Multi-Sensor Lane Verification for Stable Real-Time Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing lane detection systems for autonomous vehicles face challenges in achieving precise and stable localization, especially when using differential GPS, and require improved methods for real-time detection of lane markings.

Innovation Solution

A method and system for lane detection that involves generating a ground truth for lane markings, training a confidence module to assess the reasonableness of fitted lane markings, and using a combination of sensors like GPS, IMU, and LiDAR to optimize lane template fitting and detection.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If differential GPS is used for localization, then the system becomes affordable, but the precision of localization is not desirably stable

Engineering Contradiction:
Improvelocalization precisionVSAvoidlocalization stability
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent combines multiple sensors (camera, LiDAR, IMU, GPS) into an integrated sensing system. The camera and LiDAR capture visual and depth information of lane markings, the IMU provides motion data, and GPS provides location data. By merging these sensor inputs, the system achieves stable and precise localization that compensates for the instability of GPS alone.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent introduces an intermediary processing system that fuses data from multiple sensors. The confidence module acts as an intermediary that evaluates the reliability of detected lane markings by comparing multiple sensor inputs and historical data, thereby stabilizing the localization precision without requiring expensive hardware.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If multiple sensors are used for lane detection, then the precision and reliability of detection is improved, but the device complexity increases

Engineering Contradiction:
Improvelane detection precisionVSAvoidsensor system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent designs a multi-functional processing system that handles multiple sensor types (camera, LiDAR, IMU, GPS) through a unified architecture. The same processing modules (hit-map generation, lane marking detection, confidence evaluation) process data from different sensors, reducing the need for separate processing chains and minimizing overall system complexity.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The confidence module performs self-evaluation of detected lane markings by analyzing the consistency of multiple sensor inputs and historical detection results. This self-service mechanism automatically filters unreliable detections without requiring external validation, simplifying the system architecture while maintaining high precision.

Inventive Principle:
Principle #25Self-service

3Productivity

If real-time lane detection is implemented, then the detection speed is improved, but the measurement precision may be compromised

Engineering Contradiction:
Improvedetection speedVSAvoidlane marking accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The patent pre-generates hit-maps from historical lane detection data and stores them for quick reference. When detecting lane markings in real-time, the system compares current sensor data against these pre-processed hit-maps, significantly reducing computation time while maintaining detection accuracy through the confidence evaluation module.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system dynamically adjusts the detection process by using the confidence module to evaluate each detected lane marking in real-time. Based on the confidence score, the system can quickly accept or reject detections, enabling real-time response while ensuring precision through adaptive validation rather than fixed processing thresholds.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP4553599A1Verification module system and method for motion-based lane detection with multiple sensors
Publication Date: 2025.05.14 CREATEAI INC
  • EP4553599A1 patent drawingFigure 1
  • EP4553599A1 patent drawingFigure 2A
  • EP4553599A1 patent drawingFigure 2B

AI summary

A method of lane detection for a non-transitory computer readable storage medium storing one or more programs is disclosed. The one or more programs include instructions, which when executed by a computing device, cause the computing device to perform the following steps comprising: generating a ground truth associated with lane markings expressed in god's view; receiving features from at least one of a hit-map image and a fitted lane marking, wherein the hit-map image includes a classification of pixels that hit a lane marking, and the fitted lane marking includes pixels optimized based on the hit-map image; and training a confidence module based on the features and the ground truth, the confidence module configured to determine on-line whether a fitted lane marking is reasonable, using parameters that express a lane marking in an arc.