CN-122018512-A - Self-adaptive generation method for robot home track based on intelligent control

Abstract

The invention discloses an intelligent control-based self-adaptive generation method of a robot home track, which relates to the technical field of intelligent control and comprises the steps of collecting environment control data, establishing a corresponding relation between a local coordinate system and a global occupation grid map coordinate system, generating an initial rectangular home-returning area, executing unilateral contraction and contralateral compensation by combining obstacle conflicts to obtain a course coupling home-returning area, fusing wall shielding, gradient change and signal attenuation to execute integral offset correction on the course coupling home-returning area, screening candidate home gates on the corrected boundary to obtain an optimal home gate, constructing a multi-factor cost map, segmenting and generating a self-adaptive home-returning track, and controlling the robot home-returning based on the self-adaptive home-returning track, and detecting position and attitude deviation, dynamic obstacle and signal change in real time until accurate stop is completed. According to the invention, through the integrated control from the generation of the robot home region to the accurate stop, the stability and the stop accuracy of the robot in the home process under the complex environment are improved.

Inventors

• WU BIN
• ZHAO HU

Assignees

• 上海筱珈数据科技有限公司

Dates

Publication Date

20260512

Application Date

20260414

Claims (10)

1. 1. The self-adaptive generation method of the robot home track based on intelligent control is characterized by comprising the following steps of: Collecting the current position, the stop point position, the base station entering direction, the courtyard boundary, the obstacle distribution, the real-time signal distribution and the body motion constraint of the robot to form environment control data; based on the environment control data, establishing a corresponding relation between a local coordinate system and a global occupation grid map coordinate system, generating an initial rectangular home region, and executing unilateral contraction and contralateral compensation by combining obstacle conflict to obtain a course coupling home region; Based on the heading coupling home region, carrying out integral offset correction on the heading coupling home region by fusing wall shielding, gradient change and signal attenuation, and screening candidate entrance doors on the corrected boundary to obtain an optimal entrance door; constructing a multi-factor cost map based on the optimal entry gate, fusing obstacle risk, attitude adjustment cost, communication reliability and residual electric quantity, and generating a self-adaptive home track in a segmented manner; And controlling the robot to return home based on the self-adaptive home-returning track, and detecting pose deviation, dynamic obstacle and signal change in real time until accurate stopping is completed.
2. 2. The method for adaptively generating the robot returning track based on intelligent control according to claim 1, wherein the forming of the environmental control data is to uniformly write the current position, the stop position, the base station entering direction, the courtyard boundary, the obstacle distribution, the real-time signal distribution and the body motion constraint of the robot into the global occupation grid map to form the environmental control data.
3. 3. The method for adaptively generating the home track of the robot based on intelligent control according to claim 2, wherein the step of creating the initial rectangular home area by establishing the correspondence between the local coordinate system and the global occupation grid map coordinate system based on the environmental control data comprises the following steps: Based on the environmental control data, establishing a local coordinate system, and establishing a corresponding relation between the local coordinate system and a global occupation grid map coordinate system by utilizing a rotation matrix; Determining principal direction half-length constraint and lateral half-width constraint according to the corresponding relation, courtyard boundary, obstacle distribution, real-time signal distribution and organism motion constraint; based on the principal direction half-length constraint and the lateral half-width constraint, determining the half length and the half width of an initial rectangular home region to obtain a size parameter; And according to the size parameters, mapping four vertexes in the local coordinate system into the global occupied grid map coordinate system by utilizing the rotation matrix, and connecting to form a closed rectangular boundary to obtain an initial rectangular home region.
4. 4. The adaptive generation method of robot home track based on intelligent control according to claim 1, wherein the performing of unilateral contraction and contralateral compensation in combination with obstacle conflict to obtain a heading coupling home zone comprises the steps of: Dispersing the initial rectangular home region into a plurality of grid units, judging whether each grid unit meets the passing condition one by one, and marking the grid units which do not meet the passing condition as conflict grids; Based on conflict grids, determining rectangular sides with conflict grid sets as conflict sides, and calculating conflict depth of each conflict side; Gradually moving the conflict edge to the stop point position in the direction perpendicular to the conflict edge based on the conflict edge and the conflict depth until the moved rectangular boundary no longer corresponds to the conflict grid, so as to obtain a middle home region; calculating the size difference value of the middle home region relative to the initial rectangular home region, and expanding outwards in the direction of the rectangular edge opposite to the conflict edge according to the same grid step length until the compensated rectangular boundary reaches the expandable upper limit or the preset size requirement is restored, so as to obtain a candidate home region; And determining the rectangular boundary after the unilateral contraction and contralateral compensation is completed as a final boundary based on the candidate home region, and obtaining the heading coupling home region.
