Lidar Parallax Object Detection via Linear Beam Deviation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Autonomous vehicles face challenges in accurately detecting and recognizing three-dimensional objects due to the limitations of radar and camera sensors, particularly in low-light conditions and the unmanageable volume of data produced, which hinder effective navigation.

Innovation Solution

A lidar sensor system utilizing a line generator to produce a linear beam, a photodetector to detect reflections at an angle, and computer-implemented modules to determine object attributes such as distance, size, shape, and orientation through parallax principles, enabling precise object recognition.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If radar is used for object detection, then detection range is improved, but object detail information is lost

Engineering Contradiction:
Improvedetection rangeVSAvoidobject detail information
Core Design Contradiction:
SpeedVSLoss of information

Solution Approach 1:

The patent combines radar and lidar sensors into an integrated sensor system. The radar provides detection range and object location, while the lidar adds detailed three-dimensional structural information. The data fusion module merges these complementary data sources to achieve both long-range detection and detailed object characterization.

Inventive Principle:
Principle #5Merging (Combining)

2Loss of information

If cameras are used for object recognition, then object detail information is improved, but data volume becomes unmanageable

Engineering Contradiction:
Improveobject detail informationVSAvoiddata volume
Core Design Contradiction:
Loss of informationVSQuantity of substance

Solution Approach 1:

The patent extracts only the essential three-dimensional structural information from the lidar point cloud data using plane fitting algorithms, rather than processing complete high-resolution images. This extraction approach captures critical geometric features (planar surfaces, edges, vertices) while significantly reducing data volume compared to full camera image processing.

Inventive Principle:
Principle #2Taking out (Extraction)

3Loss of information

If cameras are used for object detection, then object detail information is improved, but performance in low-light conditions deteriorates

Engineering Contradiction:
Improveobject detail informationVSAvoidperformance in low-light conditions
Core Design Contradiction:
Loss of informationVSReliability

Solution Approach 1:

The patent replaces camera-based optical detection with lidar-based active illumination detection. The lidar system uses its own light source to illuminate the scene, making it independent of ambient lighting conditions. This substitution of passive optical detection with active illumination enables reliable object detection and three-dimensional reconstruction in low-light, nighttime, or varying light conditions.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

4Loss of information

If traditional lidar with spinning laser emitter is used, then three-dimensional information is obtained, but system complexity is increased

Engineering Contradiction:
Improvethree-dimensional informationVSAvoidsystem complexity
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The patent segments the three-dimensional scanning function into multiple stationary laser emitters positioned at different locations and angles, rather than using a single spinning emitter. Each emitter captures a portion of the three-dimensional space, and the data is fused to create complete three-dimensional information. This segmentation eliminates mechanical moving parts while achieving the same three-dimensional mapping capability.

Inventive Principle:
Principle #1Segmentation

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

The system effectively determines object attributes with high accuracy, improving navigation capabilities in various lighting conditions and reducing data complexity, enhancing the performance of autonomous vehicles.

Implementation Method 1

a line generator comprising an optical element configured to produce a linear beam of detectable light

Methodology Applied
Scientific EffectLaser: Laser

Implementation Method 2

produce a linear beam of detectable light along a first plane

Methodology Applied
Scientific EffectLight: Light

Implementation Method 3

detect a reflection of the linear beam from the object

Methodology Applied
Scientific EffectReflection: Reflection

Implementation Method 4

a photodetector tuned to detect a reflection of the linear beam

Methodology Applied
Scientific EffectPhotoelectric Effect: Photoelectric Effect

Implementation Method 5

the photodetector is displaced from the optical element such that the photodetector detects the reflection along a second plane at an angle to the first plane

Methodology Applied
Scientific EffectParallax: Parallax

Data Source

PatentUS11467255B2Lidar system for object detection and recognition
Publication Date: 2022.10.11 TURBINE X TECH CORP
  • US11467255B2 patent drawing
  • US11467255B2 patent drawing
  • US11467255B2 patent drawing

AI summary

A lidar sensor system and method use the parallax phenomenon to determine one or more attributes of an object, such its distance, size, or shape. An optical element illuminates the object with a linear beam along a first plane. A photodetector is displaced from the optical element so that it detects the reflection of the linear beam from the object along a second plane, which is at an angle to the first plane. A computer-implemented line detection module determines a measure of deviation of the linear beam reflection from a vanishing point of the first plane. A computer-implemented analysis module determines the attribute based on the measure of deviation.