Vehicle Emergency Trajectory Selection Based on Human Driving Behavior
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Highly automated vehicles face challenges in reacting appropriately in critical situations without human intervention, as current systems lack a universal approach to mimic human driving behavior effectively, leading to ethical dilemmas in accident scenarios.
Innovation Solution
A method for controlling vehicles that reads collision signals and selects an emergency trajectory based on human driving behavior patterns, using a set of reference trajectories associated with different drivers, which can be preselected, randomly chosen, or adapted based on driving conditions, to output control signals for steering, braking, or engine control, ensuring a human-like reaction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a predefined human reaction is used during unavoidable accidents, then the vehicle reaction corresponds to a human reaction, but the system lacks adaptability to different driving situations and drivers
Solution Approach 1:
The patent segments the emergency response into multiple reference trajectories, each representing different human driving behaviors in specific situations. Instead of using a single predefined reaction, the system divides the response options into multiple categorized trajectories that can be selected based on the specific accident scenario, thereby maintaining human-like behavior while improving adaptability.
Solution Approach 2:
The system dynamically selects from multiple reference trajectories based on real-time driving-environment conditions and the specific collision scenario. This dynamic selection mechanism allows the vehicle to adapt its emergency response to different situations while maintaining consistency with human driving behavior patterns, resolving the contradiction between reliability and adaptability.
2Adaptability or versatility
If multiple reference trajectories are stored for different drivers, then the number of available emergency trajectories increases, but the device complexity increases
Solution Approach 1:
The patent uses reference trajectories that are copied from recorded human driver behaviors in various emergency situations. Instead of implementing complex real-time decision-making algorithms, the system stores pre-recorded trajectory copies that can be quickly retrieved and executed, thereby increasing adaptability without proportionally increasing device complexity.
Solution Approach 2:
The system performs preliminary action by pre-recording and storing multiple reference trajectories representing different human driving behaviors before actual emergency situations occur. This pre-prepared set of trajectories allows the system to quickly respond to emergencies without requiring complex real-time computation, thus increasing versatility while managing device complexity.
3Ease of operation
If reference trajectories are preselected cyclically, then the selection process is simplified, but the emergency trajectory selection becomes dependent on the random moment of collision signal generation
Solution Approach 1:
The patent implements periodic action through cyclic preselection of reference trajectories. The system cycles through the available reference trajectories in a predetermined sequence, and when a collision signal is detected, the currently preselected trajectory is used. This periodic cycling simplifies the selection process while ensuring that different trajectories are considered over time, balancing ease of operation with reasonable reliability.
Data Source
AI summary
A method for controlling a vehicle includes reading in a collision signal that represents an imminent collision of the vehicle with at least one collision object, selecting, based on the collision signal, an emergency trajectory that represents a human driving behavior associated with a human driver, and outputting a corresponding control signal to guide the vehicle along the emergency trajectory.