CN-121979208-A - Robot automatic recharging control method and intelligent mowing robot

CN121979208ACN 121979208 ACN121979208 ACN 121979208ACN-121979208-A

Abstract

The application relates to the technical field of robot recharging control, and provides an automatic recharging control method of a robot and an intelligent mowing robot, wherein the method comprises the steps of monitoring whether an algorithm switching triggering condition is met in real time in the process of aligning and positioning based on a target positioning algorithm; and when the trigger condition is determined to be met, re-determining a target positioning algorithm from other positioning algorithms, and performing alignment positioning based on the new target positioning algorithm, wherein the trigger condition comprises one or more of incomplete interface docking exceeding a preset time threshold, interface docking failure times exceeding a preset time threshold, or abnormal positioning data. Based on the method, continuous positioning can be kept under various complex scenes such as visual failure, too far distance or local environmental change, and the success rate of automatic recharging and environmental adaptability are improved.

Inventors

ZHANG XIAOLIN
HUANG XINYONG

Assignees

深圳市园睛智能科技有限公司

Dates

Publication Date: 20260505
Application Date: 20260115

Claims (10)

1. The automatic recharging control method of the robot is characterized by being applied to the robot, and comprises the following steps: When the charging interface moves into the guiding range, determining a target positioning algorithm from a plurality of positioning algorithms, aligning and positioning based on the target positioning algorithm, and adjusting the pose in the guiding range according to the positioning result so as to complete the butt joint of the charging interface; Monitoring whether the algorithm switching trigger condition is satisfied in real time in the process of aligning and positioning based on the target positioning algorithm, re-determining the target positioning algorithm from other positioning algorithms when the algorithm switching trigger condition is satisfied, aligning and positioning based on the new target positioning algorithm, wherein, The triggering condition includes one or more of incomplete interface interfacing exceeding a preset time threshold, interface interfacing failure times exceeding a preset number of times threshold, or locating data anomalies.
2. The control method according to claim 1, wherein the method for performing alignment positioning based on the target positioning algorithm, and performing pose adjustment within the guiding range according to a positioning result, comprises: monitoring whether the device enters a preset alignment positioning area in real time; re-determining target positioning data based on the target positioning algorithm if entry into the alignment positioning area is determined; the pose is adjusted based on the target positioning data.
3. The control method according to claim 2, wherein the re-targeting data based on the targeting algorithm comprises: acquiring a plurality of groups of positioning data in a preset positioning period based on the target positioning algorithm; Maintaining a sliding time window, and calculating the deviation between each group of positioning data and the mean value of the time window to which the positioning data belong; if the deviation is greater than a first preset deviation threshold, determining that the positioning data is an abnormal value; Removing the abnormal value; and carrying out weighted average on the reserved positioning data to serve as the target positioning data.
4. The control method according to claim 3, wherein the acquiring a plurality of sets of positioning data within a preset positioning period based on the target positioning algorithm includes: And in the preset positioning period, when the effective input value required by the target positioning algorithm is acquired, the target positioning algorithm is kept still, and the plurality of groups of positioning data are acquired according to a preset time interval.
5. The control method according to claim 2, wherein a plurality of the alignment positioning areas are defined in the guide range, each of the alignment positioning areas being divided according to a distance from the charging stake.
6. The control method according to claim 1, wherein the method of re-determining the target positioning algorithm from the other positioning algorithms includes: selecting the highest preset priority in the other positioning algorithms as a new target positioning algorithm, or The method comprises the steps of obtaining real-time positioning data determined based on each positioning algorithm, determining the similarity of positioning results of each positioning algorithm based on the real-time positioning data, and determining the highest similarity of the positioning results as a new target positioning algorithm, wherein the real-time positioning data is obtained by executing calculation of each positioning algorithm in real time after moving to a guiding range, and the similarity of the positioning results is determined based on intra-group similarity or inter-group similarity.
7. The control method according to claim 1, wherein the image recognition result dependency degree of each positioning algorithm on the charging interface positioning identifier is different.
8. The control method according to claim 1, wherein the method for alignment positioning based on one of the positioning algorithms comprises: acquiring RGB images of surrounding environment; invoking a deep learning segmentation model to identify a charging interface positioning identification region in the RGB image so as to obtain pixel coordinates of the charging interface positioning identification region; Generating point cloud data of the charging pile; extracting a point cloud subset of the charging interface positioning identification area from the point cloud data according to the pixel coordinates; Preprocessing the point cloud subset, performing plane fitting, and extracting the central point of a fitting plane and the long-side and short-side directions of the charging interface positioning mark; and determining the pose and the position of the charging interface positioning mark in a camera coordinate system by using the center point, the long side direction and the short side direction, and converting the information of the charging interface positioning mark in the camera coordinate system into relative pose information of a charging pile charging interface through coordinate conversion.
9. The control method of claim 1, wherein the positioning data anomalies include one or more of a number of positioning hops exceeding a preset hopping threshold, a failure to obtain a valid input value, or a fuselage positioning outside of the guidance range.
10. Intelligent mowing robot, characterized in that it realizes automatic recharging based on the control method of any one of claims 1 to 9.

