# US 2024/0201700 A1 · Robot task execution method and apparatus, robot, and storage medium

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2024/0201700 A1
- Document ID: US20240201700A1
- Status: Published application
- Kind: A1
- Application number: 18/571,017
- Filed: 2022-06-15
- Published: 2024-06-20
- Earliest claimed: 2021-06-18
- Assignees: Beijing Indemind Technology Co., Ltd.
- Applicants: Beijing Indemind Technology Co., Ltd.
- Inventors: Dongkun YAN; Shuaishuai WANG; Dong FAN; Ziwen LIU
- Web page: https://patentgrep.com/patent/US20240201700A1

## CPC

- G05D 1/2297: {positional data taught by the user, e.g. paths (G05D1/2295 takes precedence)}
- G05D 1/2464: {using an occupancy grid}
- G05D 1/2295: {defining restricted zones, e.g. no-flight zones or geofences}
- G05D 1/6482: {by dividing the whole area or space in sectors to be processed separately}
- G05D 2105/10: for cleaning, vacuuming or polishing
- G05D 2107/40: Indoor domestic environment
- G05D 2109/10: Land vehicles
- G05D: SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES

## Abstract

The present application discloses a robot task execution method, apparatus, robot and storage medium. The method comprises: acquiring a training trajectory and an environment map in a training mode; generating a target region for tasks to be performed by a robot based on the environment map and the training trajectory, wherein the target region is a maximum envelope region in which the robot can complete tasks autonomously; controlling the robot to traverse the target region until the robot completes the tasks to be performed. By adopting the above technical solution, the robot can perform tasks stably and efficiently in various environmental regions, thereby being able to be applied to various application scenarios.

## Claims

### Claim 1

A robot task execution method, wherein, the robot task execution method comprises:
acquiring a training trajectory and an environment map in a training mode;
generating a target region for tasks to be performed by a robot based on the environment map and the training trajectory, wherein, the target region is a maximum envelope region in which the robot can complete the tasks autonomously;
controlling the robot to traverse the target region until the robot completes the tasks to be performed.

### Claim 2 (depends on 1)

The method according to claim 1, wherein, the step of generating a target region for tasks to be performed by the robot comprises: expanding the training trajectory within the environment map to obtain the maximum envelope region in which the robot can complete the tasks autonomously.

### Claim 3 (depends on 2)

The method according to claim 2, wherein, the step of expanding the training trajectory within the environment map to obtain the maximum envelope region in which the robot can complete the tasks autonomously comprises:
importing grid information of M layers of grids on a left side of the training trajectory and N layers of grids on a right side of the training trajectory into a newly established coverage region map, wherein, a one-time coverage width of the robot is equal to M+N+1, M and N are both integers greater than or equal to 1;
extracting an inner envelope and an outer envelope of a coverage region within the coverage region map, overlaying an inward expandable region of the inner envelope and an outward expandable region of the outer envelope, thereby obtaining the maximum envelope region in which the robot can complete the tasks autonomously.

### Claim 4 (depends on 1)

The method according to claim 1, wherein, the step of acquiring a training trajectory and an environment map in a training mode comprises:
when controlling the robot to start moving from an initial position, establishing a global coordinate system based on the initial position, and constructing a global map based on the global coordinate system;
during the movement of the robot, recording the training trajectory of the robot, and recording information of at least one target position on the training trajectory and information of business operation respectively corresponding to each target position;
binding relevant information in the global map, wherein the relevant information comprises: the training trajectory, the information of at least one target position, and the information of business operation respectively corresponding to each target position, which are bound to the global map.

### Claim 5 (depends on 4)

The method according to claim 4, wherein,
before the step of generating a target region for tasks to be performed by the robot, the method further comprises: receiving a first request instructing the robot to autonomously traverse the target region; and loading the global map and the relevant information bound to the global map;
the step of controlling the robot to traverse the target region until the robot completes the tasks to be performed comprises: re-executing a traversing-type path planning within the target region; and the robot autonomously tracking a path planned by the traversing-type path planning, and for each target position among the at least one target position, the robot performing the business operation corresponding to the target position when the robot reaches the target position.

