Static Object Detection via Adaptive Resolution Data Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing detection methods for static objects in automotive systems face limitations due to high memory and computational resource requirements, particularly when handling different resolution levels and bandwidths needed for various driver assistance functions, such as parking and autonomous driving.
Innovation Solution
A detection method that progressively compresses environment data from sensors into multiple data sets (determination, cluster, segment, and enveloping structure) based on geometric characteristics, allowing for adaptive resolution and bandwidth reduction, enabling efficient processing and resource utilization by providing specific data sets to receiving functions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high resolution environment data are processed for autonomous driving functions, then detection precision is improved, but memory consumption and computational effort increase exponentially
Solution Approach 1:
The patent segments environment data into multiple resolution levels (first resolution level with high detail, second resolution level with reduced detail). Different receiving functions can selectively access appropriate resolution levels, allowing high detection precision for autonomous driving while reducing overall memory consumption by not storing all data at maximum resolution.
Solution Approach 2:
The patent applies local quality by providing different data qualities to different functions: high-resolution data for autonomous driving where precision is critical, and lower-resolution data for warning functions where extreme precision is less critical. This optimizes the balance between detection precision and resource utilization.
2Measurement precision
If high resolution environment data are processed for autonomous driving functions, then detection precision is improved, but computational effort increases exponentially
Solution Approach 1:
The patent segments computational processing into multiple resolution levels. Receiving functions process data at appropriate resolution levels rather than always processing high-resolution data, significantly reducing computational effort while maintaining necessary detection precision for each specific function.
Solution Approach 2:
The patent applies partial action by processing only the necessary portion of environment data at high resolution. Not all receiving functions require maximum resolution, so processing data partially at different resolution levels reduces overall computational effort while maintaining sufficient precision where needed.
3Adaptability or versatility
If multiple resolution levels are maintained for different receiving functions, then adaptability is improved, but device complexity increases
Solution Approach 1:
The patent segments environment data into multiple resolution levels and provides appropriate segments to different receiving functions. This segmentation approach improves adaptability to different function requirements while managing system complexity through a structured, hierarchical data organization rather than requiring completely separate processing pipelines for each function.
Data Source
AI summary
A detection method for detecting static objects in surroundings of a vehicle, and a vehicle.