Passenger-Eye 3D Modeling for Accurate In-Vehicle Surrounding Views
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current autonomous driving vehicles struggle to accurately project a three-dimensional representation of their surroundings onto on-board screens, making the visual environment seen by passengers differ significantly from the actual environment viewed through windows.
Innovation Solution
The method involves acquiring coordinate points of obstacles in a vehicle coordinate system, determining the passenger's eye position, establishing an eye coordinate system, converting obstacle coordinates, and calculating visualization distances based on the observation angle to perform three-dimensional modeling, ensuring the on-board screen display aligns with the real-world view.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional 3D modeling methods are used to project the surrounding environment onto on-board screens, then the visualization can be displayed, but the visual environment seen by passengers differs significantly from the actual environment viewed through windows
Solution Approach 1:
The patent transforms the coordinate system from a vehicle-centric reference frame to an eye-centric reference frame by changing the origin point and coordinate axes. This parameter change in the coordinate system allows the 3D model to be projected from the passenger's actual viewing position, making the virtual visualization align with the real environment seen through windows.
Solution Approach 2:
The patent creates a virtual copy of the surrounding environment using 3D modeling techniques, then projects this copy onto the on-board screen. By using the passenger's eye position as the projection origin, the virtual copy accurately replicates what the passenger would see in reality, rather than creating a distorted vehicle-centered representation.
2Ease of manufacture
If the coordinate system is established based on the vehicle center, then the modeling process is simplified, but the projection does not match the passenger's actual viewing perspective
Solution Approach 1:
The patent changes the fundamental parameter of the coordinate system origin from the vehicle center to the passenger's eye position. This single parameter change transforms both the coordinate axes and the projection methodology, enabling accurate representation of the passenger's viewing perspective while maintaining a systematic approach to 3D modeling.
3Ease of operation
If the 3D model is projected from the vehicle center, then the implementation is straightforward, but the visualization distance and observation angle do not match the passenger's actual perception
Solution Approach 1:
The patent modifies the projection parameters by setting the origin at the passenger's eye position rather than the vehicle center. This changes the calculation of visualization distances and observation angles, ensuring that the projected 3D model reflects the actual geometric relationships as perceived by the passenger, thereby preserving depth perception accuracy.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
The present disclosure discloses a method and apparatus for three-dimensional modeling, and relates to the field of intelligent transportation and autonomous driving. A specific implementation solution includes: acquiring coordinate points of obstacles in a surrounding environment of an autonomous driving vehicle in a vehicle coordinate system; determining a position of eyes of a passenger in the autonomous driving vehicle, and establishing an eye coordinate system using the position of the eyes as a coordinate origin; converting the coordinate points of the obstacles in the vehicle coordinate system to coordinate points in the eye coordinate system, and determining a visualization distance between the obstacles in the surrounding environment based on an observation angle of the eyes; and performing three-dimensional modeling of the surrounding environment, based on visualization distance between the coordinate points of the obstacles in the eye coordinate system and the obstacles. This embodiment may make the surrounding environment of the autonomous driving vehicle seen by a passenger through an on-board screen closer to the surrounding environment seen through the vehicle window.