# US 2025/0146835 A1 · Systems and methods for estimating lane boundaries using a slicing model

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

- Number: US 2025/0146835 A1
- Document ID: US20250146835A1
- Status: Published application
- Kind: A1
- Application number: 18/504,693
- Filed: 2023-11-08
- Published: 2025-05-08
- Earliest claimed: 2023-11-08
- Assignees: Woven By Toyota, Inc.
- Applicants: Woven by Toyota, Inc.
- Inventors: Shunsho Kaku; Jeffrey M. Walls; Ryan M. Wiesenberg
- Web page: https://patentgrep.com/patent/US20250146835A1

## CPC

- G01C 21/3819: {Road shape data, e.g. outline of a route}
- G01C 21/3833: {characterised by the source of data}
- G06V 10/82: using neural networks
- G06V 20/588: {Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road}
- 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)
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

Systems, methods, and other embodiments described herein relate to estimating lane boundaries using a slicing model with road data for generating maps. In one embodiment, a method includes computing three-dimensional (3D) representations that are discretized from acquired data about road edges associated with driving lanes. The method also includes deriving discrete and lateral slices of the road edges using a slicing model, the road edges are connected in a road graph that describes a mapped area. The method also includes extracting features from the lateral slices individually using a neural model for forming a histogram to estimate lane boundaries about the driving lanes. The method also includes generating a map by linking the lane boundaries individually along the road edges.

## Claims

### Claim 1

An estimation system comprising:
a processor; and
a memory storing instructions that, when executed by the processor, cause the processor to:
compute three-dimensional (3D) representations that are discretized from acquired data about road edges associated with driving lanes;
derive discrete and lateral slices of the road edges using a slicing model, and the road edges are connected in a road graph that describes a mapped area;
extract features from the lateral slices individually using a neural model for forming a histogram to estimate lane boundaries about the driving lanes; and
generate a map by linking the lane boundaries individually along the road edges.

### Claim 2 (depends on 1)

The estimation system of claim 1 further including instructions to output by a decoder confidence values and boundary placements for the lane boundaries through counting compressed data within bins of the histogram and the bins are each associated with one of the lateral slices, wherein the neural model includes the decoder.

### Claim 3 (depends on 2)

The estimation system of claim 2, wherein the instructions to generate the map further include instructions to translate the map using an inverse distance to the lane boundaries from the confidence values and the boundary placements.

### Claim 4 (depends on 1)

The estimation system of claim 1, wherein the instructions to extract the features further include instructions to:
process detected keypoints of the driving lanes using matrix calculations for different channels from the acquired data, wherein the matrix calculations factor kernel size per layer of the neural model and the different channels fill gaps of the lateral slices; and
fitting non-linearities from the matrix calculations to estimate the lane boundaries.

### Claim 5 (depends on 4)

The estimation system of claim 4, wherein the layer of the neural model includes trace points of a vehicle within one or more of the lateral slices and the trace points identify lane directionality.

### Claim 6 (depends on 4)

The estimation system of claim 4 further including instructions to select the features using the neural model according to a distance between compressed data in bins of the histogram.

### Claim 7 (depends on 1)

The estimation system of claim 1, wherein the instructions to generate the map further include instructions to link the lanes boundaries individually according to two or more of the lateral slices being adjacent for at least one of the road edges, wherein the two or more of the lateral slices satisfy a feature clarity for the road edges.

### Claim 8 (depends on 1)

The estimation system of claim 1 further including instructions to aggregate the features within the histogram, wherein the features include one of adjacent slices, confidence values, distance between vehicles, a boundary type, a lane offset, a surface type, a time-of-day, and a heading that are associated with different channels from the acquired data.

### Claim 9 (depends on 1)

The estimation system of claim 1, wherein the lateral slices are subsections of the road edges having fixed dimensions that are scalable longitudinally and data within the histogram varies according to the fixed dimensions and the scalability.

### Claim 10

A non-transitory computer-readable medium comprising:
instructions that when executed by a processor cause the processor to:
compute three-dimensional (3D) representations that are discretized from acquired data about road edges associated with driving lanes;
derive discrete and lateral slices of the road edges using a slicing model, and the road edges are connected in a road graph that describes a mapped area;
extract features from the lateral slices individually using a neural model for forming a histogram to estimate lane boundaries about the driving lanes; and
generate a map by linking the lane boundaries individually along the road edges.

### Claim 11 (depends on 10)

The non-transitory computer-readable medium of claim 10 further including instructions to output by a decoder confidence values and boundary placements for the lane boundaries through counting compressed data within bins of the histogram and the bins are each associated with one of the lateral slices, wherein the neural model includes the decoder.

### Claim 12

A method comprising:
computing three-dimensional (3D) representations that are discretized from acquired data about road edges associated with driving lanes;
deriving discrete and lateral slices of the road edges using a slicing model, and the road edges are connected in a road graph that describes a mapped area;
extracting features from the lateral slices individually using a neural model for forming a histogram to estimate lane boundaries about the driving lanes; and
generating a map by linking the lane boundaries individually along the road edges.

### Claim 13 (depends on 12)

The method of claim 12 further comprising:
outputting by a decoder confidence values and boundary placements for the lane boundaries through counting compressed data within bins of the histogram and the bins are each associated with one of the lateral slices, wherein the neural model includes the decoder.

### Claim 14 (depends on 13)

The method of claim 13, wherein generating the map further includes:
translating the map using an inverse distance to the lane boundaries from the confidence values and the boundary placements.

### Claim 15 (depends on 12)

The method of claim 12, wherein extracting the features further includes:
processing detected keypoints of the driving lanes using matrix calculations for different channels from the acquired data, wherein the matrix calculations factor kernel size per layer of the neural model and the different channels fill gaps of the lateral slices; and
fitting non-linearities from the matrix calculations to estimate the lane boundaries.

### Claim 16 (depends on 15)

The method of claim 15, wherein the layer of the neural model includes trace points of a vehicle within one or more of the lateral slices and the trace points identify lane directionality.

### Claim 17 (depends on 15)

The method of claim 15 further comprising selecting the features using the neural model according to a distance between compressed data in bins of the histogram.

### Claim 18 (depends on 12)

The method of claim 12, wherein generating the map further includes linking the lanes boundaries individually according to two or more of the lateral slices being adjacent for at least one of the road edges, wherein the two or more of the lateral slices satisfy a feature clarity for the road edges.

### Claim 19 (depends on 12)

The method of claim 12 further comprising:
aggregating the features within the histogram, wherein the features include one of adjacent slices, confidence values, distance between vehicles, a boundary type, a lane offset, a surface type, a time-of-day, and a heading that are associated with different channels from the acquired data.

### Claim 20 (depends on 12)

The method of claim 12, wherein the lateral slices are subsections of the road edges having fixed dimensions that are scalable longitudinally and data within the histogram varies according to the fixed dimensions and the scalability.