# US 2025/0081958 A1 · Automatic obstacle avoidance method and system of pesticide application 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 2025/0081958 A1
- Document ID: US20250081958A1
- Status: Published application
- Kind: A1
- Application number: 18/465,685
- Filed: 2023-09-12
- Published: 2025-03-13
- Earliest claimed: 2023-09-12
- Assignees: Tobacco Research Institute of CAAS
- Applicants: Tobacco Research Institute of CAAS
- Inventors: Xiuguo WANG; Zhongfeng ZHANG; Guangwei REN; Xiao ZHENG; Wenjing MU; Tong LIU
- Web page: https://patentgrep.com/patent/US20250081958A1

## CPC

- A01M 7/0089: {Regulating or controlling systems (the delivery being related to the movement of a vehicle B05B9/06)}
- G01C 21/20: Instruments for performing navigational calculations (G01C21/24, G01C21/26 take precedence)
- G05D 1/0214: {in accordance with safety or protection criteria, e.g. avoiding hazardous areas (monitoring the location of vehicles within a certain area, e.g. forbidden or allowed areas, in traffic control systems for road vehicles G08G1/13)}
- G05D 1/0246: {using a video camera in combination with image processing means}
- G05D 1/0274: {using mapping information stored in a memory device (navigation using map-matching G01C21/30)}
- G05D 1/2464: {using an occupancy grid}
- G05D 1/622: Obstacle avoidance (predicting or avoiding probable or impending collision of road vehicles B60W30/08)
- G05D 1/644: Optimisation of travel parameters, e.g. of energy consumption, journey time or distance
- G05D 1/648: Performing a task within a working area or space, e.g. cleaning
- G05D 1/693: for avoiding collisions between vehicles
- G05D 2107/21: {Farming, e.g. fields, pastures or barns}
- A01M: CATCHING, TRAPPING OR SCARING OF ANIMALS (appliances for catching swarms or drone-catching A01K57/00; fishing A01K69/00 - {A01K99/00}; biocides, pest repellants or attractants A01N); APPARATUS FOR THE DESTRUCTION OF NOXIOUS ANIMALS OR NOXIOUS PLANTS
- G01C: MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY (measuring liquid level G01F; radio navigation, determining distance or velocity by use of propagation effects, e.g. Doppler effects, propagation time, of radio waves, analogous arrangements using other waves G01S)
- G05D: SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES

## Abstract

Provided are an automatic obstacle avoidance method and system of a pesticide application robot and a storage medium. The method includes: acquiring a plan view of a target pesticide application operation area and acquiring path information; acquiring start point information and end point information of a pesticide application robot and environmental information of the target pesticide application operation area, performing preliminary path planning, and acquiring an optimal path; causing the pesticide application robot to arrive at a specified operating point along the optimal path, and performing environmental perception by machine vision, determining obstacle point information and updating the plan view of the target pesticide application operation area with the obstacle point information; and correcting the optimal path with a current location of the pesticide application robot and an area to which a pesticide has been sprayed based on the updated plan view of the target pesticide application operation area.

## Claims

### Claim 1

An automatic obstacle avoidance method of a pesticide application robot, the automatic obstacle avoidance method comprising:
acquiring a plan view of a target pesticide application operation area and acquiring path information based on the plan view of the target pesticide application operation area:
acquiring start point information and end point information of a pesticide application robot and environmental information of the target pesticide application operation area, performing preliminary path planning based on the start point information, the end point information, and the environmental information, and acquiring an optimal path based on the preliminary path planning:
causing the pesticide application robot to arrive at a specified operating point along the optimal path, and performing environmental perception by machine vision during pesticide application operation, determining obstacle point information based on passability of the pesticide application robot, and updating the plan view of the target pesticide application operation area with the obstacle point information; and
correcting the optimal path with a current location of the pesticide application robot and an area to which a pesticide has been sprayed based on the updated plan view of the target pesticide application operation area.

### Claim 2 (depends on 1)

The automatic obstacle avoidance method according to claim 1, wherein the acquiring start point information and end point information of a pesticide application robot and environmental information of the target pesticide application operation area and performing preliminary path planning based on the start point information, the end point information, and the environmental information specifically comprises:
establishing a raster map of the target pesticide application operation area based on the plan view of the target pesticide application operation area and the environmental information, and presetting a speed constraint and a steering constraint of the pesticide application robot and a maximum safe distance of the pesticide application robot to an obstacle:
performing the preliminary path planning with a D* algorithm and constraint information based on start point information and end point information on the raster map, and setting a priority queue based on the end point information to perform backward searching in the raster map:
when an obstacle node is detected during searching, correcting path costs of neighboring raster nodes, and placing the neighboring raster nodes into the priority queue again until the current location of the pesticide application robot comes out of the priority queue; and
selecting from the neighboring raster nodes a node having a minimum path cost for connection to the start point information and then performing the preliminary path planning to generate an optimal path.