5. 5. The method for adaptively generating the robot home track based on intelligent control according to claim 4, wherein the integrated offset correction is performed on the heading coupling home region based on the heading coupling home region by fusing wall shielding, gradient change and signal attenuation, and comprises the following steps: extracting wall body distribution, gradient distribution and signal intensity distribution around a heading coupling home region, judging the unbalance degree of the heading coupling home region on two sides of the entering direction of a base station and in the front-back direction, and determining a limited direction; based on the limited direction, the heading coupling home returning area is gradually translated along the direction far away from the limited direction until the corrected home returning area meeting the requirements of wall avoidance, gradient adaptation and signal reliability is obtained.
6. 6. The method for adaptively generating the robot returning track based on intelligent control according to claim 5, wherein the step of screening candidate entrance gates on the modified boundary to obtain an optimal entrance gate comprises the following steps: based on the corrected home region, taking a rectangular edge of one side of the base station in-position orientation and rectangular edges of two sides adjacent to the base station in-position orientation as candidate boundaries; Dividing a plurality of continuous boundary sections along the candidate boundary according to a preset length, and judging passing conditions of adjacent grids corresponding to each continuous boundary section to obtain a candidate entry gate; Based on the candidate entrance doors, the passing distance from the current position of the robot to each candidate entrance door, the direction adjustment quantity from the candidate entrance door to the stop point position, the obstacle clearance in the vicinity of the candidate entrance door and the signal reliability in the vicinity of the candidate entrance door are calculated respectively, and the optimal entrance doors are obtained by sequencing.
7. 7. The method for adaptively generating the robot returning track based on intelligent control according to claim 6, wherein the constructing of the multi-factor cost map is to determine a start grid, a first target grid and a second target grid based on optimal entering doors and environment control data, and assign cost values to passable grids in a global occupied grid map based on obstacle risks, gesture adjustment costs, communication reliability and residual electric quantity, so as to obtain the multi-factor cost map.
8. 8. The method for adaptively generating the home track of the robot based on intelligent control according to claim 7, wherein the step of generating the adaptive home track by segments comprises the following steps: based on the multi-factor cost map, performing A x algorithm search by taking the current position as a starting point and taking the optimal entry gate as an end point to obtain a first section of track; Generating a second section of track grid by grid in the navigation coupling home region based on the first section of track, and judging whether the second section of track meets the passing condition corresponding to the environment control data or not until the second section of track meeting the passing condition is obtained; and splicing according to the driving sequence based on the first section track and the second section track meeting the passing condition to obtain the self-adaptive home track.
9. 9. The method for adaptively generating the robot home track based on intelligent control according to claim 8, wherein the method for controlling the robot home based on the adaptive home track detects pose deviation, dynamic obstacle and signal change in real time until accurate stop is completed, comprises the following steps: Determining a track tracking sequence, a positioning switching area and a stopping end point grid according to the self-adaptive home track, the current position, the optimal positioning gate and the stopping point position; Based on the track tracking sequence, controlling the robot to move along the self-adaptive home track, and switching to the next target grid after the current position enters the neighborhood of the current target grid to obtain a track tracking result; continuously detecting pose deviation, dynamic obstacle and signal change based on the track tracking result to obtain a track effectiveness judging result; Based on the track effectiveness judgment result, a running track close to the stop point position is obtained or a self-adaptive home track is regenerated until accurate stop is completed.
10. 10. The method for adaptively generating the robot home track based on intelligent control according to claim 9 is characterized in that the step of acquiring the running track close to the stop point position or regenerating the adaptive home track based on the track effectiveness judgment result is that the robot is continuously operated along the adaptive home track and driven into the stop point position along the follow-up track of the optimal entrance gate when the track is effective, and the step of stopping the robot movement when the track is invalid, rescreening the optimal entrance gate, reconstructing the multi-factor cost map and regenerating the adaptive home track.

