Multi-Resolution Vehicle Environment Sensing for Semi-Automated Driving
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for vehicle environment sensing in semi-automated driving face challenges with limited computing power and the need for real-time accuracy, particularly in handling emergency situations.
Innovation Solution
A method and device that evaluate sensor data in a first lower resolution for overall processing, with defined focus regions in a higher resolution, using neural networks or traditional methods, to reduce computing power while maintaining accuracy in critical areas.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sensor data is evaluated in full resolution to maintain accuracy, then measurement precision is improved, but computing power requirements increase
Solution Approach 1:
The sensor data evaluation is segmented into two resolution levels: full resolution evaluation for defined focus regions and reduced resolution evaluation for remaining areas. This segmentation allows the system to maintain high measurement precision in critical regions while reducing overall computing power requirements by processing only essential areas at full detail.
Solution Approach 2:
Different evaluation resolutions are applied to different spatial regions of the sensor data. Focus regions containing potentially relevant objects are evaluated at full resolution to ensure high measurement precision, while other regions are evaluated at reduced resolution to conserve computing power. This local quality approach optimizes the balance between accuracy and computational efficiency.
2Productivity
If sensor data is evaluated in reduced resolution to save computing power, then productivity is improved, but measurement precision deteriorates
Solution Approach 1:
The evaluation process is divided into focus regions and non-focus regions. Non-focus regions are processed at reduced resolution to improve processing speed and productivity, while focus regions maintain full resolution to preserve measurement precision. This segmentation enables the system to achieve real-time processing capabilities without sacrificing accuracy in critical areas.
Solution Approach 2:
The system applies different evaluation qualities to different spatial locations based on their importance. By identifying focus regions that contain potentially relevant objects, the system maintains high measurement precision locally where needed while using reduced resolution processing in other areas to maximize overall processing speed and productivity.
3Reliability
If full resolution evaluation is used for all regions, then reliability is improved, but use of energy increases
Solution Approach 1:
The sensor data is segmented into focus regions and non-focus regions for differential processing. Focus regions are evaluated at full resolution to maintain reliability for emergency situations, while non-focus regions use reduced resolution to reduce energy consumption. This segmentation strategy ensures that critical safety functions maintain high reliability without the excessive energy cost of processing all data at full detail.
Solution Approach 2:
The system applies high-quality full resolution evaluation locally to focus regions where reliable detection is critical for safety, while using lower-quality reduced resolution evaluation in other regions to minimize energy consumption. This local quality differentiation allows the system to maintain emergency handling reliability only where absolutely necessary, optimizing the trade-off between safety and energy efficiency.
4Use of energy by moving object
If reduced resolution evaluation is used for all regions, then energy consumption is reduced, but reliability deteriorates
Solution Approach 1:
The evaluation process segments data into focus and non-focus regions, applying full resolution processing to focus regions to maintain reliability for emergency detection, while using reduced resolution for non-focus regions to reduce energy consumption. This selective approach ensures that energy savings do not compromise safety-critical detection capabilities.
Solution Approach 2:
The system maintains high evaluation quality (full resolution) locally in focus regions where reliable object detection is essential for safety and emergency handling, while using lower quality (reduced resolution) evaluation in other regions to reduce overall energy consumption. This local quality strategy ensures that reliability is preserved where it matters most without the excessive energy cost of universal full-resolution processing.
Data Source
AI summary
Technologies and techniques for sensing the environment of a vehicle driving with at least semi-automation. The environment of the vehicle is captured via at least one sensor, wherein sensor data captured by the at least one sensor are evaluated by an evaluation device via at least one evaluation method in a first resolution. The captured sensor data in at least one defined focus region are evaluated by the evaluation device via the at least one evaluation method in a second resolution, the first resolution being lower than the second resolution. The evaluation results are combined and output. Aspects also relate to a device for sensing the environment of a vehicle driving with at least semi-automation.