Autonomous Emergency Braking Predicts Pedestrian Trajectory
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous emergency braking systems provide warnings and perform braking based solely on the current position of pedestrians, leading to unnecessary startles and excessive braking, as they do not account for pedestrian movement, potentially causing false alarms and inefficient braking.
Innovation Solution
An autonomous emergency braking system that predicts the relative position of pedestrians by tracking their movement using a vehicle speed detector, pedestrian position detector, and electronic control unit, calculating a time to collision (TTC) and adjusting warnings and braking levels based on predicted positions relative to the vehicle.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If warnings and emergency braking are performed based only on current positions of pedestrians, then collision prevention is achieved, but unnecessary warnings and excessive braking occur when pedestrians laterally move
Solution Approach 1:
The system performs preliminary action by predicting the future position of the pedestrian at the time of potential collision (TTC time) before the actual collision occurs. The ECU calculates where the pedestrian will be when the vehicle reaches the pedestrian's current position, and bases the warning/braking decision on this predicted position rather than the current position, thereby avoiding unnecessary warnings when the pedestrian is likely to move laterally out of the path.
Solution Approach 2:
The system applies dynamics by transitioning from a static decision-making approach (based on current pedestrian position) to a dynamic approach (based on predicted future position). The ECU continuously updates the pedestrian position prediction by tracking lateral movement, making the warning/braking decision adaptive to the pedestrian's motion rather than fixed on their current location.
2Reliability
If autonomous emergency braking is performed regardless of driver braking, then pedestrian collision is prevented, but driver convenience deteriorates due to excessive braking
Solution Approach 1:
The system performs preliminary assessment by predicting the pedestrian's future position before executing braking. This preliminary prediction action allows the system to determine in advance whether braking will be necessary, avoiding unnecessary braking operations that would reduce driver convenience while still preventing collisions when truly needed.
Solution Approach 2:
The system uses feedback by continuously monitoring the pedestrian's lateral movement and updating the position prediction. This feedback mechanism allows the system to adjust its braking decision based on the pedestrian's actual motion patterns, ensuring braking is applied only when the predicted trajectory indicates a genuine collision risk, thereby maintaining driver convenience.
Data Source
AI summary
An autonomous emergency braking system includes a vehicle speed detector, a pedestrian position detector, a pedestrian information storage portion and an electronic control unit which receives the vehicle speed and information of the position of the pedestrian detected by the pedestrian position detector, calculates a time to collision (TTC), predicts a relative position of the pedestrian compared with the vehicle at a point in time after the calculated TTC passes by tracking a moving trajectory of the pedestrian from a time series change in the position of the pedestrian stored in the pedestrian information storage portion, and performs autonomous emergency braking control depending on the predicted relative position of the pedestrian.