Road Marking Accuracy Control for Autonomous Vehicle Positioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle control systems struggle to accurately determine the state of compartment lines on other lanes or outside the measurement range of external sensors, especially when there are nearby vehicles, leading to incorrect state recognition and inability to recognize faded or complex road markings.
Innovation Solution
An output device that collates road marking detection results with map information to acquire accuracy information and outputs control information to ensure the accuracy of road marking detection is equal to or higher than a predetermined value, allowing the vehicle to avoid low accuracy road markings and approach high accuracy ones, thereby improving position estimation and route selection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If external sensors are used to detect road markings, then position estimation can be performed, but detection accuracy deteriorates when there are nearby vehicles or when road markings are faded or complex
Solution Approach 1:
The patent introduces map information as an intermediary to mediate between the vehicle's detected road markings and the actual road marking states. By collating detection results with pre-stored map information, the system can infer the states of road markings on other lanes and outside sensor ranges, overcoming the limitations of direct sensor detection in the presence of nearby vehicles or faded markings.
Solution Approach 2:
The system performs preliminary actions by pre-storing map information about road markings in advance. This allows the vehicle to have prior knowledge of road marking locations and patterns, which is then used to collate with real-time detection results and infer states of road markings that are currently undetectable or poorly detected.
2Loss of information
If the vehicle relies solely on current lane road markings for position estimation, then detection is simpler, but the vehicle cannot recognize road markings on other lanes or outside measurement range
Solution Approach 1:
The patent makes the detection system multi-functional by enabling it to not only detect road markings in the current lane but also to infer the states of road markings on other lanes and outside the measurement range. This is achieved by combining real-time detection with map information collation, allowing the same system to provide both localized detection and broader environmental awareness.
3Measurement precision
If the vehicle changes lanes frequently to maintain high detection accuracy, then position estimation accuracy improves, but driving efficiency and productivity decrease
Solution Approach 1:
The system performs preliminary action by pre-acquiring and storing map information about road markings along the route. This allows the vehicle to plan its driving path in advance, selecting routes where road markings provide sufficient detection accuracy without requiring frequent lane changes, thereby maintaining both position estimation accuracy and driving efficiency.
Data Source
AI summary
An onboard device 1 is provided with an own vehicle position estimation unit 17 and an autonomous driving control unit 18. The own vehicle position estimation unit 17 estimates the own vehicle position by collating a detection result of a road marking by a lidar 2 with map DB 10. The autonomous driving control unit 18 acquires, from the map DB 10, the road marking information including detection accuracy information Idet indicative of the accuracy of the collation with respect to each road marking. On the basis of the detection accuracy information Idet, the autonomous driving control unit 18 outputs, to an electronic control device or an information output unit 16 of the vehicle, information for controlling the vehicle so that the accuracy of the collation is equal to or larger than a predetermined value.