Description

Robot automatic recharging control method and intelligent mowing robot Technical Field The application relates to the technical field of robot recharging control, in particular to an automatic recharging control method for a robot and an intelligent mowing robot. Background Along with the popularization of service robots, an automatic recharging function becomes standard, and the automatic recharging function refers to the capability of the robots to return to a charging pile and charge automatically when the electric quantity is insufficient, which is a key link for guaranteeing continuous and stable work of the robots. Currently, the automatic recharging function of a robot depends on a single positioning algorithm, which results in poor positioning reliability. In particular, single sensor positioning algorithms are susceptible to interference from environmental factors, for example, vision algorithms are greatly affected by illumination, characteristics are lost due to overexposure and blurriness of strong light, positioning is inaccurate, and ultrasonic algorithms are prone to generating false signals due to obstacle reflection. In addition, a single positioning algorithm is difficult to adapt to complex and changeable scenes, and the performance of the single positioning algorithm is reduced due to environmental layout changes, weather influences and the like. Meanwhile, the precision and the robustness of the method are difficult to be compatible, the positioning drift is easy to occur when the method faces to noise, data loss and other anomalies, and the fault recovery capability is weak. Along with the widening of the application field of robots, the recharging positioning requirements are improved, the limitation of a single positioning algorithm limits the wide application of the robots, and improvement is needed. Disclosure of Invention In order to solve the technical problems of a single positioning algorithm, the embodiment of the application provides an automatic recharging control method of a robot, which is applied to the robot and comprises the steps of determining a target positioning algorithm from a plurality of positioning algorithms when the robot moves to a guiding range; the method comprises the steps of carrying out alignment positioning based on a target positioning algorithm, carrying out pose adjustment in a guiding range according to a positioning result to finish docking of a charging interface, monitoring whether an algorithm switching trigger condition is met in real time in the alignment positioning process based on the target positioning algorithm, and when the algorithm switching trigger condition is met, re-determining the target positioning algorithm from other positioning algorithms and carrying out alignment positioning based on a new target positioning algorithm, wherein the trigger condition comprises one or more of incomplete interface docking exceeding a preset time threshold, interface docking failure times exceeding a preset times threshold or abnormal positioning data. Based on the technical scheme, by applying various positioning algorithms, monitoring the abnormal condition of the target positioning algorithm in real time, switching the algorithm in time when the abnormality is found, and reselecting a new target algorithm from non-target positioning algorithms, the fault recovery capability is improved, and as the requirements of different positioning algorithms on input values are different, the dynamic change of the actual application scene can be adapted by specific selection of the algorithms in combination with the actual application scene, so that the environment adaptability is improved. In one implementation, the method for aligning and positioning based on the target positioning algorithm and adjusting the pose in the guiding range according to the positioning result comprises the steps of monitoring whether a preset aligning and positioning area is entered in real time, re-determining target positioning data based on the target positioning algorithm when the entering of the aligning and positioning area is determined, and adjusting the pose based on the target positioning data. Based on the technical scheme, the alignment positioning area is divided, alignment positioning is controlled to be carried out again in the area, and then pose adjustment can be carried out after a stable and accurate positioning result is determined, so that frequent invalid adjustment caused by unstable algorithm in the process of positioning and adjusting is avoided. Based on the method, a plurality of alignment positioning areas can be divided according to the actual situation of the application scene, and the positioning and the pose adjustment in stages can be realized, so that the positioning accuracy and the fault tolerance are improved. In one implementation, the re-targeting data based on the targeting algorithm includes obtaining multiple sets of positioning data