Vehicle Hatch Collision Avoidance via 3D Environmental Modeling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle imaging systems struggle to effectively detect and avoid collisions with objects above the vehicle when the hatch is opening or closing, especially in parking scenarios where the object is out of the camera's field of view.
Innovation Solution
A vehicle vision system utilizing multiple cameras and sensors to capture and process images, generating a 3D environmental model, and storing data on structures and objects around the vehicle, which allows for the determination of potential hatch collisions and automatic control to stop or reverse the hatch to avoid contact with detected objects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the hatch is opened or closed without object detection, then the operation is simple and fast, but collision with objects above the vehicle cannot be avoided
Solution Approach 1:
The system performs preliminary scanning of the environment above the vehicle using cameras and sensors before the hatch opening/closing operation. Data about objects in the sweep path is collected and stored in advance, allowing the control system to make informed decisions about hatch operation without adding complexity to the actual opening/closing mechanism.
Solution Approach 2:
The patent introduces an intermediary control system that acts between the hatch actuator and the physical hatch mechanism. This intermediary layer processes sensor data, determines object locations, and controls the hatch operation accordingly, separating the detection function from the actuation function and reducing overall system complexity.
2Reliability
If cameras are used to detect objects, then collision avoidance is improved, but objects above the vehicle out of camera view cannot be detected
Solution Approach 1:
The system transitions from two-dimensional camera images to three-dimensional spatial understanding by combining data from multiple cameras and sensors. The control system constructs a 3D model of the environment above the vehicle, allowing detection of objects that may be outside the direct field of view of any single camera by inferring their position from multiple perspectives and sensor inputs.
Solution Approach 2:
The patent employs multiple types of sensors (cameras, ultrasonic sensors, radar) that can detect objects through different physical principles. Each sensor type has different strengths and coverage areas, and their combined use provides comprehensive detection capability that compensates for the limitations of any single sensor, including objects that may be obscured from camera view.
3Reliability
If the system stops the hatch to avoid collision, then safety is improved, but the opening/closing operation time increases
Solution Approach 1:
The system applies partial stopping or reversal of the hatch operation only when and where necessary to avoid collision, rather than stopping the entire operation. The control system can allow the hatch to open or close freely in areas where no objects are detected, while applying corrective action only in the specific region where a collision risk exists, minimizing the overall time impact.
Solution Approach 2:
The patent implements real-time feedback control where the system continuously monitors sensor data during hatch operation and dynamically adjusts the hatch speed and position. If an object is detected in the sweep path, the system provides feedback to stop or reverse the hatch; if the path is clear, the hatch operates normally, optimizing both safety and speed through continuous adaptive control.
Data Source
AI summary
A vehicle hatch clearance determining system includes a plurality of cameras disposed at the vehicle and a control having a processor for processing image data captured by the cameras. During a parking event, and responsive to processing of captured image data, the system detects objects, gathers information pertaining to detected objects exterior the vehicle and generates an environmental model based at least in part on the gathered information. With the vehicle parked, the system determines location of the vehicle relative to at least one detected object of the environmental model and determines if the model includes an object in a region that would be swept by a hatch of the parked vehicle when the hatch is being opened or closed. Responsive to determination that the object is in the region that would be swept by the hatch, the system stops movement of the hatch to avoid impact with the object.