# US 2026/0198416 A1 · Autonomous machine having vision system for navigation and method of using same

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

- Number: US 2026/0198416 A1
- Document ID: US20260198416A1
- Status: Published application
- Kind: A1
- Application number: 19/550,729
- Filed: 2026-02-26
- Published: 2026-07-16
- Earliest claimed: 2020-07-02
- Assignees: The Toro Company
- Applicants: The Toro Company
- Inventors: Alexander Steven Frick; Michael Jason Ramsay; David Arthur LaRose; Stephen Paul Elizondo Landers; Zachary Irvin Parker; David Ian Robinson; Christopher Charles Osterwood
- Web page: https://patentgrep.com/patent/US20260198416A1

## CPC

- A01D 34/008: {for automated or remotely controlled operation}
- G05D 1/0253: {extracting relative motion information from a plurality of images taken successively, e.g. visual odometry, optical flow (determining position or orientation from images G06T7/70)}
- G05D 1/2437: {Extracting relative motion information}
- G06T 7/215: Motion-based segmentation
- G06T 7/246: using feature-based methods, e.g. the tracking of corners or segments
- G06V 20/10: Terrestrial scenes (scenes under surveillance with static cameras G06V20/52; scenes perceived from the exterior of a vehicle G06V20/56; scenes perceived from the interior of a vehicle G06V20/59)
- G06V 20/56: exterior to a vehicle by using sensors mounted on the vehicle
- G06V 20/64: Three-dimensional [3D] objects
- A01D 2101/00: Lawn-mowers
- G06T 2207/10028: Range image; Depth image; 3D point clouds
- A01D: HARVESTING; MOWING
- G05D: SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

Vision systems for autonomous machines and methods of using same during machine localization are provided. Exemplary systems and methods may reduce computing resources needed to perform vision-based localization by selecting the most appropriate camera from two or more cameras, and optionally selecting only a portion of the selected camera's field of view, from which to perform vision-based location correction. Other embodiments may provide camera lens coverings that maintain optical clarity while operating within debris-filled environments.

## Claims

### Claim at position 1

1-17. (canceled)

### Claim 18

An autonomous ground maintenance machine comprising:
a housing supporting a ground maintenance implement;
a propulsion system;
ground support members configured to support the housing on a ground surface of a work region, at least one of the ground support members operably coupled to and configured to be driven by the propulsion system;
a plurality of vision sensors defining different respective fields of view (FOVs) relative to the autonomous ground maintenance machine, the plurality of vision sensors including a first vision sensor configured to capture image data within a first FOV;
a controller communicatively coupled to each of the plurality of vision sensors, the controller configured to:
select, from among the plurality of vision sensors during operation of the autonomous ground maintenance machine, the first vision sensor as a localization vision sensor, wherein the first FOV includes a vertical FOV divided into a first zone and a second zone separated by a virtual boundary; and
determine a vision-based pose of the autonomous ground maintenance machine relative to the work region by matching features in the image data captured by the first vision sensor within the first zone to features associated with a previously identified three-dimensional point cloud (3DPC) used to define the work region.

### Claim 19 (depends on 18)

The autonomous ground maintenance machine of claim 18, wherein the controller is configured to match the features in the image data captured within the first zone without matching features in image data captured within the second zone to the features associated with the 3DPC.

### Claim 20 (depends on 18)

The autonomous ground maintenance machine of claim 18, wherein the first zone is located vertically above the second zone.

### Claim 21 (depends on 18)

The autonomous ground maintenance machine of claim 18, wherein the virtual boundary is adjustable by the controller.

### Claim 22 (depends on 21)

The autonomous ground maintenance machine of claim 21, wherein the controller is configured to adjust the virtual boundary based on an estimated pose of the autonomous ground maintenance machine.

### Claim 23 (depends on 18)

The autonomous ground maintenance machine of claim 18, wherein the controller is configured to generate visual odometry data based on image data captured within the second zone.

### Claim 24 (depends on 18)

The autonomous ground maintenance machine of claim 18, wherein the plurality of vision sensors further includes a second vision sensor configured to capture image data within a second FOV, wherein the second FOV includes a vertical FOV divided into a first zone and a second zone separated by a virtual boundary.

