Vehicle Path Selection Using Map and Camera Reliability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle control systems face challenges in accurately determining travel lanes, especially when map information is outdated or the reliability of camera images and map data is low, leading to potential difficulties in selecting appropriate travel paths during automated driving.
Innovation Solution
A vehicle control system that includes a first prediction path generator based on map information, a second prediction path generator based on external environment information, and a reliability determining unit that selects a travel path based on the divergence and reliability of these paths, ensuring safe hands-off driving by generating a third prediction path that intermediates between the first and second paths when divergence occurs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the system uses only map information for path prediction, then the path prediction is stable, but the reliability decreases when map information is outdated
Solution Approach 1:
The system merges map information-based path prediction with camera image-based path prediction to create a combined prediction system. When map information becomes outdated, the camera-based prediction compensates for the information loss, maintaining overall system reliability through the integration of multiple information sources.
Solution Approach 2:
The system uses leading vehicle information as feedback to validate and adjust path predictions. By observing the actual travel paths of leading vehicles, the system can detect when map information is outdated and adjust its prediction reliability accordingly, using real-world vehicle behavior as a feedback mechanism to correct outdated map data.
2Adaptability or versatility
If the system uses only camera images for path prediction, then the path prediction adapts to current road conditions, but the measurement precision decreases when camera reliability is low
Solution Approach 1:
The system combines camera image-based path prediction with map information-based prediction. When camera reliability is low, the system integrates the less reliable camera data with more reliable map data, maintaining measurement precision through data fusion while still adapting to current road conditions through the camera component.
Solution Approach 2:
The system prepares for low camera reliability scenarios by having map information as a backup reference. When camera precision deteriorates, the pre-stored map information cushions the impact of poor camera quality, ensuring continuous reliable path prediction without complete dependence on camera images.
3Reliability
If the system generates multiple prediction paths, then the path selection reliability improves, but the device complexity increases
Solution Approach 1:
The system segments path prediction into distinct modules: map information-based prediction, camera image-based prediction, and leading vehicle information-based validation. Each module independently generates path predictions, and the system selects the most reliable prediction based on validation results, reducing overall system complexity through modular segmentation.
Solution Approach 2:
The system dynamically changes the reliability parameter of different prediction sources based on leading vehicle information. Instead of maintaining equal complexity for all prediction paths, the system adjusts which prediction path to follow based on real-time reliability assessment, simplifying the decision-making process while maintaining high selection reliability.
Data Source
AI summary
A vehicle control system includes a first prediction path generator, a second prediction path generator, a leading-vehicle-information acquiring unit, a divergence determining unit, a reliability determining unit, and a travel path selector. The first prediction path generator generates a first prediction path of a vehicle based on map information and positional information of the vehicle. The second prediction path generator generates a second prediction path of the vehicle based on external environment information. The leading-vehicle-information acquiring unit acquires leading vehicle information. The divergence determining unit determines whether a divergence of a predetermined amount or more has occurred between the two prediction paths. The reliability determining unit determines reliability of each prediction path based on the leading vehicle information in a case where the divergence has occurred. The travel path selector selects a travel path of the vehicle based on the reliability of each prediction path.