Description

Self-adaptive generation method for robot home track based on intelligent control Technical Field The invention relates to the technical field of intelligent control, in particular to a robot home track self-adaptive generation method based on intelligent control. Background With the development of autonomous mobile devices such as yard service robots, cleaning robots, mowing robots and the like, robots return to a docking station and complete autonomous docking after completing work tasks, and the robots become important research contents in the field of mobile robot control. The existing conventional method generally generates a home-returning path by adopting a mode of combining a fixed return channel, global path planning or local obstacle avoidance control based on a positioning result, an environment map and preset stop point information, and controls a terminal stop process by combining the orientation of a charging base station, obstacle distribution and a communication state so as to realize the navigation, the approaching and the stop of the robot from an operation area to a stop area. However, the conventional method still mainly faces two adapting problems in practical application, namely firstly, a home-returning path and a parking area are generated according to fixed boundaries or static channels, when wall shielding, gradient change and irregular obstacle coexist, consistency between a path direction and a terminal entering direction is insufficient, secondly, cooperative consideration of communication reliability, residual electric quantity and posture adjustment cost by conventional path planning is weak, and therefore the adaptive capacity to dynamic environment change in the track execution process is limited. Disclosure of Invention The present invention has been made in view of the above-described problems occurring in the prior art. Therefore, the invention provides an intelligent control-based self-adaptive generation method for the home track of the robot, which solves the problems that the coupling between a path and the tail end in-place direction is insufficient and the self-adaptive adjustment capability of the track is limited under the multi-constraint condition in the existing home process of the robot. In order to solve the technical problems, the invention provides the following technical scheme: The invention provides an intelligent control-based self-adaptive generation method of a robot home track, which comprises the steps of collecting the current position, the stop point position, the base station entering direction, the courtyard boundary, obstacle distribution, real-time signal distribution and organism motion constraint of a robot to form environment control data, establishing a corresponding relation between a local coordinate system and a global occupied grid map coordinate system based on the environment control data, generating an initial rectangular home-returning area, executing unilateral contraction and contralateral compensation by combining obstacle conflicts to obtain a course coupling home-returning area, executing integral offset correction on the course coupling home-returning area based on the course coupling home-returning area, fusing wall shielding, gradient change and signal attenuation, screening candidate entering doors on the corrected boundary to obtain an optimal entering door, constructing a multi-factor cost map based on the optimal entering door, fusing obstacle risk, gesture adjustment cost, communication reliability and residual quantity, and segmenting to generate the self-adaptive home-returning track, and controlling the robot home-returning based on the self-adaptive home-returning track, and detecting gesture deviation, dynamic obstacle and signal change in real time until accurate stop is completed. The method for adaptively generating the robot home track based on intelligent control is characterized in that the forming of the environmental control data refers to uniformly writing the current position, the stop position, the base station entering direction, the courtyard boundary, the obstacle distribution, the real-time signal distribution and the body motion constraint of the robot into a global occupation grid map to form the environmental control data. The invention provides a robot home track self-adaptive generation method based on intelligent control, which is characterized in that based on the environmental control data, a corresponding relation between a local coordinate system and a global occupation grid map coordinate system is established, and an initial rectangular home region is generated, and the method comprises the following steps: Based on the environmental control data, establishing a local coordinate system, and establishing a corresponding relation between the local coordinate system and a global occupation grid map coordinate system by utilizing a rotation matrix; Determining principal direction half-length