### Claim 25 (depends on 24)

The autonomous ground maintenance machine of claim 24, wherein the controller is configured to generate visual odometry data based on at least one of image data captured by the first vision sensor within the second zone of the first FOV or image data captured by the second vision sensor within the second zone of the second FOV.

### Claim 26 (depends on 25)

The autonomous ground maintenance machine of claim 25, further comprising an odometry sensor configured to generate odometry data corresponding to movement of the at least one ground support member, wherein the controller is configured to determine whether the at least one ground support member has slipped based on a difference between the visual odometry data and the odometry data.

### Claim 27 (depends on 24)

The autonomous ground maintenance machine of claim 24, wherein the controller is configured to determine a vision-based displacement of the autonomous ground maintenance machine based on movement of features between sequential images captured within at least one of the second zone of the first FOV or the second zone of the second FOV.

### Claim 28 (depends on 27)

The autonomous ground maintenance machine of claim 27, wherein the controller is configured to determine the vision-based displacement based on optical flow between the sequential images.

### Claim 29 (depends on 24)

The autonomous ground maintenance machine of claim 24, wherein the controller is configured to select, from among the plurality of vision sensors after displacement of the machine, the second vision sensor as the localization vision sensor.

### Claim 30 (depends on 18)

The autonomous ground maintenance machine of claim 18, wherein the controller is configured to determine that image data captured by the first vision sensor fails to satisfy a localization-quality criterion and, in response, select a second vision sensor of the plurality of vision sensors as the localization vision sensor.

### Claim 31 (depends on 30)

The autonomous ground maintenance machine of claim 30, wherein the localization-quality criterion comprises at least one of a threshold number of feature matches or a threshold recognition score.

### Claim 32

An autonomous ground maintenance machine comprising:
a housing supporting a ground maintenance implement;
a propulsion system;
ground support members configured to support the housing on a ground surface of a work region, at least one of the ground support members operably coupled to and configured to be driven by the propulsion system;
a first vision sensor configured to capture image data within a first field of view (FOV);
a second vision sensor configured to capture image data within a second FOV that is different than the first FOV;
a non-vision-based sensor configured to provide location data; and
a controller communicatively coupled to each of the first vision sensor, the second vision sensor, and the non-vision-based sensor, the controller configured to:
determine an estimated pose of the autonomous ground maintenance machine relative to the work region based on the location data provided by the non-vision-based sensor;
select, from among the first and second vision sensors during operation of the autonomous ground maintenance machine, the first vision sensor as a localization vision sensor, wherein the first FOV includes a vertical FOV divided into a first zone and a second zone separated by a virtual boundary;
determine a vision-based pose of the autonomous ground maintenance machine relative to the work region by matching features in the image data captured by the first vision sensor within the first zone to features associated with a previously identified three-dimensional point cloud (3DPC) used to define the work region; and
determine an updated pose of the autonomous maintenance machine relative to the work region based on the estimated pose and the vision-based pose.

### Claim 33 (depends on 32)

The autonomous ground maintenance machine of claim 32, wherein the controller is configured to determine the updated pose by correcting the estimated pose based on the vision-based pose.

### Claim 34 (depends on 32)

The autonomous ground maintenance machine of claim 32, wherein the controller is configured to select the first vision sensor as the localization vision sensor based at least in part on the estimated pose.

### Claim 35 (depends on 32)

The autonomous ground maintenance machine of claim 32, wherein the controller is configured to match the features in the image data captured within the first zone without matching features in image data captured within the second zone to the features associated with the 3DPC.

### Claim 36 (depends on 32)

The autonomous ground maintenance machine of claim 32, wherein the controller is configured to generate visual odometry data based on image data captured within the second zone.

### Claim 37 (depends on 32)

The autonomous ground maintenance machine of claim 32, wherein the second FOV of the second vision sensor includes a vertical FOV divided into a first zone and a second zone separated by a virtual boundary, and wherein the controller is configured to generate visual odometry data based on image data captured within the second zone of the second FOV.
