Method for robot repositioning

A method for repositioning a robot, comprising the following steps:

• Step S1:
• detecting, by the robot, an obstacle, and entering step S2;
• Step S2:
• walking along an edge of the obstacle, judging whether a path along the edge of the obstacle meets a condition of determining that the obstacle is an isolated object or not, when the path does not meet the condition of determining that the obstacle is the isolated object, entering step S3, and when the path meets the condition of determining that the obstacle is the isolated object, entering step S4;
• Step S3:
• adjusting a walking angle of the robot to leave the obstacle and continue walking, and when another obstacle is detected, re-executing step S2;
• Step S4:
• determining that the obstacle is the isolated object, recording an along-edge path walking along an edge of the isolated object, judging whether the presently recorded along-edge path is similar to an along-edge path that is stored before or not, when the presently recorded along-edge path is not similar to the along-edge path that is stored before, entering step S5, and when the presently recorded along-edge path is not similar to the along-edge path that is stored before, entering step S6;
• Step S5:
• determining the recorded along-edge path walking along the edge of the isolated object as a stored along-edge path, and jumping to step S3;
• and
• Step S6:
• determining the along-edge path that is stored before and similar to the presently recorded along-edge path as a reference positioning path, determining a first local map where the present along-edge path is located and a second local map where the reference positioning path is located, overlapping the first local map and the second local map with a same shape and size, determining a grid cell at a same location in the reference positioning path and the present along-edge path as a positioning cell, overlapping the reference positioning path in the second local map, in the present along-edge path in the first local map as at least one positioning cell, and replacing a grid coordinate value presently detected when the robot is in each positioning cell with a grid coordinate value of a corresponding grid cell in the reference positioning path;
• after step S1 and before step S2, the method further comprising the following steps:
• Step S11:
• determining a corresponding grid coordinate value when the robot detects the obstacle;
• Step S12:
• determining an along-edge path stored by the robot in a preset time period counted back from a present time point;
• Step S13:
• judging whether the grid coordinate value determined in step S11 is the same as or adjacent to a grid coordinate value corresponding to the along-edge path determined in step S12 or not, when the grid coordinate value determined in step S11 is the same as or adjacent to the grid coordinate value corresponding to the along-edge path determined in step S12, entering step S14, and when the grid coordinate value determined in step S11 is not the same as or adjacent to the grid coordinate value corresponding to the along-edge path determined in step S12, entering step S2;
• and
• Step S14:
• adjusting the walking angle of the robot to leave the obstacle and continue walking, and when another obstacle is detected, re-executing step S11,
• wherein the adjacent mentioned when two grid cells corresponding to the two grid coordinates have a common edge or an angular point, they are considered adjacent.