### Claim 3 (depends on 1)

The automatic obstacle avoidance method according to claim 1, wherein the performing environmental perception by machine vision during pesticide application operation, determining obstacle point information based on passability of the pesticide application robot, and updating the plan view of the target pesticide application operation area with the obstacle point information specifically comprises:
acquiring video image data during the pesticide application operation, preprocessing the video image data to extract a color feature and a texture feature, and performing edge segmentation on the preprocessed video image data by the color feature and the texture feature:
acquiring a binary image of the video image data by the edge segmentation, acquiring obstacle information in an ambient environment during the pesticide application operation, and acquiring contour data of an obstacle based on the binary image:
determining the passability of the pesticide application robot based on the contour data of the obstacle, size information of the pesticide application robot, and a preset safety distance; and
in case of determining that the pesticide application robot fails to pass through the obstacle, performing obstacle avoidance path planning, and updating the obstacle point information in the plan view of the target pesticide application operation area.

### Claim 4 (depends on 3)

The automatic obstacle avoidance method according to claim 3, wherein the determining the passability of the pesticide application robot based on the contour data of the obstacle, size information of the pesticide application robot, and a preset safety distance specifically comprises:
acquiring a size parameter of the obstacle based on the contour data of the obstacle, and comparing the size parameter of the obstacle with a maximum width or a maximum ground clearance of the pesticide application robot:
when the size parameter of the obstacle is greater than the maximum width or the maximum ground clearance, determining that the obstacle is impassable; and
when the obstacle is passable, determining a tilting probability of the pesticide application robot, and when an inclination angle of a body of the pesticide application robot is determined to be greater than a preset inclination angle, indicating that the pesticide application robot is prone to tilting and determining that the obstacle is impassable.

### Claim 5 (depends on 1)

The automatic obstacle avoidance method according to claim 1, wherein the correcting the optimal path with a current location of the pesticide application robot and an area to which a pesticide has been sprayed based on the updated plan view of the target pesticide application operation area specifically comprises:
performing obstacle avoidance path planning based on a relative position of the obstacle point information in the updated plan view of the target pesticide application operation area, acquiring current location information of the pesticide application robot, taking the current location information as a start point of the obstacle avoidance path planning, and setting a maximum pesticide application radius reachable point after an obstacle point in the optimal path as an end point of the obstacle avoidance path planning:
acquiring an area to which a pesticide has been applied before the obstacle point in a current path segment of the pesticide application robot in the optimal path, marking the area to which the pesticide has been applied, and determining whether an area to which the pesticide is not applied exists within the target pesticide application operation area based on the marked area:
if the area to which the pesticide is not applied exists, performing the obstacle avoidance path planning preferentially on the area to which the pesticide is not applied, and carrying out the pesticide application operation in the area to which the pesticide is not applied, and if no area to which the pesticide is not applied exists, performing the obstacle avoidance path planning according to a shortest path principle:
acquiring an overlapping area of an operation area corresponding to an obstacle avoidance path in the obstacle avoidance path planning with the marked area:
when the pesticide application robot is located in the overlapping area, stopping the pesticide application operation, and increasing a preset speed increment on the basis of an original operating speed to pass through the overlapping area; and
updating the area to which the pesticide has been applied according to the obstacle avoidance path in the obstacle avoidance path planning, and correcting the optimal path based on the area to which the pesticide has been applied and a location of the pesticide application robot after obstacle avoidance.

### Claim 6 (depends on 1)

The automatic obstacle avoidance method according to claim 1, further comprising:
when a plurality of pesticide application robots are present in the target pesticide application operation area, displaying optimal paths and real-time location information of the plurality of pesticide application robots in the plan view of the target pesticide application operation area:
determining whether collision occurs or whether pesticide application operation areas overlap according to the optimal paths based on current location information and motion speed information of the plurality of pesticide application robots:
when detecting that a current pesticide application robot is to collide with a target pesticide application robot, setting a waiting time of the current pesticide application robot based on size information and motion speed information of the target pesticide application robot, and causing the current pesticide application robot to wait in situ for the waiting time until the target pesticide application robot passes:
if the target pesticide application robot has no certain motion trajectory, performing the obstacle avoidance path planning with a possible motion area for the target pesticide application robot at next point of time as an obstacle area, the current location of the current pesticide application robot as a planning start point, and a nearest point in the optimal path of the current pesticide application robot after the obstacle area as an obstacle avoidance end point; and
when operation areas of the plurality of pesticide application robots have an overlapping area, taking the overlapping area as the obstacle area for any pesticide application robot and performing secondary planning on the optimal path thereof in combination with the current location information thereof.