### Claim 6 (depends on 5)

The method according to claim 5, wherein,
the step of recording the training trajectory of the robot comprises: sequentially recording the position information of each path point on the training trajectory of the robot according to a temporal sequence in which the robot traverses the path points;
the step of re-executing a traversing-type path planning within the target region comprises: for all target points corresponding to the information of at least one target position, dividing multiple target points into a set of target points respectively to obtain at least one set of target points, according to a distance between the target points on the training trajectory and/or according to information of association between the business operation respectively corresponding to each target position, and determining a traversal sequence for target points in each set of target points among the at least one set of target points according to the temporal sequence; when re-executing the traversing-type path planning, performing point-to-point path planning on the target points in each set of target points according to the traversal sequence for target points in each set of target points, or maintaining the training trajectory corresponding to each set of target points and the traversal sequence for target points in each set of target points in a re-planned path.

### Claim 7 (depends on 6)

The method according to claim 6, wherein, the step of dividing multiple target points into a set of target points respectively, for all target points corresponding to the information of at least one target position, according to a distance between the target points on the training trajectory and/or according to information of association between the business operation respectively corresponding to each target position comprises:
determining distances between every two closest target points on the training trajectory respectively for all target points corresponding to the information of at least one target position, and when the distances between every two closest target points among a plurality of adjacent target points are all less than a first predetermined distance threshold, classifying the plurality of adjacent target points to be target points that satisfy a first predetermined condition;
disassembling business functions of the robot to obtain at least one business operation, dividing business operations with association relationship into the same business function group, establishing a first business function library including one or more business function groups in advance, matching the recorded business operation respectively corresponding to each target position with the first business function library, performing grouping on the business operation respectively corresponding to each target position according to a result of the matching, and classifying target points that correspond to target positions corresponding to business operations that are determined to belong to the same business function group after the grouping to be target points that satisfy a second predetermined condition; and
dividing the target points that satisfy the first predetermined condition and/or satisfy the second predetermined condition into a set of target points.

### Claim 8 (depends on 4)

The method according to claim 4, wherein, after the step of acquiring a training trajectory and an environment map in a training mode, the method further comprises:
receiving a second request instructing the robot to autonomously traverse target positions;
loading the global map and the relevant information bound to the global map;
identifying one or more target positions that the robot needs to autonomously traverse and that correspond to the second business request;
adding the one or more target positions and business operations corresponding to the one or more target positions into a to-be-executed list;
performing path planning on target points corresponding to each target position in the to-be-executed list to obtain a motion trajectory for the robot to traverse the target positions;
the robot tracking the motion trajectory planned by the path planning, and when the robot reaches each of the one or more target positions, the robot performing business operations corresponding to each of the one or more target positions.

### Claim 9 (depends on 8)

The method according to claim 8, wherein,
the step of recording the training trajectory of the robot comprises: sequentially recording the position information of each path point on the training trajectory of the robot according to a temporal sequence in which the robot traverses the path points;
the step of performing path planning on target points corresponding to each target position in the to-be-executed list to obtain the motion trajectory for the robot to traverse the target positions comprises:
dividing multiple target points into a set of target points respectively to obtain at least one set of target points, for the target points corresponding to each target position in the to-be-executed list, according to a distance between the target points on the training trajectory and/or according to information of association between business operation respectively corresponding to each target position in the to-be-executed list, and determining a traversal sequence for target points in each set of target points among the at least one set of target points according to the temporal sequence;
when performing the path planning, for each set of target points among all the divided sets of target points, performing point-to-point path planning on each target point according to the traversal sequence for target points in each set of target points, or maintaining the training trajectory corresponding to each set of target points among all the sets of target points and the traversal sequence for target points in each set of target points in a planned path.

### Claim 10 (depends on 9)

The method according to claim 9, wherein, the step of dividing multiple target points into a set of target points respectively, for the target points corresponding to each target position in the to-be-executed list, according to a distance between the target points on the training trajectory and/or according to information of association between business operation corresponding to each target position in the to-be-executed list comprises:
determining distances between every two closest target points on the training trajectory respectively for the target points corresponding to each target position in the to-be-executed list, and when the distances between every two closest target points among a plurality of adjacent target points are all less than a second predetermined distance threshold, classifying the plurality of adjacent target points to be target points that satisfy a third predetermined condition;
disassembling business functions of the robot to obtain at least one business operation, dividing business operations with association relationship into the same business function group, establishing a second business function library including one or more business function groups in advance, matching the business operation respectively corresponding to each target position in the to-be-executed list with the second business function library, performing grouping on the business operation respectively corresponding to each target position in the to-be-executed list according to a result of the matching, and classifying target points that correspond to target positions corresponding to business operations that are determined to belong to the same business function group after the grouping to be target points that satisfy a fourth predetermined condition;
dividing the target points that satisfy the third predetermined condition and/or satisfy the fourth predetermined condition into a set of target points.

