Vehicle Trunk Opening by Computer Vision and Multi-Action Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for automatically opening vehicle trunks require users to physically interact with the vehicle and may pose difficulties for individuals who cannot perform certain actions, such as standing on one leg, and do not allow for hands-free operation or safe distance activation.
Innovation Solution
A method using sensors to detect a registered user's presence, facial gestures, body movements, and voice commands from a distance, allowing the trunk to open automatically while the user is approaching the vehicle, incorporating a combination of camera, microphone, radar, sonar, and LiDAR sensors to identify and track facial and body landmarks, and verify user identity through stored profiles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If existing automatic trunk opening systems are used, then the trunk can open without physical contact, but the user must still get close to the vehicle and perform specific actions that may be difficult or impossible for some users
Solution Approach 1:
The patent replaces mechanical interaction systems (physical contact, specific body positions) with computer vision-based detection systems. The trunk can be opened using facial recognition, hand wave detection, or voice commands from a distance, eliminating the need for difficult physical actions like standing on one leg.
Solution Approach 2:
The system provides multiple interchangeable methods for triggering trunk opening (facial recognition, hand waves, voice commands, approach detection), making it adaptable to users with different physical capabilities and ensuring universal accessibility regardless of user condition.
2Ease of operation
If the trunk opens automatically when the user is close, then convenience is improved, but the user must move back to avoid being hit by the opening trunk door
Solution Approach 1:
The system detects user intent in advance through facial recognition, hand wave detection, or voice commands before the trunk begins to open. This allows the user to remain in a safe position while the trunk opens automatically, eliminating the need to move back and avoiding potential injury from the moving door.
3Reliability
If multiple sensors and verification steps are added, then security and accuracy are improved, but the system complexity increases
Solution Approach 1:
The patent combines multiple sensor types (cameras, microphones, radar, LiDAR) and multiple detection methods (facial recognition, hand wave detection, voice commands, approach detection) into a unified trunk opening system. This integration improves reliability through multiple verification steps while managing complexity through coordinated sensor fusion.
4Reliability
If the system requires the user to perform specific actions, then accidental activation is reduced, but users carrying items cannot perform hands-free operations
Solution Approach 1:
The system replaces mechanical action requirements (hand gestures, body movements) with alternative detection methods that work while carrying items. Facial recognition, voice commands, and approach detection allow users to trigger trunk opening without using their hands, enabling hands-free operation when carrying groceries or other items.
Data Source
AI summary
A method for opening a trunk of a vehicle is provided, comprising: receiving sensor data from at least one of a plurality of sensors mounted on the vehicle; determining that a person is present within a predetermined distance of the vehicle based on the sensor data; determining that the person is a registered user of the vehicle based on the sensor data and stored user data; determining that the person performs an acceptable combination of actions; and responsive to determining that the person performs the acceptable combination of actions, controlling a trunk door actuator to open the trunk of the vehicle. The acceptable combination of actions includes at least two of the following actions: approaching the vehicle, performing a predetermined facial gesture, performing a predetermined body gesture, and performing a predetermined voice command.