Ground Detection Using Dual Resolution Laser Mesh Filtering

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing ground detection methods for unmanned vehicles are not quick, accurate, or comprehensive, particularly in filtering out ground points from laser point clouds, which affects obstacle detection and tracking.

Innovation Solution

A method involving projecting laser point clouds onto high and low resolution meshes to filter candidate ground points and perform ground fitting, using algorithms like RANSAC, with the filtering done in the high resolution mesh and fitting in the low resolution mesh to improve stability and accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If ground detection is performed using traditional methods on laser point clouds, then the detection can be completed, but the detection speed is slow and accuracy is insufficient

Engineering Contradiction:

Improveground detection speedVSAvoidground point detection accuracy

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The patent divides the ground detection process into two distinct stages: a high-resolution mesh processing stage for filtering and candidate selection, and a low-resolution mesh processing stage for final fitting. This segmentation allows each stage to optimize for its specific function, improving both speed and accuracy overall.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent applies different mesh resolutions to different parts of the processing pipeline: high-resolution mesh is used locally for filtering operations where precision is critical, while low-resolution mesh is used for fitting operations where computational efficiency is prioritized. This local quality differentiation resolves the contradiction between speed and accuracy.

Inventive Principle:

Principle #3Local quality

2Measurement precision

If high resolution mesh is used for ground fitting, then detection accuracy improves, but computational complexity and processing time increase

Engineering Contradiction:

Improveground fitting accuracyVSAvoidcomputational complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the processing workload by using high-resolution mesh only for filtering operations (where accuracy matters) and low-resolution mesh for fitting operations (where computational efficiency matters). This segmentation reduces overall computational complexity while maintaining accuracy where needed.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent applies partial action by using high-resolution mesh only partially (for filtering) rather than entirely (for both filtering and fitting). This partial application of high-resolution processing achieves necessary accuracy without the full computational burden, resolving the contradiction between accuracy and complexity.

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS11313951B2Ground detection method, electronic device, and vehicle

Publication Date: 2022.04.26 APOLLO INTELLIGENT DRIVING (BEIJING) TECHNOLOGY CO LTD

AI summary

Embodiments of the present disclosure provide a ground detection method and apparatus, an electronic device, a vehicle, and a storage medium. The method includes: projecting a laser point cloud obtained to a high resolution mesh and a low resolution mesh respectively; filtering out candidate ground points in the high resolution mesh; and performing ground fitting based on the candidate ground points in the low resolution mesh.