### Claim 7

An automatic obstacle avoidance system of a pesticide application robot, the automatic obstacle avoidance system comprising:
a memory and a processor, wherein the memory comprises an automatic obstacle avoidance method program of a pesticide application robot, and the automatic obstacle avoidance method program of a pesticide application robot, when executed by the processor, performs the following method:
acquiring a plan view of a target pesticide application operation area and acquiring path information based on the plan view of the target pesticide application operation area:
acquiring start point information and end point information of a pesticide application robot and environmental information of the target pesticide application operation area, performing preliminary path planning based on the start point information, the end point information, and the environmental information, and acquiring an optimal path based on the preliminary path planning:
causing the pesticide application robot to arrive at a specified operating point along the optimal path, and performing environmental perception by machine vision during pesticide application operation, determining obstacle point information based on passability of the pesticide application robot, and updating the plan view of the target pesticide application operation area with the obstacle point information; and
correcting the optimal path with a current location of the pesticide application robot and an area to which a pesticide has been sprayed based on the updated plan view of the target pesticide application operation area.

### Claim 8 (depends on 7)

The automatic obstacle avoidance system according to claim 7, wherein the acquiring start point information and end point information of a pesticide application robot and environmental information of the target pesticide application operation area and performing preliminary path planning based on the start point information, the end point information, and the environmental information specifically comprises:
establishing a raster map of the target pesticide application operation area based on the plan view of the target pesticide application operation area and the environmental information, and presetting a speed constraint and a steering constraint of the pesticide application robot and a maximum safe distance of the pesticide application robot to an obstacle:
performing the preliminary path planning with a D* algorithm and constraint information based on start point information and end point information on the raster map, and setting a priority queue based on the end point information to perform backward searching in the raster map:
when an obstacle node is detected during searching, correcting path costs of neighboring raster nodes, and placing the neighboring raster nodes into the priority queue again until the current location of the pesticide application robot comes out of the priority queue; and
selecting from the neighboring raster nodes a node having a minimum path cost for connection to the start point information and then performing the preliminary path planning to generate an optimal path.

### Claim 9 (depends on 7)

The automatic obstacle avoidance system according to claim 7, wherein the correcting the optimal path with a current location of the pesticide application robot and an area to which a pesticide has been sprayed based on the updated plan view of the target pesticide application operation area specifically comprises:
performing obstacle avoidance path planning based on a relative position of the obstacle point information in the updated plan view of the target pesticide application operation area, acquiring current location information of the pesticide application robot, taking the current location information as a start point of the obstacle avoidance path planning, and setting a maximum pesticide application radius reachable point after an obstacle point in the optimal path as an end point of the obstacle avoidance path planning:
acquiring an area to which a pesticide has been applied before the obstacle point in a current path segment of the pesticide application robot in the optimal path, marking the area to which the pesticide has been applied, and determining whether an area to which the pesticide is not applied exists within the target pesticide application operation area based on the marked area:
if the area to which the pesticide is not applied exists, performing the obstacle avoidance path planning preferentially on the area to which the pesticide is not applied, and carrying out the pesticide application operation in the area to which the pesticide is not applied, and if no area to which the pesticide is not applied exists, performing the obstacle avoidance path planning according to a shortest path principle:
acquiring an overlapping area of an operation area corresponding to an obstacle avoidance path in the obstacle avoidance path planning with the marked area:
when the pesticide application robot is located in the overlapping area, stopping the pesticide application operation, and increasing a preset speed increment on the basis of an original operating speed to pass through the overlapping area; and
updating the area to which the pesticide has been applied according to the obstacle avoidance path in the obstacle avoidance path planning, and correcting the optimal path based on the area to which the pesticide has been applied and a location of the pesticide application robot after obstacle avoidance.

### Claim 10 (depends on 1)

A computer-readable storage medium, comprising an automatic obstacle avoidance method program of a pesticide application robot, wherein the automatic obstacle avoidance method program of a pesticide application robot, when executed by the processor, performs the steps of the automatic obstacle avoidance method of a pesticide application robot according to claim 1.

### Claim 11 (depends on 10)

The computer-readable storage medium according to claim 10, wherein the acquiring start point information and end point information of a pesticide application robot and environmental information of the target pesticide application operation area and performing preliminary path planning based on the start point information, the end point information, and the environmental information specifically comprises:
establishing a raster map of the target pesticide application operation area based on the plan view of the target pesticide application operation area and the environmental information, and presetting a speed constraint and a steering constraint of the pesticide application robot and a maximum safe distance of the pesticide application robot to an obstacle:
performing the preliminary path planning with a D* algorithm and constraint information based on start point information and end point information on the raster map, and setting a priority queue based on the end point information to perform backward searching in the raster map:
when an obstacle node is detected during searching, correcting path costs of neighboring raster nodes, and placing the neighboring raster nodes into the priority queue again until the current location of the pesticide application robot comes out of the priority queue; and
selecting from the neighboring raster nodes a node having a minimum path cost for connection to the start point information and then performing the preliminary path planning to generate an optimal path.

