Context-Aware ISA Speed Limit Fusion for Map and Sign Conflicts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Intelligent Speed Adaptation (ISA) systems that rely solely on positioning or situational-awareness subsystems face challenges in determining accurate vehicular speed limits, particularly in scenarios with outdated maps, map matching errors, rural areas, and complex environments where speed limit signs may be obstructed or multiple signs are visible, leading to incorrect speed limit applications.
Innovation Solution
An ISA system that combines both positioning and situational-awareness subsystems, using deep neural networks to process images from cameras and determine contextual speed limits based on the vehicle's location and environment, applying dynamic fusion rules to identify legal speed limits and adjust them according to context profiles and priorities, ensuring accurate and safe speed control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If ISA systems rely solely on positioning subsystems to determine speed limits, then the system can operate without additional sensors, but the system may implement incorrect speed limits due to outdated maps, map matching errors, or missing road segments
Solution Approach 1:
The patent combines positioning subsystems (GPS/GNSS with map databases) and situational-awareness subsystems (computer vision with traffic sign detection) into a unified ISA system. The system merges location-based speed limit determination with visual recognition of posted speed limits, allowing cross-validation and fallback mechanisms to improve reliability while maintaining reasonable system complexity
2Adaptability or versatility
If ISA systems rely solely on situational-awareness subsystems to determine speed limits, then the system can adapt to real-time conditions, but the system may fail to determine speed limits when signs are obstructed, not visible, or in complex environments with multiple signs
Solution Approach 1:
The patent introduces context estimation as an intermediary layer that processes raw data from both positioning and situational-awareness subsystems. The context estimator analyzes the driving environment (urban, rural, highway, construction zone) and uses this contextual information to weight and fuse data from both subsystems, improving reliability in scenarios where one subsystem may fail
3Reliability
If ISA systems use both positioning and situational-awareness subsystems, then the system can improve speed limit determination accuracy, but the system complexity increases due to integrating multiple subsystems and processing methods
Solution Approach 1:
The patent segments the ISA system into distinct functional modules: positioning subsystem, situational-awareness subsystem, context estimator, and speed limit determination module. Each module performs a specific function, and the context estimator acts as a coordinator that fuses inputs from positioning and vision systems. This modular segmentation manages system complexity by organizing components with clear interfaces and responsibilities
Data Source
AI summary
A method for limiting a speed of a vehicle includes determining the vehicle's current location; capturing images of the environment in which the vehicle is traveling using a camera; processing the captured images using one or more deep neural networks; based on an output from the one or more deep neural networks, determining a context under which the vehicle is traveling and determining a seen posted speed limit if any applicable speed limit sign is sighted by the camera; in accordance with speed limit identification (“SLI”) logic, identifying a legal speed limit for the vehicle either based on the vehicle's determined current location or a determined seen posted speed limit; applying a speed policy to the identified legal speed limit to derive a practical speed limit; and determining a contextual speed limit by which to limit the vehicle's speed based on the practical speed limit and the determined context.