### Claim 11 (depends on 1)

The method according to claim 1, wherein, after the step of generating a target region for tasks to be performed by the robot, the method further comprises:
determining an outer boundary of the target region and determining a restricted area corresponding to the tasks to be performed;
setting restriction fences on the outer boundary and in the restricted area.

### Claim 12 (depends on 1)

The method of claim 1, wherein, the step of controlling the robot to traverse the target region until the robot completes the tasks to be performed comprises:
dividing the target region into multiple sub-regions;
controlling the robot to traverse each of the sub-regions to perform tasks until the robot completes the tasks to be performed in the target region.

### Claim 13 (depends on 12)

The method according to claim 12, wherein, the step of controlling the robot to traverse each of the sub-regions to perform tasks until the robot completes the tasks to be performed in the target region comprises:
determining a sequence of sub-regions for the robot to perform tasks in each of the sub-regions;
controlling the robot to traverse the sub-regions one by one according to the determined sequence of sub-regions until the robot completes the tasks to be performed in all the sub-regions.

### Claim 14 (depends on 12)

The method according to claim 12, wherein, the step of controlling the robot to traverse each of the sub-regions to perform tasks until the robot completes the tasks to be performed in the target region comprises:
S1: determining a nearest sub-region relative to the robot based on information of an initial position of the robot;
S2: controlling the robot to traverse the nearest sub-region to perform tasks;
S3: after completing the tasks in the nearest sub-region, determining the positioning information of an end point where the robot completes the tasks, and determining the next nearest sub-region relative to the end point;
S4: executing the steps S2 to S3 in a loop until the robot completes the tasks to be performed in all the sub-regions of the target region.

### Claim 15

(canceled)

### Claim 16 (depends on 1)

A robot, comprising a memory and a processor, wherein,
the memory is configured to store computer-executable instructions;
the processor is configured to execute the computer-executable instructions stored in the memory to cause the robot to perform the method of claim 1.

### Claim 17 (depends on 1)

A computer-readable storage medium, wherein, the computer-readable storage medium has computer-executable instructions stored therein, and the computer-executable instructions, when executed by a processor, causes the processor to perform the method of claim 1.

### Claim 18 (depends on 16)

The robot according to claim 16, wherein, the step of acquiring a training trajectory and an environment map in a training mode comprises:
when controlling the robot to start moving from an initial position, establishing a global coordinate system based on the initial position, and constructing a global map based on the global coordinate system;
during the movement of the robot, recording the training trajectory of the robot, and recording information of at least one target position on the training trajectory and information of business operation respectively corresponding to each target position;
binding relevant information in the global map, wherein the relevant information comprises: the training trajectory, the information of at least one target position, and the information of business operation respectively corresponding to each target position, which are bound to the global map.

### Claim 19 (depends on 16)

The robot according to claim 16, wherein, after the step of generating a target region for tasks to be performed by the robot, the processor is further configured to execute the computer-executable instructions stored in the memory to cause the robot to
determine an outer boundary of the target region and determine a restricted area corresponding to the tasks to be performed;
set restriction fences on the outer boundary and in the restricted area.

### Claim 20 (depends on 16)

The robot according to claim 16, wherein, the step of controlling the robot to traverse the target region until the robot completes the tasks to be performed comprises:
dividing the target region into multiple sub-regions;
controlling the robot to traverse each of the sub-regions to perform tasks until the robot completes the tasks to be performed in the target region.

### Claim 21 (depends on 17)

The computer-readable storage medium according to claim 17, wherein, the step of acquiring a training trajectory and an environment map in a training mode comprises:
when controlling the robot to start moving from an initial position, establishing a global coordinate system based on the initial position, and constructing a global map based on the global coordinate system;
during the movement of the robot, recording the training trajectory of the robot, and recording information of at least one target position on the training trajectory and information of business operation respectively corresponding to each target position;
binding relevant information in the global map, wherein the relevant information comprises: the training trajectory, the information of at least one target position, and the information of business operation respectively corresponding to each target position, which are bound to the global map.