Adaptive Brake Release Based on Inferred Road Category
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle safety systems struggle to effectively manage automatic braking interventions after an initial collision, as they do not adequately consider the varying road categories and their impact on the initial distance of a hypothetical following vehicle, which can lead to increased collision risks or severity.
Innovation Solution
A method that determines the road category based on vehicle data, such as speed, acceleration, and steering patterns, to adapt the initial distance calculation for a hypothetical following vehicle, allowing for a more precise release condition of the brake after an initial collision, thereby reducing the risk or severity of subsequent collisions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If the brake is released after an initial collision to reduce severity, then the collision consequences are reduced, but the risk of subsequent collision with following vehicles increases
Solution Approach 1:
The brake release decision is made dynamic by inferring road category from vehicle data and using it to adapt the initial distance parameter in the braking model. The system transitions from a static brake release approach to a dynamic one that adjusts based on inferred road conditions, allowing optimal brake release decisions for each specific situation.
Solution Approach 2:
The initial distance parameter in the braking model is changed adaptively based on the inferred road category. Different road categories (urban, rural, highway) have different typical initial distances of following vehicles, and the system adjusts this parameter accordingly to optimize both collision severity reduction and subsequent collision prevention.
2Device complexity
If a hypothetical following vehicle model is used to determine brake release conditions, then the system works without surrounding sensors, but the accuracy of collision risk assessment is reduced
Solution Approach 1:
Instead of using actual sensors to detect following vehicles, the system creates a hypothetical copy or model of a following vehicle with characteristic behavior patterns. This virtual model allows the system to simulate and predict following vehicle behavior based on typical initial distances for different road categories, enabling brake release decisions without additional physical sensors.
Solution Approach 2:
The system uses its own vehicle data (speed, acceleration, steering patterns) to infer road category and determine brake release conditions, rather than relying on external sensors to detect other vehicles. The vehicle essentially serves itself by using its operational data to make safety decisions about brake release timing.
3Reliability
If the initial distance is adapted to road category, then the brake release timing is optimized, but the complexity of determining road category increases
Solution Approach 1:
The system determines road category by analyzing its own vehicle operational data (speed, acceleration, steering patterns) rather than requiring external infrastructure or complex classification systems. The vehicle uses its own behavior patterns as fingerprints to identify which road category it is traveling on, simplifying the overall system architecture.
Solution Approach 2:
The vehicle data collection and analysis system serves multiple functions: it monitors vehicle operation for safety reasons, characterizes driving behavior, infers road category, and supports brake release decisions. This multi-functional approach avoids the need for separate dedicated systems for each function, reducing overall complexity.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method and device for avoiding a possible subsequent collision and for reducing the consequences of an accident resulting from a collision are proposed, in which, after an initial collision of the vehicle with a further road user has taken place, an automatic braking intervention by a vehicle safety system is triggered according to a braking model, in that vehicle data is obtained on the basis of the driver's own driving behaviour, the vehicle data which is obtained is buffered in a memory, the category of the road being travelled on at that time is inferred from the buffered vehicle data, an initial distance is determined as a function of the determined road category, and the triggering condition of the brake is determined using this initial distance.