Vehicle Road Slope Detection via 3D Camera Reconstruction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Autonomous vehicles face challenges in detecting road slopes, leading to poor control performance as they cannot accurately determine whether they are on an up-sloped, down-sloped, or non-sloped road, affecting their ability to adjust parameters like the gas pedal and brake effectively.

Innovation Solution

A method and apparatus that utilize a camera to capture multiple images, perform three-dimensional reconstruction, and determine a gravity direction vector within the three-dimensional model, allowing the vehicle to detect road slope types and values, thereby providing necessary reference information for precise control.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Extent of automation

If autonomous driving technology is used without road slope detection, then the vehicle can operate autonomously, but the vehicle control precision deteriorates due to inability to detect road slopes

Engineering Contradiction:
Improveautonomous driving capabilityVSAvoidroad slope detection accuracy
Core Design Contradiction:
Extent of automationVSMeasurement precision

Solution Approach 1:

The patent introduces an intermediary system consisting of a camera and three-dimensional reconstruction algorithm to detect road slopes. The camera captures images of the road surface, and the reconstruction algorithm processes these images to calculate slope information, serving as a mediator between the autonomous driving system and the physical road conditions.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces traditional mechanical slope detection devices with an optical-based system. Instead of using mechanical sensors or inclinometers, the system uses a camera to capture images and computational algorithms to determine road slope, substituting mechanical measurement with optical sensing and digital processing.

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

2Device complexity

If traditional road shape prediction methods are used, then the device complexity is reduced, but the road information detection capability deteriorates due to inability to accurately detect slopes

Engineering Contradiction:
Improvedetection device structureVSAvoidroad slope information
Core Design Contradiction:
Device complexityVSLoss of information

Solution Approach 1:

The patent extracts road slope information from ordinary road surface images through three-dimensional reconstruction. Instead of requiring specialized slope detection sensors, the system extracts the needed slope information from standard camera images by processing the spatial relationships between detected road features across multiple images.

Inventive Principle:
Principle #2Taking out (Extraction)

3Ease of operation

If the vehicle cannot detect road slopes, then the control system is simpler to operate, but the vehicle control performance deteriorates due to lack of slope awareness

Engineering Contradiction:
Improvecontrol system operationVSAvoidvehicle control reliability
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The system enables the vehicle to self-detect road slope conditions automatically without requiring manual input or complex operator intervention. The camera and reconstruction algorithm work autonomously to provide slope information to the control system, making the vehicle self-aware of its operating conditions while maintaining simple operation.

Inventive Principle:
Principle #25Self-service

Data Source

PatentEP3675041B1Method and apparatus for determining road information, and vehicle
Publication Date: 2023.08.09 APOLLO INTELLIGENT DRIVING (BEIJING) TECHNOLOGY CO LTD
  • EP3675041B1 patent drawingFigure 1~2
  • EP3675041B1 patent drawingFigure 3
  • EP3675041B1 patent drawingFigure 4

AI summary

Embodiments of the present disclosure provide a method and an apparatus for determining road information, a vehicle and a computer readable storage medium. The method includes: triggering (101) the camera to capture M images during driving, M being an integer not less than 2; performing (102) three-dimensional reconstruction based on the M images to obtain a three-dimensional model; determining (103) a first vector of the three-dimensional model, the first vector indicating a gravity direction in the three-dimensional model; and determining (104) the road information of a road where the vehicle is located based on M spatial positions and the first vector, the M spatial positions being spatial positions of a center of the camera in the three-dimensional model at M time points and the M images are captured by the camera respectively at the M time points, the road information including a slope type and/or a slope value.