Automated Driving Safety System Lane Segment Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing complexity of automated driving systems poses a significant computational burden, particularly in determining all possible routes and intersections between vehicles in complex driving environments, leading to high resource utilization and potential safety risks.
Innovation Solution
A safety system utilizing a lane-based coordinate system and model-based approach to reduce computational complexity by filtering irrelevant obstacles and determining relevant lane segments, allowing for efficient collision risk assessment and safety operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all possible routes and intersections between vehicles are determined in complex driving environments, then collision detection completeness is improved, but computational burden and resource utilization increase significantly
Solution Approach 1:
The patent segments the continuous space around the vehicle into discrete lane segments in a lane-based coordinate system. This segmentation allows the system to focus computational resources only on relevant lane segments rather than calculating all possible routes in the entire environment, thereby reducing computational burden while maintaining collision detection completeness.
Solution Approach 2:
The patent applies local quality by determining relevance of obstacles based on their spatial relationship to the vehicle's current lane segment. Only obstacles located in relevant lane segments (those that could potentially intersect with the vehicle's path) are considered for collision detection, while obstacles in irrelevant lane segments are discarded. This local filtering approach reduces computational complexity without compromising safety.
2Productivity
If the number of obstacles to be considered is reduced through filtering, then computational effort decreases, but risk of missing relevant obstacles increases
Solution Approach 1:
The patent performs preliminary action by pre-determining the set of relevant lane segments based on the vehicle's current position and intended path before obstacle classification. This preliminary identification of relevant spatial regions allows the system to efficiently filter obstacles while ensuring that no potentially relevant obstacles are missed, as the relevance criteria are established in advance based on geometric relationships.
Solution Approach 2:
The system implements feedback by continuously updating the set of relevant lane segments as the vehicle moves and its intended path changes. The relevance determination is dynamically adjusted based on the vehicle's current state, ensuring that the filtering process remains accurate and no relevant obstacles are discarded due to outdated relevance criteria.
Data Source
AI summary
In some aspects, a safety system may be configured to receive vehicle position data indicating a position of a vehicle, determine a first lane segment in a lane coordinate system based on the vehicle position data, the first lane segment being a lane segment in which the vehicle is located, determine a relevant set of lane segments based on a safety-range from the first lane segment, determine or receive obstacle position data indicating a second lane segment in the lane coordinate system, the second lane segment being a lane segment in which an obstacle is located, and classify the obstacle either as a non-relevant obstacle in the case that the second lane segment is not included in the relevant set of lane segments, or as a relevant obstacle in the case that the second lane segment is included in the relevant set of lane segments.