timestamp is greater than the second specific time value, the general controller is further configured to control the driver corresponding to the first point to be operated to interpolate N second points to be operated after the first point to be operated, so that the N second points to be operated are used as the first point to be operated in the next path planning process, and the operation motion path of the robot is re-planned;
and the driver is further configured to interpolate, by the controller, N second points to be operated after the first point to be operated according to a result that the moving timestamp determined by the controller is greater than the second specific time value, so as to take the N second points to be operated as the first point to be operated in a next path planning process, and to re-plan an operation movement path of the robot, where N is an integer greater than zero.
4. The robot work motion path planning system according to claim 3, further comprising a sensor for detecting and recording spatial position points of the M first points to be worked during the sampling teaching; wherein M is an integer greater than zero;
the master controller is further configured to calculate a position distance between two first to-be-operated points corresponding to adjacent operation time, obtain an expected speed in a set unit time, and obtain the moving timestamp according to the position distance and the expected speed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110027874.6A CN112847370A (en) | 2021-01-11 | 2021-01-11 | Mobile robot path planning method based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110027874.6A CN112847370A (en) | 2021-01-11 | 2021-01-11 | Mobile robot path planning method based on deep learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112847370A true CN112847370A (en) | 2021-05-28 |
Family
ID=76002029
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110027874.6A Withdrawn CN112847370A (en) | 2021-01-11 | 2021-01-11 | Mobile robot path planning method based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112847370A (en) |
-
2021
- 2021-01-11 CN CN202110027874.6A patent/CN112847370A/en not_active Withdrawn
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP5311294B2 (en) | Robot contact position detector | |
| US8626341B2 (en) | Apparatus and method for adjusting parameter of impedance control | |
| CN103147577B (en) | Control method, equipment, system and construction machinery for multi-joint mechanical arm support | |
| CN105458878B (en) | Robot system for performing force control | |
| EP1758003B1 (en) | Method for controlling systems provided with redundant actuators | |
| CN108422420B (en) | Robot system having learning control function and learning control method | |
| CN107553484B (en) | Robot operation motion path planning method and system | |
| US20090125146A1 (en) | Method of and Apparatus for Automated Path Learning | |
| JP5236596B2 (en) | Processing robot system | |
| CN105122160A (en) | Numerical control apparatus | |
| US11766780B2 (en) | System identification of industrial robot dynamics for safety-critical applications | |
| JP2005103681A (en) | Robot system | |
| CN109551485B (en) | Motion control method, device and system and storage medium | |
| JP2003236787A (en) | Drivingly controlling method and drivingly controlling device | |
| CN111438687A (en) | Determination device | |
| JP2010238174A (en) | Positioning device and positioning method | |
| US20210039256A1 (en) | Robot control method | |
| JP5904445B2 (en) | Robot controller | |
| CN112847370A (en) | Mobile robot path planning method based on deep learning | |
| JP6088601B2 (en) | Robot controller that suppresses tool tip deflection in robot with traveling axis | |
| CN111136652A (en) | Speed reducer system, manufacturing method thereof, correction data generating method, correction data correcting method, robot, and robot | |
| KR100389010B1 (en) | Control Method of PC based Gantry Robot | |
| JPH0825260A (en) | Deflection correcting method for industrial robot | |
| CN205734875U (en) | Inertial navigation formula robot teaching equipment | |
| Lei et al. | Vision-based position/impedance control for robotic assembly task |
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 | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20210528 |
|
| WW01 | Invention patent application withdrawn after publication |