Vehicle External Environment Recognition with Adaptive Vertical Scanning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing external environment recognition systems for vehicles face high processing loads due to increased detection points, leading to inefficiencies in data processing and potential increases in system complexity and cost.
Innovation Solution
The system intermittently emits electromagnetic waves in a horizontal and vertical direction to acquire point cloud data, adjusting angular resolution based on object size and distance to optimize detection points, reducing the number of points needed for recognition processing while maintaining accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If scanning and emitting electromagnetic waves in horizontal and vertical directions to acquire point cloud data, then external environment recognition accuracy is improved, but processing load increases
Solution Approach 1:
The patent applies local quality by setting different angular resolutions for different vertical directions. Specifically, a first angular resolution is used for scanning the road surface area, while a second angular resolution (coarser than the first) is used for scanning areas above the road surface. This allows high detection precision where needed (road surface) while reducing processing load in other areas (above road surface), thereby resolving the contradiction between recognition accuracy and processing complexity.
2Area of stationary object
If increasing detection points for comprehensive scanning, then detection coverage is improved, but system size and cost increase
Solution Approach 1:
The patent implements local quality by applying different scanning resolutions to different spatial regions. The road surface area is scanned with a first angular resolution ensuring comprehensive coverage, while areas above the road surface are scanned with a coarser second angular resolution. This differential approach maintains adequate detection coverage across all areas while reducing the total number of detection points, thereby controlling system size and cost.
Solution Approach 2:
The patent applies partial action by not uniformly applying high-resolution scanning to all directions. Instead, it uses finer resolution (first angular resolution) only where necessary (road surface area) and coarser resolution (second angular resolution) elsewhere (areas above road surface). This selective approach achieves sufficient detection coverage without the excessive processing burden of uniform high-resolution scanning, thus controlling system complexity.
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
This approach reduces the number of detection points required for recognition processing, minimizing processing load and system size, while maintaining accuracy in object detection, thus enhancing the efficiency and cost-effectiveness of the system.
Implementation Method 1
acquire point cloud data including three-dimensional position information of detection points on a surface of an object based on a reflected wave from the object
Data Source
AI summary
An external environment recognition apparatus includes: an in-vehicle detection unit configured to acquire point cloud data of an object around the subject vehicle; and a microprocessor. The microprocessor is configured to perform: recognizing a object based on the point cloud data; determining a first angular resolution in a vertical direction based on a size in the vertical direction of a first object on a road apart from the subject vehicle by a first distance based on a vehicle speed; determining a second angular resolution in the vertical direction based on a size in the vertical direction of a second object located at a predetermined height from a road surface apart from the subject vehicle by a second distance; and controlling the in-vehicle detection unit with the first angular resolution or the second angular resolution for each angle in the vertical direction.