Monocular Vehicle Distance Estimation Without LiDAR

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current methods for determining vehicle proximity, such as using LIDAR equipment, are expensive and inaccessible to many, necessitating a cost-effective alternative for computing vehicle distances using visible light cameras.

Innovation Solution

A computing platform equipped with a visible light camera processes video footage to calculate longitudinal and lateral distances to other vehicles using deep learning algorithms and perspective transformation techniques, enabling efficient and scalable distance estimation for vehicle control systems.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If LIDAR equipment is used to determine vehicle proximity, then measurement precision is improved, but device complexity and cost increase

Engineering Contradiction:

Improvevehicle proximity measurementVSAvoidequipment complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent replaces the mechanical/optical LIDAR system with a computational vision system using visible light cameras and deep learning algorithms. The system substitutes physical distance measurement hardware with software-based monocular depth estimation that processes video footage to calculate longitudinal and lateral distances between vehicles, thereby reducing device complexity while maintaining measurement capability

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent creates a virtual 3D representation of the physical scene by generating a virtual camera pose and projecting 3D bounding boxes of vehicles into the 2D image plane. This virtual model copying approach allows distance estimation without direct physical measurement, reducing hardware requirements while preserving measurement accuracy

Inventive Principle:

Principle #26Copying

2Measurement precision

If LIDAR equipment is used to determine vehicle proximity, then measurement precision is improved, but cost increases

Engineering Contradiction:

Improvevehicle proximity measurementVSAvoiddeployment cost

Core Design Contradiction:

Measurement precisionVSEase of manufacture

Solution Approach 1:

The patent employs inexpensive visible light cameras instead of costly LIDAR sensors. The system uses standard camera hardware that is widely available and significantly cheaper than LIDAR equipment, combined with software-based depth estimation algorithms to achieve cost-effective vehicle proximity measurement suitable for widespread deployment

Inventive Principle:

Principle #27Cheap short-living objects (Disposable)

Solution Approach 2:

The patent replaces expensive physical LIDAR measurement systems with a computational approach using affordable camera hardware and deep learning software. This substitution eliminates the need for costly specialized sensors while maintaining the ability to estimate vehicle distances through algorithmic processing of visual data

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Productivity

If deep learning algorithms are used for distance estimation, then productivity is improved, but device complexity increases

Engineering Contradiction:

Improvedistance estimation speedVSAvoidprocessing system complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent performs preliminary actions by pre-training deep learning models on large datasets of vehicle images and depth information. The models are prepared in advance with learned features and parameters, enabling rapid real-time inference during actual operation without requiring complex processing during runtime, thus improving productivity while managing system complexity

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent introduces intermediate representations including virtual camera poses, 3D bounding boxes, and projected image planes as mediators between the input video footage and the final distance estimation. These intermediate structures organize complex processing into manageable stages, improving computational efficiency and productivity while structuring the complexity in a tractable manner

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS20240343248A1Dynamic distance estimation output generation based on monocular video

Publication Date: 2024.10.17 ALLSTATE INSURANCE COMPANY

AI summary

Aspects of the disclosure relate to a dynamic distance estimation output platform that utilizes improved computer vision and perspective transformation techniques to determine vehicle proximities from video footage. A computing platform may receive, from a visible light camera located in a first vehicle, a video output showing a second vehicle that is in front of the first vehicle. The computing platform may determine a longitudinal distance between the first vehicle and the second vehicle by determining an orthogonal distance between a center-of-projection corresponding to the visible light camera, and an intersection of a backside plane of the second vehicle and ground below the second vehicle. The computing platform may send, to an autonomous vehicle control system, a distance estimation output corresponding to the longitudinal distance, which may cause the autonomous vehicle control system to perform vehicle control actions.