### Claim 12 (depends on 10)

The computer-readable storage medium according to claim 10, wherein the performing environmental perception by machine vision during pesticide application operation, determining obstacle point information based on passability of the pesticide application robot, and updating the plan view of the target pesticide application operation area with the obstacle point information specifically comprises:
acquiring video image data during the pesticide application operation, preprocessing the video image data to extract a color feature and a texture feature, and performing edge segmentation on the preprocessed video image data by the color feature and the texture feature:
acquiring a binary image of the video image data by the edge segmentation, acquiring obstacle information in an ambient environment during the pesticide application operation, and acquiring contour data of an obstacle based on the binary image:
determining the passability of the pesticide application robot based on the contour data of the obstacle, size information of the pesticide application robot, and a preset safety distance; and
in case of determining that the pesticide application robot fails to pass through the obstacle, performing obstacle avoidance path planning, and updating the obstacle point information in the plan view of the target pesticide application operation area.

### Claim 13 (depends on 12)

The computer-readable storage medium according to claim 12, wherein the determining the passability of the pesticide application robot based on the contour data of the obstacle, size information of the pesticide application robot, and a preset safety distance specifically comprises:
acquiring a size parameter of the obstacle based on the contour data of the obstacle, and comparing the size parameter of the obstacle with a maximum width or a maximum ground clearance of the pesticide application robot:
when the size parameter of the obstacle is greater than the maximum width or the maximum ground clearance, determining that the obstacle is impassable; and
when the obstacle is passable, determining a tilting probability of the pesticide application robot, and when an inclination angle of a body of the pesticide application robot is determined to be greater than a preset inclination angle, indicating that the pesticide application robot is prone to tilting and determining that the obstacle is impassable.

### Claim 14 (depends on 10)

The computer-readable storage medium according to claim 10, wherein the correcting the optimal path with a current location of the pesticide application robot and an area to which a pesticide has been sprayed based on the updated plan view of the target pesticide application operation area specifically comprises:
performing obstacle avoidance path planning based on a relative position of the obstacle point information in the updated plan view of the target pesticide application operation area, acquiring current location information of the pesticide application robot, taking the current location information as a start point of the obstacle avoidance path planning, and setting a maximum pesticide application radius reachable point after an obstacle point in the optimal path as an end point of the obstacle avoidance path planning:
acquiring an area to which a pesticide has been applied before the obstacle point in a current path segment of the pesticide application robot in the optimal path, marking the area to which the pesticide has been applied, and determining whether an area to which the pesticide is not applied exists within the target pesticide application operation area based on the marked area:
if the area to which the pesticide is not applied exists, performing the obstacle avoidance path planning preferentially on the area to which the pesticide is not applied, and carrying out the pesticide application operation in the area to which the pesticide is not applied, and if no area to which the pesticide is not applied exists, performing the obstacle avoidance path planning according to a shortest path principle:
acquiring an overlapping area of an operation area corresponding to an obstacle avoidance path in the obstacle avoidance path planning with the marked area:
when the pesticide application robot is located in the overlapping area, stopping the pesticide application operation, and increasing a preset speed increment on the basis of an original operating speed to pass through the overlapping area; and
updating the area to which the pesticide has been applied according to the obstacle avoidance path in the obstacle avoidance path planning, and correcting the optimal path based on the area to which the pesticide has been applied and a location of the pesticide application robot after obstacle avoidance.

### Claim 15 (depends on 10)

The computer-readable storage medium according to claim 10, further comprising:
when a plurality of pesticide application robots are present in the target pesticide application operation area, displaying optimal paths and real-time location information of the plurality of pesticide application robots in the plan view of the target pesticide application operation area:
determining whether collision occurs or whether pesticide application operation areas overlap according to the optimal paths based on current location information and motion speed information of the plurality of pesticide application robots:
when detecting that a current pesticide application robot is to collide with a target pesticide application robot, setting a waiting time of the current pesticide application robot based on size information and motion speed information of the target pesticide application robot, and causing the current pesticide application robot to wait in situ for the waiting time until the target pesticide application robot passes:
if the target pesticide application robot has no certain motion trajectory, performing the obstacle avoidance path planning with a possible motion area for the target pesticide application robot at next point of time as an obstacle area, the current location of the current pesticide application robot as a planning start point, and a nearest point in the optimal path of the current pesticide application robot after the obstacle area as an obstacle avoidance end point; and
when operation areas of the plurality of pesticide application robots have an overlapping area, taking the overlapping area as the obstacle area for any pesticide application robot and performing secondary planning on the optimal path thereof in combination with the current location information thereof.
