Vehicle Door Opening Prediction Using Sensor Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In urban traffic areas, especially in side streets, pedestrians often walk along parked vehicles and open vehicle doors, posing a risk as existing automated systems only reduce vehicle speed upon detection, failing to anticipate and react to the opening of vehicle doors effectively, which can lead to dangerous situations.
Innovation Solution
A method using sensor units like cameras, lidar, and radar to detect pedestrians and parked vehicles, calculate the probability of a door being opened, and adjust the trajectory or speed of an approaching vehicle to avoid collisions by predicting the door's impact on the lane width, allowing for proactive braking or trajectory adjustments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the system classifies objects and calculates probability of door opening, then reaction accuracy is improved, but computational complexity increases
Solution Approach 1:
The control unit segments the processing task into distinct stages: first classifying detected objects into categories (pedestrian, parked vehicle, moving vehicle), then calculating probability only for relevant combinations (pedestrian near parked vehicle), and finally determining appropriate reactions. This segmentation avoids unnecessary computations for all detected objects and focuses processing only on potential door-opening scenarios
Solution Approach 2:
The system applies different processing levels to different detected objects based on their classification. Parked vehicles near pedestrians receive detailed probability calculation and reactive measure preparation, while other objects receive standard detection processing. This local quality approach concentrates computational resources on high-risk scenarios rather than uniformly processing all detected objects
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The method effectively prevents collisions and enhances road safety by anticipating and reacting to the opening of vehicle doors, reducing the risk of accidents and providing comfort through early braking or obstacle avoidance.
Implementation Method 1
The at least one sensor unit may be, for example, one or more cameras, at least one lidar sensor, at least one radar sensor and the like
Implementation Method 2
The at least one sensor unit may be, for example, one or more cameras, at least one lidar sensor, at least one radar sensor and the like
Data Source
AI summary
A method initiates a reaction of a first vehicle to a person in the environment of a second vehicle. The person is on a road on which the first vehicle is travelling and the second vehicle is detected. Measurement data determined by at least one sensor unit are received by a control unit. The control unit carries out a classification and registers the evaluated data as a person and as a second vehicle. Motion vectors of the person are determined and the expected movement of the person is calculated. A position and width of the vehicle doors of the second vehicle are determined or estimated on the basis of the measurement data. A probability of the person opening a vehicle door is calculated. A reaction of the first vehicle is initiated by the control unit depending on the calculated probability.